The AM Forum

I am working on a compact intermediate power GG 813x2 amp and I want to use an ANTEK 8t800 transformer for the B+
This transformer has two 800vac @500mA windings and I want to use bridge rectifier(s) to get about 2200vdc for the 813s

I have seen two articles on 813x2 amps using this tranny.
One hooks the 800vac windings in parallel and feeds a voltage doubler. (800x 2.8=2240vdc unloaded (1800vdc under load)
The second one hooks the 800vac windings in series to get 1600vac and feeds a bridge 1600 x1.4 =2240vdc (about 2100vdc under full load).

I am certain that the voltage doubler will “work”, but I hate the things in principle because they have poor regulation, so I want to avoid them…

The bridge is great, giving a nice stiff voltage, but hooking two 800vac windings buried in the toroid makes one nervous because even though ANTEK promises well over 3.5kv insulation between primary and secondaries, there is no spec between secondaries.

I figured out a possible way to hook up two 800vac bridge circuits and hook them in series and trying to find out if it has been considered elsewhere, I found the circuit below in a forum.

Has anyone else considered this arrangement to limit winding to winding peak voltages and avoid an arcing episode?
Opinions and suggestions desired- please… 73, Mike

PS: I had not thought of hooking to the center of the capacitor stack, but was planning to hook the negative of the second bridge directly to the positive of the. First one that has it’s negative at ground. Which way is best??
If I wanted half voltage (1100vdc) , I would hook that node to it’s own filter caps.

It might be worth a try to call Antek and see if you can find a good guy in the engineering department who would hipot secondary-to-secondary and give you an unofficial result.  Antek is a real company, so I suppose that their listed value includes a good safety factor.
Years ago, making high power/high voltage RF components, we had a DC Break...basically a 14 inch, 50 ohm, coaxial blocking capacitor that was rated and sold at 80 kv.  When the customer came for acceptance testing I took it to 80kv and held it for the required time.  The customer then wanted to see it go to 100 kv.  We made him sign the acceptance sheet so that he owned it and would have to pay for repairs if it smoked.  I ended up taking it to 150 kv..  He wanted to go even higher but I told him it was time for lunch and we weren't going further.  Period.  The point is that if you find the right person, things can happen.
73 de Norm W1ITT

It is amazing that this question has been up all day and no one has made any suggestions, so I will offer my opinions.....  Edit:  Looks like either I type slowly, or Norm types faster!

The two bridge rectifiers connected in series, as in your drawing, are not really any better than putting the transformer secondary windings in series, and feeding one bridge rectifier.  The voltage across the secondary is the same in either case.  The only possible advantage is that the peak inverse voltage across each diode in bridges is half what it would be with a single bridge across both secondary windings in series.

I think the voltage doubler gets a bad rap because it is basically two half-wave rectifiers, with outputs in series, so the ripple, even though it is 120 cycles, is greater than that of a full wave rectifier.  But load regulation is not necessarily any worse.  It has that reputation because of the many cases where a transformer intended for a full-wave bridge is used in a doubler, thus doubling the load current, causing more heating in the transformer windings, thus offering poor output regulation.

Transformers designed for single wave doublers generally have larger secondary wire, thus lower resistance, and less voltage loss (and heat) than transformers designed for the lower current load of a bridge rectifier.

Single phase voltage doublers are commonly used for transmitter and linear amplifier supplies, because they can actually provide better load regulation than a center-tapped full wave rectifier, by virtue of only half as many turns on the secondary winding, allowing for larger wire size, and less voltage loss (heat).  But I actually prefer the full wave voltage doubler.  It requires a few more diodes, and a couple more capacitors, but it actually provides better voltage regulation, less ripple, and much less peak voltage on the transformer secondary than does the single phase voltage doubler.  Eight diodes, same number as two bridges, will do the job.  
The two Antek transformer secondary windings may be placed in parallel, providing a larger overall wire size, and less I2R loss (heat) than using them in series.  One other factor that affects the transformer efficiency (heat) is whether current is being drawn during each half cycle.  The center-tapped full wave rectifier uses each half of the secondary only half the time, thus the current during that time, for the same power, is double, yielding double the heat loss in the winding.  So it is best to use all the windings, all the time, for most efficient transformer performance.  The single phase voltage doubler does this, but at the expense of higher secondary peak voltage with respect to the primary and core, but the full wave voltage doubler does not.

