The AM Forum

Its a modest start. We cut and installed 2x10 planking for rack floor. It can be held in place with screws. The parts can be fastened to  it with lag bolts. It was amazing that the planks fit side by side. Only the length had to be cut and a couple of notches. A block and tackle was used to move the plate transformer and variac into position. In the end I think the capacitor and choke will go under the Powerstat. It has not been possible to touch up the paint the rack, and its not a big deal as it only has surface corrosion a few spots at the bottom where it can be seen and this can be fixed later.

2x10s on top of the rack with a chain  held up the block. The big transformer was on a dolly and it was not very hard to roll it up to the rack and as it was pulled up it slid into place Only one choke and capacitor has to be used in this. The choke is 30H@1.3A and the capacitor is 31.25uF@9600V. About the choke, the tube's data sheet says 2A max but as an amplifier the would only happen on in CW condition, so 1.3A should be enough. Not sure how to calculate that right now but it is a great choke and should be fine in here.

Just some rigging. Might not be so great so don't do this at home.

.

The Powerstat was originally set up with the sections in series for 460V 28A. It is made of two 230V 28A sections and can be rewired for 230V 56A by putting them in parallel which is the more common arrangement. I have to find a balancing coil for it. I like the unique handwheel on this one. It has some burs on the inside making it hard to slide on the shaft so some filing might help. The Powerstat has a Ward Leonard 600 Ohm potentiometer on the main shaft. It's rated 600V and 2.62A t0 0.4A. I think this means 2.64A max and 0.4A when the whole element is used but it's not clear. There is a concentric 1/4" shaft running through the main shaft and an Ohmite 50 Ohm 2.12A potentiometer at the very back. The shaft can't protrude because the handwheel is not on all the way due to the shaft burs. There is supposed to be a big knob on the front for the Ohmite, but it has not been found yet. The Powerstat has to be moved forward. its not mounted yet. Need to find a special nut for it and also a die to run over one damaged bolt head on the front.

Wow! That's BIG STCHUFF !

You'll easily be a contender in the Heavy Metal Rallies.

What'cha planning on running, tube wise, etc. from it?
Still thinking of using the 3cx3000a7 previously mentioned?  

OMG - what's this a new VOA site? This going to be single band or multi band?

If you care about aesthetics this might be a good time to mask off and do touch up paint to the bottom front faces of the racks??

but this looks like a fairly "manly" amp for manly men...  ;D

Hope you have it where you want it?
Also depending on your thinking, you may wish to bond a ground wire to the stuff on the wood.
Otoh, shorts to the core of the iron will not effect the operation, but could effect you if that happened.

                      _-_-bear

That was my thought too.....A tube(s) with handles being used??? Serious stuff going here.

Stout stuff there Patrick!

Think that the tube will need Lifting Eyes.
Looking forward to progress reports and pics.

Have Fun,   GL,   Vic

Here's a couple of the new VOB station.

(VOB  Voice Of Bakersfield)   I'm close enough to VOA Delano that I'm gonna consider myself facility B.

50 ohms at 200 Watt CCS glitch R assy.

12 kva ccs Dahl supply

3cx3000.

This one works OK,  the next will be the same size xformer (I have 2 here)  with a 4 by 5.

Looking good Pat,  can't wait to hear it!

--Shane
KD6VXI

Beauty IRON, gentlemen!
Looking forward to seeing more great photos as progress occurs!
73
Frank
KJ4OLL

It'll be the 3CX3000A7. There is only 100A service to the lab and the HVAC has to run, so it leaves about 60-70A for everything else. Up against a limit, I guess. Beside, +6dB capability is plenty to assure long component life and hopefully low distortion.

The transformer is a conversion to single phase from a 3 phase unit of an RCA Ampliphase if I recall right. BTA-10G? H? I wish I had the manual for the unit it's from to study. The secondaries in series should be about 6KVAC, rating at least 2 Amps CCS maybe a little more. 120V input made something like 3600V at no load.

I'm going with the classic choke input in compliance with the 'Electronic Renaissance'. As for the tube there are none in glass that seem practical for do this.

I'm somewhat of a Rat Builder. The term "Rat" is usually applied to vehicles but anything can be ratted.
It's somewhat of an art and does not mean poor work, bad parts, or low performance. There is usually at least a little rust somewhere. 8-)
No disrespect to those who put in more hours to make shiny things and replicas of classic apparatus.

I don't know how to get a pretty finish on something like this or match the color. The sides are black and doors blue, all are in similar condition.

But not to get ahead of myself. Many people start something like this and not all of them finish it, and even more talk about it and never start for lack of motivation, time, cash, etc. Jacob's help one day each weekend should keep me motivated. His energy and motivation must spill over to me when we are working. Never seen a kid with such a positive attitude and so ready to get to work. Kid.. 25 years old and with a wife that loves electronics. His favorite pursuit is modulated tube type Tesla coils so we will work on that at times. Two way street.

It's good to see the VOB station, its Dahl iron is perhaps more compact, but about the same 12-13KVA.  Interesting, 4KV @3A from that one, perhaps a capacitor input filter? It's very interesting and should have a topic with pictures as well.


p.s. not noticed til today, the way one set of notches came out on the flooring 2x12s. Look like anything?

Yup.   I'm going cap input,  should be good for 6.4 kv,  unloaded.   I'm expecting 6+ kv loaded.

I actually have a pair of the xformers.   One an ordinary Pete Dahl and another mag cap Pete Dahl.

Took one xformer in on a trade for labor.   Can't beat that price.

I'll post some more pics later.   Sitting on the side of the road waiting on a wrecker.   In a big truck.

Gonna be expensive,  thank goodness it's not my dime.

It appears that the saw blade had knowledge that this rig will be Amplitude Modulated......

A Milwaukee 'Sawzall' with a wrecking blade.
tool tip: sawdust on the concrete floor helps when sweeping up everything else, lowers dust, and leaves it looking clean.

Hey Pat,

Before you fire that bad boy up, you better call Comanche Peak and tell them to bring up the reactors into the red! I'll know you're on the air when my lights flicker.

Mike/KE5YTV

It will no doubt be months.. plenty of advance notice. Each time it gets worked on, I will try and post some pics.

Right now the status is:
Need to get a die to run over the threads on a Powewrstat bolt which I think is 3/8-20 as the 7/16-20  die goes on it loosely but the 3/8 dies in other TPI 24 and 16? do not fit. I have a nut the right size and have to go to the hardware store to find out exactly what it is.

Need make sure I have a drill for that size, have to drill the brackets a bit as they seem to be from a different Powerstat.

Buy some of that angle iron with the holes in it, retrieve some unistrut as well, there is 900# of it stashed in Gainesville TX. The Powerstat is just sitting there and needs to be mounted properly.

If there is some cheap tool for clearing burs out of a hole, that would be nice, otherwise a round file and elbow grease. The burs are at the inside rims of the setscrew holes. Dont know how they came to be there but seem original.

Once the variac is mounted I can temporarily mount a breaker and use them together to test out the 3 PH blower on single phase 240V -see how well the capacitor start setup works. Need a potential relay for that too. It'll do until a VFD shows up to make three phase from single.

So many little details.

Ahh yes I think the model transmitter the transformer is from should be a BTA-10L, or possibly the BTA-10J which is older.

Today the front of the Powerstat was fixed in place. Much thought and checking was needed to see how this should fit. It took a few hours to get it done with hardware on hand.

One of the long threaded rods that goes from front to rear in the Powerstat had a messed up thread end. This was chased with a 3/8-24 die to restore it.

The right side of the rack had a gap in the rails from the factory. This is inconvenient. The lower section of those rails were moved up to face at the powerstat so a panel can be put there.

Inside the rack halfway to the back was another (third) set of rails that had supported some unknown item. These were removed and sawed to half height. Two parts are used to back-set the panel that holds the front of the Powerstat. The remaining halves can be used elsewhere.

Because the Powerstat is mounted recessed, the only front panel hole needed will be for the shaft. This makes it much easier.

The possible weakness is that only four rack bolts are used to support the front of the 100 LB Powerstat. More fasteners can be added if needed and they can be 1/4" grade 5 hardware.

This Powerstat was originally set up for 460V @ 28A. It was done by connecting two 230V 28A units in series. These will be reconnected in parallel for 230V @ 56A. I need an "equalizing coil" yet. It is required to force equal currents in parallelled Powerstat sections.

RAT it is!

Looks like you may have named the amp?

Clandestine Rat?

Btw, where can I find more pix of trucks like that??
And, WHY DIDN'T I THINK to build a car/truck like that??
I had a nice 55 Ford pick up at one time, but the idea of making it
all pretty was too much... too much money, too much time. If I had only realized
I could have gone RAT!!

Paint?

Paint whatcha got! That's what masking tape is for.
Or the paint store will match paint, then you can spray it with a spray gun and air compressor...

Why match the paint, if ur going RAD RAT!!

                       _-_-bear

That may be Candelstein Rat.. Radio Candelstein, Flagship Station of the NRK Radio Network.

RAT should maybe be an acronym in this case?  Hmmm...

Goog images "rat truck".   "rat bike" also gives results.

Anyway, Last Sunday update: was off for Mothers Day.

Radical Atomic Totalizer?

Radio Attack Transmitter?

Ridiculous Amplifier Tenderizer?

Very nice build!

Since the powerstat and the 3cx3000A7 seem to be major components, one by weight and the other by function, looks like the amps' name might honor them.

"PowerStash 3000"

The 3k not only honors the tube but a decent, loafing output.

Hmmm..
Not sure how to work the powerstat in, but it's just a handwheel. Haha later on, maybe I'll drop a note to Superior. I got a nice reply after telling them how pleased I was that the 60's Stabiline here was performing flawlessly without regard to age.

Now RAT 3000 I like.. For one, it does not use the peak output as a brag, like the cheap stereos of the 1960's. it'll have to wait until this is built and tested tho'  ;)


Today, I quote Dr Smith: "The PAIN the PAIN!!" I lifted nothing but spent a lot of time on the floor and close to the ground making and fitting and unfitting and remaking and bolting brackets and bolts and warshers and all that kind of stuff. What a wreck I made of myself. Without my young apprentice, this would not be possible. He is like the son I never had, but if I had a family, would want. My luck in that regard however would stick me instead with the son that that trashes the car and won't get a job. So this is good.

The Powerstat was moved back an inch to a perfect fit so the handwheel is 1/4" from the front panel. Just enough rheum for the fine 0-100 round scale to fit behind the handwheel.

Now this is not really an heroic handwheel, or is it? Three spokes and a hub does not a wheel make. What is this human interface device called, properly?

What I would prefer, but is not to be found, is a big lever that moved up and down on a radiused panel mounted escutcheon. These were sometimes used with Ward-Leonard rheostats for stage lighting controls. The lever pulls a chain like a bicycle chain, this is transmitted by pulleys to a gear on the shaft of the rheostat, so why not a powerstat?
(black and white images)

However, I'd like a longer scale/bigger radius on the lever than shown in the panel if possible. Those scales probably are about 8" long, and the lever tips may be traveling 18".

Point is that it's not lighting which would use a coarse positioning, but a Powerstat, made for greater accuracy. The scale should be 0-100 and be able to be set easily to within 2 points or so. Those that have operated old Ward-Leonard rheostats know that the levers could be hard to position precisely, slow to push. In high school we had a stage lighting console with similar resistive devices but the individual escutcheon was about 12" long, low to the panel with a large radius, with the handles protruding about 6" from that. Real easy to move.

Hey it's art.

OK now having wiped away the drool from looking at those banks of controls, here are the pics of the progress today.

To improve the strength of the front mount, two holes were drilled in the supporting panel and instead of rackmount screws and nuts, grade 5 1/4" bolts were used, with washers on each side and also the nice 'nylon insert' nuts. These can be used instead of lockwashers. They may not stay as tight but they will not vibrate off or work loose to where the nut falls off and are popular on drive shafts. In this job it probably does not matter which is used, nut and lockwasher - vs - nylock nut. It only matters the strength of the fastener.

Mounting the back of the Powerstat was an interesting exercise and also painful for arthritis and back issues to kneel behind the rack and reach in to do this fitting up the pieces.

A thicker angle iron could have been used such as the side supports at the rear, but it was as well to use the flat metal because all the holes lined up well and 3/8" hardware was used between the X braces and the angles. the angles were fastened to the rack by 1/4" hardware, and believe it or not the holes did line up nicely.

Where a bolt head or nut  was small compared to the hole (some oblong holes) in the angle, a large fender washer was used to provide a flat surface. This is important because it prevents the edge of a nut or bolt head from going down into the end of an oblong hole, which would cockeye the bolt, or the tension on the bolt, when tightened.

The front to back fit of the angle iron position vs the rear of the Powerstat "big hexagonal nuts" just happened to be perfect. One flat strap was placed upwards left to right on the front side of the angle and the other was placed downwards left to right on the backside. These had to come in and out a few times to trim the ends at angles for a good fit without sharp triangular tabs protruding into the next space, or just to make them fit inside the angle. These little details consumed much time.  I was glad Jacob was helping me. He would trim the ends as I asked, then hand me the next piece, and wrench one end of fasteners, etc. Tell ya what he ain't afraid of work.

Now it may not look very strong in the back end, but believe me the triangular arrangements here, I am confident, will support several times the weight, and should support half of a 100 LB Powerstat until the next person owns the amp, at which point I won't care any more. The triangle arrangement came about due to the effort to make the holes like up. A flat 'hoey metal' piece was bolted loosely to the Powerstat's support spot, and rotated until there was a clear hole at each end. It may be silly but that is how the geometry was determined. It would not have been used if it were unacceptable. I guess there are two schools of construction. One starts with mechanical CAD and precise machine tools. The other starts with a lot of imagination and the ability to stare at something until the imagination calls a few simple solutions to mind.

The only ghetto part of today's work is the stack of washers taking up the space between the 3/8" grade 8 bolt head and the big hexagonal nut. This is the support for the back of the Powerstat. I promise to eventually get a shorter fastener. I believe it is strong enough however and with a lockwasher it will not come loose. I want to add another X or a single angle bracket to give it two support points.

 

More pictures of some details.

As for making the Powerstat level, we looked for the level fo 20 minutes, then the bright idea flashed on: The level is the rack. The front and back of the top of the powerstat were compared to holes in the rack using an old panel as a carpenter's 90 degree angle. So yeah this is level front to back with the rack to about 1/32"

It might be asked where the giant potentiometers went. For now, the too-long bolt with all the washers has made it difficult to mount that back on. It will fit but the rotary coupling does not engage well. It never did engage well. The big pot also needs cleaned up and lubricated. The Powerstat BTW responded very well to being cleaned and lubricated. It's very easy to turn. The giant Ward-Leonard potentiometer may be another story because its contacts are held in tension by a powerful spring. It may never be really easy to turn. But should a 14' diameter pot be easy to turn? that is a good question.

Also, although it's nice to have, what would be the purpose of a huge pot like that ganged to a powerstat in an amplifier?  

The plan, the way it is working out now, is to build the HV power supply first, then the mains cable and stuff so incrementally things can be tested.

Until next Sunday..

My Harris uses pots for feedback of where the roller inductor is.

Inductor spins 10 times,  geared 1:1 to a pot that indicates the position of inductor.

Use a motor driven shaft,  and with a simple comparator circuit,  you can have presets of the powerstat (or inductor,  in my case).

Think remote control bud.

--Shane
KD6VXI

Loving it already - a 55 year old 600 ohm 100W pot as a position feedback pot. Get into that real analog servo stuff there! True renaissance.

It is a great idea. Doesn't matter if the pot's giant, pot's a pot. It's big enough to handle some current so even a simple relay controlled position or something like a 115VDC gearmotor could work. Hadn't thought about remotes but others have said they did it because of blower noise. That is a great idea how to use the pot!

The overall layout seems to be coming to mind too. I learned the hard way restoring the COL Tucker, that one should not try to skimp on space or build a very compact equipment unless one has a lot of time for detailed engineering, knows all about thermal issues and everything else. It's a fine rig but there's no room to add anything and the OM did not put a mod reactor in it. Lesson to me: need one rack, use two.

The plate transformer and blower can sit in the left hand rack bottom. Above them, there will be tons of room for the RF section and the control circuits, filament supply, etc. I do not want to try and cramp that because I will make more design mistakes and compromises and it'll be hard to work on.

The filter cap and choke can sit in the right hand rack below the Powerstat. It should help weight distribution, them poor casters are suffering on the left side. Above the powerstat should be the window for the MV rectifiers. They don't take much depth. Behind them and above them, plenty of space for a built in tuner. That wasn't strictly planned, but it looks like Frank had some luck with the idea on his 3cx3000 and there will be that space.

Thinking about casters, now would be a good time to replace all of them while the plate iron can still be easily removed with the block and tackle. It will only be 10X as much work later.

Yesterday's work. It does not look like much, but getting a couple panels to fit so the exact position of the Powerstat panel could be known, and then making up that panel to look nice was time consuming. a large hole was made with a punch in the white panel, the panel mounted in the rack over the Powerstat, then then the round scale put on, its mounting holes center punched, removed the scale, drilled, and tapped the holes, then put it all back.

This turned up, a coaxial filament choke, but it is for 10-80M, and more like the size for a 3-1000 or 4-1000. I still have the filament choke from the TMC 10KW amp to use on this one. Found a possible power cord. It's 8 Gauge, may or many not be enough. I have to look it up. Also, some nice red push buttons could be used for the power switch but nothing like that has been decided.

Some 70A 250VAC breakers. A set would be perfect for the mains input. They run on DC to 60Hz. I stripped these RCA MV rectifier sockets off a BC-5, but I don't have the 575A tubes I wanted to use. I thought I had some, but didn't find them. I have plenty of 673s and 4 new Johnson sockets.

Not sure, but I think that "safe vault" adjuster for the variac sort of needs a slick looking and clearly visible pointer mounted off the bottom of the central part that affixes to the shaft?

I'm thinking a pointer with a round ring as an indicator at the end that goes over the scale area?

Brass would be easy to fabricate, and I think the hobby shops may sell brass rings.
But epoxy and aluminum or even plastic will work too... just need the arm and a way to
mount the ring on the end.

Classy.

                            _-_-

I like the looks of the safe handle and i see the shaft to the safe handle is insulated from the internals by the U-joint connector...but I think I would prefer a nonconductive steering wheel....Good luck...neat project..

There was no pointer on the Powerstat when I found it. One of the handles was the pointer. It may be possible to slip some heat shrink tubing over the handles. Black for two of them and white for the indicating one. It could be possible to make a brass pointer from flat sheet stock or some odd piece of hardware and tap holes in the rear of the wheel hub to mount it.

The 'safe handle' wheel was on the Powerstat when removed from the original equipment. It's a very good point about it being conductive. It is insulated by a ring of about 3/8" wall thickness from the brush-carrying metal disks of the unit. I should make sure the shaft is grounded.

The U joint in the back for the huge potentiometer is insulated mainly because that is the pot's wiper. That has not been reinstalled yet. It's not decided about it.  It is fun to talk about alternate ways to turn the Powerstat, but simple is going to be best. Remote control or cool looking mad scientist style levers and other stuff would complicate this whole thing and with rational thought I would prefer it to be simple because of the realistic time factor to finish it.

We want to replace the casters. These swivel on ball bearings but have no ball bearings on the axles, and some kind of plastic wheels. The constraint is that they are right in the corners and close to the skirts of the rack so going bigger is not likely. Better is possible.

Exactly!!

But I suggested rather than a simple ---> pointer use a circle that is wide enough to show the numbers in the center as the indicator at the end of a pointer :   ----o  like that. 

                _-_-bear

I see. I didn't exactly understand where the brass ring was suggested to go. It would be a nice touch to have it encircle a mark or two on the dial as i think you suggested. Brass is good stuff. It solders easily.

The most interesting home made pointer I have seen had a 'self focus' grain of wheat lamp under the tang of the pointer that it up the marks being pointed to. It was not on a variac but on a large rheostat with a handwheel.

BTW the correct Superior Electric T5587 paralleling choke was found on ebay for $60.

This information could help others - it is simply a roll of steel strap 1" wide wound into a toroid shape with an ID of 2.5" and OD of 4". The windings are bifilar with each having 12 turns of #8. The insulation is rated 600V and 105 deg C. The current rating is 28A per side, same as the Powerstat.

Well here comes the weekend! more to do! Saturday I go to the hardware store and then take Mom to church, then Sunday = all day for the project. Time management!

OK last week the circuit breakers were found, some 6 GA wire, Hg relays, and lastly the paralleling choke from eBay is staring me in the face. The selection of breakers is pretty good, all are DC to 60hz, and there are many sizes dual and single from about 10 to 70A.

Those breakers were used before on a 12/24VDC bus and are still wired for that but the wire tying them together needs to be redone because it is not insulated and too close together as well as made on only two #12 solid conductors together. Its a PITA since the 6 GA is so stiff.

If lockwashers (cut washers) are used with the breaker terminal nuts, could an aluminum strip be used as a bus? My understanding is that it is OK as long as the connection can not become loose.

Their mounting has also to be considered because they are in a couple of boxes that mount to the front surface of wall or rack, and so it 'sticks out'. One has to be chosen. Looking for a way to mount it "back-set" so it is somewhat flush.

some idea for the heavy gauge input wiring is attached.

Before that, I want to remove the casters and set the racks on some 2x4 boards until new ones can be had.

Will be a full day, hope it can all get done. Jacob is supposed to come over early Sunday to help.

Not everything wished for got done but does it ever? Several things were not available and other things did not fit and had to be modded or maybe fit into the realm of hackery. It's not an issue - it's turning out to be a lot more fun to try and get as much of it from the 'junk box' as possible. There's a large junk box here so why not use i!
 
Mounted some breakers in a nice old Rockwell panel. It was tricky because the panels on-hand mount at the rear and 'stick out' from a rack. To make it flush, rack rail angle was screwed to the internal horizontal supports of the rack cabinet. Only one hole matched up on each side and it seemed like a problem at first. I don't yet have the crimp-on ring terminals for #6 that will be attached to the terminal studs of the breakers and the Powerstat. (The wire will be soldered after crimping). The one-hole mounting means that the panel can be tilted up to get access to the top of the Powerstat to finish the job once the electrical connectors are had.

