The AM Forum
March 29, 2024, 03:58:40 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1] 2   Go Down
  Print  
Author Topic: Transmitter Cabinet layout?  (Read 14022 times)
0 Members and 1 Guest are viewing this topic.
KI4YAN
Member

Offline Offline

Posts: 161


« on: November 12, 2017, 07:28:28 PM »

I have finally gotten my radio room built, and my rack cabinet rolled into the corner where it will sit. I've had some of these parts for almost 10 years now, and things have been put together, taken apart, put together, taken apart a few dozen times...

I am working on trying to get things organized and actually get the framework of this cabinet going, and am looking for advice on how best to manage airflow in and out of the cabinet, how to deal with RF and AC in and out of the cabinet, and ideas on how to mount some of the heavy iron. I haven't built a door for the back of the cabinet yet.



It's a 1950's DuKane intercom rack, the rails are 19" wide but they were originally not drilled or tapped. Equipment was just stuck in the rack and holes drilled willy-nilly. Some of them line up, some don't. The plan is to lay the rack on it's back and set panels in until the front face is full, preferably the panels in their final resting places, and drill & tap the rails for the correct screws in the correct spacing-if I need to weld up and grind back a hole, I can do that easily.



I have accumulated a small pile of blank steel rack panels-these are the ones that are bolted into assembled racks during shipping to help them keep their shape, so some of them are 20g, some are 18g thick. There's also a 1/8" thick aluminum one in there and another chunk of aluminum that's 1/8" thick and 20" wide-I'd have to shave an inch off to make it into a rack panel.

This is the HV transformer, it's 5" wide and 7" tall, and about 4" thick, as far as the core dimensions go. 1KVA at 2300V.



The modulation iron is roughly the same size-it's a Kenyon T-495.

Over the years, the plan has slowly evolved, and I've finally nailed down some specifics. I have my RCA Ham Tips exciter built, the VFO is temperature stablized (after several years of goofing with it) and the drift isn't too bad. I also have the little SS rig I have been working on, and plan to have an auxiliary jack that will have keying and audio outputs so I can use it as an exciter as well.

The final amplifier will be a 4-125A, modulated by 4-65A's. I considered a pair of 4-125's modulated by a pair, and also a pair of 4-65's modulated by a pair, but this configuration seems to take best advantage of the parts I have already, and I can crib directly from the Eimac data pages, where data for the 125 by 65's is layed out and listed, with grounding recommendations and all. I also have a few ARRL handbooks that have similar final amplifiers.

Tonight, I am thinking about how to route airflow through the cabinet, and how to keep the cabinet stable on the casters. Airflow seems like I might put some vents in the bottom panel, and vents in the top of the cabinet or in the top panel. Stability seems to demand putting the heavy iron as low in the rack as I can get it, but then I wonder about putting the modulation transformer down next to the power transformers-worried about hum pickup.

Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #1 on: November 12, 2017, 08:31:12 PM »


Opcom comments
BTW sorry for the text size, it was not able to be corrected, no idea why it happened.

I have finally gotten my radio room built, and my rack cabinet rolled into the corner where it will sit. I've had some of these parts for almost 10 years now, and things have been put together, taken apart, put together, taken apart a few dozen times...

I am working on trying to get things organized and actually get the framework of this cabinet going, and am looking for advice on how best to manage airflow in and out of the cabinet, how to deal with RF and AC in and out of the cabinet, and ideas on how to mount some of the heavy iron. I haven't built a door for the back of the cabinet yet.

If you are absolutely sure everything will fit in the cabinet, that is great, and a fine time to work on the cabinet and the mains input cable/fixtures for it, and cooling.
It is unconventional but if you anticipate a dusty floor then it is not a bad idea to intake the air at the top and exhaust at the bottom. It works well as long there's enough airflow to create gentle turbulence. It also gives an opportunity to push air down onto the final RF section if that's to be at the top. It also presumes a closed cabinet except for the inlet and outlet. That's a real nice looking cabinet!




It's a 1950's DuKane intercom rack, the rails are 19" wide but they were originally not drilled or tapped. Equipment was just stuck in the rack and holes drilled willy-nilly. Some of them line up, some don't. The plan is to lay the rack on it's back and set panels in until the front face is full, preferably the panels in their final resting places, and drill & tap the rails for the correct screws in the correct spacing-if I need to weld up and grind back a hole, I can do that easily.