I could not find my drawing of that circuit, but I located a similar one on AMfone here:  
http://amfone.net/Amforum/index.php?topic=25120.0
and I attached the relevant image to this post.  That other post may not provide the best explanation of the circuit, but it can be discussed in much more detail here.

The diagram shows a single output capacitor, but there is no reason two or more could not be used, with appropriate equalizing resistors, of course.

I used this circuit when rebuilding the HP-23A power supply for my SB-101, and it works very well.  Neither end of the transformer has more than the peak secondary voltage, with respect to primary or core, at any time.

Thanks Norm and Rick.
Well, I had to wait a while, but the advice does make food for thought.
Rick, that 4-8 full wave doubler is a fascinating circuit, and this 8T800 with the 800vac windings in parallel will have a low resistance (17-18 ohms).
I followed the links to an article that goes into a lot of detail on how the 4-8 works
It uses 5 half cycles to complete a circuit cycle and it seems to have the potential to outperform the doublers I see in commercial amps- I think- because I cannot be sure.
Fascinating circuit explained here:

http://rawfire.torche.com/~opcom/psu/4x8_power_supply.html


I admit that my attitude toward voltage doublers is based on prejudice and unfounded intuition and I have built a lot of bridges up with great success and they gave me at least 1.3 x rms voltage under full load.
Still- I am giving this some thought…You almost convince me, but I have ome more play on tje bridge idea, even though I seriously doubted that my Rube Goldberg double bridge was going to stress the windings any less than a refular series winding with a bridge.

Norm, I like your idea of making the call. The guy that used the 8T800min a series winding with a bridge said that he had called and received assurances that it was ok to do it and he had not had an arc over, as of publication.

I also discovered this sheet on the ANTEK listing for the 8T800 on Ebay, showing that the series winding was ok, or at least, it was illustrated.
See below:
Note that they specify that wires 2 and 3 are connected and the note at the bottom indicates polarity of windings. I do not know if this indicates a higher insulation for flash over or not, however, so I will make that call.
The circuit us in the top left corner.
73, Mike

rawfire.torche.com is a site on which I was kindly allowed some space many years ago, glad to see Chuck's site is still up. So many have disappeared.

My current site (below) has the article with the simulation result. Good to have things in more than one place!
https://bunkerofdoom.com/lit/4x8/index.html

I am also interested in the results of the call. I am using an 8T650 with the windings in series into a Harbach half wave doubler. I upped the specs on the caps to 390uf at 500V to give me a full 4kv rating on the cap bank. I get 3600V unloaded and 3250V loaded by the Drake L4B.

The thing that worries me is that the leads that come out of the transformer are rated at 600V even though the transformer is rated at 650V rms AC per winding. Not good.

What are your leads rated at? 600v for 800v windings is even more worrisome.

John

Hi there,

I have done this exactly many many times and it works perfectly.
It is series connected of two or in my case more dc supplies.
The way you have designed it is the absolutely right one.
I have 5 different transformers and I connect them this way in order to get their total dc output voltage.
I can always decide how many of them will be activated from their primary side in order to get different output voltages.
Go on this and it will work very well, no doubt.
In your case however using one transformer with two secondaries I don't see any benefit to use two bridges.
It could also work well with the standard design, the secondaries in series (in the correct phase) and only one bridge.
With the today insulation material 1600vac is not a big deal. For sure the insulation of a today transformer will outstand it.


Stefano
 

Thank you for the encouragement, Stephano.
I have used bridges in numerous center tapped transformers going back to old TV pulls and never had one melt down.

I was unfamiliar with the insulation in these new Toroids, but I just remembered where I saw the 813x2 amp that uses the 8T800 with the windings in series and a bridge rectifier.
The article is in the March/April issue of ER Magazine.

I have included a screenshot of the power supply circuit in this post.