The breaker panel will be held in place vertically (held down to prevent it pivoting up) by the rectifier chassis. An old 1U 10 base T hub fit the bill to act as a chassis for the rectifiers. Once emptied of its guts, this is a light and strong box. The rack mount 'ears' are not shown, not in place because the sockets are not yet mounted, no holes drilled yet. There will be a 1U panel to cover the 10BT thing so it does not look like a junk box part. I just wanted to see what this will look like before doing the work. Instead of being at eye level, they are a bit lower, as is the Powerstat and the circuit breakers. This is better to observe from a typical seated operating position.

The rack space above the rectifiers is reserved.

The rectifier sockets are Johnson part number 123-2026-200. If anyone knows the insulation rating for these I would appreciate it. because it is a bridge rectifier the peak voltage could be fairly high.

There are some 1" tall ceramic standoff insulators that can be used with the sockets in case the voltage rating of the sockets alone is approached too closely.

The project amp should also get this set of nice 'hash chokes' taken from an RCA BC-5.

A bunch of really nice #6 wire was salvaged. This is 105 deg C rated, plenum rated, gasoline resistant.

The rack was lifted up onto some blocks and all the casters were removed. They are not too bad looking but these are very well-worn and filthy. They do not have ball bearings in the wheel axles but the swivels are ball type. If it were for something lighter I'd want to just clean them and relube, but I want to find some better ones for this. The originals all swivel and have 2.5" wheels of some hard rubber or some kind of resilient plastic.

Does anyone have advice on what features a caster would have that would make it easy-swiveling? Other than ball bearings in a race, I don't know. I am not a caster expert at all. This dual rack is probably going to be 600 LBs on the left side and 400 on the right side.

All comments welcome! This is the largest piece I have ever tried to build from scratch and I've seen what others have built on here.

Lastly, Jacob's Tesla coil is coming along. It's experimental 70KHz RF-driven type with audio modulation.

I convinced him to ditch the microwave transformers and use some 500VA potential transformers instead. They are not the best thing but were free and make 2400VAC. These were tested before and have good regulation with the 'burden'.

Two modified tube-type regulated supplies are to be tried with this. One is for the grid to make bias and provide the 70KHz drive signal. The other is for the screens and should be the 'screen modulator'.

Way kuhl so far!
Say, what's the light green meter panel from?
Reminds me of the Birdwell (div of SSC /Raytheon) pwr supply panels I used to use at Nevada NTS back in the 60's.  All our stuff including the logging trucks used that color.

The casters provided on my "surplus store" cabinet were never rated to support the weight I provisioned, and ultimately collapsed.
 
Mcmasters had a replacement that was an exact fit for the mounting method, @ a load rating that worked.

http://www.mcmaster.com/#casters/=12mw2mr

Wish I had changed them "before" filling up the cabinet  ::)
73
Frank
KJ4OLL

The light green panel is a Lambda tube type regulated power supply. It will be modified to serve as a bias supply and 50-100KHz RF grid driver for three 4-400s. There is another regulated supply of about 500V coming. This will screen modulate the 4-400s and give the audio signal to the RF discharge. We hope it will be clean sounding. These kinds of mods are not complicated.

It's turning out that by two persons cooperating to do two projects, both projects are easier to do because help is always available. An extra head and pair of hands goes a long way.


What you experienced is what I am trying to learn from. Once that plate iron was in, the rack acted like it was on feet, not wheels.

Thanks for that link! If I don't find them locally they have what is needed - just need to measure. This rack has skirts so the diameter of the wheels is limited. The stock wheels are 2.5" diameter and 1" wide with sleeve bearings. I prefer the larger wheels and better bearings because they swivel easier and give a little more clearance. Clearance is important if the gear has to be rolled over a threshold etc. Racks also need to be able to be moved around to sweep the floor because all kinds of dust and junk collects under them.

Looks like the McMaster-Carr 9908T19 should work. They are only $12.45 each. There was found no other caster that would fit, and be easier to push. This one says it is more about durability than ease but the amp is not going to be rolled around much.

Today's work does not look like much was accomplished but it took some time to map out and drill the holes for the rectifier sockets so it would be straight and neat. It all has to come apart again for wiring but the major mechanical stuff is worked out.

The picture of the front with rectifiers is what it looks like when seated.

I still have a question about those Johnson sockets and their voltage rating. No catalog so far has given an answer.
Seeing that it is a bridge, two will be at B+, but two will have the AC on them and the peak voltage could be more like 8500V to GND. The clearance from the pins to the metal chassis is just 3/8" and I think it is marginal.
Should I add some short standoffs under the sockets?

The rectifiers are easily spaced in the width of the rack. They fit well with centers 2.5" apart. I had a panel with a large cutout for the viewing window over this but it seems to be hiding somewhere.

The rack-front picture s the view while standing. Looks like a good compromise. Mounting the rectifiers too high would have made them hard to see when operating, which to me is the more important view.

The third image is Jacob's Tesla project. Today it was a mechanical layout day to fit the main units in and for the big power supply at the top, a set of rails were made from punched angle iron. Two right-to-left pieces underneath help support it. The power supplies still have to be modified and most of the stuff behind the panels wired up. It all fits. There is another chassis for the tubes, not sure where they are going. Maybe on top.

Patrick,

According to this chart,  you should be fine....

0.175"                7000         
0.250"                9000       
0.350"                11000
0.500"                13000

Your looking at a .375 inch spacing,  if it's 3/8.

You should be good for slightly over 11kv.

Not a lot of room for dust mites,  hair or dander,  but it should work.

--Shane
KD6VXI

Set your HV gaps a bit bigger than the leg spread on your local spiders.  Trust me...

Dennis

Hi Patrick,

I wanted to put casters under my 20V-3 and was given some friendly advice, yours FOC. If your considering 4 swiveling casters consider this. Most swiveling casters place the load ahead of the axle. The taller the wheel, the higher the possibility of a tip over. Consider 2 swiveling wheels at a narrow end of the rig and the 2 fixed at the other. A tip over is much less likely. Things are looking good. What type of rectifiers did you settle on?

Mike

Thanks!  It is that marginal spacing. The small spiders in the lab are 1" across.  I am counting on dust bunnies too so it may be some insurance to add the insulators. My concern is not having that huge transformer arc to chassis. May be carnage and scorch marks/holes from something like that.

Today I managed to order casters and picked up some ring terminals for the wiring, hope they fit.

McMaster-Carr had a nice lady (not a machine) answer the phone and she for one thing knew exactly what I was asking about and for the next thing said that if the casters did not fit I could return them, and then she took my order. That is service, rare today.

I learned a good lesson on that with the Tucker transmitter when I added a steel plate and casters to the bottom. It originally had none. The probaility of tipping seems to be related to how much weight there is going up. The goal for that on this job is to put as much of the real weight in the center bottom of each rack and avoid heavy transformers and the like being up high.

I have to use 8 casters in this one because it is two racks together. Most of the weight, all of the big power supply stuff, will be right in the bottom and in the center front to back, as best I can arrange that. The exception is the powerstat and it's below waist level.  The original casters have 2.5" diameter wheels. the replacements are 3" diameter and wider having dual wheels. They are a little taller therefore but not by much. Each is supposed to be rated 300 LBs.

When raising the unit on 2x4s to remove the wheels, it was found there is barely room to shove a 2x4 under the rack skirts. Without anything in the racks but the powerstat, It was very hard to tip it enough to slip another 2x4 additionally in, so I think I am good on balance but we shall see when heavy transformers are mounted in the bottom and the RF stuff is put in.

The MV tubes are type 673 and will be a bridge. I still want to use the solid state rectifier stacks from a large BC rig to back up the MV tubes but it's not decided. It will be a matter of room. There should be plenty.

If you do not need to put racks flush and side-by-side then one way to get casters under them, increase the stability somewhat and not lift the unit very high is to use a set of "outriggers" - usually placed left and right and bolted to the rack on the outside of the side panels (or to the rails). Heavy angle iron is possible, but rectangular box stock is maybe more stable and stronger...  since you can extend these rails front and back by a bit you get a wider base.

                      _-_-

Today no pictures because it looks the same as a few days ago. It was a big job to get the casters to fit. The holes fit, but when the rack was loaded the caster's chassis that holds the wheel and swings around would interfere with the inverted U-channel under which the casters mount. Some slight mod of the U-channels was done with a rather large set of Channel-Lock pliers. Now it all moves smooth as butter! The whole process though had a few gyrations.. the 400LB transformer was put in, then pulled out, then finally put back in. That's the third time it has been in the rack, so its the charm!

The circuit breaker panel was bolted in place semi permanently. All the terminals can be reached and the breakers can be replaced from above. There is no way to make it a simple rack mount but it's rock solid.

The MV rectifier chassis got some 1" ceramic stand-offs under the sockets. It also got a set of four hash chokes from a BC-5. Each of those perches on its own 3" tall conical stand-off. A couple of the hash chokes needed repair to the mechanical mount, a strip of metal. They have a slot in them to accept the soldering of the end of the choke wire. They are prone to tear at this point if abused, and these were. But they deserve to be used and displayed and are right behind the rectifiers.

The equalizing choke was also mounted beside the powerstat. All this took about 7 hours.

About tipping, there is no way for one person to tip this forward or backward. The weight in the bottom is too much for that.

You have probably seen this before, I'll mention it anyway. The 20V-3 uses 575A VR tubes. They punch a hole thru the chassis and mount the socket below using ceramic standoffs. The metal tube guide and twist-lock of the socket is just flush with the chassis top. Makes a neat chassis and would scratch the separation item off the to do list.

I've seen that before and it is very attractive. Well the work is already done, insulators installed. I didn't have the special card paper disks that go between ceramic parts so I made substitutions from a piece of panel switch hardware and a rubber grommet. They seem to work great together to maintain screw tension and prevent ceramic to ceramic contact that leads to looseness over time. The screws under the chassis use cut washers to keep the screw from loosening.

I found a Delrin  rod and turned it to the size needed. Like you, I puzzled over the need for them (the cardboard washers) and figured if the engineers saw a need for them, that was enough for me. Delrin isn't the best choice for the job but it is holding up fine.

 Bill Carns, N7OTQ had an interesting problem with his 20V-2. It was a hidden arc within a ceramic standoff. A crack had developed in the middle of the standoff between the ends of the threaded  bolts. The arc, and the arc damage was totally hidden. We think the previous owners gave up and if it wasn"t for Bills determination the transmitter might have been scrape.

What do you plan to use for GTO wire?

Mike

I always thought the paper washers were to keep the ceramic from cracking during initial installation, especially antenna cone insulators on the mostly hollow Johnson matchbox and the like.

Sort of intuitively knew it would help keep the screws under tension, but thought that would only be immediate protection from over-tightening during installation. Certainly the paper would soon lose its first period elasticity.  The spring action of metal locknuts are the primary tension actor in the long run.
Paper 'insulates' the brittle ceramic from not-so malleable aluminum or steel panels or chassis during installation torqueing.

Early use was thin cork discs.

Later builds used translucent plastic discs.

Sp.  "Delrin"

73DG

GTO wire would be 15KV outdoor rated 14 GA stranded with PVC blend insulation. There is a spool here. It has to be found.
http://prosites-llcwccd.homestead.com/GTOINFO.html

Any special areas will be wrapped either with 3M self-fusing Ethylene Propylene Rubber tape or covered by some military spec 2-3 mil polyester? dielectric tape. It's been used on a tesla coil of Jacob's to reduce arcing. The 3M '23' tape is super for insulating. the milspec stuff is slick and can also be put on the bottom of an RF deck and on the brackets it sits on, to slide in and out easily. Been tested on the 70 LB Tucker RF deck!

http://solutions.3m.com/wps/portal/3M/en_US/NA_Communication_Technologies/Home/Products/~/Scotch-Rubber-Splicing-Tape-23?N=8694168+3294318588&rt=rud

Thanks Dennis, I corrected the mistake. ;D

Today some wiring was done on the primary side. The wire is #6 THNN stranded and not very flexible, and all of the crimp on lugs were soldered after crimping as is the practice here so it took a long time. THNN wire has a PCV insulation and a nylon jacket The wire has 19 strands.

The Powerstat sections were wired in parallel with the outputs going into the equalizing coil. From there the red lead goes to the 240V terminals on the plate transformer and the black lead from the "0" tap on the plate transformer goes back to the common terminals on the Powerstat.

The two primaries are wired 'out of phase' according to the terminal boiard however that is the correct way because the secondaries must be out of phase, not in phase as they would have been in the original 3-phase application.

The Line side of the powerstat has only the one side of the line "L1" done. It is going to a triple pole mercury contactor. There is another contactor for the other line "L2" that will control the Common input of the Powerstat.

more. That is all #6 wire.

The white wire ties hold the wire in place on the rack and are not very tight so there is no chafing. There is ample room in the center section where the racks are bolted together for running all the wires. The space is about 2-3" wide.

The triple 4-400 deck for the Tesla coil is coming along. The sockets are being wired and the 5VCT 40A filament transformer was mounted to a wood reinforcement on the backside of that shiny panel. The cooling is going to be a trick. Probably a sheetmetal plenum will be made on the blower and some hoses will be run to the 1" PVC pipes coming down from the air system sockets. There may not be enough pressure. It will have to be measured.

This rack is the power amp section. The other rack shown before is the control, power supply, and modulators.

No pics of last Sunday's work but didn't get so much done. The other side of the line to the Powerstat was wired to the remaining 3x mercury contactor and the order of the circuit breakers was sort of decided on.

The Mercury contactors had to be moved rearward a little to give more room to the bottom of the MV rectifier chassis. Things will move much faster maybe after all this heavy gauge wiring is done. Its slow because it takes much time to solder the terminal lugs onto the large wire.

Heavy wiring about done. Power cord made. 50A should be more than enough to run the tube full bore though the power transformer can handle more it won't be called upon for excesses. The smaller wire of #14 can be used for primary on almost everything else because of low current  consumption. The blower may be an exception when starting. The red and green give a festive christmas time look. #14 may seem small but at 20A it is 1V for every 10 FT round trip, and it won't be seeing anywhere near 20A on 240V. The breakers may be as shown from right to left main, blower and controls, filament, plate. The breakers have marked Line and Load sdes, so that is being observed. the Line / inputs are at the top.

 

The filament transformer is 208V so a 230-208 autotransformer is being looked at. It is designed for stepping up 208 to 230 but is there any reason it can not step 230V down to 208V?

Filer name plates shown, and dug out the manly SS diode banks.

On the choke, it is rated for the inductance at 1.3A. The max current on the amp should be 2A but that is the single tone or CW rating and AM will be lower.

I think we're well past DO-cased parts at this stage.

next some mechanical work has to be done to mount the transformer. It needs some angle iron to strengthen the planks so the weight is distributed. Same for the other side with the choke and capacitor.

What did those 1n3495 diode strips come out of?
Pretty beefy!

And your the wing nut king. ;)

Now we are waiting for the "IT'S ALIVE!!!" scream...

73DG

Those are from a BC transmitter to my knowledge. Maybe someone can say which one. The wing nuts let a person quickly change out a bank of rectifiers. Almost as easy as replacing a tube.

I was tempted to plug it in and play with the powerstat, a case of 2500 Ohm 225W resistors, and a clamp on ammeter. Rollicking good time like a German fun house.

Repaired the h/p M.O. for Jacob's Tesla today, then got to some ampwork.

There was a situation where the floorboards sagged in the center. Being supported only at the outer edges it might have been expected of them. A cold rolled steel angle iron of 0.125" thickness and 2" width was proved out. Just placing the Plate transformer on them straightened up the floor pronto. So, time to buy some lag bolts and do some drilling.

Nearly thought there might be some blasting, but avoided it by observing all safety precautions and taking a healthy spoonful of fear of arc flash. At least I had the decency to plug it in myself instead of cajoling someone else into doing it. I am a decent sort.

The 240 was routed to the main breaker in the rack panel and from the second breaker flying leads were fastened to the Mercury Contactors' energizement coils.

The result was that the main breaker became the on/off switch, the second breaker became the control for the power to the Powerstat, which is also controlled by the Plate breaker. None of this matters right now because all of them are rated the same 70 Amps, but it made me have to turn on three breakers to apply the juice.

The energizing current for the plate transformer turns out to be 5A @ 218V. The secondary made 6430V with no load. There is a noticeable humming buzz from the plate transformer, I guess it is normal on a large unit. With the rack enclosed it should not be noticeable.

The 3 phase blower was played with a bit. I'm leaning towards a VFD over the usual method.  Starting current's a monster.

Jacob liked the buzz from the Plate transformer and I had to agree completely, a sweet and endearing sound! - Mellow and relaxing, yet not without a hint of ball bearing pie with a slice of clack. - So we recorded it as a sample being brought up on the powerstat as well as with the sudden application of 240V and the Ungh/Bonk/Huzz type of sound we all know and love. A couple of times it shot the clamp-on's needle to the top of the 50A scale so the peaks must have been a little larger than desirable. Soft start will be a must on this just out of politeness to the rest of the house. Maybe I can post the samples later once Jacob sends them over. I am sure you will like them.

BTW for those that don't know, Hg contactors are nearly silent. No kerchunk with those.

pr0n of the 4KV Tesla power supply and modulator is shown as well. Top to bottom: Audio modulator and screen supply, RF driver and bias supply, Plate voltage control, Plate supply and filament control. There are three plate transformer + rectifier sets, one for each 4-400. The DC filter will be in common.

We are not drunk on power. We just like electricity. These are all completely normal and wholesome Sunday afternoon activities for today's nuclear family. God bless America! One of few places in the Free World where you can make whatever you want!

Nice work Patrick and great attention to detail. I thought I was good but I never thought of matching color shrink tube over crimp on connectors. Good job.

Thank you. It was on hand so why not? Truthfully the other end of the cable is plugged into a 50A outlet so the tubing gave me a little better chance to avoid a short. I got the tubing years ago, bought several packs in different diameters each with an assortment of red, blue, green, yellow, white, black, and clear. The 8" pieces, two in each color, this was a good investment. I want to go back and put some on the other leads too.



Someone suggested that the buzz and 5A draw is because I am running the transformer on single phase, not 3 phase wye 208V or 240V, meaning that I'm putting too much voltage into the primary.

Instead of 240V across each primary winding he suggested I use 139V. That is, divide 240V by 1.7333 to arrive at the voltage that would have been there normally on the primary winding assuming the wye input.

Now, the transformer says 208 and 240V on its connections.

Putting 120V to the 208V taps  is the same as putting 139V to the 240V taps. I'm not explaining well but what I am trying to say is:
208V wye = 120V from phase to neutral
240V wye = 139V from phase to neutral

I do not have the schematic of the transmitter it came from so I do not know what the hookup options are.

The small wires from the breaker panel to the back of the rack were done, as well as the wiring from the Hg contactor coils. These are going to a small terminal strip at present. The way is now clear to start building up the rectifier circuitry starting with the MV rectifier chassis installation and wiring the filaments mounting up the Dahl FWB filament transformer then the silicon stacks. Time to get out the 15KV wire for that stuff. I have lots of red and green 600V wire, hardly any black or white in the appropriate gauge so.. maybe it will be done by Christmas.

The magnetizing current issue with the plate transformer was checked by measurement. A graph shows how it is low then increases more quickly as the line voltage goes above 136VAC. This is the possible saturation issue and the cause of the hum from the unit. The vibration can be felt on the rack. Unfortunately the HV AC is too low to do much with unless the saturating and humming is allowed, so trying to think of remedies for it.

One graph shows the primary voltage vs. current (no load) normalized to 1.0. the next shows the deltas between voltage increments and current increments as the voltage is raised through a range. I'd appreciate any suggestions. The chart is the measurements.

One idea could be to add a separate additional transformer and bridge making 1KVAC@2A with a separate bridge and put that between the low (-) side of the main supply being built, and GND (B-) Two 240 x 480V 1KVA control transformers in series might do the job.

Another idea could be to wind some turns around each of the well-insulated windings of the existing transformer, there is room. Put them in series with the existing secondaries. Might be risky. Don't know the turns per volt yet.

net net is I need 5600VAC.

On the Tesla side, the other project, some work was done with the Lambda 481-C supply to modulate it for use as a grid driver. It's not correctly amplifying the 70KHz carrier signal yet but it did self-oscillate at 145Khz making a 400V p-p signal across a 50K load so it looks good. The self oscllation was able to be lockes to the generator. It was found that the reference regulator was igniting and extinguishing during large output excursions. The reference regulator and error amps need their B+ changed, from a source linked to the regulated output, to an unregulated point so they are not directly powered by the DC output loop.

update on the plate transformer - -it has two primaries and two secondaries. A test can be done with one primary then the other, to check the current and secondary voltage.

On the power supply mod, the paragraph about it would be clearer with this schematic.

The problem with the high current on the transformer has been found and fixed.

When the transformer was reassembled from a three phase unit into a single phase unit, the 208 and 240V wires for one primary were put on the terminals reversed. The terminals as previously shown are on a large insulating board. This was not very evident because the primaries should all be made the same, that is, wires from right to left should be the same taps. However, it was seen that as the taps progress from inside to outside, they are not in the same order right to left on both bobbins. The 240V wire is closest to the core, then one layer after, is the 208V wire.

The testing of the transformer was done as shown in the diagram, hence the unusually high current due to one winding having more turns in use than the other.

All well & good, that must've happened in shipping ::)

73DG

That out of the way, more was done. The 240V and +11V taps gave right at 5660V secondary with the full 220V coming from the Powerstat. In the summer it tends to be 220V not 240V here. The primary current is just 1.75A which seems fine on a 20KVA transformer. I previously misquoted the KVA rating at 13-14KVA because I did not see the label properly. The rating was reduced to 20KVA from an even higher value when the transformer was reworked to single phase. So it's a 20KVA unit, fine with me as someone may want to use a larger tube in the future.