Perfect! Done than several times and on steel Bud racks so full and heavy that a block and tackle from the ceiling beam was required to stand them upright! Watch out for pinching/severing fingers when putting the final panel into place, that is the only dangerous part of the exercise.



I have accumulated a small pile of blank steel rack panels-these are the ones that are bolted into assembled racks during shipping to help them keep their shape, so some of them are 20g, some are 18g thick. There's also a 1/8" thick aluminum one in there and another chunk of aluminum that's 1/8" thick and 20" wide-I'd have to shave an inch off to make it into a rack panel.

This is the HV transformer, it's 5" wide and 7" tall, and about 4" thick, as far as the core dimensions go. 1KVA at 2300V.



The modulation iron is roughly the same size-it's a Kenyon T-495.

Over the years, the plan has slowly evolved, and I've finally nailed down some specifics. I have my RCA Ham Tips exciter built, the VFO is temperature stablized (after several years of goofing with it) and the drift isn't too bad. I also have the little SS rig I have been working on, and plan to have an auxiliary jack that will have keying and audio outputs so I can use it as an exciter as well.

The final amplifier will be a 4-125A, modulated by 4-65A's. I considered a pair of 4-125's modulated by a pair, and also a pair of 4-65's modulated by a pair, but this configuration seems to take best advantage of the parts I have already, and I can crib directly from the Eimac data pages, where data for the 125 by 65's is layed out and listed, with grounding recommendations and all. I also have a few ARRL handbooks that have similar final amplifiers.

Tonight, I am thinking about how to route airflow through the cabinet, and how to keep the cabinet stable on the casters. Airflow seems like I might put some vents in the bottom panel, and vents in the top of the cabinet or in the top panel. Stability seems to demand putting the heavy iron as low in the rack as I can get it, but then I wonder about putting the modulation transformer down next to the power transformers-worried about hum pickup.

Cooling is twofold then, cooling for each power tube and then general air clearing of the rack.

If you have the room to allow a 1U space under the lowest panel/chassis, then the bottom floor of the rack could be an air outlet/inlet easily. If I may suggest that one of those 1U fan panels will slide right in. If equipped with the usual 9 screaming TA450 fans, some DC fans would be much quieter. You only need 100-200CFM. A couple rows of holes or a screened slot 1-2" tall in the rear door at the top could be the outlet/inlet for the cooling air. I am guessing you will have a small blower with hoses arrangements or two small chassis-mounted blowers to cool the modulator and RF power tubes.



Why I mention it is from experience with internal blowers but no cabinet blowers in the AM transmitter here. The builder put a blower under the 4-1000 and I added one under the 3-500Z modulators when I installed those and removed the previous tubes. There was no forced air to clear the cabinet and the set got extremely warm, having only heavily screened louvered back doors. The convection was nowhere enough. The subchassis of the bias and screen supplies and everything else inside there was being heated by the big tubes' circulating waste heat to where it was all very hot.

The problem was totally solved after adding a 400CFM low speed centrifugal blower from a VAX6000 on top of the rack where there had been a heavy aluminum panel with many very very small perforations. This kind of fan is not just the most common thing, you will probably want something else, although it is about 4" deep, exhausts sideways 360 degrees for 'zero clearance' between fan and wall, would fit perfectly in/on the backside of the door of a small rack.



That blower is 24VDC but runs silently at 14VDC and does a wonderful job. This set is tall, yours is not and I assume you don't want to put anything on top. The point of that is a large low speed blower run at low voltage, or a bunch of low speed high pitch fans in a 1U tray should be so quiet that you will hear the internal blowers more than the cabinet blower.



[/size]
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #2 on: November 12, 2017, 08:36:20 PM »

sorry about the mess.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #3 on: November 12, 2017, 09:16:05 PM »

A few thoughts that are my initial impulses.

4-65s are a bit light.
Use the 4-125s, you can always go up to 4-250s or 4-400s... same sockets.

You said you can weld. It might be easier and simpler to buy pre-drilled strips that are for
racks from one of the rack manufacturers. The biggest ones are Bud and Premier. You may
find others online, since pro audio people use them to go to wood or 'glass road cases.

Have you decided on your B+ voltage? Aka, cap input or choke input PS?

                    _-_-
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #4 on: November 12, 2017, 11:03:39 PM »

I thought about 125s by 125s, but then I look at the power limits-375 carrier is the legal limit for me, 250W carrier isn't even 2db down there...not even half an S-unit on receive.