ER Magazine is a fantastic resource for Construction of equipment and restoration.
Unfortunately, QST and CQ are more light operators magazines and QEX gets into stiff that is less interesting to guys on AM-fone.
Fortunately, the old issues of Ham Radio Magazine, QST, 73 and and QEX and other legacy publications are available online at World Radio History.com and Archives.org
Or the art would die out.

John, I wrote ANTEK a message, but have not heard from them. I may call them tomorrow.
73, Mike

Patrick, I had not made the connection between you and Bunker of Doom.
Wow! Cannot tell you how many times that site has provided very useful and rare knowledge. Many thanks for your contribution to saving this stuff for the future.
73, Mike

Mike, I looked at the Antek spec sheet for the 8T800 transformer, and it looks to me like you are good to go putting the secondary windings in series to feed a bridge rectifier.

Following is an excerpt taken from this link:
https://www.antekinc.com/as-8t800-800va-transformer/


"The 800VA toroidal transformers are commonly used in the high-end tube audio amplifier and RF transmitter. They have static shield between primary and secondary coils to improve the isolation and noise interference. This toroidal transformer has a very low magnetic leak. They are specially designed to work on all standard 115V or 230V at 50Hz or 60Hz. These transformers have heavier gauge wires then the normal requirement to avoid the copper lost during the full power output. The dielectric test is more than 3500V in between primary and secondary coils. Please see the test data for short circuit and open circuit. In most of the cases, this transformer can be output 20% more power from its rating at 60Hz power source without any problem. This transformer comes with 2 rubber pads and all mounting hardware."

The 3500 volt dielectric test would indicate to me that there should be no problem with 1600 volts into a bridge rectifier.  But after reading their text, I continue to wonder what they do with the "lost" copper!

It is interesting that they use larger than necessary wire for the windings, and that should help significantly with regard to dynamic voltage regulation.

You did not mention whether you are using an oil cap, or a bank of electrolytics.  Either should work well, but I have often seen too few capacitors in series for a comfortable voltage margin.  I know it hurts to lose more microfarads by adding more caps in series, but it is much better than seeing them fail in a chain reaction if one goes belly-up.  I do not believe a 5 or 10 percent margin is enough, but 20 or 30 percent makes me more confident.  Think of how much overvoltage is applied to the remaining caps if one becomes leaky, or fails altogether.  We cannot guarantee an equal voltage distribution based upon matched equalizing resistors, as that does not take into account different leakage, causing imbalance between the various caps.  For 3600 volts, I personally would not use less than ten 450 volt capacitors.

You mention laughing about bonehead moves, and your perfectly functional schematic with two bridge rectifiers in series reminds me of a perfectly innocent error, based upon lack of understanding, that was made by ME when I was six years old.  I had dad's battery charger, consisting of a variac, transformer, and a large selenium bridge rectifier, and I wanted more voltage.  Knowing that batteries in series produce additive voltage, I thought, why not series connect the output of two huge bridge rectifiers?  I proceeded to connect the input AC terminals of both bridge rectifiers in parallel, feeding both from the same transformer.  When I started to connect the rectifier outputs in series, well, let's just say it was more than spectacular.  Next came a theory lesson from dad, and another transformer to fix my error.  Lessons like that are not quickly forgotten, but there was no laughing, and no yelling.  Just part of dad mentoring, and I remember vividly how he stated that we learn more from mistakes and failures than we do from easy successes.

The old GE Master Pro base station and repeater AC power supply had a +250 volt power supply and a second +400 volt power supply that were stacked with the 400-volt supply using the +250 volt bus as ground. Those things ran for decades without issues. I used them for external B+/HV power supplies for all type of radio projects and never had any issues. Will attach a picture.