The power cable could not stay connected to the main breaker for testing because the rectifiers had to be put in. A run of #6 was added between the breaker and a temporary terminal block in the back, and the power cable was moved there. The ground from the power cable to the rack frame was also added and the neutral is now also conveniently available for more test-as-you-build fun.

The Mercury vapor rectifier chassis is now mounted in its place and the PWD "575A Bridge" filament transformer was put on a 1/8" steel panel and mounted sideways inside the rack between the front and rear rails. Despite steel being used, the heavy filament transformer will wobble or oscillate the panel a little if jarred. I would like to find an "L" shaped material of the same length as the panel and mount it on the back at both ends and the center to stop that. It's not really a problem but I don't like things moving.

I am happy to find out the real rating of the transformer is 20KVA!  The regulation should not be changed and be around 2-3% from tube idling current to full current. The ripple might be about 12v or 0.2%.

PWD filament transformer for 575A bridge. Each tube 5V@10A. The 673 is the same tube except the base.

mercury vapor Rectifier hookups
gas rectifier hookups

RCA TT-4:
1.) When two gas rectifier tubes have their filaments connected to the same filament supply the filament shield or anode return side of the filaments must go to the same filament supply terminal.
(Note that some old-design or low-power rectifiers of these types do not have the shield feature. It is still wise to connect like pins together. -ed.)

2.) For gas rectifier tubes in parallel, it is necessary to use an impedance in series with the anode.

In the case of a resistor, it should create an average voltage drop of not less than 50 Volts.

In case of an inductor, a value of 1/6 henry should be used for 50-60 cycle operation. Inductances are preferred because they minimize power losses and help to limit the peak anode currents.

Center tapped inductors (interphase reactors) can be used as stabilizing elements for pairs of parallel tubes. These inductors assure simultaneous starting and equal division of current.

3.) For gas rectifier tubes in parallel, corresponding filament terminals must be connected together. Failure to do this will cause unbalanced voltage drops in the tubes and make it necessary to use undesirably high stabilizing impedances.



I have also read somewhere that when two gas rectifiers are used without a center tapped filament transformer, the filament shield (anode return) pin must be used as the B+ output terminal. It only makes sense. -ed.

If anyone has the schematic for the Sorensen 500-BB "Nobatron" tube type regulated DC power supply, we would appreciate it. It is to be modified as the screen supply for the Tesla coil. A smilar model schematic might do as well.

Some thought given to soft start and mercury vapor warm-up and parts on hand. There's a box of really cool Freeman time delay relays here. They are pretty old but elevator grade. They need the the electrolytics replaced after what must be a few decades, the schematic was traced out. Some cap-goop can be seen around the back capacitors.

These are really classy - 7 pin pillar base and 0.5 to 60 seconds. Made in Fort Worth, Texas. The only sockets I have are phenolic types from old radios. I would prefer ceramics or heavy duty plastic ones.

If that falls through there are some Potter Brumfield AMF types that take octal sockets. Two are 5 seconds and one is adjustable to 180 seconds.

more

more

It might make sens to use the newer PB units. Some matching octal sockets were found. Still, if I can get the right 7-pin ones I would prefer that.

A bulkhead mount power cable connector was found in a box of junk. This may be the solution to not having a pesky power cord permanently attached to the back of the amplifier.

Be careful Pat. If I understand the idea is to mount the receptacle on your transmitter and use that to attach and remove power. If so the power cord would have the wrong gender. Your extension cord would have the potential of having exposed power pins. I know the pins have a skirt surrounding them but the skirt is just to protect the pins from damage. Just a thought Pat, not a lesson.  ::)

I have to agree with Mike.   Those Hubbell connectors are designed to have chassis mounting on both ends.   You hook up the appliance first,  then hook it to 'shore power'.

That last line should give ya an idea of who they are designed for.   Extension cords for marine.   At least,  that's what I was told when I pointed the same thing Mike did to a boss once.

In your application,  since you know the danger,  you should be fine.   BUT,  another disconnect,  maybe the clothes dryer or stove outlet style on the other side would be a GREAT Idea.

--Shane
KD6VXI

How about that. Well you guys are right. Maybe I can find a less impressive and more safe connector just in case someone else is eventually involved. Sure is pretty though.

The idea was a detachable power cord. There will certainly be a stove/dryer plug, NEMA 14-60 at the wall. The outlet is already installed to run the Tucker.

I've seen the reverse of that polarity - holes on the cable end and pins on the bulkhead end. My M35A2 project used that sort in a 100A size. What I recall most about it was the price.

It's needed to get the controls and soft start working before going forward with the power supplies. Instead of waiting to find possibly unobtainable 7 pin ceramic sockets (like for the 1625) and refurbishing the 1-amp Freeman relays it was decided for now to install good quality standard 50's vintage US-made Octal sockets on-hand and the 10-amp Potter & Brumfield time delay relays. One is a fixed delay of 5 seconds and one is variable 1 to 180 second unit. The 5 second unit should be able to be modified for 1 second and that is what is wanted for soft start, more or less. The variable one is for mercury vapor warm-up timing. The relay types might change later but for now that's what's going on.

From left to right on the panel is the filament transformer relay, the soft start relay, the MV rectifier filament warm-up delay relay, and the interlock relay that prevents anything from happening until the airflow switch closes. A keying relay might be added. It's maybe just a matter of interrupting power to the soft start relay coil. A key jack or terminal can be used but the circuit is 120V and it would be better to use a lower voltage and isolate it from the line.

There is room on the panel for terminal strips. One will be the 10-position one previously shown that connects to the control and filament breakers and the coils of the main and soft start plate contactors.

The sockets were mounted on the surface of the panel rather than below it because the panel is thick compared to a chassis and sinking the sockets below it caused interference between the bodies of the relays and the heads of the fasteners. Countersunk fasteners could have been used but were not on hand.

Because the two time delay relays require the sockets to be at 90 degree rotation from each other, it should be impossible to plug in the relays into the wrong sockets. Their pinouts are also different.

Fascinating power plays!
I've been watching your progress from the start and have learned a lot,  Even in detail, e.g.,Your Lambda power supply even has a neon "fuse out" neon lamp indicator across a fuse.  I'd never thought of that.

All Just amazing. But I did modify (t=rc) a P.B. 10 amp octal time delay relay for plate voltage timing specs--. A long time ago for my twin 8875 amp before updating whole thing  with GM3SEK triode control board along with matching power supply mods.

Lambda never did cut corners. The neon across the fuse is built into the fuse holder's cap. Very nice! Somewhere there's a box of triple fuse holders with neons in the caps - but the trouble is making the precise 3-phase holes they need in the panel.

Glad to hear that the timing adjustment was possible on a similar relay.

The sockets on the panel have not been wired yet. There should be some surface mount Octal sockets coming. A row of those would look pretty nice and let me use a couple more relays as well for example a keying relay with a 12V or 24 V coil. The real advantage in surface mounted sockets is having the terminals above the relay panel because that's where the rest of the terminals will be. I need to find some DIN rail as well because a bunch of new-in-box AB terminals for that showed up. It would look nice to have them instead of the black terminal strip.

The schematic for the power supply area (attached) has changed a lot since the work is under way. The timing relays are rated 10A so that eliminated intermediate relays. There will be a couple more relays in the power supply for keying etc. but this scheme here should make it impossible to bypass the rectifier filament delay or soft start no matter what order the breakers (as front panel switches) are operated.

I need some advice on picking the inrush resistance.

Right now it looks like an 4 to 10 Ohm resistor may do to limit the inrush to no more that the 'full power' current while letting the supply charge up most of the way.

The tube is going to be unbiased, and should draw 230mA at 4800V, and that is why the voltage won't charge up all the way with a large value inrush resistor.

I had not given that much thought. To what percentage of full voltage should the power supply charge through the soft start resistor? 80%?
A small resistance lets it charge more, but too large a resistance may cause a surge when the soft start resistor is bypassed because the cap is not well-charged..

Some values for inrush seem to be like the numbers below. It's not the easiest thing to calculate accurately but I think it's close.

The concern is the large current (wattage) spike in the inrush resistor. The wattage averaged over 1 second is reasonable but the peaks are nuts! The simulation says the 31uF cap should charge through the 30H choke in much less than one second. A minimum of 0.1 second with no inrush protection and up to half second with a 14 Ohm resistor. It seems fairly proportional in between but my calculations are not perfect.

What really determines the peak wattage the resistor can take? I suppose the mass of its wire has to do with it. Can this be averaged over 1 second without blowing the resistors to pieces? If so, great! a couple of 200-300W resistors could do.
A ceramic heater element was considered but those have a pretty high resistance when cold and does not seem so suitable for inrush limiting in something this size. Maybe a bank of 1KW projection lamps could serve due to their very wide resistance change. There is a box of them here and they probably would stand up to the duty. I'm looking for guidance on this. Don't mind being creative.

The normal bleeder load (tube idling) voltage should be 5020. -the first line represents that. The cap is storing 380J at most. (31uF/5KV)
It is a 4800VDC @ 2A supply and that is 56A@240V so it's a drawing little more current than on some units.

- inrush resistor 0 ohms
- peak inrush 145A from simulation
- peak resistor power 0W,
- peak secondary current 5A,
- HV 5020V (100%)

- inrush resistor 4 ohms
- peak inrush 60A
- peak resistor power 8441,
- peak secondary current 2A,
- HV 4670 (93%)

- inrush resistor 8 ohms
- peak inrush 30A
- peak resistor power 5710,
- peak secondary current 1A,
- HV 4172 (83%)

- inrush resistor 14 ohms
- peak inrush 17A
- peak resistor power 3824,
- peak secondary current 0.62A,
- HV 3670 (73%)

- inrush resistor 20 ohms
- peak inrush 12A
- peak resistor power 3160,
- peak secondary current 0.41A,
- HV 3160 (63%)

Did I not just post the same question,  on a 3cx3000 project?

Deja Vu?

I believe YOU where the one who broke it down and gave me the maths I was looking for!   Lol

--Shane
KD6VXI

On the subject of ceramic heater elements,  this is Exactly what Henry used in my 3cx3000 based 2000D (single phase,  one heater element)  and the 1500D (8877), also factory single phase.

This is for B plus.   Filament is it's own power supply with a large ceramic glazed wirewound left in line continuous for both inrush and setting of fil V.

They do work.   These only have 4 uF of capacitance along with the choke.

--Shane
KD6VXI

I forgot about that. Maybe I should go back and look at those elements. Too many variables floating around in my head I guess.

The RF stage is going to be built up from a 3CX3000 Henry RF generator. At least that will be the tube mounting deck.

The RF thingie on the back of a Henry RF power generator - this picture is not of mine but on one the e-place. This would be the power metering device?

Can these be repaired? mine is in bad shape and has some parts missing (no 'bird?' slugs of course.. long gone), Anyone got a diagram of this thing?

That is a generic Bird line section.  Not much of a diagram, as the slug wiper hooks to the screw-on RG-58 receptical center pin.

Disassemble yours, either ultrasonic clean or soda blast and reassemble.  The slugs for that are pricy.

73DG

Now to divert temporarily to the other project, Jacob's hi-fi musical tesla coil, which is important not just beacause it is cool but also because we are together doing both projects and progress has to be made on both.

So at this point, to briefly describe the coil - it is a regular coil but driven by parallel 4-400 tubes in class C (like any transmitter). The coil is not run by RF feedback oscillation, but will be driven at the grid.

The screen voltage is 400-500V supplied by a regulated unit. The screen voltage will be modulated to put audio on the system. That seems fairly easy, its power supply is the kind that is pretty easy to modify. The plate supply is about 3000-4000V.  The grid supply will need to be somewhat regulated and provide a swing from about -200 to +200V. The supply we have is OK but isn't lending itself well to the mods. For this reason we will be rebuilding part of it.

Basically the 4-400 grid is a problem because it has to be biased and also driven. The frequency is 70KHz so it is almost audio, yet audio gear is not going to work for this frequency.

The choices we want to consider are the two diagrams. Both use 6 parallel 6l6s, they are already in the power supply.

One is to use the tubes as a pass device, the other use them as a voltage amplifier. Both would have about a 1K resistor as the 'load'. Would either of these drive a 4-400 grid well? It seems so, due to the low resistance of the 1K load.

Hi Patrick,

Great work,  you may have found 7-pin sockets,  and the ones in the following Link probably come from Asia ...   but:

https://www.tubesandmore.com/products/socket-7-pin-ceramic-1625

Real,   vintage sockets are probably in the junque boxes that many of us have.   Know of none in stock here.

73   Have Fun!   Vic

I'm not sure your ultimate goal (as much off the shelf / Homebrew / what),  but......

This seems like a perfect spot for Frank's (GFZ)  mosfet driver,  no?

--Shane
KD6VXI

I have moved ahead with Octal sockets, not wishing to delay progress. There are enough now to build the electronic brain for the amp project.

The Tesla project is aligned to vacuum tube technology, so no Mosfets for now.

just looking at drive choices schizmatic .... may need some refinements ... need a second filter choke for each bipolar power supply or dc voltage and ripple amount will differ...the 4-400 will need up to +290 V swing above bias setting so I would go to capacitor input anyway.... need to run both driver sections as a cathode follower remembering that grid current will be provided on + going swing .... a single 6l6 should be able to provide the 25 or so mA needed but a higher voltage tube would be better, say a russian 5881 or el34....you can raise the cathode resistor value to avoid unnecessary current dumping .... definately put the 6l6 on separate fil supply due to large swing.... make each 4-400 grid negative bias separately adjustable via the  cathode follower... interesting

Here's a more accurate diagram of the 6L6s and shows the three 4-400s. The bank of 6L6s is already wired like so in the power supply. The + and - 220VDC supplies are hypothetical - -one for bias and one for a high current source of DC to be used as drive to the 4-400's. The 6L6s already have a isolated filament voltage. The cathode follower is shown - I agree this would be more responsive, low impedance. Also, the basic version of the load is shown, a Tesla coil. That is already almost made.

The bias of -220V should be enough to keep the 4-400s cut off. the three pots - could be tapped resistors or big WW pots, could control the drive to each of the 4-400s. Would this work in lieu of trying to control the bias individually?

The +220VDC should be enough to provide the +290V swing above the bias setting of -220VDC.

I do not understand the current dumping statement. Do you mean the discharge of the capacitance of the grids of the 4-400s?

hi Patrick ....tnx for including the schemat soze can see what you are trying to do ... the 6l6 grids will need to go 65 V or so more negative than the cathode to cut off the tubes to set the condx shown .... this means an additional bias supply for what is shown ....no biggie but is needed .... big V swing for the 6l6 grids .... good use for single ended step up interstage xfmr.... the 47 ohm cathode resistors help the six pack 6l6 share the load ... the drive pots to the 4-400's as shown are not the most effective way to perform this job and are what is driving the need for the six pack 6l6 as the pot values are low to get the required drive currents .... it would be better to move the balancing adjustment to the 6l6 grid side like done in Arrl handbook screen modulator 6l6 cathode follower ckt

The idea came from this image.

1K was a random value but considering six 6l6s it could be handled. But it is not necessary to have too much current in the resistor- the 4-400 grid should only draw current when the grid is positive and that comes from the cathode follower. I still don't understand about current dumping and the cathode follower's cathode resistor. At 70KHz, what is there to happen?

I suppose my idea is that with the 4-400's all cut off by a voltage of -220V or so, the only difference/alignment need, as I understand it, should be their individual merit, that is the amps per grid volt applied. I can see that if i use two 6L6 per 4-400 grid, some balancing can be done. But that is more of a refinement after some very basic tests. For now I don't want to worry about it, and just pretend there is only one 4-400 driven by a giant 6L6 cathode follower.

So, about the cathode load resistor, the thing I had in mind is like the attached image. The cath. resistor in there does not really have to take any current - whenever the cathode follower is not conducting, it just supplies bias to the driven grid. Ok that is audio, and class B. 77KHz and class C bias is not a far jump.

The whole article which is about a driving a push pull class B stage is thoroughly explained starts on page 138 here in the AGH-AM (128MB large file)
http://69.36.164.11/kb/goth/AGH-AM-027p.pdf

Wouldn't this work OK?  Driving the 6L6 and supplying it with bias could be simplified - It looks like the 6L6 could be driven without grid current.

wow that was a big file    :o ....lotsa good ole stuff there..

yes, I think that is a refinement from the first idea ... since the cathode follower is dc connected to each 4-400, I would split it up and I think a single 6l6 should easily take care of a single 4-400 but this will require 3 driver transformers ... the sticky wicket is the grid swing required for the 6l6 cathode follower ...roughly 350 Vp ....no big deal since the cf should not draw grid current, just use a step up transformer that is good at 77 kHz (probably a toroidal) ... another possibility is to split up the carrier (77 kHz) and audio to different tube sections (grid modulated vs screen modulation) .... this would tend to simplify things

the 15 K resistors would dump much less current than the aforementioned 1 K resisotrs ... use just low enuff to keep the waveforms clean

was thinking of a 6BQ6 for driving the 6L6 grids. Such plate swing is real easy for that tube and it needs only a low screen voltage. It makes more sense to drive the grids like that than to try and float a low voltage (70V p-p) grid driver on the 6L6 cathode line. At 77KHz it gets difficult to keep good waveforms due to capacitances and coupling to large floating power supplies. Transformer would probably be OK but not too willing to go that route yet.

Here's what I came up with for bias and a voltage amplifier. I'm sure it's in no way final but may be a place to start. There should be a way around two plate transformers. Maybe the "economy" power supply hookup. The lower-most transformer is for bias and can be low current.

So, the 6BQ6 used class A as a voltage amp, 6L6s biased off unless drive is present, same for the 4-400s. The text on the left gives the voltages relative to the cathodes with no signal applied.

I didn't work on anything this weekend. Was too tired after excessive WED-FRI work stuff.

looks like some good work here ....one question .... how does the 6bq6 screen get down to -32 V quiescent ? .... maybe another resistor to negative supply ? I don't have much experience with tubes operated with bipolar supplies ... I can see that the cathode is way negative....

 may need a bit of grid resistance in the 4-400's to equalize their grid currents


n.b.  been thinking about this today (now 5 pm) and the screen voltage  for the 6bq6 could be just -32 v ..... maybe a zener diode  or mv 0A2 to regulate it .... a bit of bypassing as well for cleanup 

It's a good idea to regulate the screen voltage. An 0A2, referenced to the -220V rail, which is at the bottom of the 6BQ6 cathode resistor, would give a 112V screen voltage. This may be low but the goal is to run only 15mA plate current in the 6BQ6, so it could be sufficient. Returning the VR tube to the cathode would mess up the bias considering the VR tube should pass 10-20mA depending on screen current. The 6BQ6 bias will have to be lower with 112V on the screen instead of 150V but I don't have any idea how low, because I do not have any screen voltage curve for the 6BQ6/6CU6.

I agree on equalizing the 4-400 grid currents - that may be the least of the worries if wiring is too long as they like VHF.

On a different note - -some friends dropped off a nice looking Systron Donner spectrum analyzer. I have not had one of that brand before but it looks like it does everything and then some. Covers 0.01 to 40GHz. Said to be working but in need of periodic de-ox-it on some of the controls. Must be old, maybe 1971? no picture of this seems to appear on the www.

Model 763  ? - yes.

PN 110225
NSN 6625011224858 possibly 6625-01-122-4858
110594

The units are marked as:
4701A sweep unit
4745B preselector
4809BK tuning unit
1335A variable persistence display (HP)

Friday night!! what does that mean? Tomorrow Saturday I can do some more work on the RF amp..

Last time, basically I worked on Jacob's stuff - but now thanks to you all input I think we have a decent schematic for him to work with and he can start to build that from the Lambda DC supply.

1.) The nice industrial grade surface-mount Octal relay sockets came in, along with some 1-Ohm 100W resistors to help make up an inrush limiter. The goal is to wire up the control system as much as possible. Then the tube HV rectifiers need hookup and the SS rectifiers need mounted up on panels.

2.) I spent a lot of time in 'Front Panel Designer' drawing the cut-out shapes for the big old Westinghouse/RCA meters I want to use. The project's supposed to be low budget 'junkbox' but there's ample justification from a time and effort standpoint to have a panel machined with the unusual cutout shapes. I think there are 8 meters and four will just fit across a 19" panel. It may give the amp that Classic Transmitter look.

3.) Lastly, I have been looking at VFDs and really don't know which brands are better/worse, except that Allen Bradley is one of the best and those with no names are of lower quality.

The blower motor's 1HP. Possibly an inexpensive 2HP Chinese VFD for $110 would be reliable as an Allen-Bradley unit rated just for the 1HP requirement?
http://www.ebay.com/itm/NEW-1-5KW-2HP-VFD-7A-220V-SINGLE-PHASE-VARIABLE-SPEED-DRIVE-VSD-DRIVE-INVERTER-/141441065981?hash=item20ee8b63fd:g:rswAAOSwv2JXwO2z

Hi,
I used a TECO-Westinghouse FM50-101-C VFD on mine, works great.
http://www.factorymation.com/Products/FM50_115V/FM50-101-C.html
73
Frank
KJ4OLL

Interesting on the VFDs.

Any residual receiver noise,  and if so,  how'd ya fix it?   Ferrite?


--Shane
KD6VXI

You might consider these for your project Pat.

Themz seckzi!


--Shane
KD6VXI

Those are super! The HV should be 4800, I would not dare re-scale nice meters like that. Or is it acceptable?  They definitely predate the 3CX3000 era!

I thought W7TFO held the worlds collection of those!  Good to see others can appreciate beautiful meters!

I will look into that brand of unit. the price is about the same. The nice ebay seller going to e-mail me the manual on his (no name) goods.

I really like that blower you have there, have heard they are fairly quiet. How is it working out in the amplifier? Do you measure the pressure to get the CFM yet?


But the 1HP Dayton-powered beast is here and may as well use it. The biggest noise from the blower is what sounds like ball bearings. They don't sound as though bad but maybe about dry. Time also to investigate that and I hope there may be a grease zerk on the motor ends.

The air noise is high but I hope the blower can run at a fractional speed for just the one tube. I think it used to cool a couple of 20KW tubes.