I had planned on capacitor input, with a diode bridge. The power transformer isn't centertapped, so it needs a bridge rectifier. That would give 3250V, so I started look for a choke for choke input...would give me 2070V. I am thinking maybe run the transformer on a variac and cap-input for 2500V output, is the correct answer.



I have five of these for filter capacitors, though. They're rated 115uF at 2300V, and were removed from defibrillator units. It's very tempting to run 2kV and give up some power, but I could also put these in series for 4600V rating:



If I put them in series, they'd be ok for 4600VDC at 57.5uF each. And I'd have one left over. What I don't have is a filter choke yet-have been more focused on other things and minding my bank account than seeking out one right now. I figure everything else has come cheap and easy if I waited long enough, the choke will too.
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #5 on: November 13, 2017, 07:13:35 AM »

Well, speaking from only my personal point of view, I'd not go through the trouble to build a large transmitter with less capability than the "legal limit".
Certainly I'd want >100% positive peak capability for the modulator.

The 4-65s are a bit small... really.

Otoh, the Collins 30k transmitter is ~250w rated and puts out a nice signal. (4-250s, iirc)

Imo, a variac is a very good idea, especially in the case of output tubes with greater than the "legal limit" potential.

At 2kv the 813 is maybe a better tube choice. You might want to look at the "K1JJ" 813 design that has been built by
quite a number of AMer's...

Suggest that you check into those Aerovox caps, see if you can find the mfrs specs?
The reason I suggest that, is that these caps may not be designed to handle high ripple.
They're rated very high for the size package, which is terrific, but check into the ripple handling thing.
That would make them excellent for a "second position" in a "pi" filter in a power supply.
And, no, don't run them close to the limit... unless ur 100% certain that the PS voltage is incapable of
rising, no load, above the max rating.

                              _-_-bear

Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #6 on: November 13, 2017, 10:48:51 AM »

Bear bear me to it, but beware of the photo flash style caps (which is what those are commonly called).

In the early days of the internet, stories galore about people blowing power supplies up with those style caps.  The CBers bought them in droves.

Turns out they would tolerate about 60 pct of published ratings and any high current demands would make them heat up and kaput city.

They are designed to charge up once, dump their stored energy and then sit until called upon to do so again.  They are NOT designed to be run near limit nor to be called on to continuously filter.

Two, in series, I'd call good for maybe 3kv, thereabouts.

--Shane
KD6VXI
Logged
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #7 on: November 13, 2017, 02:18:26 PM »

The Collins 30k was a single 4-125a modulated by a pair of 75th, for 250W AM output. RCA's BTF-250 was a 250 carrier FM transmitter, again a single 4-125a, driven by another 4-125A in the doubler.

I am aware of the legal limit thoughts, and plan on making sure the RF deck components can accommodate a 4-250 instead of the 4-125a. I just plan to run the 4-125a, because the 125s are available and cheap-the 250s not so much. The ability to run both by switching the taps on the mod transformer is appealing.

The thought has occurred to run the 125s in parallel too.
Logged
KU8L
Member

Offline Offline

Posts: 64


« Reply #8 on: November 13, 2017, 03:20:34 PM »

Sweetwater Sound has reasonable priced rack rails in a variety of lengths.    Ft Wayne area if I recall.

Curt
KU8L
Logged
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #9 on: November 13, 2017, 11:52:40 PM »

I am familiar with all the rack manufacturers-I used to work with them on a daily basis. The rails in this rack arn't totally ruined...just a little "swissed". I layed the rack down today and started laying in panels to find out how many holes were in the way/not positioned correctly/were correct and only one existing hole was mis-drilled. Won't be a problem. There would be a lot of cutting and grinding to remove the old rails anyway, as they are folded as part of the outer sheet of the cabinet, not a separate part.

However, I also finally got an answer from Aerovox on those capacitors...they are only good for 80 hours of service at the full ratings as filter capacitors. Not good enough. So, since I have no filter capacitors OR filter choke, I'm back to the drafting table on the power supply...the lower voltages are all present and accounted for in the transformer/filter cap stocks.

Assuming I can find a 4-250 in usable condition, I could modulate one with 4-125's and only slightly exceed the mod iron's ratings. The downside is that the 4-65a's will make ALMOST the same power-within 30 watts!-as the 125s at 2500V in modulator duty at the same voltage.

If it weren't for the voltage issues, I could set up a single 125 modulated by two, and swap the 250 in later, by configuring the power supply as a capacitor-input filter. This would produce 3250V, which I'd have to drop down to 2700V for the modulator, so I'd need a 200W, 3000R resistor, and even then the voltages would soar in the modulator on key-up due to the drop in current.