😂😂😂.
Ok, Rick- you got me in the revolutionary method you used to boost those Selenium  rectifier bridges…
I remember those old stack rectifiers in tv sets. Silicon rectifiers and later- schottkys would change everything- but that idea was better in your imagination than when implemented, either way…😉
I really wish that I had not done worse, but I did. I almost killed myself several times growing up either on my Honda motorcycle or on high voltage supply mistakes, but I will save that for another day….
Your Dad did well. He reminds me of my Uncle who employed me on his farm growing up. One day when I was 13, I was using his Tiller and the thing quit and would not start. I decided that I should take apart the carburetor and fix it and surprise him with my mechanical skills. I got it apart into about 30 pieces and then my fever cooled enough that I started into a panic because I was not sure how to put it back together.
While ai was sitting in the dirt in the barn, my uncle came in with his top hand and saw me, walked over, looked around and said calmly, “what are you doing Mike?”
I was praying for a quick death by this time and felt I deserved torture instead amd I said, “I was fixing the tiller”. He shook his head and walked away, saying over his shoulder to his assistant, “help him put it together”. Bad Eye Robertson guided me on how to assemble and adjust the carburetor and get it running and I learned so much that I was able to repair those engines all my life. My uncle never mentioned my mistake, but my gratitude for his not blowing his top and guiding me to learn makes me grateful to this day. Your Dad was such a Man and we are lucky to have had them there for us, Rick.

Anyhow, the plan is tentatively to stack 6-560uF @450volt electrolytics with resistors across them for my filter. Five times 450 is 2250 and the peak voltage is 1600 x 1.414 = 2262vdc
Six will give me about 18 percent more than minimum. I agree that a lot of headspace is best, specially with electrolytics. I may stack 8 in a 2x4 arrangement to allow 3 to fail… I over-design most things, so nothing new here. I still have 70uF if I stack 8 of them. What do you think?

One thing, though gives me a brain worm-a little.
My 2x4K amp will be powered by a 3400vac tranny in a bridge. This will give at least 4800 volts if I have 230vac on the primary, though this tranny has 4 taps and I will use the full primary to minimize the voltage.
Also, I have a 30Amp variac that will be in one leg of my 240volt line, so I can adjust the imput voltage from about 125vac to 250vac worst case.
My plan is to never bring,that,variac up more than enough to get 4500 volts out of my supply.
The question is this:
I have 3-10uF @5kV oil caps for my filter.
If I am always starting at bottom and bringing up voltage slowly, is 500volts head space enough for my oil caps. They are nice Bosch units and not old and no PCBs, etc.
I would never run this close using electrolytics, but figured oils will be ok with 10% head room. Is this reasonable?73, Mike

Wow! GE Master Pro? That brings back memories. I never worked on a base station, but serviced a lot of repeaters with 4cx250 or 8122? Finals in the field. Plenty of opportunity to kill yourself while standing in mud and making sure the HV was killed…😉
I did mostly mobile radios and the Master II was popular in the late 70s with the discrete bipolar amps. What a noghtmare- those things had to go back to the shop bench!
My favorite old transmitter was a 39.5mhz low band base that,was an old Prog Line unit. It had a 100TH in the final and it was like a star in there when the mic was keyed. 😎 seems like it put out 100 watts FM on the low Band, though not sure after  40+years.
The most hated mobile radio by GE was the TPL. It was GE’s first attempt at usimg circuit boards and seems like they were phenolic. Total junk. The Master II seriesmwas much more reliable once you got the amp balanced and adjusted,
I also service old Motorola MOTRAN mobile radios. Fantastic radios. I field serviced them after scrapimg the dirt off. Seems like they used 5894s for the VHF band.
Got my start servicing mobile radios and repeaters. Life and tech has changed a lot since the late 70s. 73, Mike

Well Mike, you bring back more memories from the distant, and recent past.
My dad hated having to try to start small engines that we had to mow the lawn with an electric reel mower.  I cannot remember how many times I ran over the power cord and had to stop and splice it AGAIN!  I learned about gasoline engines the same way you did, loved them.  I taught my dad how to deal with them, but he said that job was all mine.

As to the motorcycles, the bug bit me when I was stationed at Lowry AFB, Aurora CO.  I rented a Honda, took it up Mt. Evans, the highest paved roadway in the US, then the next day up to the top of Pike's Peak.  My wife did not want me to ride, but she departed company after 38 years of marriage, and that was my Cue to get a two-wheeler.  I got a 2006 Honda Shadow 650, then an additional bike, a new 2012 Yamaha 950 VStar, that sat in the showroom for a year, so I got it for half of list.  Rode both up until last year.  Never had a close call, but I took the safety course before starting to ride.  At 75, last year I decided I should no longer ride, due to weakness in my legs after years of chemo.  Both bikes sold, end of adventure.