The 'electronic brain' panel now has six Octal relay sockets, more than needed I hope. I also took a 20-25 place screw terminal strip and mounted it for the junction point. All the 120 and 240 'control' stuff except the plate transformer and what's in front of it should be able to be connected there and it ought to make troubleshooting simpler.

The wire terminal strips are like the picture and https://www.amazon.com/Separable-Connector-Positions-Terminal-Barrier/dp/B00DUX005Y  but way older and of USA manufacture. They are asymmetrical meaning one side has a 'foot' with a screw slot and the other end accepts the next section. They just fit together.

One 'end piece' is needed for each complete assembly. There is no limit I guess to the number that can be stacked together but from just this one experience I recommend a screw be installed about every 8 sections for rigidity. These took #6 hardware and cut washers were used so the screws won't loosen.

I had a choice of spring loaded (WAGO) type where the wires are just inserted, the common very popular non-modular barrier strip style where crimp/solder lugs are held by screws, or modular screw type that clamps down right onto the wire ends. The box is full of many kinds and some are DIN mount.

I chose the non-DIN modular screw type because no crimp/solder lugs are required and several wires can be inserted and the screw tightened onto them, and they can be unhooked as needed easily for work. The ones I have are good for 10A 600V. I didn't want to mess with DIN rail.

The planned position for the panel is recessed in the lower half of the rear of the rack - and well away from all the HV stuff. Servicing should be safer. Something learned from the Tucker transmitter - when the back doors are open and interlock bypassed, everything from 115V to 3500V is right there. Something to avoid I guess - don't build safety problems into a piece of equipment.

It cooled a pair of 4CX10,000's and a quad of 8122's.

About 600CFM.

73DG

And I have a couple of those centramax blowers that are single phase..

Hi,
WOW you have really made some good progress on that amp!

The blower is a three phase Rotron Centrimax, found NIB on eBay for $100.
I wanted a single phase, but they were all many times the $$$.
Still cost less to buy the three phase and a $99 VFD!

As you probably know, the VFD can run the blower @ whatever frequency you need to obtain shaft RPM /pressure desired.

Used a manometer to verify that pressure meets the EIMAC 3CX3000A7 requirements.

https://www.youtube.com/watch?v=MNo1uH9-q38

The blower is quiet, but the volume of air required by the EIMAC is loud.
I have to roll the amp out behind the shack, operate it remotely.

Blower pressurizes plenum I fabricated, w/ choke and cathode drive inside.
Used a pressure switch to kill the HV and filament heater if air pressure is lost, so as to protect things.

73
Frank
KJ4OLL

My blower is from two 4CX10,000's and a quad of 8122's?

 :)

I got some real work done this morning. The controls panel is partly done, that is the soldering part. the rest is wiring with screw terminals and wire lugs. Previously a black terminal strip was shown, but it is was for testing and not going to be used in its current place. The wires going to it will go to this panel. Just a change.

A hole had to be made to bring the wires from the right three Octal relay sockets to the front of the panel to connect to the new terminal strip. A grommet should be used but some care was taken to allow the bundle of wires to pass through the center of the hole and not touch the sides. The wire is #14 solid so it's self supporting. Some clamps may be added later but it seems unnecessary.

Some other things that may or may not be used have turned up. Some were just discussed, like the huge 'panel lamps' and the small industrial rotary snap action switches. The panel lamps take a 2" hole and use Edison base lamps. There are no markings on the lamps and all are open.

These switches may do for a 'test' mode or whatever, if necessary.

A nice 24VDC coil relay should  be OK for keying control, and another time delay was found. It can be disassembled and hacked for other time ranges if needed.

This and the next couple are just parts found for possible use, no decision has been made.

These RF or high voltage switches should be usable. Does anyone know what they are from and what voltage they will take? Are they up to 4800VDC? How about use in the RF tank for switching a coil or cap in/out? They are pretty anyway.

Next, a GE rectifier stack. This monster is pretty old and I wonder if anyone knows what it may be from?

Lastly this UTC 54874 item. says in and out, maybe audio.

This imposing switch (as well as the GE rectifiers, audio thing, and HV shorting switch) was dug out of the garage. There is a large pile of stuff there for many years. I have always wanted to use this switch. For scale the escutcheon is about 5" square.

The high voltagse shorting/discharge switch is something that ought be standard on all high power gear. Anyone know what it is from?

Its purpose is to short out up to three HV  terminals to ground when a door or cover is opened.  Hopefully, before that happens the usual mains interlock will remove input from the power supply!

OK well I should do some wiring in the amp tomorrow (wow, later today..) maybe mount up the inrush resistors.

Those rotary switches with the red phenolic and porcelain insulators were typical in WW2 US Navy transmitters by Westinghouse and RCA.

They will easily handle your RF or HV loads.

I have one of those Dwyer liquid manometers as previously shown, will forward.

73DG

The rectifier stack looks similar to those used in cathodic protection of pipelines, typically 20 to 60 volts at several amps. May also have been used in RR signaling, etc.
-very choo choosih looking  ;D

I'm encouraged by the info on the switches. Maybe I can add a 'tune' position using the center tap of the power transformer as a selection for half plate voltage.

With the liquid manometer, how does one prevent the liquid from evaporating or becoming contaminated?
There may be a usable Magnehelic dry type manometer unit here. I got a bunch of them from in 2"-5" ranges ST when they decommissioned the fab. There are a couple of them mounted on my furnace. It amuses the HVAC guy, but it tells me when to change the filter.

They are a bit large maybe 5" diameter face. With care it might mount behind the MV tubes so it can be seen in that chamber but won't make the front of the amplifier look strange due to the incorrectly styled bezel. Alternately it could be mounted opposite a 5" CRT modulation scope. But that accessory has not even been considered yet, so much to do first.

I made my own manometer for testing blowers a few years ago. it's been very useful. It's a U shaped clear flexible tubing 1/4' diameter ID on a meter stick. The tubing need removed and cleaned out, hence the above question about the Dwyer unit.

Very interesting on the rectifier stack. Glad it was not 'tested' for what it looked like to me - a HV stack. The diodes themselves seem elaborate with the cup and a glass bead on each. Maybe I can ask on a railroad board assuming there are those.

Last afternoon I realized this thing has a lot of wires in it and I was having issues looking at the schematic, then looking at what is in front of me and making the mental images match.. I made this picture to help me get the right wires in the right places. No fancy CAD here. Good thing, I forgot to do the coil wiring for the control interlock relay. With a week between sessions, things get forgotten.

today went well. Added the missing leads. also figured out the places for jumpers to go, as each relay has its own terminal positions on that block. It is maybe more trouble to build it that way, bigger block and all, but it allows easier troubleshooting or easier changing to a different kind of relay. time to redo the schematic section again (ugh)  to reflect the larger terminal block, that is very time consuming. It's best to do it this way, as there is one place for all the control wires to meet up and each wire has its place. It's also mounted at a height in the rack so a person can sit on a chair and work on it. Thinking about some old man having to work on it some day.. mounting complicated panels near the floor or at the top is for the birds! Just make it easy to work on.

I did not get the regular sequence of events to happen, as another jumper needs added and there is a wiring mistake preventing the filament timer from starting. Look at it this way, there were no smoking wires or blown breakers or anything like that. I am sure the 70A breakers will pop, but don't want to ruin anything finding out.

The sequence should be:
control circuits powered (the interlock prevents working until that is OK)
blower on
air switch closed (I am faking it)
filament power applied
30 second delay
plate power applied through soft start resistor
1 second
soft start resistor bypassed -full power.

If power is lost or any breaker except the control breaker turned off, all stops and the 30 second delay is enforced. The control breaker powers the blower directly so it can be left on as desired. The filament breaker can also be on with no plate voltage, as long as the air switch gives its testimony.

Keying is by cycling the main plate relay, which makes the soft start happen again. - so every time it is key-down, there is a 1 second delay. Sorry CW folks! Maybe later an electronic TR switch (with chain mail undershorts!) can be made and break-in methods can be tried, but that can wait until the amp is working correctly.

The blower motor was revisited, this one is actually a 1.5HP unit. 1000CFM@4" SP. I believe we are well into the overkill range here. There may not be enough airflow through the tube and the blower motor may overload. I don't understand that part of blowers yet but if the air is stopped off, the motor will take too much current. Worst case at 4KW dissipation and 10,000 FT the tube needs 185CFM @ 2.55", and at sea level 127CFM @ 1.4"

Jacob's Tesla project is moving along. He has rewired the old lambda supply to be as like the schematic previously discussed as possible. Dang he's raided my power resistor stash too. Thanks to all that cared enough to help on this non radio project. He has been learning a lot, especially how to think when modding something from one schematic to another very different one. If we were in a video game he would have gotten the next level experience in hands on electronics today. He said something like 'It suddenly became clear once I was working on it'. I'm proud of him for this, so may young people can't be bothered to learn anything worthwhile or spend time thinking and doing. I suppose I'm old fashioned but the choices I see for youngsters today are to:
1.) fill the brain with carpet remnants and Jello Jigglers^(tm).
2.) fill the brain with powerful analysis and reasoning skills.
Hey kids! Pick one!

And thanks to all that are advising me as this amp project goes along. I know it is taking a long time but it's sincerely appreciated.  :)

The manometer I used has a fill plug that is replaced after, sealing the unit.
The liquiid supplied is of the specific viscosity for the mechanism.
Don’t know about the evaporation rate, but the meter has an adjustable zero point, so it can be “calibrated” for losses over time.
Also, twice the volume of liquid needed is supplied with the unit, so if in future years I can’t zero any more due to diminished volume, all I have to do is remember where I put that bottle of red stuff.

But it really does not seem to be an issue, as once the pressure is set, unless the blower has a fault or a major air leak develops, the manometer use is a one time thing.
I could remove the manometer and still be assured the volume/flow numbers for the 3CX3000A7 are good, and remain where I set them.

The fail-safe is an electrical switch that is operated by blower pressure.
So if anything occurs to reduce the air pressure, the pressure switch shuts down the filament and HV instantly.

Very powerful blower you have there!
If there is anything else needing airflow, and you have determined the blower has excess capacity, consider ducting some of the flow elsewhere.
I used some of the excess air to cool the rectifier.
73
Frank
KJ4OLL

Perhaps someone will drop off a theater organ.  :o 

A pressure switch is a great idea - an alternative to the traditional airflow switch using a vane. Are there any disadvantages to either method?

Perhaps if the air blows the tube from the socket, the pressure switch will prevail in the safety contest.

Dennis, I think your Dwyer would be very useful here if the correct magnehelic is not found - it also does not require me to punch a huge hole. I nearly broke my arm on the CRT hole for the Tucker.

I tried and tried to get a vane switch to work.
Attempt #1 (fail)
https://www.youtube.com/watch?v=AxLVPkwmFzM

Attempt #2 (fail)
Worked fine until I put the plenum cover on, then the vane started oscillating in a way that interrupted the microswitch circuit.
Tried baffles, etc. No luck eliminating the vane oscillation.
https://www.youtube.com/watch?v=YuudZ3Pcnvk

Then one of my many Elmers, who retired from the commercial HVAC business, gave me a very colorful rant about how much he hated vane switches, and how many service calls are due to the vane switch problems, etc.

Thus the change to a “DWYER 1910-00 LOW DIFFERENTIAL PRESSURE SWITCH 0.07 to 0.15 W.C.”
 No problems so far!

73
Frank
KJ4OLL

I see, the air switch required more flow than was present. I noticed that the flow as you said was enough for 2KW dissipation. A level of cooling for 4KW is wanted in this one so that the option is available for CCS operation at full power. After seeing the air switch videos, I understand now about the amp being very noisy and being moved away from the operating position when it is to be used.

The blower on-hand - it is noisy enough on the premise of airflow. Unknown whether a more reasonable size blower will be enough pressure-wise considering the couple of twists and bends in the tubing that will have to be made. 4" tubing should be able to be used for the high pressure side, but it may not be required to be that large.

I think most of the air noise in the 1000CFM blower is coming from the blower intake. the db meter has not been used to try and check, as of yet. I wonder of the 3CX3000 anode helps quiet the noise from below the tube as the air passes through it.

There's a unique opportunity to experiment with  silencing techniques regarding the air flow. if I understand right, those tend to 'consume' pressure like a resistor drops the voltage as current flows through it. Inside HVAC ducts there is some felt like material, whether it is for noise or condensation or what, I don't know. Noise also seems to come from velocity and impingement and rough surface transitions.

The point is that industrially, much louder things are silenced pretty well. Large industrial blowers can have silencers attached - sometimes they are bigger than the blower, but these reduce the otherwise horrendous noise. Modern trailer mounted portable generators in the 100KW size are very quiet compared to older ones. Looking inside, there is what seems to be a simple labyrinth of some sort through which all of the air flows, Air goes in, and comes out carrying the exhaust with it.

If an air stream of z area and y CFM makes z dB of noise, and the velocity is lowered by expanding the area to >>z, and the air forced to move through a couple of sound absorbing baffles, would that no quiet it a lot? An air horn on a blower intake works too, but with what principle I don't know.

If the racket can be cut by even 10dB it is well worth trying. It's a new discipline to try and learn something about, especially when so many comments relate to blower noise and even mounting blowers outside to reduce noise.

some fun info comparing popular "QRO" tubes:
3CX3000
4KW dissipation = 127CFM @ 1.4" water (at 10K FT altitude, 185CFM @ 2.55")
(not stated but this probably assumes the extra margin for filament 375W and grid 225W dissipation)

4CX5000A in SK-300A socket
4KW diss. = 145CFM @  1.1" water
6KW diss.  = 230CFM @ 2" water (SK-300 socket 200CFM @ 2.5" water)
(The above assume a margin of 1200W extra for filament 770W and screen 200-300W)

In the drawing below, it does not take into account completely the differences between the two tubes because the 4CX5000 has a slanted inlet at the bottom of the anode and the 3CX3000 has a straight vertical inlet. The main difference seems one of pressure when mounted in the sockets, which are very different, and everything else, but look at the area and volume of the cooling air flow in the straight, enclosed, tubular section of the anode. How interesting. Here is maybe a more interesting way to look at the similarities of cooling those two tubes:

If we may digress back to the Tesla project, which consumed much time last Sunday, I re-drew the class A voltage amp schematic with measured voltages so it should be very clear.

There is a serious problem which we do not seem to be able to get past. The frequency response is poor. it is pretty flat from about 400Hz to 4Khz, then gradually drops off, and at 15KHz, there is almost nothing. The 6L6s and all the rest of things not needed for the stage to work have been removed and the measurement is being made at the point shown, with no load, no grids, etc. to mess it up. This is to isolate the circuit in which the trouble appears.

The voltage gain at 1KHz is about 9. Measured 90V p-p out with 10V p-p in. The tube is not really made for voltage gain, just for handling HV switching. The tube definitely will work at 70KHz, as that is the approximate frequency of the half wave present in the resonant deflection circuit in a TV set.
I tried a 50K plate load, got about the same thing with less gain due to a lower plate voltage.

The voltage from the generator stays constant.
There is no signal higher than a few mV measured at the cathode, nor at the screen, which are the two degenerative points.

Putting a 10mH inductor in series with the plate load resistor helps, but only when the inductor is the only load present is there any signal of consequence and that is about 60V p-p. It can be tuned with a 1000pF air variable cap.

But none of that RF-style plate tuning circuit stuff should be needed at <100KHz, right? And, if the thing had any ultrasonic-frequency gain at all, it would give much more than 60V p-p when resonated! It should not be needed, this is not a 160M amp.

I do not understand this at all. It goes against expectations, very odd issue. I would expect the gain to fall off but at a much much higher frequency. I thought about peaking circuits, but surely not at this low frequency.

Tube is
grid to plate: 0.6pF
Cin G1 to all: 15pF
Cout p to all:7pF

The drive source is a tube H/P oscllator, rated for 50 ohms and unloaded, so I do not think the Miller effect is to blame and the 10V p-p appears at the grid at 77KHz.

There is something perhaps very obvious going on but I must be too dumb to see it. Could anyone advise me on this?

The progress on the RF amp, BTW:
Got all the control relays wiring sorted out and working. There had been a couple of wiring errors. There is still one exception but it does not affect the basics and should be pretty simple to figure out.

The blower relay is working to prevent filament power from being applied unless the air switch is closed.

The time delay for the filaments, including MV rectifier heating, is working so that the plate power can't be turned on until the Mercury is heated for the correct seconds (1 to 150 can be set).

The soft start is working and whenever the plate is enabled, the soft start is enforced. It's 5 seconds but it can be modified to 1 second.

It was found that if the interlock relay was pulled, the blower and filaments stayed on. It means that the neutral is being found somewhere for both of the time delay relay coils. Seeing these take very little current, it would not take much. The schematic has to be reviewed to find the cause.

But anyway the good news is that the controls work, and the filaments and HV section can start being wired up. It's great news!

hi Patrick .... gain dropoff with inc F .... might try a screen bypass cap, electro paralled by poly ...see if that helps....

what happened to bipolar supply?

The supplies are still there, the schematic was redrawn with just one to make it as simple as possible to find the trouble. Have not had time to try the caps yet.

In the meantime Jacob tried a solid state driver. Lost one 2n3055 already. However the thing is as shown, 12V DC supply, and 6V p-p in, 100V out wih 3200 Ohm load, >500V p-p with no load and the tuning cap across the secondary. That is a microwave oven inverter transformer there, the drive is darlington circuit, single ended.

On the amp, got some work on the rectifiers done.

Eimac HR-8 plate connectors were handy. Black #12 solid core wire goes from there to the tops of the hash chokes. The crimp on ring terminals make this simple. The stiffness of the wire reduces movement. It's only 600V insulation in this link but there is nothing close for arcing. An insulating rigid support such as a Lexan rod or bar may be run across the rack just above the hash chokes plate leads and secure them there. The chokes tend to wobble in a resonant manner when the rack is moved. They are not as substantially mounted as I would have thought being from an RCA BTA-5.

The SS rectifiers will be wired in series with the plates of the tube rectifiers. It will make a 31KV per leg bridge. To the right of the SS rectifiers modules is maybe some space for a bleeder and HV meter multiplier, etc.

I was not going to mount the SS rectifiers up that high because I wanted that space for a 50 ohm to balanced transmission line matcher, but they would not fit anywhere else without <2" clearance and I did not want them near the 'electronic brain' panel because that may have to be accessed with the TX on from time to time. Also with the tube rectifier height, the space above them for the matcher was smaller than wished.

Whatever transmitter these SS rectifier banks are from, it is good because the length of each module is such that when they are center mounted vertically between two 1U panels, the mounting holes on the 1U panels line up with those in the rack.

There is enough room on the other side for the RF amplifier chassis and the matcher. It's not the best thing but it will have to do.

Are those 575A's Patrick? Talk about unobtaineum! I had a hard time finding a matching pair for the 20V3. 

Mike

Mike, They are 673s which are the same except use the Super Jumbo socket instead of the Jumbo socket. There are maybe some 575A here and if they turn up I will let you know. At this point I'm a 673 shop.

Understand about matching with the 20V3. That's a very cool transmitter.

So, looks like tomorrow will see some more work done. Hopefully it will be the filament wiring to the 673s and then if time permits the wiring to the HV transformer, choke, and filter cap. A bleeder still has to be made up with insulators etc. and a temporary HV voltmeter set up. Can't wait to light up the 673s.

It is slow going but it takes care to avoid creating dangerous situations while also keeping cost down. Here is an example of the work of someone else showing why a schematic should not be just copied into hardware without understanding of what's going on physically. The handiwork is very nice and little expense seems spared, but I'm concerned about the bleeder resistor bank for the high voltage power supply. I think it is supposed to be between 7KV and 10 KV on that with a 100uF filter across it.

Today's work. For one thing it is a landmark. Today, Mercury was vaporized.

Finished understanding and marking on paper the terminal barrier strip on the 'electric brain' panel. There was a problem because of constant interruptions and I forgot what I did. Time to write it all down.

Wired up the mercury vapor rectifier filament transformer today. it was more of a chore than thought.. about 4 hours work.

Each tube filament has its own pair of wires going to the transformer. 14 GA stranded wire rated 15KV outdoor duty was used. Luminous sign wire.

The tubes' wires are color coded with 1/4" wide bands of heatshrink tubing. It is therefore possible to troubleshoot easily, with 8 wires going from the tubes to the transformer. The 'cathode shield' pin of each tube is further identified by a white HST band.

The cable of eight of those 15KV wires was not very pretty, could not be made 'squared off', so it was run through the 'wall' of the rack, through a modified rack panel made of insulating material. Between the two bolted-together racks is a space about 3" wide. So, the insulating material 'wall section' hides the wire. The wires come back out just to the left of the row of MV tubes. This was a great help to making a tidy and orderly appearance. It also cost nothing but labor, in keeping with the "junkbox/rat" theme.

Today's pictures showing the MV tubes show the operation with filaments on and Mercury vapor cycle running. I am very pleased but tired. Next logically should be the high voltage work.

I have been thinking about adding 120 or 240V LED indicator circuits inside the amplifier, across each controlled item so that the state of any section can be known at a glance. This would be for a troubleshooting aid. Why not? Modern systems based products have many lights inside to show what the relays and VFD and PLC are doing. Most of the few LEDs needed for this beast could simply be mounted in the 'brain' panel. Later for that stuff.

and

lastly,

The insulating panel cut outs. I hope to eventually enclose the MV rectifiers in a box of some kind with a little forced ventilation. Just an appearance thing. It would be nice to have the inside of it black so when it is looked at, the tubes and glow is seen but not the whole inside of the rack there.

When I rebuilt the 20V-3 it was important to me to keep it original. That decision meant MV tubes a priority. The reward is to watch the violate vapor dance with modulation peaks. rewiring the MV sockets was an experience I don't care to repeat. I lost track a few times of the circuit polarity and studied the schematic until I was totally confused. ;D In the end the wiring was right! a proud moment. I replaced all of the GTO wiring and even learned to knot and string the loom like the original. It is not visible to anyone but me. Like you, looks are important!