I am currently hunting 4kV rated filter capacitors, and contemplating selling the Eimac tubes or trading them for something happier at lower voltage-I could run the transformer at half voltage by wiring the primary for 230V and feeding 125v to it, producing a 1600V/430mA output, but then I can only get 500VA from the 1000VA unit.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #10 on: November 14, 2017, 02:21:29 PM »

If it weren't for the voltage issues, I could set up a single 125 modulated by two, and swap the 250 in later, by configuring the power supply as a capacitor-input filter. This would produce 3250V, which I'd have to drop down to 2700V for the modulator, so I'd need a 200W, 3000R resistor, and even then the voltages would soar in the modulator on key-up due to the drop in current.

Hi Jacob,

Sounds like a fun project.

A series resistor feeding the HV for the modulator would create severe distortion due to the varying drop across it as the modulator tube current swung around under modulation. Using a class B or AB1 audio stage needs as stiff a high voltage as possible to insure cleanliness. A power resistor  would work fine with a class A or class C stage (CW, FM or unmodulated) that uses constant current, but not well for "current swinging" linear AB or B operation.   Plus it would be a huge waste of power in the form of heat.

The usual way is using a big Variac or better yet, use a lower tap on the plate transformer if available.  Even a large 240 to 240 isolation transformer usually has taps to allow the drop you need efficiently.

Also, if you use tetrodes like the 4-125A, etc., be sure the screen and grid voltages are well regulated. The screens do need to draw current for their cleanest service as well as putting out rated power. Usually a simple zener regulated screen supply is OK. For the grid bias, use a string of forward biased diodes or a zener in the filament transformer center tap to get the proper idling current. This gives rock solid bias regulation.

You can also tie the grid and screen together, drive them p-p, and use the bias diodes in the CT as a complete system. This is the easiest and will be very close to the tetrode config for cleanliness.  Your audio driver needs to be pristine with no transfomers, if possible. The modulation transformer has plenty of undesirable phase shift to handle if you plan a negative audio feedback loop, which you should have. (If the mod xfmr is included in the loop)  Usually just one transformer is the max in a NFB loop or it may become unstable.  

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #11 on: November 19, 2017, 02:36:02 PM »

Now that I have to find new filter capacitors AND filter chokes, the question of cap input/choke input comes up. With modern diodes and capacitors, is there significant advantage anymore to one over the other? Aside from output voltage, clearly.
Logged
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #12 on: November 19, 2017, 02:53:28 PM »

Now that I have to find new filter capacitors AND filter chokes, the question of cap input/choke input comes up. With modern diodes and capacitors, is there significant advantage anymore to one over the other? Aside from output voltage, clearly.

A choke input filter will give you better voltage regulation.  Especially if the rig is plate modulated using class B modulators.  I think your xfmr will produce about 2.1KV with choke input, probably even more with today's 125vac line voltage.

You can increase the output voltage by adding another xfmr in series with the HV xfmr.  Example, using another xfmr with a 400 volt winding and its own set of diode rectifiers, connect the diode output directly to the (un-grounded) negative terminal of the HV FWB  diode rectifiers.  Your output voltage will increase by the 400 volts.  The current rating of the 400 volt xfmr has be the same as the HV xfmr.

The VA load on the PS is divided proportionally across the two xfmrs.  Makes for a much better supply than using a cap input filter supply and a Variac.

You can lower the output voltage by simple shutting off the primary of the 400 volt xfmr without disconnecting any other connections.

The secondary winding of the HV xfmr has to be capable of running 400 volts higher above ground.  Should be zero problem with your Electro xfmr.

Fred
Logged
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #13 on: November 19, 2017, 07:06:52 PM »

That presents an issue-2100v would be fine for a plate modulated 4-125a, and would work fine with the class AB1 modulators. But it would be a little low for a 4-250a-but would still work, didn't the globe king 500 run around that?

I can always boost the voltage by using the 140V tap on the input Variac.
Logged
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #14 on: November 19, 2017, 07:21:26 PM »

That presents an issue-2100v would be fine for a plate modulated 4-125a, and would work fine with the class AB1 modulators. But it would be a little low for a 4-250a-but would still work, didn't the globe king 500 run around that?

I can always boost the voltage by using the 140V tap on the input Variac.