Back to the power supply.  For your 2260 volt rig, 6 caps provide a 16 percent margin 2262/2700  at 93 uF, while 8 caps give you 37 percent margin at 2262/3600, and you still have 70 uF.  If you have them, and they fit in the enclosure, I would opt for the 8-pack.  With that margin, they will probably never fail, and run cooler.  An engineer would say use the minimum needed to get by.  The accountant would come along and subtract one, maybe even two.  But you are only building one rig, and you are the one to fix it, and clean up the mess if the caps go like a chain of firecrackers.  The difference between an optimist and pessimist?  Optimist sees glass have full, pessimist sees it half empty, while the engineer sees a glass that is 50% larger than necessary to hold the liquid.

To put it in perspective, I ran a pair of 4-400As in a linear I built in my barracks room at Davis Monthan in Tucson.  I used a UTC S-48 HV transformer, rated at 1300 volts, 500 mA, 650 VA output, where a choke input supply configuration was recommended.  I used a stack of HV diodes like used in the F4C Phantom radar, only needed four to make a bridge.  I fed it with a 10 amp variac from the 120 VAC line.  A 200 watt Kodak Photoflood bulb was in series for tune-up.  My filter consisted of four 14 uF 2000 volt GE Pyranol oil caps in series parallel, providing 14 uF at 4KV..  Not knowing better, and liking to live on the edge, I ran them up to the full 4000 volts, and ran the linear at 250 mA peaks, for 1KW indicated power input.  Even though the transformer center tap was intended to be grounded, no one told him, so he didn't complain.  Those capacitors were previously used by dad for a photography strobe light, charged slowly via an 816 MV tube, and discharged all at once, many times.  That supply never failed, all the components still working.  The amplifier build was May 1968.

I realize this is many fewer uFs than you plan to use, but all signal reports were excellent.  More might have given better regulation and a bit lower IMD, but never any hum reported.  More Ufs needed at lower voltage, as you are doing, but I believe 70 is more than enough.

The three 10 uF 5KV Bosch caps should do marvelous duty at 4500 to 4800 volts.  That is plenty of capacity at that voltage.  I do not know how the Bosch compare with the GE caps I used, so staying under the 5000 volt rating is good practice.  At least you do not have to worry about balancing the voltage across many capacitors in series.  The difference between 4500 and 4800 volts will not be heard or noticed at the receiving end, and the variac lets you avoid going above 4500.  If one cap shorts due to overvoltage, the end result WILL be heard at the receiving end.  Need I say more?

PS  I have not forgotten your request for more info on my Frankenstein T-60, and will get around to sharing more soon.

Rick,
I have to hand it to you. I rode into my 20s and after getting married stayed with the van or my Suburban.
My Uncle Glen rode all his life and was in a Honda club that toured all over the country. It is the only way to ride “safely”, because people who drive cars either do not see bikes or figure they do not rate the normal respect one deserves on the road.
Actually, in college, I rode in a bicycle club on 10 speed peddle bikes and they are really dangerous unless you know what you are doing. These days, there are a lot of bike paths and lanes, but I got literally blown off the road aomce by an 18 wheeler hauling coal, who missed me by 2 feet and was going over 70.
I had some close calls on my 305 Honda too. I woke up in a cold sweat for several years reliving some close ones… I was full of ice water while it happened and only got the shales later. Funny how time slows down in those situations, bit it is essential to be out there. Still- I loved it and remember a lot of great rides with friends in a group of 4 or more staggered. With those guys, we got space and respect and nobody cut us off.
I laughed at your Dad hating lawn mower engines. As a youth, I was a gear head and loved getting dirty, but as an old man, I hate getting grease on me. I still service my Troy-bilt tiller with that venerable Tecumseh T60 engine that I bought in 1982. It still starts on one pull in the spring, except two years ago. They don’t make tjose great cast iron engines any more. I am forced to repair my Yanmar YM165D tractor too and my ZTR, because mechanics are not around that can service them.
Last time I rebuilt a real carburetor was in 1982 when I rebuilt a two barrel Holley on a Dodge Omni.
I still have a dwell meter and feeler gauges somewhere, but do not own anything olde enough to use them. It might be why I love Venerable old tubes and big iron for Radio. When I was 30, I had the bases of at least 50 tubes memorized, but most young people never heard of any of them… Glad a few other old Geezers remember it and even use that knowledge, even though my grid is leaking these days in my Class C memory…😉