Mike

Today no physical work was done but the inrush setup was decided. The low values I wanted which would allow the unit to fully charge in a few cycles would not be acceptable with this filter. It would be OK with a capacitor-only filter. A compromise with a larger value to soften the rate of change of magnetic field in the 30H choke looks like it would only require a delay of about 0.2 second and charge to 90% of the voltage with the tube running idling current so it's no great sacrifice.

Because the power supply uses so a large choke in order to give best regulation, the inrush will cause a high voltage pulse at the output of maybe 8800-9500V (third file) unless the current during inrush is limited to a small fraction of the operating current. It looks like a resistance of 1800-2200 Ohms in series with the plate transformer secondary would do. This with the 25:1 turns ratio in the transformer means a 63-86 Ohm resistor as the inrush limiter in the primary. It could be small like 220-300W with a delay between re-starts of 5-6 seconds, but in case of a failure where the soft start resistor is not disengaged then a 600W rating is minimum.

This brings up possible pulsing issues when the drive is turned on and off or varied widely, etc. but with the tube having a huge 225W grid there may be options. Audio rates don't look like an issue but jumps in level from cutoff to idling, and from idling to high power  could be.

Does anyone have experience with momentary excursions above the 3CX3000 rated plate voltage? The datasheet talks about pulse width modulation and 15KV, and this is why I want to study the Power Rock manual, to understand.

I've run the 3cx3000A7 at 11.5 kv.   It would pump out 20kw plus,  pep.   This also used a bit higher filament voltage when keyed,  and slightly lower when unkeyed.  This was mobile,  with a couple three phase alternators.   I wouldn't suggest running it that way.......   BUT,  I don't think you'll have ANY problems with that spike.

The current incarnation has 4kv on the high tap of the xformer.   I've been able to run near 7.6kv, dropping to 7.2 keyed,  with no latent problems.

Hope this helps,  Pat.

--Shane
KD6VXI

That's pretty encouraging. It does help, thanks for a direct account about this. Well 20KW might melt the copper off the copperweld. Heck, the 6-7KW that my little project might be capable of could endanger the OWL. But I'll be running the boring 1500W levels.

There are plenty of amplifier videos where the tube is run at overly high voltage according to a panel meter.
The swinging meters in some of those videos compete with the amp's builders for Guiness Book lying.

I believe the tube needs to be gettered after sitting 10+ years, before trying it. Somewhere on this board is a good topic on that. I should also finish my 4KW CCS dummy load. It's like the previous one, a bunch of chip resistors, microwave type, on a huge copper heat sink. Power supply first though.

Looks like all is clear to proceed on the power supply.  :D Unfortunately I am traveling from TX to CA in the company truck and the next two weekends won't have any activity. I will be getting some training on commercial UAV maintenance and operations. I may stop at Murphy's on 10/15 if time allows.

Hmm.   Murphys,  UAV action...   Someone's going to San Diego!   Lol.   Make it up Bakersfield way,  lemme know.

I do not recommend running the tube that way at all.   It was just as a been there,  done that thing.   It was a 30 second burst,  so to speak.

Continental also marketed the power rock as another model,  I believe a 315-R1 rings a bell.

VE7RF has heat tested the 3 and 6000 tubes and found the diss ratings can be increased if you employ a chimney design that allows adequate airflow.   I believe he got a neat 50 pct increase on the 3cx6000, but I could be off.....   Been years since we talked about it.   Kind of like the yc156 being a 15000 tube with a 5000 radiator.


All about heat removal with the filament and grid the 3k and 6k have.   The CBers proved that.

--Shane
KD6VXI

I'm really impressed by this project- wish I was capable of doing this. So I have to ask - what is the projected completion date? I can't wait to hear it on the air.

@KD6XVI I wish I had the time to do it and there are more surplus places in the area too, but since we announced our certification we already have many local people wanting to get their drones and UAVs serviced.

@WA2SQQ Thank you for saying so. I have many changing life-duties and no defined completion date. Maybe sometime next year. Basically the work is done about 4 hours a week. That may seem like a lot but it took 20 minutes to find and then about 2 hours to cut, paint, and install the insulating panel. Some things have to be made out of of other things and it's a job sometimes.

While in CA I visited Murphy's Surplus. The store has been re-organized and a visit is recommended. I had called him the week before and said I was visiting, asked about Saturday hours, etc. Fortunately I was able to get there on Friday later afternoon and avoid an extra hotel night, that is, I can't expect my boss to pay for my fiddlepharting around.
The traffic in CA was horrible as afternoon turned into rush hour. I spent over 3 hours doing 10-20MPH average. Imagine dropping your colleague off at the John Wayne airport at 1:30 PM, then being in that traffic and watching the precious hours of Murphy's business day tick tock away. I called and he said he would wait for me. That is what I call customer service!

I found some of these which I think are "finned strip heating element"s. Each measured 29-30 Ohms. Horrendous indestructible wattage I imagine. I bought three and there are several more. They are about 18-20" long. Price was just $5 each. The image is from a new one online. These are dirty and a bit oxidized from use but should clean up. They may be good for an indestructible inrush resistance. They may be 1KW size from an electric "HVAC furnace". The brand may be "wattco" or "hotwatt" http://www.hotwatt.com/stripmfgf.htm  and possibly on page 8 of this PDF: http://www.hotwatt.com/images/Web%20Page%20June%202010/Strip%20%20and%20Finned%20strip.pdf The bodies are stainless but the fins are steel and those are what really needs cleaned up.

I also picked up a partly cannibalized SONAR unit. 7" P7 CRT, magnetic deflection, the HV board seems intact but most of the other boards are gone. Its unusual in that it is a civilian or commercial unit and has a 360 degree display, a PPI display like a RADAR set. I think it could make a fine scope clock or something else, though it if were complete I had in mind an open air SONAR or a LIDAR.

Back to the amp, because the rectifiers are taking up room at the top on that side, the tuner probably won't fit there. I have a 19" panel with two CRTs in it and it may be worthwhile to consider a dual modulation scope, one the regular kind with sweep on the H axis and RF on the V axis, and the other set up for trapezoid where the incoming signal is detected to make audio, and that is plotted against RF. Alternately there is an HP 1300 type display or two around here. These are rack mountable 20MHz scopes with 9 to 11 inch rectangular CRTs. Lots of options.

The tuner should fit on the other side above the RF section. It all remains to be seen. If I am out of room I may have to relocate the solid state rectifiers. Wanting to include a high powered antenna matcher is one reason two racks were chosen.

Amp stuff today.

As mentioned I procured three large heater elements from Murphy's in El Cajon, and a more suitable blower. The blower seems to make 1.25 inches of water pressure according to my home made manometer. I didn't have time yet to refill the nice one I got from Dennis that takes special fluid.

This is to say that the water in one column moved 1.25" when the blower was running as shown in the picture so I hope that's the way to measure it at 1.25". It has plenty if air volume available. I had to cut two small windows in the cardboard test block because when it was blocked off and not flowing any air the blower was stalling noisily and the pressure was varying. When delivering air it is nice and quiet. If there is no impediment to the air at all it 'overloads' that is the motor current goes way up. This is the first time I have witnessed and played with this phenomemon but it is mostly common on high performance blowers and very interesting. In the test it ran 2A. The nameplate says 7.6A and that would be with maximum air flow. The air flow will depend on the duct and what can be shoved through the tube. It's interesting that the pressure stays constant as the air flow is varied over the range of the blower. It would be boring old stuff to fan experts but it was interesting to play around with and see first hand.

That blower came from a Bauer 5kW AM TX.  It kept 2 4-1000's and a 4CX5,000 happy.

73DG

The blower is perfect for this project. The use for the heaters is the inrush limiting. Some cleanup was needed due to a little rust on the fins and this could have been done with a wire brush but it was decided that with them in series for 90 Ohms total the fins could be removed and there would still be no danger of the units overheating if the soft start relay failed to close. The heater bodies are stainless steel. 240V across 90 Ohms is 640W or 213W for each unit. It's a lot of resistance but a very soft start is wanted to minimize the HV pulse toward the end of the capacitor charging. It should charge in about 200ms. If the pulse is not an issue these can be using in any combination of 30 Ohms each to speed things up.

Removing the fins allows many more options for mounting. The fins are easily removed by grabbing the edge with a channel-lock pliers and twisting sharply so they break. The units can be supplied without fins and clamped to a vessel for surface heating, so in an extreme case they could be clamped to a rack for dissipation. This is not necessary here and they were mounted in a convenient way. The terminals are 3/4" from the side panel of the rack when it is installed. These giant heater units were just $5 each at Murphy's Surplus. There are a few more of these there at this time as well as some of other kinds and also some long quartz elements that would do a similar job. It beats paying big bucks for a resistors of similar specifications.

The last picture shows how much more room there is for the new single phase blower. It might be moved to the other side to be under the RF section and the large filter cap moved where the blower is now shown.

First the text, what was done and why, then some pictures from Sunday.

Last weekend no images were taken. Despite some sort of evolving diagram, the soft start resistors were put across the wrong mercury contactor. This kind of thing happens when something is left along for a week at a time. It took a couple hours to redo that wiring.

Jacob made up a house-type 120VAC receptacle with mounting plate and ran some wire there for the blower motor. We thought it would be best to use a receptacle and plug for the blower so that if it ever needs service, it can just be unplugged and removed rather than having to unsolder or unscrew wire terminals while crouching low to the floor. I don't do that very well any more. The 'extra' side of the receptacle can feed other cabinet cooling fan(s) and or disco lights to be determined.

Today some difficult decisions were made resulting in a big time-out for planning and measuring. The result of that was a major and probably final  re-arrangement of components. This was on top of finally connecting the two 'floor' planks on each side together with stout "L" iron and mounting the plate transformer. The final result is that next weekend, barring problems, we can start running the high voltage wiring.

The solid state rectifiers were removed from the inconvenient location at the top rear of the right cabinet and moved to the lower front of the left cabinet. The space at the top of the right cabinet was re-claimed for whatever lofty purpose. Meters, tuner, scope, whatever ends up there after the RF section is built.

The blower was moved from the the left cabinet to the right cabinet and mounted with the scroll case facing forward. A thickness of sound absorbing material to be determined should be put on the backside of the lower right rack panel, and a cone or some other air-turning device put in front of the blower. This may eliminate some of the blower noise as long as the blower is not too close to the panel. There is a convenient path from its output for a flexible duct to the left cabinet and vaguely planned RF chassis location.

The filter capacitor was moved to the front of the right cabinet and sits front-to-back beside the blower.

The filter choke now sits at the back of the right cabinet. An insulating device has to be made later to prevent acidents in case the electronic brain has to be troubleshot with the unit on, but I am thinking about a set of lights to indicate the status of the controls so that troubleshooting is made simpler.

A piece of 1.5" "L" iron is screwed into the floor planks on the right side at the front. This connects the 'floor' planks on the right side together to make them stay level with each other (another at the back will be added) and this front "L" may have to also support the front rack panel because there are no rack rails at the bottom of the right cabinet. The cabinetry is an AMCO modular enclosure and is now their legacy or FX line of Electronic Packaging enclosures, so some rails may be able to be had. The "MXL mounting channels" look like the right thing. The fact that this is still made today says it is a timless design.

There is enough room next to the solid state rectifiers for a bleeder stack. There just happens to be an insulating board with ten 100K 100W resistors on it from a cannibalized 13KV 2A regulated power supply. These are Ohmite and their spec is that the unit will take the voltage required for its rated dissipation. In a 100K 100W resistor this would be 31mA and 3162V. The resistors are designed for convection cooling but maybe a fan will be used to remove heat from the area and increase the safety margin.

Possible bleeder resistances using the board with good safety margins are 250K@20mA, 100K@50mA, 50K@100mA, 40K@125mA.

Bleeders : HV @ bleeder load
250K: 6620V@20mA 132W, 13W per R (10 pcs)
100K: 6021V@60mA 361W, 40W per R (9 pcs.)
50K: 5400V@108mA 583W, 72W per R (8 pcs.)
40K: 5188V@129mA 669W, 70W per R (10 pcs)

Normally the RF tube's idling current will hold the power suply in its good regulation area. There's a balance to be struck between the danger that the tube may not light or work, causing the voltage to increase too much, and the wasting of hundreds of Watts in the bleeder. A bleeder has to be there but maybe there is a way to detect a tube-not-running condition and shut off the HV. That would avoid a 500W+ bleeder. (That said the Tucker TX only makes 700W carrier/2800W PEP and it has 200mA of bleeders; 680W!! Big bleeders: not 'fraid of 'em)



If anyone wants the 3 phase plate transformer from the previously mentioned cannibalized power supply let me know. I don't know the primary or secondary voltages but the connection was fullwave bridge of 18 Varo VC80 8KV 1A rectifiers (3 per leg) with a 1uF capacitor. It's in a 4 FT rack with wheels (some undisclosed low-value parts still in there too) in Gainesville TX. This means it's cheap if someone wants to pick it up by appointment only.

Rear view of right cabinet. Choke is at the back, cap and blower up front. An L bracket still needs put in back here to connect the 2x12 floor boards so they are flat and level. Recall that they are supported only at the rack's 4 corners so a heavy weight makes them tilt down in the middle rather than act as one piece. When this was done on the other side with the plate transformer, the boards could be seen being drawn up level as the lag bolts were tightened. Each "L" bracket has six lag bolts 1.5" long. Holes about the same size as the lag bolts shaft diameter of 0.205" (not the ~0.25" thread diameter) were pre-drilled. I think a 7/32" (0.21875") drill was used and a 3/16 (0.1875") might have been too small, concerned with splitting the wood maybe.

Rear view of left cabinet. Silicon rectifiers mounted up front. They are 3u above the rack floor. The thumbscrews, remember are how the HV wire attaches and this has to be made easy to work on because it may need service at some future time. The bleeder unit is just standing there next to the rectifiers. It should mount up to the inside of the rack and already has standoff insulators. At the very top of the bleeder board is already a HV resistor of 20 meg. It probably went to the regulator error amp in the power supply the bleeder assembly came from. The plate transformer sits on the two "L" brackets in this side of the rack. Its weight alone made the floor straighten up but it needed mounting. The brackets were fastened to the floor with lag bolts. The transformer is held to the "L" brackets by 3/8-24 grade 8 bolts. I do not claim it will stay there if the cabinet is turned upside down but it isn't going anywhere in normal situations. There was a long and complicated procedure to do this "L" bracket and mounting job without taking the transformer back out of the cabinet. It's too tedious to describe here.

Front view of left cabinet. The space can be covered by a panel. The plate transformer is taller than the space taken by the silicon rectifiers and bleeders. The rack floor space seems very efficiently used, yet everything can be gotten to, and nameplates seen for those who like to read them. It may seem like a lot of extra HV wire would be needed to have the silicon diodes here and the mercury diodes up front but since each silicon diode stack is in series with one mercury rectifier, and the choke and filter are both close by, it only adds a few feet to the whole. An "L" bracket can be seen in the right side cabinet.

A view of the nice receptacle Jacob made up. The plug and green twisted pair go to the blower motor. The messy power cable hookup will eventually be relocated after the power supply is completely done.

Reclaimed space in the top of the left cabinet.

A possible improvement might be to rotate the silicon rectifier assy. 180 degrees so that the stacks are to the left. The extra space for the bleeder unit will then be on the right and a small amount of air from the blower can be directed at the bleeders. The hacked up picture shows the idea.

It is not too hard to make little nozzles or vanes from cardboard stiffened with epoxy or from plastic softened and bent with a heat gun or to use a fiberglass repair kit in some way.  The drawing shows the idea.  It does not have to be elaborate or expensive. . I have done this kind of thing in racks before where a single blower fed a cardboard duct and it ran down the backside of the rack with little cut out windows in it, the flaps of the windows aimed the air at the receivers and how open or closed they were determined how much air each got, so the ones at the far end got as much as the ones near the blower.

Sorry if this is tedious but these small details may help prevent heat damage over the long term. The resistors could use convection but the board they are on is subject to be being heated.
After seeing how the circuit boards in some old broadcast transmitters are brown and cooked just like the board in a cheap TV set, there's no excuse for that.

Most of the power supply HV wire was done today. Not quite ready for test.

The HV circuit wire was run in two PVC tubes because there is low voltage wiring in the center section. Its 15KV GTO #14. Rectifiers and bleeder board all mounted.

Different colored wire ties are put on the HV wires so to identify each end of the same wire.

The bleeder board is on 3" standoffs and the resistors are at least 1" from the metal of the rack. The resistors face the area where the blower goes. Right now, 9 resistors are used, each 100K 100W in a 3x3 arrangement.

There is also a mess here from this project.

Pat,
I'm very impressed with your amp. I wish I had the knowledge and parts supply to visualize that project and just start putting it together. Good job! Bye the way, your work area looks like mine. I would feel at home there.  :D

Mike
KE5YTV

Thank you Mike. The parts have been collecting for a long time. It's not so different from any amp in the handbooks except the size. Just some thing I always wanted to try.

Today is momentous because the power supply was tested. A 0-10KV HV meter was temporarily made for testing using a couple of 100M HV resistors and an old 50uA meter.

The MV rectifier filament delay and soft start worked great.

The cap has 3/8" terminals. There were no ring connectors that size so some adapters were made of pieces of plated steel strap so 1/4" hardware could be used.

The HV was 6200V max. with the 100K bleeder

"Leakage current"  ;D
That's a good one.

I read that and had to go take a leak... ;)

73DG

Just 32uA to deflect it to read 6400V on the 1.0 scale.


That meter is a good find and might be nice for a hi-pot tester later.

What is a normal current range for a tester that hams would find useful? Do they have to go to 1uA or less to be non-destructive?

This weekend most of the remaining mechanical work was done on the power supply. Brackets were made for the HV cap and the choke. An air switch is available but has no vane so one is being made. The blower - it was decided to mount it on a shock absorbing mount to reduce noise, so one was made. The blower is quiet but there is always some conducted humming sound and whine.


The blue plastic at the bottom keeps the metal bolt from abrading on the capacitor. The top section is made so the screws are held away from the cap.

The choke is held in place by a steel strap. It may look like it could be loose but it can not move at all. The air switch happens to fit on the blower perfectly.

The mount is made of some thick foam rubber sandwiched between two thick aluminum plates with a generous helping of weatherstrip adhesive. The bottom plate bolts to the floor of the cabinet. The big holes in the top plate let a socket through to tighten the bolt.

and

Today the vane was put on the sir switch. The modified paper clip soldered easily to the brass vane cut from a sheet. A cable was run from the switch along the power cable of the blower and a connector soldered on. The  blower with air switch was installed and wired in. The connectors allow the blower to be easily removed if it needs service. The blower also has a ground strap that screws to the frame of the rack. There are spare pins in the air switch connector for possible future use.

Blowers have definite ranges of pressure and air flow. This blower for example, if left free, will overload because it moves too much air. The air flow is properly restricted by the load such as the tube to be cooled. The pressure remains relatively constant with varying air flow and the current drawn by the blower increases with increasing air flow. The name plate gives the proper current for the motor. The air flow can be adjusted to give this value by means of a damper in the path of the air. It is quieter if dampers or restrictions are in the output air flow rather than at the blower inlet.

View from back. There is just enough room for the blower, capacitor, and choke. That means the space was used well!

The grounding arrangement of the blower. Note the lug end of the ground strap is soldered. All crimped connections are soldered.

This is about it for the power supply except keying, removable power cable, HV fuse, a few low current breakers, and a set of monitor lamps to indicate the progression of control and power supply start-up and soft start. The electromagnetic brain was finally tested with the blower and air switch and all is working correctly. 240V lamps or 120V lamps with resistors in series should be very reliable. LED panel lamps could be used but the ones available locally seem very cheaply made and they look like cheap plastic too. I would like to match up some jewel-lens panel lamps and use them.

Keying could be electronic for the tube but may make trouble because of power supply dynamic voltages when the HV supply is left on. It may as be impossible to have fast enough TR switches. An electronic TR switch may be useful.

The normal keying for this on voice can be by the plate supply with a soft start each time. The soft start should take less than one second to fill the capacitor. Keying the HV supply and retaining soft start is as simple as breaking one 120V wire. So that's about it on the power side of things.

This is a dual project - the amplifier and jacob's modulated Tesla coil. They are intertwined because we  are building them in the same place and helping each other. We need some help with the MOSFET driver and hope those with experience with FETs can assist with advice. Fortunately MOSFET are cheap as we have blown 2 or 3. So, to give an outline (schematic to follow) the driver design has changed from modulated tube power supplies to a push pull FET amplifier and separate grid driver and DC supplies. The frequency is only 75Khz so on the surface it would seem simple but it is different in practice.

Here's what we are experimenting with. The simplest way to drive the FETs in push pull was just to tap off the vertical amp of a scope. lots of gain that way. It works well.

The trouble we are having is that there is some 120MHz bursts of parasitic oscillation lasting 1uS or so, around the gates of the MOSFETs and we are afraid of this blowing them.
This parasitic waveform is quite large, well over the gate max volts in both directions.
We do not know how to stop this.
Does putting MOSFETs in parallel invite this?
The 75KHz drive to them from the scope amp is approximately a sine wave, but it is clipped by the zener diodes to a +/- 12V trapezoid or square. The burst of oscllation seems to be where the FET turns on as Vth is reached. It's very strange.

Ferrite beads to quench the oscillation? Maybe a couple.   On the gates of the fets.

I've had to use them on the base leads of high gain amp transistors before to damp vhf energy to keep them stable.

I got the idea from the ARRL solid state design handbook I believe.

You want a mix that is very active at the freq of interest.

--Shane
KD6VXI

My Gawd
Is this an Area 51 project? Some serious engineering for the boring 1500 watt PEP Ham radio crap.

Are you sure that this is for Earthly communications????????? Talking to aliens from another solar system?
You might see pop-up ads appear in this post, because it has surpassed 10,000 views

Good advice on the ferrite beads! There are many strange things going on with that coil experiment and the picture may help explain, the lash up is very something. The coil and 4-400 chassis/power is not shown in the picture. A 2x 811 chassis in the first picture is being used as a test pig.