Don't know anything about GK-500s never had one.  All my xmtrs are HB.  You can overheat a xfmr by applying too much voltage on the primary.  Not sure if this would be the case with the Electro xfmr. 
Logged
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #15 on: November 19, 2017, 11:46:16 PM »

Either way-I need to come up with a good filter capacitor solution.

Right now I'm looking at using a series string of 450V electrolytics, 10 caps in series. By using 100uF caps I could make 4kV, 10uf units-although they may not be very compact, they would be very inexpensive. I suspect a 500v leeway in the voltage rating should be enough?

I've also found a few chokes that might be useful, 10h and 12h, both sub 80 ohm DCR. By using them in the ground leg, would the insulation requirements be reduced? (I'm thinking insulation relative to the core)
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #16 on: November 22, 2017, 08:44:47 AM »

You'd need to parallel a number of series strings to get usable capacitance for filtering using the
10ufd 450vdc caps... The cost would go up, since you'll end up with a matrix of caps.

Email me, I may have some caps that will work for you... Cheesy
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #17 on: November 22, 2017, 12:31:15 PM »

Right now I'm looking at using a series string of 450V electrolytics, 10 caps in series. By using 100uF caps I could make 4kV, 10uf units-although they may not be very compact, they would be very inexpensive. I suspect a 500v leeway in the voltage rating should be enough?

I've also found a few chokes that might be useful, 10h and 12h, both sub 80 ohm DCR. By using them in the ground leg, would the insulation requirements be reduced? (I'm thinking insulation relative to the core)


Put the choke on an insulated board to protect it from DC-to-ground potential.   However, the voltage ACROSS the inductor (across the turns) is present whenever there is current flow. It will not matter if it's in the negative lead or insulated above ground.  Current change is what causes the 90 degree lag which generates high voltage across it. (inductive reactance)

As for the capacitors...  I would suggest getting a bank of "beer can" sized  caps, maybe 500VDC @ 100 uFd, or whatever. I would go at least 50% over the working voltage for safety. IE, a 3KV rated bank used for 2KV would be about minimum.   40uFd total is a good working number for a class B supply.  So, 6 beer cans in series X a few strings in parallel would do the trick.  

Remember that filter capacitance working with the input choke is needed to get the regulation acceptable. Poor HV regulation under modulation will cause distortion and splatter.   The class C carrier will help somewhat to draw down the peak voltage, but class B ssb alone has no mercy and regulation will be pulled down hard during voice syllables.

Personally, I use 140 uFd total, oil filled caps for my bank with a 30H choke. Rock solid and the IMD numbers are -55dB 3rd with my class A linear system. If the HV supply sagged even a little bit, all this effort would be for naught.



T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #18 on: November 23, 2017, 09:51:16 PM »

Would there be any gain from running an L-C-L-C filter, or is a single L-C section generally going to be enough?

Currently I am looking at 30-40 henries inductance and 50-60uF capacitance as being "affordable", as the pile of microwave oven transformers I've collected over the years produced 6 identical units, and I plan to harvest two HV windings and fit them onto a single core. The rough TPV is 1.1 as they were originally wound, and the output voltage is 2100V, so there should be 2300 turns per HV winding, and I know that the windings easily carry 400mA. (they are 21g wire)

I also have a big pile of transformers that need to be rewound, so I may just find the biggest stack of EI200 laminations I have and start winding turns.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #19 on: November 24, 2017, 12:03:04 AM »

Sounds like a good plan, especially starting with a big core.

Yes, a simple L-C filter will work fine with 30-40H and 50-60 uFd.  That is considered a robust supply filter by most.  

Usually with that much filter reactance, hum/ripple is not an issue. The amplifier gain is not that high.  Regulation is more the big question and potential issue with big power amps with ~13 dB gain or so.

The key problem will be how stiff you can build that power transformer so that it does not sag much under full power demand.  If you can keep the AC HV sag under 5% or better, you will be golden.   You might put a resistive load across the transformer before committing to metalwork to be sure.  Heavy wire and core mass is what counts.  