Yes, I am definitely going with 8 of the 560uF @450volt caps. They are new ones and more compact than the older aluminum ones. With 8, I only have 250v across each.
I was just thinking this morning that I bought some surplus plastic caps that are 16uF or so and high voltage. Three or 4 of those in parallel might be better and they are fairly compact. As you said, 70uF is overkill and 32uF is plenty for my 2260 supply.
I just bought the Bosch caps and they look very good. I think that I have a box full of either 4 or 8uF oils in my RV Garage and they are 5kv too. May be GE brand and no pvcs in them, though as long as,they do not leak, probably mot a issue anyhow.
Your pushing your oils to the ledge gives me confidence that my 500 volt headspace is adequate. As you say, the difference between 4500 and 4800 volts in a 2x4K rog is insignificant anyway. The plan is to run the tubes at about 1kw each out (2kw total) so,the 3rd order products are way down. Tom advised me about this because the IMD is -35dB or so, at 1000 watts per tube and -30dB at 1500 watts per tube.
The 2x813 gg amp is to drive the 2x4K amp both on SSB and on AM linear.
The plan is to drive the 813s with about 20 watts from my FLEX5000A and get 100 watts AM linear to drive the 2x4Ks to 500-600 watts carrier output and 2000 watts pep with modulation. The FLEX can give me -35 to -38dB IMD and hopefully, the 813s can give -35dB run at 100- 200 watts out to drive the big jugs.
The IMD output of the amp is never better than the driver or intermediate amp.
The goal is not to make enemies of my neighbors on the band… I credit Tom for helping me see this factor. In the deep past, I just troed to load my amps for maximum and then cut back mic gain so, the meter current stayed less than half the maximum and ask buddies to listen around the frequency for,splatter.
Of course, when splatter appears, you have already lost the war. New SDR receivers that have built in panadapters allow one to see distortion products that many may not even notice.
I have never been an appliance operator interested in plugging in a radio amd talking only. The design and construction of the stiff is the real fun in it. Guys like Tom inspire me because they keep pushing the envelope and other guys have built a lot of stuff and their advice is priceless.

I ran a 5kv GE cap with a 4kv xformer, FWB.

Yes, 6.2kv on the cap.

I don't recommend this at all.  An amp came in for a freshilening up.  Tested the cap, esr was low.  Didn't think the original builder would 'extend' the ratings like that, but he did.

That dn capacitor is still in the amp and working!

I asked the original builder, he built the amp in the early 90s.  So it's got some miles.

500v headroom on an oil filled is fine.

Just be mindful of the summer / winter dryer outlet voltage difference, though with the vatiac you have that figured out.

--Shane
WP2ASS / ex KD6VXI

😂😂😂.
Well, Shane, I thought that Rick was pushing hi 4kv cap at 4kv, but your guy lays him over completely. 😉
I was worried a little about only 500 volts headspace, but running a cap at -2.2kv is insane…

The guy was obviously suicidal with that Chernobyl overload😉😂😂
Wheww- I would not want to be in the shack if that thing arced over…

I need to say, though- I have never been able to get by with this kind of thing.
I have seen guys put up antennas that would not last a week at my house, but they stay up through violent windstorms and ice for these guys. On the other hand, if I do not over design a thing- it will go down in a gentle breeze.

Reminds me of my youth- when my little brother (I was eldest) got into anything or broke something, I got into trouble. I would say- “it was he that did it, not me” .
The answer? “He does not know better, but you do and should look out for him”.
I used to think it was just Dad, but then when I built anything or put up an antenna, it had to be done with strong supports and well waterproofed with tension relief on feedlines…

Bottom line. At least 500volts headroom is minimum on 5k caps.
Thanks for sharing.