On engineering, I'm just trying to avoid building weak links into it. If it's made like a battleship it should last a long time. I hope that people won't get bored when seemingly slow progress is made because we are having great fun doing every 'home made' part of it. Many times there are delays because something has to be made or modified and that work can't be shown, only the result, but be assured it was fun to do it. Up til now, the power supply - it's the simple part. The RF part is the hard part and yet to be started.

The second picture is just some panels starting to cover the HV areas for safety reasons while other work is done. It would not do to accidentally stumble into, lean on, or touch something hot.

Does anyone know of a kind of push-on rubberized, cylindrical shaped insulating cap to be put on and cover up stuff like choke terminals, cap terminals, bolts, etc? It's very dangerous to work on this.

A bag of small blue indicator lamps was found. These have also push buttons in them. Incandescents can burn out but if operated at slightly lower than rated voltage should last many times longer than the spec. These things have all-metal housings, high quality types, with push-in lamps of the bi-pin style. They might be used inside the back of the amp next to the 'brain' panel with the relays, to indicate progress of start-up. When something is turned on, the lamp should light. The push-button feature might be usable as a lamp test or something. It's a lot of extra wiring though.

I have not thought about overload relays but those lamps and buttons should be on the front, and be larger. The regular sized panel lamps were dug out. There may be some more but this should give enough variety and numbers. The question is whether there are enough 'jewel' type lenses that fit the enclosed panel-mount sockets.

I did not find the stash of 1" panel lamps which would be even more better. Also, no two industrial push buttons around here seem to be alike, so I may have to find some. The price of new Allen Bradley is outta sight!! So forget that! Maybe some used ones will turn up and be all the same style.

Automotive starter/alternator 90 degree rubber boots work great for insulating large HV bolts.

73DG

That's a great idea.

I enjoy pieces of vacuum hose,  cut judiciously long.   Makes for great insulation 'posts'  on your hv posts.

I've also used  if you can find them now,  vacuum port close offs.   Basically,  they where larger versions of the bat handle toggle markers that slid on.   

I'm sure those are for off road use only here in California,  but you may find them at your local auto parts store in Tejas.

--Shane
KD6VXI

Have you had any issues with some of them having perhaps high-Carbon or otherwise being slightly conductive? I would prefer them to go 90 degrees, as the terminals are almost always a '90' when the post and lug are considered.

I found this site which is possibly interesting. Maybe part of this is the desire to insulate, and partly to make certain high voltage terminals very obvious. Red is good.
https://www.electricalhub.com/battery-terminals-and-lugs/insulator-battery-terminal-boots

Something that has to be done is to shorten the 5-second delay of the Potter & Brumfield fixed delay relay to <1 second for soft start. About 1/2 second looks good. This turned out to be as simple as changing the timing capacitor. A good quality capacitor has to be used.

The current schematic is attached. Its point is the power supply and control circuits are completed aside from adding indicator lamps, that will wait til almost last. Just a 1U panel for that. The RF section is somewhat hypothetical at this point but is generally what is wanted. The ATU is not shown.

I have both colors in lots of sizes, never had a leakage problem.

73DG

I haven't with vac hose either,  but that does bring up a good point.   I've never tested for leakage!   They could be hot,  and I'd never know.

--Shane
KD6VXI

The two projects are opposite ends of the same hopefully bidirectional thing. beyond that, the attached document and the fact that there are no semiconductors in it at all above a rectifier, is the only information of the project approved for public release. Had to disclose to get funding.  ;D

(but ads I have never seen any here in ages pop up or be summoned from the depths)  ???

spark plug angle boots

Hi,
WOW, you are really making great progress on the amp! The local power Utility grid should be afraid….

A couple of years ago, my Elmer found some switches for our home-brew 3CX3000A7 @ Mouser.
Multiple color lenses, illuminated, different logic options, etc.
We went with two lens colors, (I have not yet connected the lamp power, too excited when the thing actually worked)
but here is how we implemented them:
https://www.youtube.com/watch?v=vywkoK5LKQc (https://www.youtube.com/watch?v=vywkoK5LKQc)

All this talk about putting rubber caps on high voltage power supply connections and components scares the heebee-jeebies out of me.  I have designed and built transmitters running up to 20 KV at the anode, and operated quite a few running less than half of that.  I'm not afraid of high voltage, but I respect it.
Why anyone would want to reach into a live high voltage supply completely escapes me. I admit to having put a clip lead across a door interlock, but I always operated the rig with my paws well out of the way of high voltage.  These rubber high voltage condoms would only to serve to convey a false sense of security and lead to less than fully mindful behavior.  There are few enough fellows doing good homebrew work left these days, and it would be a shame to lose one halfway through the project.

Sound advice Norm!
I spent a year rebuilding a spent Collins 20V-3 transmitter. In the process I cleaned and tested all of the substantial high voltage interlocks and safety devices. For a 1960's design this transmitter is a marvel of analog logic and mechanical implementation.
I have great confidence in the interlocks but I still use a bang stick first and use good practice meter handling. One nice thing is the large picture window on the front of the transmitter. I can hang a meter inside the TX, run the leads and close the doors, reset the interlock logic circuit and key the TX without putting my hands inside.
On another TX newly delivered. I needed a circuit test on the HV area. I opened the door and was about to reach in. Something stopped me! I rigged a Bang stick and reached in with two fingers on the wood handle. BANG! the spark almost flipped the stick out of my fingers. The rebuilder had stripped the interlocks from the TX. No excuse, I should have done an inspection first AND used proper safety but I was in a hurry.
Don't be dumb like me.
We can see that Patric Knows what he is doing. He is probably passing his experiences on to his apprentice. All I can say is NEVER be in a hurry and always take a moment to THINK BEFORE you reach in.

The spark plug silicone rubber caps are interesting.

KJ4oll/Frank, That is cool shack. with a lathe, mill and brake. That's all I saw. I can only imagine. Git that work of art on the air. How do you put up with the fan noise? Is it made to live outdoors? I saw it sitting outside in one of the online pictures.

Mike

Patrick has cats, and they can get into the smallest spaces.  He doesn't want harm to come to them, exploring vs. HV hence the insulator question.

73DG

I can't find the post about someone intending to reach into a live HV supply but I think there is one that mentioned  enclosing the HV areas because the door facing the controller may have to be open in case of troubleshooting. The rubber covers are for cosmetic and marking purposes and their use is nothing to worry about. They cost money and so might not be used, in accordance with the 'junkbox' rules.

There are these nice "HV shorter-outers" here. Eventually one should be put inside each of the two rear doors to operate after the interlock switches (yet to be rigged) turns off the HV by opening a contactor. It should normally never be needed to have the HV on with the back doors open although during the building it may be done from time to time.

It has not been decided how to handle the HV-grounding function. Shorting a 31uF cap is not a good idea. There's an idea already for the 'hv shorting stick' that there should be two terminals for it. One through a resistor bank to limit the current to non-fireworks level and discharge in a second or so, and the second terminal direct to the cap to just hang the stick on. This kind of thing also needs to be fused with a HV fuse, maybe the #30 or 28 wire type  at some HV circuit location so that any direct short will not cause carnage. There is room for a few wire-type HV fuses in different areas. I think there are enough insulators left to do them.

One rule here is not to break the plane of the rack enclosure while it is plugged in. It's a pain to have to unplug and discharge, rig the test meters and stuff  then re-plug in and test but that is the only way to avoid accidents.

The cats are not allowed in the building. It's too dangerous a place for them.

In the diagram, I am suggesting a simple way to have the cabinet interlock, HV removed, and also provide for the shorting stick's safe use (non-lightning/exploding discharge). To have a 'soft' discharge of the capacitor, a resistor has to be put in series with the discharge switch. The resistor will be subject to the 4000-5000V of the HV supply, so a typical single "200W tubular ceramic" of a low enough resistance to bleed off the supply when used is out of the question. Fortunately the bleeder assembly from the Henry generator is intact. It has what look like four 100W resistors in series, all mounted to a metal plate with ceramic standoffs. The may serve the 'soft discharge' purpose. The diagram shows 3 fuses. I have no objection to more than one if it makes the unit less likely to blow something in case of trouble. They cost so little to make.

It's time to start thinking more the RF section. The basic chassis will be made from a Henry RF generator. It was not quite complete so I can't find a model number. Its power supply is 3-phase, apparently 208V between phases because that is the tap used on the filament transformer.

Looking at web pages, I don't see how to tell which it may be, the 2000D or 2500D. The 3000D uses an HN connector, the smaller units use N connectors.
The RF deck says Henry 13.56MHz, CAT # 9600-0050, SN# 28-E4-90. From all that I don't know. Maybe it is not so important.

The RF deck has already a 13.5MHz grounded grid RF amp installed complete with PI-L tank and PI network input matching. This one needs little but cleaning and the replacement of one of the ceramic standoffs that support the filament 'socket' under the chassis. The unit includes a coaxial filament choke, provisions for metering signals, and apparently the 'standby' bias resistor, cathode fuse (1.5A) and safety resistor, and other goodies. As much of this will be used as makes sense.

There are options for something like this, in order of increasing usefulness and trouble required:

1.) Put on 20 meters (14Mhz), or other single band including 50 or 144Mhz.
2.) Multiband HF amplifier on ham bands.
3.) HF amplifier covering 2-30Mhz in 8 or 9 bands, or better to include 160M.
4.) HF+6 amplifier, 1.7 to 30Mhz in 8 or 9 bands, plus 50-54Mhz in "high gear". -very ambitious.

I want to try for the 4th option. If the 50MHz tank is built in the existing chassis, and the HF tank coil is mounted right above the tube anode, this may be possible. The biggest thing may be keeping lead length short and stray capacitance lowest. I already learned about parasitics from COL Tucker's transmitter and its 20" piece of wire from the anode to the RF choke!

At the least, the PI coil on hand should provide coverage of all the ham bands because it was from a full-coverage hf transmitter.

Some pictures of this stuff. 5 pcs. UCX 375pF 14KV caps. The schematic calls for larger sizes but let's see what happens.
 
This transmission, 37 turns of the 1/4" shaft turns one coupler 9 turns and the other 18 turns. It is maybe for using a roller coil and vacuum cap together?

some insides of the starting point

some insides of the starting point and also the filament transformer. It has 5 taps for 280-240V input

A couple of the caps have an unfinished look to the outside of the plates. How to know of this is bad or what? They were cranked open til the inside could just barely be seen and that part looks great. Could they have been overheated? Do some of them have a less than perfect appearance?

You should also have a 240 to 120 isolation xformer in there that has a couple taps.

This allows it to run on 'two hots and a ground'  safely.  You can save the expense of the neutral wire run.

I elected to bring neutral over in my runs for future proof.   Never know what will end up in the shack.

I was able to get a 40 meter tank coil in one of those rf decks.

--Shane
KD6VXI

--Shane
KD6VXI

That isolation transformer looks like it's good for 750VA. Very nice.. Thanks for the tip on it. I thought it was just another multi winding unit for the rest of the guts in the generator - which are mostly missing.

A helper brought it up that a Tungsten type filament should be soft started and gave me some links to papers on this and the chart. It goes for Tungsten and Thoriated Tungsten. "W" being the operative element.

http://www.w0btu.com/miller-larson_effect.html

http://www.r-type.org/pdfs/thoria05.pdf

https://www.photonis.com/uploads/lit...rica-Paper.pdf

http://www.tubecollectors.org/federa...beapps(70).pdf

http://www.cpii.com/docs/datasheets/79/3CX15-000H3.pdf

ITT recommends limiting the current to 150% for medium size tubes and 250% for large ones. Of course this could mean 50-100KW as a medium one.. But the issues outlined in the papers about stresses of inrush should be taken seriously. Smaller tubes like the 813 may not suffer as badly but in the boxes of old tubes I get from time to time there have been have had a few of those and many other old Tungsten type tubes with open filaments.

How can one divine the temperature vs current figures for a Tungsten filament? One answer is a chart in one of the publications. There is also a volume resistivity specification for almost every conductive element.

The region of interest for soft-starting is the warm-up between 250-300K where the filament resistance may be 10% of the operating value and 1400K to 1700K where it may be 1/2 to 2/3 of operating  value.

Any startup that is gentler than just throwing the switch is an improvement. Even the quick turning up of a variac is good. MOVs designed for inrush current limiting are used by some and others use a step start. One question is how to have a gradual startup and avoid varistor type elements as they tend to wear out due to their natural state being one of 'hot' - maybe no worse than a couple of transistors unless transistors are well-oversized.

I think a 75 Ohm resistor on a 200-220V primary may limit the inrush current to around 150%.

Soft-starting tube filaments via a time ckt isn't really necessary in my opinion, nor is it done in most commercial transmitters of your power level.

Proper technique is to use a big wirewound pot in the primary of the fil trans, rather than a Variac....the pot senses the inrush current and provides the desirable 'ramp-up'. 

The pot should provide the target voltage at 60% rotation or so, and then you have a bit of leeway to increase emission as time goes by.

73DG

I'd like to use a resistance with a more substantial portion of power-handling than the 10 ohm one. I'm stuck with a 200 or 208V transformer tap and an actual 220V line voltage so with about 2A, the max R is 10. The line voltage varies a bit as well.

For fine adjustment I'm leaning toward an Amertrans line voltage adjustment variable autotransformer. It does +/- 15% and has a worm drive so it's easy to set precisely.

It would be very fine to have a smaller handwheel in the middle of that big one. There was a concentric pot setup before.

The filament transformer for the 3CX3000 was mounted where the leads should reach across to the RF chassis in the adjacent rack. Its 10 Ohm resistor can be next to it. 
It could have been mounted in that rack but it would have either taken up otherwise free vertical space or had to be between the HV transformer and solid state diodes.

Here's a Jennings VMMC-1000 is a 20KV unit covering 50pF to 1000pF. I hope there is some way to use it for the tuning. 50pF is a lot. There is an older article for an 813 amp on 10M that uses a tapped 10M coil and puts the tuning cap on the tap because the 813 plate capacitance is high on 10M. Wonder if the scheme would work to the reverse.. if the tuning cap's minimum value is a bit large. What to do there?

The two coils there are from a 10KW transmitter and one should be large enough work in a high-C circuit without too much loss. Hope they can be used.

The schematic is just to recall that 813 output section.

the pi coils are huge. The 1-turn loop is about 9" diameter.

fil choke

Patrick,

Figure out how much C you want to take away.

Convert C to Xc.

Xl is given by the product of  Xc*-1.  Convert Xl to L.   Add this between Ctune and the pi inductor.

That's one way.

Method two.   Find Xc and Xl as above.   Tap the pi tank coil at Xl.   Now,  use a GDO or VNA and resistor backfed to design your tank.

Method 3.

Use a L-Pi.  Ua 4 to 600 ohm image impedance.

The last ten  meter 3k I did performed ten to fifteen with a standard 25 pf minimum cap.   I did NOT use any trickery like above.   It will NOT work at the fm portion of ten,  but the owner doesn't go above the ssb portion,  so it fit him fine.  I designed the input and output tanks for 25 mhz.   1 of 2 and 8 respectively.

--Shane
KD6VXI

Looks like that may work out fine. A small L in series with the slightly too-big minimum capacitance for the highest band will have little effect at lower frequencies right?


So it is coming to the layout. Because the parts are large and cumbersome, the Paint program was used again to made drawings which are then 'pasted' together to get different views.

The first drawing 1 looks OK from a standpoint of short connections and wastes some vertical space, but the antenna matcher stuff should still fit above.

The second 2 is better on space but the pi coil to anode is 11 inches. Maybe the RF chassis can be raised higher to shorten this to something reasonable.

The third #3 is most compact but the anode connection is only an inch from the cabinet wall so that's not a good idea.

Putting a bit of Xl in series between Ctune and the pi circuit only serves to cancel out some c.   Xc-Xl=new Cmin.   Subtract the same amount from CMax of Ctune and that is your new max value.  Usually your taking 5 to maybe 15 pf.   Not even noticeable when you try for 80 metros.

Make sure there isn't a resonance.   

--Shane
KD6VXI

If I understand this correctly,
What I want is to be rid of 25pF.
At 28MHz that 25pF has Xc of 227.5 Ohms
227.5 Ohms  XL at 28MHz is 1.294uH
Because the reactances are equal that is a resonance there but as for where else a resonance may show up due to this added L, it has to be checked for, but it should not get out of the series circuit of just the added L and the tuning C.

Yes.   

I've not had a problem with resonances on any of the networks I've built.....   But,  I also take a bit more C out than I need.   That way Xc is never = to Xl.   

Use heavy gauge wire.   I used qtr inch tubing.

--Shane
KD6VXI

Today that old Henry chassis was cleaned out pretty well. I left the two capacitors for now. An insulator had to be replaced on the tube socket as it was broken and that is when I discovered the messy way the socket and grid collet had been assembled onto the chassis.  The filament collet plate is mounted off center of the main hole in the chassis and the grid collet is also off center nut in the other direction, although it can be slid around a little. The tube fits OK and the only contact is with the collet springs but the sloppiness a little disappointing. Must have been made on a Monday or Friday.

The capacitance of the tuning and loading caps in the 13.56MHz Henry chassis was measured. The tuning cap looked like 20-30pF and the loading cap 40-100pF. These values are just the bare caps, without the various doorknob style padding caps that were present. It remains to see what to do with them, whether either can be used.

The filament transformer has several taps on it and a rotary switch was found that should be useful there. It was so dirty but is a kind similar to what is used in RF tank switching so even if it is not good enough for that any more it should be OK for this. Some GC control and switch cleaner with 1-1-1-Trichloroethane cleaned that up right away. The red stuff, hoarded for a reason.

This chassis arrangement is looking good. Adding a 4" chassis below the Henry chassis makes room for the large coaxial filament choke that will replace the smaller Henry one. It also puts the anode closer to the PI coil.

I found a switch on the fil xformer was unnecessary.

The original generator was designed for commercial service,  so 208, not 220.  On 240, you'll see North of 9.5 volts with 240 on 1-2.  The one I'm working on for a friend now with 244 volts on the 250 volt tap,  still makes 8.8 volts measured at the input to the fil choke.

The fil xformer is tapped for 208, 220, 30, 40 and 50.   I'm using the 250 tap and see 8.0 VAC.  If one wanted more finite adjustments,  you could use the 208 to 110 step-down xformer since it has multiple taps as well.   Series them.

How much does that rack weigh?   I've alway wondered,  they are battleships!

--Shane
KD6VXI

I found that out.. 10VAC there. so the next tap or two needs chosen as you have done.

The switch is just for variation in AC line depending on season. The line voltage varies a lot in the shack when the a/c kicks on and off, and can be from 210 to 240V at different times. The service there is only 100A and I'm also at the end of the transformer's run in the neighborhood.

I want to find a "230V" output sola type regulator, 500VA, to handle the filament under these wide variations.

The rack itself, the double, is only about 150 lbs. It's very strong and made of steel but its open construction makes it fairly light. The doors and sides add quite a bit more. As it is now with the power supply basically finished, it's probably pushing 900 lbs sans doors.

Spent some time figuring out the existing current metering stuff in the Henry chassis. It's being kept intact and should feed external meters OK.

Pat,

Do you have the schematics for the control circuitry?  You may need to build op amp meter drivers,  depending on your meter sensitivity. 

I've found the diodes in the bridge for the fil circuit metering to be...   Barely adequate?   Which was because the chokes feeding them,  as well,  where barely enough in Uh to keep rf out of the bridge. 

Here's a pic of the one here.   Have engraved placards on order,  but the 'Dymo'  tape helps to remember what's what.   It has a grounded grid and a variable bias position,  bias voltage up into the 50s, so I can cut plate current off.

Also has a ptt in out,  and a FTT or Flip To Talk.

--Shane
KD6VXI

FAIR sold those Sola regulators.

I don't remember them being cheap or lightweight.

What about a power conditioner,  Pat?   That might be an easier find from an IT shop.

Basically,  a SOLA box.

--Shane
KD6VXI

I have the few manuals that are online. The controls on this one seem to be just relays, etc. There is a remote control section on one variant manuals (Randex) but this unit apparently did not include it.

The metering on this looking at its schematics used no op-amps but yes indeed my 'nice' meters are according to Murphy's Law not going to be the correct mA readings. Usually I've attacked this kind of problem by changing the shunt values within reason. Any reason that could cause troubles?

I didn't think about adjustable bias yet. With the grid grounded and the cathode being a high current path, what did you do to make that work in a variable manner?

That's the trouble, they are not cheap. I'm open to either the transformer-like package or the cased units sold as conditioners. The 230V units are less common. I'll check Fair. There are some on ebay, not sure if I'll get 'stuck' with a bad one -no one wants to guarantee any of it. Still $100 average price and the way some of them overcharge for shipping is pushing the budget.

Shipping isn't charged in the fees.   Hence them having a low sales price and an extreme shipping.

Sounds like you have the same as my 2000D.  That one has a door for the front panel and was designed for a remote controlled sputtering head......   Whatever that is.  I also have one that has the rf deck and fil xformer changed.   That one has full metering,  etc.  And a 4-250 driver stage.  With an 8877 final.....!   Needless to say,  stoked about the power supply for that one!

The first bias circuit I used was a string of 15 diodes.   As others found,  it's not really necessary.   But,  looks cool,  and provides for the ability to change things.   I like that.

I'm using a variant of the W4ZT bias circuit with a modification to use 3 pass transistors instead of one.   This will work for tubes up to a single 10 triode,  according to GM3SEK.   This rf deck may have a 3x6 installed if the 3 ever goes soft according to the end user.   He says a single Darlington would be sufficient for a single 3000.  For a YC156 (5k) or larger,  more pass xisters.

--Shane
KD6VXI

The tube is the limiting factor on this project. I don't think I'll ever have to replace it from wearing it out.

Today I guess the RF deck could be said to have been started. Been so busy the last couple months. The front to back partition on the Henry was removed since there will be no output signals under the chassis. The coaxial filament choke from a GPT-10 was mounted. A 4" tall chassis will be inverted to serve as a bottom cover on the RF deck. This leaves enough room for the large filament choke and probably a tuned circuit. The 13.56MHz Henry had its own filament choke but it was lower inductance than wanted.