Usually a heavy #6 wire feed from the 240AC mains will be a great source to start with.  Since drop problems are cumulative in the final number, make it so the only factor is core mass. Everything else is easily controllable.  You appear to have most things covered already.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #20 on: November 25, 2017, 11:05:09 AM »

By all means use a choke input. Additionally a resonant filter will give best regulation over the audio envelope cycle.
For break in, the resonant setup will be well worth the trouble and reduce B+ transients.
Ran into a real world experience of the difference between a choke input followed by a big cap vs a resonant choke input followed by the same big cap and the transients were much more controlled with the resonator. I did not understand what I was seeing so I put it in ltspice and it looks pretty clear to me. As always others with more experience/other opinions please comment/correct.

henry used:
5H & 0.3uF
8H & 0.02uF ?
11H & 0.1uF
11H & 0.2uF

these seem to be ok but for the 8H one, but it's Henry right? This was confusing. The 8H value above must be either an error in the book or it's not intended to resonate at 120Hz.

Some information suggests to resonate at a slightly higher frequency than the line, such as 125H for 120Hz ripple, etc.

120Hz:
5H & 0.352167uF
8H & 0.220104uF
11H & 0.160076uF

125Hz:
5H & 0.324557uF
8H & 0.202848uF
11H & 0.147526uF

all the decimals are not needed, just for show, but if it is too many Hz off the resonating quality is diminished. The actual testing showed that the resonance changes as the choke value changed with DC current.

My opinion is that the '125Hz' suggestion was because choke ratings are given at full load but the most critical regulation (resonance) point would be at no load or bleeder load or tube idling load, the user must decide. The capacitance should be chosen for those regions.


* nonresonant filter.png (45.17 KB, 1052x971 - viewed 414 times.)

* resonant filter.png (44.27 KB, 1052x971 - viewed 443 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #21 on: November 25, 2017, 12:49:17 PM »

Yes, a resonant  "pre-filter" is a great idea. I've never tried one but see their value.  Once the supply is on, the dynamic regulation effect seems worthwhile.

Somewhat related, I control the initial spikes and unruly nature of HV supplies by T/R keying the 240 AC primary with a step start.  Using a 10 ohm 200 watt power resistor in series with the 240VAC primary slowly brings up the HV the first 500 mS and then shorts itself out of the circuit with a relay.  

The reason is three-fold:  I like the HV off when in receive for safety reasons. I like the bleeder current off to conserve power during receive. And, the in-rush charging current to the whole system is mild, thus easier on the SS rectifiers, filter caps and no spikes to raise havoc in the amplifier.  My 4KV supply starts at a smooth attack 3300V the first 500 mS and then jumps  to 4KV. The last 700V is hardly abusive. Perfect for my use.


T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #22 on: November 26, 2017, 07:48:15 PM »

Right now I am pretty tied up with work this coming week-but hopefully this weekend I'll have time to scrub the bottom of the rack back out and get the HV transformer bolted down, and the control/interlock transformer wired in and start wiring the interlock circuits for things like the rack door.

I think it'll be worthwhile to have interlock switches mounted for each of the transmitter's chassis decks-the intent is to build an RF deck on rails, so the rack cabinet screws can be removed, the cabinet slid out, and tubes changed/settings tweaked without having to get behind the rack. Interlocks would be vital for that kind of feature on any HV wiring, so I think any section that is on sliding rails will have to have one.

The only issue with having the modulator deck slide out of the rack is the weight of the modulation transformer-the Kenyon T495 is not a lightweight! (even though it is undersized for the job)

Logged
KI4YAN
Member

Offline Offline

Posts: 161


« Reply #23 on: December 01, 2017, 03:42:36 AM »

Not much to report on this I'm afraid, only that the rack rails have been (mostly) re-drilled to EIA spacing, and the few errant holes welded up, ground back, and re-drilled to suit.

I set up a 17x10x3 chassis and started to look at stiffening, and parts layout. I've pulled down an old 100VA transformer to rewind for a 5V power transformer, to heat the final amp. The grid input tuning will be via an MB-40L national link-tuned unit. Mine is the fixed-link model. I still haven't found a "correct' plate tuning cap, but I do have the big 200pf/section, 2 section EF Johnson with a 0.125" plate spacing. (supposedly good for 4kV) A tapped coil wound from copper tube, and possibly silver plated, combined with one of the big 400v rated variables (one measures in fractional uF with all the sections tied in!) should complete the RF deck.

Now, as far as the clamp tube for the RF final, would a small sweep tube work? a 6BQ6 or similar should easily hold off the 400V screen supply, or maybe it's 2:30AM and I am out of my mind.
Logged
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #24 on: December 01, 2017, 12:49:47 PM »

I think a 6Y6 is usually use for a clamp tube.  I think that's what I used in my 813 xmtr.  A clamp tube is only needed if you use all grid leak bias on the final.
Logged
Pages: [1] 2   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.07 seconds with 19 queries.