Mike,

Yes to all your comments about riding safety.  A group of cyclists are hard to miss.  When not riding in a group, I tried to stick to the rural two-lane roads.  Problem with the four-lanes is the fact that so many people try to thread the needle, switching back and forth without looking first, that is scary.  One of my good friends at HP was killed by a wild pickup driver doing just that, and Mark was on a Gold Wing.

I had a Troy Built Horse for many years, but I hated the Tecumseh, it was always hard to start.  I was given a Honda engine that fit perfectly and was quieter, so out went the T.  Now my brother is gardening with the horse.  I do not mind getting greasy at all, but working on the engine under the hood gives me a backache.  And why do I need to pull the throttle body and the entire induction system from my Nissan Maxima 3.5L just to service the spark plugs and coils?  When I serviced it, I found two different makes of plugs, three in the front were easy access, and a different brand than the inacccessible back bank.  What does that tell you?

Back to the focus of this thread, Yes, I believe you will be happy that you used all eight of the caps.  Which is better, 8 units, working conservatively, or two good spares on the shelf?  Good for what?  Just the look of a bank of 8 is so much better than a six-pack.

Two linears in cascade, each running well below their headroom limit, always provide much better IMD performance.  Tom went even one better, and used the best IMD-rated tube for the intermediate amplifier, I believe it was something like a 4CX350.  For the intermediate amplifier, I prefer a grid-driven class A, as opposed to a cathode driven stage.  I am wondering if a pair of 813s is overkill for the intermediate stage, but if ya got em, run em.

I have a Hermes Lite II which I plan to use as a receiver and a VFO for the AM rig, providing true transceiver tuning convenience.  If you can zero-beat the received signal, then you are on the net frequency.  But now I am actually considering the construction of a solid-state linear for that rig, based upon a recent design of WA1QIX.  It will throw a lot of heat, but that ain't all bad this time of year!  I am so tired of many nights in the low 20s, and living near HotLanta, that should not be happening.  I live in the SOUTH!

So many projects, so little time.   If I could get finished with the final version of the Audio Processor project, and get the artwork verified, I could spend uninterrupted time on the 4CX3000 Class-A series modulated rig.  Talk about a shack heater, that will be 100 percent efficient in turning dissipation into cozy shack warmth!  I am thinking that modulator will be paired up with a couple Eimac 304-TLs in push-pull.  They will work well at 2KV, and provide good modulation headroom with a 5000 volt supply!

By the way, I recall Ray, KA3EKH, and AMLover both commented on stacking the rectified and filtered outputs of multiple HV secondary windings.  Yes, that is very common, and if you look at the early Tektronix 500 oscilloscopes, in fact all the way through the 535, 535, and 555, they all stacked the regulated outputs to provide the many voltages needed in the instrument.  If it worked on the communication equipment, and the Tek instruments, it must be A-OK.

Rick,
Sorry that you did not bond with Tecumsehs. I have only had this 6hp T60 and maybe I got lucky.
I did have a problem one year when I forgot to drain the tank and carburetor bowl and the alky-gas corroded my jets and after fighting and cleaning with mixed results I got a chinese knockoff and installed a shutoff valve and a filter in the line, plus I was able to find a source for gas without alcohol and all has been smooth since that ethanol us hydroscopic and rots traditional fuel systems… not a fan! 🙄

I saw Steve’s post on the stack of 11cN90s as a linear for AM.
It was something he had prototyped, but have seen nothing else since.