A 2" ID piece of PVC pipe holds the filament choke and the ends of the choke were adjusted to reach the filament input terminals and also the socket. This cost me less than a turn.

The pipe is mounted by some rack brackets and #10 thread stock. The threads go through the PVC pipe and the pipe is clamped to them by some washers modified to fit the inner and outer curve of the pipe to reduce stress from any jiggling of the unit. This arrangement uses cut-washers so the nuts will stay in place well.  The two large holes in the PVC pipe allow a nut driver to be inserted to work with the fasteners of the pipe.

The filament voltage monitoring board has to be relocated and a power resistor put back in place then the Henry amp's original 'monitoring' wiring can be re-used.

The filament choke is at least 1.5" from any metal. It's the best that could be done there.

more pictures.
My assistant remarked that this part of the job, the amplifier chassis, seems to already be a lot harder than the power supply. I could only say, we have not seen nothin' yet, this is the simple part.

I have a question - what type ferrite rod material should be used to cover 2-30Mhz?

The reactance of the filament choke will be low at 160M so the material should be active there.

If it drops off in adding inductance as the frequency increases, that may be ok, right? The coil as it is measured about 4uH.

next, will 2" ferrite 'donut' cores slip into a 2" PVC pipe? can those be used for this same purpose?

Patrick,

   Off the top of my head, I'd say type 43 material for 160m, and 31 for the higher bands.

One option that may or may not be relevant would be to isolate the filament transformer from ground, and put a common mode RF choke on the filament transformer primary wires. This might work for 160, 80m, and then above that the existing air wound filament choke takes over.

You would still need a single wire choke for the filament centertap to ground or zener wire, but that would only need to pass an amp or two.

Question: Could this AMP be a driver for a larger amplifier project yet to be built?  ;D

Jim
Wd5JKO

Here are some "before and after" numbers my choke.
I stuffed it with type 43 ferrite rods to get more inductance.

That filament choke looks like it came from a GPT 10K .....it is coaxial......for a 4X5...TMC mounted it on a fibre glas rod...

Jim/ WD5JKO:

Your avatar picture above of a car radio with "All Transistor" reminded me of something from 1962. Back in those days kids walked around with the new 2 transistor radios listening to local AM rock and roll stations.   The first question was always, "How many transistors does yours' have?" As time went on, it would increase from two transistors to four, then six, then ten as radios became better.  

Then one day I met a girl who answered, " Mine is ALL transistor!"  It said so on the radio. We didn't know what to think except that she had trumped us all no matter how many we had.  Little did we realize that even the older 2 transistor radios were "ALL Transistor."    :-)

T

About 55 years later she got "trumped", but this time it was for saying, "what difference does it make"!  :P

I do like that Avatar. I sent that radio carcass to an AMFONE member who was wanting to experiment with permeability tuned circuitry. I bought that radio to get the audio output transformer which was a tapped inductor for the single ended class A PNP germanium transistor (2N176?). It seemed to act as a good modulation transformer for my Retro75.

Jim
Wd5JKO

I swear it's not a driver for a larger stage. I have only 100A service to the lab so with the Texas heat and HVAC requirements that leaves me about 50A for playtoys.

The choke is indeed from a GPT-10. It seemed appropriate to use a higher rated choke. The Henry's driver was rated 100W. The operating conditions in the tube datasheet call for up to 400W of drive. The GPT-10's driver is rated 1KW.

To figure out how many ferrite rods will fit in the 2" section of PVC, there is a formula somewhere but this web site has a calculator on it. 
http://www.engineeringtoolbox.com/smaller-circles-in-larger-circle-d_1849.html

Material 43 or 31 Ferrite rods seem costly. Am I missing something?

Lastly I don't know that much about ferrite material. Will type 43 material be overly lossy or dissipative at 28 or 50 Mhz? Or will it just have a reduced effect?

The original Henry choke will handle an 8877 for drive.

Ferrite load it for 160 to 10.  As is good for 20 and up, maybe 40.

You can use different mixes in your various stuffed rods to equalize the bandwidth.

--Shane
KD6VXI

I thought about a mix of ferrites and also about filling 1/2 or 2/3 of the length of the filament choke with the ferrites and leaving the rest free to help in case the material was messing it up for the highest frequencies. -There will be an attempt to have 6 meters as well as HF.

These ferrite rods were found and bought so it's just the waiting now. I have no idea what they are and there seems to be no data available from the salvage resale company except that they are Russian, which at least means the quality will be good.

The offer seems fortuitous because exactly 8 of them will fill in the 2" ID of the PVC pipe supporting the choke.
http://www.engineeringtoolbox.com/smaller-circles-in-larger-circle-d_1849.html

Wow, can you believe it's been almost a year already? So much time devoted to this, it's shaping up nicely at least for you.

I finally got an account on here; this is your apprentice by the way.

To any and all who may be wondering about the side project mentioned in this thread (audio modulated vacuum tube tesla coil) I've had to largely depart from my design and much of what I had constructed because the 75khz grid driver remained plagued with parasitic oscillations.

During the attempts at constructing this, I began thinking about the possibility of combining a plate modulator and a plate power supply; basically a much larger version of the grid modulator I had been constructing with a ferrite transformer from an inverter-microwave oven power supply.

This solid-state supply would drive a ferrite transformer at the resonant frequency of the tesla coil secondary, the tube would be operating at half-wave, and the audio modulation would be applied at the grid.

I'll make a thread about my current thoughts on this project and come back to post a link to it.

Ahh yess my not-so-evil apprentice. It's best to use a separate topic. The Tesla coil and driver thereto are indeed radio frequency gear for practical purposes and the amplitude modulation does make it an AM transmitter. That is why the work right now is being done in a metal building. No guarantees how faraday-cage-ish but give it the big W for now.

The ferrite rods showed up for the filament choke. Anyway, some experiment will be done with a VFO to see what the reactance of the choke is with the rods in and try to find out if they are causing a loss at high frequencies due to this unknown material, etc. It will be kind of an elaborate setup I think. I do not have a lot of experience with ferrites.

A piano hinge was found to permit the chassis bottom to be opened and closed for testing and later after the input circuit is done and tested the shielding of the filament area/wind chest can be fixed up.

Measurements with a Boonton LC meter at 1MHz show 21uH when 6 or 7 rods are put in the PVC pipe supporting the filament choke. As many rods as in the KJ4OLL was not possible due to the support for the choke. The 7th rod didn't make much difference at all.

7 rods fit snugly once the SCH40 2" pipe was slit lengthwise to allow it to be expanded a hair, otherwise it was very tight and I did not like that because ferrite can crack. Each rod was partly sleeved with heatshrink tubing to avoid abrasion and 7 were slipped in the tube tightly. Heat shrink tubing was put over the 2" PVC pipe and recovered. The rods were held in the cylindrical shape by some heat shrink tubing on the pack of them. This was 4" tubing.

Once the rod assembly was done, the lengthwise slit on the pipe was opened with screwdrivers and the rod pack was slipped into the PVC pipe and the screwdriver/spacers removed. Then the heat shrink tubing was used to cover the PVC tube. Looks nice.

Because the rods were 10" long, a new 11.5" long piece of PVC pipe was required. It all mounts the same. 

So here are some figures on the inductance with 0 to 6 rods, and based on that, what I calculated for reactance over the frequency ranges to be tried, 160-10, plus 6M.  It would be possible to resonate the filament choke but with 7 ferrite rods inside, stray capacitance takes over above 8MHz.

It should have added to this some low Q LC circuits for the bands. Not sure where to find design examples for sanity checks.

I should apologize to interested parties for the lack of progress and pictures. I have not given up on it or become disinterested. My parents' estate of which I am co-exec., and the additional process of re-homing Mom's Cats (http://bunkerofdoom.com/momscats/index.html) leaves me only the occasional weekend.

pictures of the choke etc.

The 'socket' end of the choke was a brain teaser. What I did is very ugly but the resistance from the end of the choke to the socket is 2.5 Milliohms and it's mechanically solid and very strong. A clamp-on-and-solder type of copper strap arrangement could have been used but this worked as well and was simple to put lugs on the tube socket end.

The 'transformer' end of the choke has very thick copper strap with holes. To get it to connect to the existing filament power feed-through capacitors' terminals in the chassis, one was bolted directly to one capacitor and the other connected with a #6 jumper.

The original filament voltage monitoring board was slightly relocated. It is close to the outer conductor of the filament choke but this is the 'cold' end just inches from the feedthrough caps. It should not bother things but we will see.

A test was done with the MFJ259B. A 50 Ohm NI resistor was attached to the tube socket and RF put in to the cathode connection. The SWR and the R and X  are plotted in the chart. It seems satisfactory.

haha yeah 11m was included in the chart for my little joke, but only fussbudgets would care as it's a 1.75-30MHz and hopefully 6M amplifier. The band switch and plate coil don't match ham bands but are from a general coverage unit so it does not matter what tests were done as long as coverage of the ham bands is included. There was only one 'bad' area where the SWR rose to 2, and it was at 46Mhz. No operation there (or on 11M for those who may scowl).

There are rusty fasteners in some pictures. These will at least be cleaned up at some future time so the connections are good and there are no unwanted diodes..

to be continued..

here is the way it was tested. Comments or suggestions are welcome. 50 Ohms is conveniently the datasheet Z for the tube driving.

Jameco sent me a catalog and inside are Dialight brand products. I am amazed at how inexpensive most classical style jewell type lamp holders are.

http://www.jameco.com/Jameco/catalogs/c171/P25.pdf

Pat,

I just love this thread, and the incremental updates as you move along.

You touch on many issues with your RF input circuit. Every designer has to go through the process of selecting the options of tuned versus untuned input while weighing the pros/cons of each method. The tube makers often publish a grounded grid input impedance for a set of operating conditions. In this case that value is conveniently 50 ohms. The part that is often ignored is that the tube loading is every half cycle when the tube is running class B.
This uneven loading is why a broadband solid state driver might have trouble driving a big amplifier like this when the big amplifier RF input is untuned. Of course, a driver like a 100 watt tuned amplifier such as a DX-100 would have no trouble driving this amplifier.

Over the years I have seen circuits from the 1960's, and more recently discussed on some of the amplifier reflector boards where they take a stab at the 1/2 cycle loading when using an untuned input. Some have proposed using a power supply high vacuum diode to load the other 1/2 cycle. Others have proposed using a high speed silicon diode. I suppose the diode could be terminated into a resistor that matches the tube RF input impedance.

I am not saying you should or should not make any changes, but for purposes of discussion, I thought to bring up the topic.

Keep up the great progress.

Jim
Wd5JKO

Never tried the fast diode. What diode could switch at 30MHz much less 50MHz and handle the current? The peak current would possible be a few amps.

I tried a 5Y3 in the cathode circuit of a 3-1000 a long time ago. It seemed to work fine but was driven by a 2x 6146 amp and its pi-network output, so I can't say how well the 5Y3 worked.

The current in the cathode circuit of a large GG tube amp looks like 1-2A if you take the drive wattage and the impedance and do the I2R math, but I'm sure that's not very accurate because peak values can be much more and it's a half wave.

A tube with a resistance of 50 Ohms is not so easily found, or is it?

Maybe something to throw on the table could be a high current low impedance triode like the 6AS7/6080, 6336, or even the monstrous 7242.
Plate resistance:
6AS7/6080 = 280 Ohms
6336 = 200 Ohms
7242 = 82 Ohms

I'd like to explore a tube or diode.

I'm leaning toward a low Q tuned circuit as is traditionally used in GG amps but have little experience in choosing values for that.

I'd avoid ferrite beads in this situation.

Small resistors, about 100ohms as close as possible to the actual gate leads. Aka "gate stoppers".
Then small pf caps from the source to gate. Sometimes pf + resistor.

Should kill the unwanted HF or VHF parasitics.

Just reading backwards now...

   Would sure be interesting to do a 2 tone test and try something like that. A lower impedance diode would be a damper diode like a 6AX4, or maybe 2 in parallel. Looking at the diode curve, at 70v it will pull around 800ma, so R=E/I = 88 ohms. The impedance will be dynamic with drive level, just like it is with the G-G RF tube.

Jim
Wd5JKO

Perfect! as my friend Henry SWL also needs to know this stuff since several FETs were sacrificed!

That's a great idea! It would seem to fit the bill. What about the plate dissipation? I don't recall the 5Y3 ever getting red in the face but it's a higher impedance job.

Two things need to be done.

1.) come up with a nice flywheel for the cathode circuit.

2.) look much more closely at the GPT-10K tank coil - because it's from something else. I harbor no fantasies that it will be plug and play. This seems to be the harder part so I have been working on that.

The next post shows what was found out about using the GPT-10K coil with the 3CX3000 and offers some options to consider.

To re-introduce it, the big beautiful general coverage GPT-10 tank coil is made to last!

The coil has an integral bandswitch and is designed for 2-28MHz and its inductance in uH for each of its 9 bands in are:

13.750 - 2-3 MHz
9.750 - 3-4 MHz
5.850 - 4-6 MHz
3.750 - 6-8 MHz
2.250 - 8-11 MHz
1.500 - 11-15 MHz
0.950 - 15-19 MHz
0.600 - 19-24 MHz
0.225 - 24-28 MHz

Now I have to point out that this may contain all kinds of errors, so don't take my word for it. The big point discussed in here is about the Q of the plate circuit.

It was designed for the 4CX5000 operated in GG mode but with normal electrode operating conditions G2=1200V, G1=negative 300V, running 7500V @ 1.5A.

This alone, just looking at an E/I result gives 5000 Ohms. Not the plate load but a comparison figure.

(The Eimac data calls for AB1 and 1.9A plate current @ 10KW output, but the GPT-10K uses cathode drive and 1.5A (1.75A max) plate current. The Eimac data calls for 1.9A but it makes no sense to quibble when the equipment manual gives application data.)

The amplifier application here is the 3CX3000 in GG mode running 4800V @ 1.7A.
The dividend of these values gives a figure of about 2800 Ohms.

From this, it is easy to see that for an equivalent L/C ratio or "Q", the plate coil from the GPT-10K has inductance values that are quite higher than those desirable for the 3CX3000 application.

I believe that tediously calculated values for actual plate load impedance in both cases would further bear this out so I skipped it for now.

For this thought-experiment, which should become a real experiment, the lowest band of interest, 2-3MHz, is considered. It also needs to be expanded down to 1.7MHz to cover 160M.

The Q will vary a lot over this range and that is why this band is a good choice to use in order to consider how to use the coil.

Using the PI-EL program, which seems as accurate as any other, some "Q" numbers on the GPT-10K bandswitched plate coil, 3CX3000, and 50 Ohm output are:

1.7MHz: 15
3MHz: 7.8

If a Q  of 7.8 is not acceptable, then the inductance of 13.75uH is too large. If the bandswitch tap were moved down one position to 9.75uH, these "Q" numbers are obtained:

1.7Mhz: 22.2
3MHz: 12.1

It looks like a high Q is probably acceptable for this coil because it is made 1/2" silver plated copper tubing and it will not need to handle 10KW, but by the limits of the tube, only about 6KW. The equivalent coil current decrease should be:

1-(SQRT 0.6) or 22.5%.

Going back to the complete set of PI-EL calculations done with this coil's series of inductances (post later), the Q seemingly improves as the frequency goes up. I didn't try to investigate this but it is interesting.

Thoughts on ways to use the coil including re-tapping the inductance as above:

1.) re-tap the inductance to values from calculations so that on each band Q=>10 is obtained

2.) Use a higher plate voltage on the 3CX3000 but the tube is rated 5KV absolute maximum and it would violate design rules.

3.) Run a lower current at "full output", meaning higher load resistance and much lower output power target. It could more closely match the original application of the coil if the maximum plate current was limited to 1A.  Output might be limited to 3500W. (the horror!) and presumably only 170W of drive to make the 3500W. This is not a bad thing but not good either as a 3500W rating is extreme stupid underkill for what is being built considering the equipment size and power supply capacity.

4.) Design pi-network for output impedance of 200 Ohms and use the intended balanced tuner to handle the ladder line (or even a correctly sized 2:1 transformer to get to 50 Ohms if that's wanted) Here is the data calculated for 50 ohms vs 200 Ohms:

50 Ohm output
1.7MHz: Q=15, Cin=732pF, L=13.75uH, Cout=3781pF
3MHz: Q=7.8, Cin=221pF, L=13.75uH, Cout=767pF

200 Ohm output
1.7MHz: Q=21, Cin=845pF, L=13.75uH, Cout=2440pF
3MHz: Q=11.5, Cin=267pF, L=13.75uH, Cout=743pF

This would be quite a transformer but there is precedent in the V.R. Stokes book and if I am correct, a transformer just like a balun works well if designed for the frequency range and impedances. Hmmm..

Pat,

I can tell you from recent experience, the 3cx3000 will take 6kv plus on it.  The xformer here was a 4kv AC output into a fwb, cap in.  That's 6400 DC.

15 reverse connected 10A10 diodes brought the ZSAC close to what a 3x3 on 4.8 kv was.  I then added a variable bias adjustment, for good measure.

If you want to bump your plate voltage up you'll be OK up to the mid 6kv range.  Some of the CB mobiles had 10 kv plus.  A lot of bias, but the tubes didn't arc.

As to Q, I like low Q.  7 to 8 on the output, 2 on the input.

--Shane
KD6VXI

Shane, those are very good points.

The HV is possibly able to go to 6KV by using the right taps.  One thing that caused my last post/question (a complicated question indeed!) was that I found the Q of the tuned circuit, with 1720 Ohms load to the anode and 50 Ohms on the output, would look good at one end of the band and rotten at the other. I guess I knew that was coming but did not realize how large a difference there could be.

On the other hand, Q of 7 or 8 might be OK, but am I mistaken to worry about how clean or efficient the AB2 amp will be with a low Q? I also worry about excessive Q causing a high frequency rol=off of the audio. I would not want the amplifier to make me down 6dB@4KHz due to wasting power by part of the signal being off resonance.. Audio bandwidth shaping, if desired, should be done in the audio control amplifier and EQ. I do not know the math on calculating that but will have to look into it.

Back to the PI network investigation, I created a spreadsheet, taking knowledge from the typical formulas in the ARRL Hand book and Radio Handbook, which agreed.
It turns out that the various online calculators and downloadable programs are not all created equal, do not all agree, and they demand that the user 'do it their way'. This is fine for straightforward building, but I wanted to learn more by plugging in a Q value, input Z and output Z, then having a spreadsheet calculate the required input capacitance, inductance, and output capacitance. This took me a couple three hours to get it right and double check it.

So, I am halfway there, with a spreadsheet that takes in plate load, output load, and Q value; and gives C/L/C for 100KHz increments from 1.5 to 60MHz. There are limits to it - very low Q values mixed with high plate loads give errors. It's probably the square root stuff and limits of the excel format. Maybe someone can figure out why.

What's wanted is to add a sheet to take inputs of L value, input Z and output Z, and from that provide the Cin, Cout, and Q. This is for those who are determined to make an existing coil work.

This file has been saved as .xls and .ods and here the xls version is, so others can participate in this numerical experiment. No macros, nothing hidden. Those things make a mess in the computer and in the mind.

Others can try it. I hope it is fun and I accept responsibility for any errors, just please mention them to me so I can fix them.

I feel you on the different calculators giving different outputs.

I've found that Tonne's and Ian G3SEK spreadsheet give the best.  With Tonne's software, you have to take stray C, Lsupp, etc into account after the numbers are spit out.  Ian's spreadsheet, on the other hand, will darn near give you exact values as long as you give it all the data it asks for.  Ie, on the 3cx3000 just built, Tonne's program gave  about 28 pf for Ctune.  After running through Ian's spreadsheet I found I needed about 8 pf Cmin to get to 29.7 .  That necessitated getting a new cap for Ctune.  Prior I couldn't tune about the tech and of ten.

You give up some harmonic cleanliness with a low q.  You gain bandwidth excursions without returning (your pass and is wider) and you also usually have a lot less peaky tuning.  A Q of 2-3 is what's regarded as OK on inputs, and that lets us hopefully tune the entire band under 1.5 or so.  But, there is very little harmonic suppression compared to a Q of 8.  And when you go up to 12, it's even better.

What I'm not real sure about is IMD in relation to tank Q.  Maybe Tom, K1JJ could chime in on that.  I'm sure he's mentioned it before, but I plead CRS now :-)

--Shane
KD6VXI

It's said the pi network is made of two L networks, as in the figure. My main difficulty is the messy equation indicated.

Hi Shane,

Generally, the higher the Q, the better the IMD figures, but the lower the efficiency.  To take an extreme case, just imagine a plate tank Q of 2 - hardly any flywheel effect to integrate the class B pulses.  A  Q of 12 is a decent compromise since efficiency is important to most hams. But if pure cleanliness and great IMD figures are most important, then:

I proved this out to myself a few years ago when building and testing my super linear amplifier chain.  I started with -75 dB 3rd order IMD from the 100mW output of the FT-1000D > 1 watt precision lab amp > 3CX-350J > 8877  > pair of 8877s.

The class A 3CX350J ran a plate tank Q of about 50.  The  8877 class A  driver ran a Q of about 30 IIRC.  I was able to get an IMD of about  -55 dB 3rd from the chain's output after tweaking everything and running it at ~ 1/2 power from max. The high Q and class A elements were most key to this excellent IMD performance.  Super clean drivers are so important. They can make or break a system.

Of course, these are old school techniques. An SDR using pre-distortion ("Pure Signal", etc)  would make all of this futile. A big, "CB-quality" solid state driver into the pair of 8877s would make more power out and still be -55 dB third order achievable....sigh.

T

As always, no free lunch.

How about adding the L, as in Pi-L...  I'd imagine same / same, since you're adding more filtering.

Predistortion....  Yeah.  I'm currently contemplating a motorcycle or ANAN... Lol.  That's ham dedication there.

--Shane
KD6VXI

Pat,

If you have a calculator or MatLab you can use these equations:

Calcs. for Tube Circuits
clc
F = 3.7e6;
twopi = 2*(355/113);
twopiF = twopi*F;
% Tube Calcs
Ip = (Value of Plate Current here);
Vp = (Value of Plate Voltage here);
Pin = Ip*Vp
Vsat = 0.075*Vp;
Pout = (expected Power Output here);
Rout = 50;
Rplate = (Vp - Vsat)/(2*Ip)
Eff = Pout/Pin
TubeC = 15e-12;
%
% Wingfield Pi-Net Equation
Qz = 12;
Qzsq = Qz^2;
Q1 = ((Rplate*Qz) - sqrt(Rplate*Rout*Qzsq - (Rplate - Rout)^2))/(Rplate - Rout)
Q2 = Qz - Q1;
Xc1 = Rplate/Q1;
Xc2 = Rout/Q2;
XL = Rplate*Qz/(Q1^2 + 1);
CPT = 1/((twopi*F*Xc1)- TubeC)
L1 = XL/(twopi*F)
CANT = 1/(twopi*F*Xc2)

some improvements.
I'll have to do some math later. Looks promising.

This is something I did Sunday, tedious, but after putting it into a table according to the coil and switch for the bands, it shows Q at the 'band edges' of the coil as-is.
The band edges that the coil was assigned can probably be changed.
Anyway the Q looks decent in most places so I think the coil will be OK.
Experiments should show how things need to be changed.
Pi-L is possible but it'll change everything.

polished some copper today. For the tuning unit.

also worked on the steel panel that will mount the filament regulator

The filament regulator is mounted. This is a 60 Lb 1KVA Sola unit. It has excess capacity and so the rectifier filaments will also be regulated. It is needed here because in the summer the mains in the shack with everything running HVAC etc. can vary from 210-240V

After the transformer A 3 Ohm 300W pot connected as a rheostat should let me get the voltage right on and allow +/- 0.15V fine adjustment and an expanded scale voltmeter setup will be used if possible. If the voltage can't be hit by choosing a tap on the transformer and adjusting the pot, then a few more Ohms may have to be added.
3CX3000 soft start: 700W
3cx3000: 390W
4) 673: 200W
-----------------
590W - an adequate load. 900W on soft start - the reson for the 1KVA rating.
The 673s don't need a soft start but the extra load on the transformer will help it run more efficiently.
This will have to be tweeked later.

So the panel is an old fashioned 5/32" steel panel for strength and is held in place with six grade 5, 1/4-28 bolts and nylock nuts. The transformer mounted with 3/8-24 bolts and grade 5 was also used there with nylocks and split washers.

The last three relays were also put in sockets. One is the filament inrush timer, the next is  for something, forgot right now, and the third is for keying and has a 24VDC coil. After having a mess of relays behind the rack due to the Tucker's non-isolated 120VAC keying circuit, going with a nice low voltage isolated circuit on this project looks attractive.

The plate supply soft start resistor was changed to 10 Ohms by putting the three resistors in parallel. It was 90 Ohms for experiments before.

So much still to do. the RF chassis will go back on the workbench but first a metal plate or set of rails has to be installed in the rack for it to sit upon and some sort of flexible hose has to be gotten and installed on the blower. Maybe dryer duct parts will work OK for that but I would prefer a smooth duct.

If you worry about hitting the limit of the SOLA while on soft start, use the unregulated 240 through the resistors, and have the contactor switch to the regaled voltage.

The soft start doesn't need to be regulated.

--Shane
KD6VXI

I considered that, but there would be a short power cut at change over because the phase difference between the input and output sides of the regulator transformer would not allow a simple relay closure across a resistor, but a break then make of a few msec via heavy duty DPDT relay would be needed. The shock to the filament might be harsher with the interruption but still way better than just turning the filament on. The nice 10 ohm resistor in series with the filament transformer is also helpful for those with the Henry parts.  This might be nit picking but every little bit helps. I was reading that the structure of the tungsten changes as it passes through a certain temperature on its way to operating temp and that is is hard on the wire. Just want to be as easy and soft as possible. The article says that tungsten is subject to recrystallization as it reaches 1200 deg C so whether this only happens when the filament is first made or whether it happens 'slightly' each time the filament is heated is not explained there, but an easy does it approach should be best. anyway the attachments seem close to a gospel on it.

http://amfone.net/Amforum/index.php?topic=42633.msg310468#msg310468 references the W4ZT bias board whic is popular.

I like that circuit but I have a complaint that it can only control current in one direction. I know I have seen it before and was puzzling over how well itworks yet is so simple. I do not have a TL431 device in LTspice, so could not examine this for myself.

When I have been doing bias supplies, I always wanted them to sink or source current.
A giant bleeder rated for 3-4x the 'bias current' in this case of the 3CX3000 amp is not practical.
This reminded me of an IC I used to mis-use.

The audio IC STK0050, a 50W 4 Ohm amplifier module from the 1980s that is very basic. It is really a few dice and a thin film 'ic' all assembled on an aluminum plate with a plastic cover. Anyway it is no longer available but the audiophiles of old Pioneer receivers have reverse engineered it and the schematic is published in a souped up version, so they make their own. Anyway I had made a few DC supplies, yoke drivers, and other things with them in the old days when they could be had for ten bucks.

The circuit can maintain a steady voltage with current flowing either way. OK well the cathode of the tube only lets current flow one way but the extra control is worth something to me. And like the W4ZT board or a string of diodes there are no giant bias bleeders. It may have other uses, such as when the screen current goes negative in some tubes, etc.

This is maybe more of just a thought experiment related to the project but the LTspice simulation looks very good so it seems worth sharing.

The setup is like for a 3CX3000. 4KV on the anode, 0-2A variation in cathode current at an audio rate 100Hz to make it easy to see.

Pictures:
1. the original STK0050 circuit (no values given) and the circuit from the audio message board, which someone made. It's a souped up version.

2. the circuit I was simulating.

3. the LTspice waveforms from that circuit.


schematic voltages:

V1 and V2 are just two 30V DC supplies in series. It's the STK0050 notmal kind of operationg voltage, except in this case the negative rail is grounded rather than the common tap.

V3 is a voltage that could come from a pot, with a range of 0 to 40V.

V4 simulates the 0-2A cathode current audio rate current by using a 4000V peak to peak signal biased 100V positive, through a 2000 Ohm resistor. This gives a brute force current waveform to be forced upon the bias supply just like wta comes off the cathode of the tube.

Waveforms:

V(output) is the bias voltage to the center tap of the filament transformer. In this simulation it is raised over time from about 1.5 to 38V then lowered back to near zero bias.

I(R1) is the current supplied by the upper amplifier the NPN stage. It varies from 0 to about 1A.

I(R12) is the cathode current through the filament transformer CT. -2A is the peak audio-rate signal, 0 is the trough of modulation, the negative peak.

I(R2) is the current sinked through the lower amplifier stage and varies from about 0 to -2A.

There are issues with this, for one thing the simulation does not take into account the reduction in cathode current signal amplitude as bias is raised or lowered. I don't see this as an invalidation but it is more like the drive was adjusted to make the same 2A peaks as the cathode bias was changed from zero bias to 38V.

OK so this may get off track but it seemed interesting enough to post.

Time to start fitting things together.

Sorry for the delays. Parents and house maintenance issues been eating weekends constantly. Hah sometimes as slow as projects go here I think If I grow old and pass on before finishing this then someone will have to take up the job!

So the tube cooling has to be done. It takes up space and so will determine what else fits where. A few 'dryer vent' items were had from the hardware store. There's a round 6" to 4" reducer that can be cut and formed to fit over the square blower outlet and transition to 4" duct size. Screws will hold it to the blower and aluminum tape should cover any gaps. A flexible but somewhat rigid 4" duct (not the plastic with coil spring type or the thin foil-sided type, but the compressed aluminum tubing that holds its shape well) should get the air to the underside or back of the RF chassis.

The lower chassis is 4" tall and I hope I can feed the air in at the back rather than from underneath. If it has to come from underneath it will be a PITA because the HV rectifier stacks are under there and I don't want the aluminum duct in the same space.

A piano hinge on-hand will probably be used to join the front of the RF chassis together and two brackets with screws should take care of the rear. There should still be room for a tuned circuit inside. Once it is together the aluminum tape can seal around the crack between the two chassis as well as any holes. The filament wires from the transformer are too short, so some more #6 will have to be added.

We also want to look at the scopes and stuff gotten recently. (http://amfone.net/Amforum/index.php?topic=42929.msg311240#msg311240) At least the HP 122AR dual channel unit should replace the single channel 120AR on the bench meter stack for audio range and general power supply stuff. Henry SWL also wants a portable tube low frequency tube scope for Tesla coil tuning. Probably a DuMont 304 may do or HP 120. There may or may not be time for scope stuff. The amp will have to come first. The sticking point is the cooling arrangement, can't do more til that is figured out.

Whatever is done there will be a few more images posted.

Your going to want to WAY oversize your blower Pat.

The corrugations in the dryer vent will kill airflow.

Much better to use pvc pipe as much as possible.  Smooth wall.  Will cut turbulence down a LOT.

Turbulence = slows airflow, causes blower to work harder and doesn't deliver as much pressure.  Also, will make a lot more noise.

FYI


--Shane
KD6VXI

this blower's last job was a 4CX5000 and a 4-1000 in a commercial transmitter. The round part of the outlet there is 4".  It's a 3450RPM capacitor run motor. I measured 1.75in s.p. H2O from it. Hope it is enough.

It's not the little blower that comes in those Henry amps, I think it should be enough. The two I have from Henry 2000W models both tested out at only 0.65" S.P. H2O. Enough to blast directly through the socket and remove 2KW heat but not much else.  IMHO those are really only good for a 4-1000 etc.

Anyway I will be putting a Dwyer Magnehelic gauge there sooner or later. Once it's plumbed up I can check the pressure drop in the air duct.

PVC is great but it's not at all a straight run, kinda horrible really and I am trying to figure that out.

choice 1. run the air in from the bottom, making the RF chassis a pain to ever remove.

choice 2. suffer a couple elbows and restrictions and plumb in the air into the cathode compartment from the back of the chassis.

got to spend a few hours on it today.

I moved the PVC pipes that conduct the GTO-15 HV cables from the rectifier tubes to the transformer and filter. Moving them 6" to the front made room for the 4" duct to easily pass through to the level where the tube chassis should be. I was lucky and only had to flip the brackets 180 degrees and there was enough loose wire so nothing was stretched. The brackets are a flattened "c" shape with a tab.

For now I put the 4" aluminum duct in. It was a little chore but fits well because it bends and stays bent. PVC will be explored later if things are too noisy or hot.

To bring the air up from the bottom, the amp chassis will have to be elevated so the duct will fit. Doing it from the back, it should fit right in. It seems like it would be easier to deal with later if it enters by the side or back. It should never need servicing but it will sure need tweeked.

This said, the RF chassis will need to be moved to the back of the rack. It's not a problem because it only contains the tube, socket, input jack, metering pickups, and the filament choke. Aside from the calibrations for the grid and plate current and filament voltage meters, there are no controls on it. The turns counters are there but not likely to be used. I think the original plan to put 6 Meters on this is too ambitious. The plate circuit components will be separately mounted but as close as
they would normally be.

I'm learning a lot on this project. It's quite different than what has been done here before. Lots of good ideas an examples from stuff already out there.

About the RF chassis - there are apparently a couple of open wires in the factory wiring loom. Glad this was checked today and not assumed it was good. All the un-used holes, a lot of them, have been sealed with that aluminum/metal tape normally used for ductwork.

Fired up the HV again to check the new inrush resistor value of 10 Ohms. It's great. The same 3 resistors were used just put in parallel.

A glitch resistor needs added. I have a box of 50 Ohm 200W units coming in the mail. They were free for the shipping! How many should be used in parallel? maybe 2 of them? I think 25 Ohms is a decent value.

I probably should be putting resistors in series to share the voltage better. There is 5KV. What do yall think about glitch resistors?

There is also a #30 wire in the plans for a HV fuse. Do I need both? Should it be #28 for a 2A plate current?

This table shows fusing current for various times. One is 32ms.  How fast should a wire fuse in the anode lead blow out in case of an arc in the tube?

https://en.wikipedia.org/wiki/American_wire_gauge#Rules_of_thumb

I used 50 ohms at 800 watts.

4 X 200 ohm resistors @ 200 watt each.

Big wire wound jobs.

This was in use with a 32 uF oil filled cap and a 12 kva ccs Dahl powered supply.

There are pics, I believe, on the thread I started about it.

--Shane
KD6VXI

That sounds pretty good to me.

Are these the ones or did you put 200W ones in? Very tidy assembly here.
http://amfone.net/Amforum/index.php?topic=41800.msg304387#msg304387

So that's about it then for the glitch resistor. I'll do the same thing or something like it but have to find space for the units as they are pretty long.

Before the RF chassis can be installed I have to know the 'desired' meters will work with the plate and grid current shunts that are present. If not then use larger shunts or smaller meters! These two meters have 5 Ohm DC resistances and need about 3mA to go full scale. Hope the others are similar in regards current and resistance.

OK well my helpful assistant has put the box of the rest of the Westinghouse 'RCA Transmitter Meters' away somewhere and after 2 hours I could not find the box.

The two shown are the RF ammeters, which I probably won't use. c'mon 20A!! 10A is 5KW on 50 Ohms, but with a tuner the Z therefore the current could be anything. No thermocouples for them anyway.

The thing is, these RF ammeter movements are not linear. I believe they would have been calibrated individually with the accompanying shunts, which are long gone never had them.
Here are the scale readings vs. mA DC through them. Used a 10K resistor and h/p 0-40V supply to get my bearings. In the end they are ~3mA full scale deflection. These (the rest of them) will clean up nicely assuming he remembers where he put them!

Some 50 Ohm resistors which arrived last week were opened and out of 17, 4 are cracked. They still check OK and maybe I will use a sleeve and epoxy to fix them later just for fun. The 'glitch' resistor can use 4 or 6 of them as in two parallel stacks, for a total of 50 or 75 Ohms. I would not want to go too much more since the resistance will drop the plate voltage with current, the resulting poorer regulation causing slight compression of modulated signal peaks. That is only worth bringing up because this effect is a cause of distortion and the plate supply already has very outstanding regulation.

An interesting factoid shows up when the broken resistors were examined. The Ohmites, which are fixed/non-tappable have what appears to be nine resistance ("novafilar" winding?) wires and the HEIs, which are ther type that can be tapped with a sliding clamp, have one wire.

Which is better for taking that huge voltage spike and surviving the 390 Joules when the 100A or 67A glitch happens? No sane way to check that I guess. Nit picking?

The glitch resistor has to be mounted insulated from the chassis because it is in series with the plate supply voltage. A glitch/bleeder combo from some other thing can be repurposed. (I think it was from a satellite uplink transmitter klystron supply, 3500W on S band.) The glass epoxy material looks a little overheated but is well within usable condition and will not be heated in its second life. A couple of thick ceramic stand-offs should mount this up once the new resistors are put in it.

Both types have similar ratings:
Ohmite 270 series:
Overload 10X for 5 seconds (335V@50R)
Body insulated to 4595V

Vishay AVT:
5% tol.
AVT-200 = 225W
Overload 10X for 5 seconds (335V@50R)
Body insulated to 1000VAC

A couple of 25uH 15A variable inductors were found. They have a shaft on each end and should be easy to use later for a tuner. They will be put aside for now.

I have some questions about the tarnish -does this act as an increased resistance or is it of little concern? Would it make sense to dip the parts in tarn-X or something?

How are these flat ribbon types units cleaned since you can't get into the coils contact area easily?

Is there a special grease for them?

There is a 'tight' or harder to turn  spot as the roller goes through a 360 degree turn. I loosened the screws that support the coil and put the roller to the tight spot, and the support moved a bit. Re-tightened in that new alignment - seemed to fix it. What causes this to happen? Someone drop the part or bang on it? They look in good condition otherwise.

The wheels in them look adjustable -notice the threads on the hubs. The small indentations on the hub 'collar' seem to be for a special wrench. One of them has a set screw in it. I don't want to fiddle around with that. Any time a special wrench is implied, it's easy to break something.

I'm told the tarnish, silver oxide, is a very good conductor so I guess I'll leave it. Lots of warnings about Tarn-X being acidic.

What is a good lubricant? Some people recommend Molycoat, a dry lubricant with Molybdenum Disulphide.
https://www.lawsonproducts.com/Lawson/Moly-Coat-Dry-Film-Lubricant/92964.lp

What will not harm the silver plating?

I noticed the 'wheels' the rollers are made of a few materials, looks like some aluminum and some steel-like or tin-like metal discs.

I really need some good advice here and have no doubt there are knowledgeable persons about this subject on the board. I searched cleaning edge wound variable inductors and got no result.

I used the wash and lube that was available at radio shack when I had to lube and clean my roller in the Harris.

Don't think it's available now, though.

--Shane
KD6VXI

Black silver oxide is nearly as conductive as clean new silver, which is why it's preferred as a coating for rotary inductors.  Skin effect currents actually flow deeper than the plating, so the silver basically acts as a preservative for the substrate material.  For the contact, the only place that connection resistance matters is where the coil and the wheel come together.  If you keep this area free of grease and yuck, you'll be just fine.  The coil surface and the wheel need to be free of burrs so that they will remain in close contact at all times. Everything else is cosmetic.  I have used Tarn-X a couple times, but the ammonia smell will knock you over.  Multiple rinses will solve any residue problems.  Calgon dishwasher detergents works well, but you will need to carefully relube after things dry out as they clean oilas well as they clean last night pork chop dinner plates..  If the ceramic is fairly spiffy and the coil isn't covered with gunge, it's OK the way it is.
Those Kintronic inductors are nice.  I have used their big brothers on medium wave systems all over the world. Take good care of them and your grandchildren will be fighting over them some day.

That's good news! Too bad it does not apply to oxidized copper.. I had to spend a few hours on a couple other coils of that material and now they are as shiny as a new still.

The only remaining thing, in two parts or aspects, is that there are a couple places where the movement/rotation is 'tighter' and the rest is fairly free; and that when I turn the shaft  through 360 degrees quickly, there are a couple of areas on a few turns where it seems contact is lost but if I go slowly, this does not happen. Considering the 360 degree rotation, the lost contact happens when the friction is least.

The picture shows the wheel. This turns apparently by friction between the two thin 'flanges' as the offset shaft moves in a circle. When the loose contact thing happens, it is always where the wheel does not turn, it stops and slides along for an inch or two. Hand in hand with this, the 'tight' places are apparently where the 'hub' of the wheel must be riding against the inside of the coil.

There is a small amount of play between the outer edges of the coil and the three grooved mounts. This may be able to be tightened up or loosened a little, and I tried this but it does not change the loose contact at the couple of places where it is an issue.

I can't see where the coil is damaged in any way, no evidence of bring dropped or hit, or anything else. What should I expect/inspect to understand this better? Is it even related to the intermittent 'loose' contact when rotating quickly?

I think I should try to slightly 'bend in' or cause the flanges to grip the flat sides of the coil harder. Bend is not the right term but how do I fix this?

It is an issue, sometimes we crank slow, sometimes we crank fast, sometimes we use a motor!

And thanks for yall's good advice on this slow moving project!

The rollers are put aside for now and the meters have been found and checked. Four similar sized meters were also found and checked.
All the DC current meter seem to have internal shunts, so they direct-read.
The RF ammeters (previously discussed) are missing the thermocouples.
The AC voltmeter has a small multiplier and relies on a larger external one.

All are Westinghouse. I have looked for a Westinghouse catalog to learn more about the meters and their part numbering schemes, but in vain.

Maybe these data points will help someone else.

image of 4 meters:
height 4" width 4.25", mounting hole about 3.75"
bakelite or plastic case
1) 4A scale, 4A full scale deflection, Type QX-37, Style PH-16164-1
2) 400mA scale, 400mA full scale deflection, Type QX-37,  Style PH-16166-1
1) 200mA scale, 200mA full scale deflection, Type QX-37,  Style PH-16182-1

image of of 14 meters:
2) RF ammeter 20A scale (ext TC not present) movement 5 Ohms, 3mA full deflection. Current not linear to scale marks. Type KT, Style N-636902
1) 20mA scale, 100mA full scale deflection, Type KX, Style XN-66146-3 100mA shows 20 on scale. Has overload relay circuits on back.
1) 200mA scale, 200mA full scale deflection presumed, Type KX, Style NY-712442-1 (movement not moving, dirty?)
1) 300mA scale, 300mA full scale deflection, Type KX, Style NY-66146-3
2) 600mA scale, 600mA full scale deflection, Type KX, Style NY-66146-6
2) 800mA scale, 800mA full scale deflection, Type KX, Style NY-66146-4
1) 2.5A scale, 2.5A full scale deflection, Type KX, Style NY-89576-1
3) 5A scale, 5A full scale deflection, Type KX, Style NY-66146-8
1) 3KVAC scale, 5VAC/1mA full scale deflection, Type KX, Style NY-66146-8, uses external multiplier

The glitch resistor bank was made up with six 50 Ohm 225W resistors to make up a 75 Ohm heavy duty resistor, but might be changed to four of the resistors for a 50 Ohm unit, in order to put the HV wire fuse on the same assembly. I have an idea how to enclose the wire so that if it explodes the pieces might be contained and also how to make it easy to replace the fuse-wire. The mounts for the resistors are two pieces of very thick epoxy-fiberglass board. One end has holes drilled with metal threaded inserts so that screws can be fit into it, and it can easily be mounted. The question is only where to mount it out of the way.

Also found a nice handwheel and these little vacuum tubes. Wonder what they are for? Notice the shield above the upper plate. It's a "discharge tube" made by Goodrich industries but I don't find any info on specs.

cage code K4358, NSN 2920-00-190-4301, 2920001904301, maybe also called "DG 71".

http://w2dtc.com/w2dtc-plasma-amp/2010-0225-hv-fuse.jpg (http://w2dtc.com/w2dtc-plasma-amp/2010-0225-hv-fuse.jpg)

I saw that, and there is a picture of another one on two cylindrical posts somewhere.  HV fuses are a great idea. I just want to enclose it for safety reasons. It might not get done so elaborately as the drawing by the time it gets made up.