I am using the 813s because they are tough and I have at least a couple of dozen of them I have accumulated. I will run a pair with 2200 volts and load them to about 120-140mA so I get 60- 70 watts carrier to drive my 2x4K rig on AM linear. If I get 25% efficiency the two 813s will be dissipating about 90-120 watts each on carrier, so they do not overheat. 60-70 watts should drive the 2x4k amp to 4500 volts at 400mA or about 2kw input and about 500 watts carrier output. The goal is 2000 watts peak with modulation I think.
The 4-1000a tubes will be dissipating about 750 watts each on carrier only.
Lots of heat- just like Steve’s FET amp! 😉.
On SSB, I would load the 813s to put out 300 watts and drive the 2x4k amp to 3kw out- peak theoretically, though Steves says 1200 per tube is cleaner, so maybe 250w from the 813s and 2500 from the 4-1ks on SSB and hope to get -35dB 3rd order IMD or thereabouts…
I always dream big and plan long and adapt as I go.…

By the way, I just remembered that I bought some 30uF at 6kv caps that were made for defibrillators in hopes of using them for hv supplies.
Do you know anthing about them?
I have attached a picture and tech data for thebones I have. They are rated at 10,000 cycles on defibrillators, but why not use them as filter caps?
Ant comments?
Very compact and they use oil and polyurethane powder, I think.
73, Mike

Interesting thread.

Mike and Rick:

All of this is very worthwhile.  Remember that the average station is running about 200 watts pep output SSB.  (my estimate)  If you run 2000W pep output, then you have to be 10 dB cleaner than the average station JUST to blend in with the 200W crowd.  Otherwise you will start to stick out and get complaints.... especially if at -25db 3rd IMD or lower.  This is like running a real world -25dB 3rd 2KW pep rig effectively producing -15 dB 3rd. (relative splatter)

Anyway, in the future, we will all probably switch to adaptive pre-distortion / Pure Signal SDR technology. The old school amplifiers like we are talking about will become instantly clean by using it. But for now...

Running the 813s in grid driven, class A service:  Compared to a pair of 813s in GG with regulated screen voltage (built in NFB) it may be a wash.   But take a grid driven pair of 813s in class A1 with additional RF NFB around it -  and the class A would win.   But more heat with the class A and an additional stage of RF for the NFB loop.

The super linear system I built about ten years ago used the 100 mW FT-1000D pre-pre-driver, (-70dB 3rd)  into a 1 watt class A lab amp, (-70dB 3rd)  into a class A  3CX-350J at 50 W (special linear purposed tube) (-55 dB 3rd)  into an 8877 running class A, (-55 dB 3rd)  and into a pair of 8877s.   At "low power" 1500W output it was showing a spectacular  -55dB third order. The RF tuning was critical to get -55dB 3rd.  I usually parked it in the DX window on 75M . A fast tuneup would yield about -45dB 3rd.  That is one clean amplifier system. Actually matched pre-distortion / Pure Signal stuff when tuned carefully with a tone and spec analyzer before each use.

Mike, if you have a stiff regulated screen voltage on the 813s and 4X1s, you should do FB.   But remember that the 813 and 4-1000A were not designed for linear service like the newer 3-500Z, 3CX-1200, etc..  But they do very FB in linear modulator service at audio frequencies.    The extra -10dB 3rd using careful cleanliness techniques can turn the 813/4X1 into a decent amplifier. (using GG, regulated screen voltage, reg grid bias, very low grid current and heavy C2 plate loading)     Clean drivers are critical. The result can never be better than the dirtiest driver stage in the whole chain.

It's not easy to break the -30 dB 3rd IMD barrier without some extra effort.    Look at the modern ham solid state amplifers going for $7K ++.  They are still hard pressed to beat the modern tube linears.  I thought by now we would have PDM linears, but that didn't happen in the ham world.  What saved our ass is SDR adaptive pre-disortion instead. I can always tell the stations running pre-distorton by their sharp skirts.  It's like when spark gap rigs were getting shamed by the new tube continuous wave DC rigs in about 1920 (~ 100 years ago) .... and finally outlawed in 1934.

Tom, K1JJ

Hi Tom,
😂😂😂
Wow, you just made me think of my best shot at low 3rd order IMD as equivalent to a rotary gap spark tramsmitter compared to a fleming valve oscillator stabilized by a quartz crystal…😳😬😬😂😂
Seriously, thanks Tom, just saw your email. Will definitely go AB1 GG with about 300 regulated volts on the screens and a switched stack of bias diodes to set my operating point for best performance as a driver.
The 2x813s will be a dress rehearsal fir the 2x4Ks setup, which will be parallel.
This SDR predistortion thing is new to me. Just downloaded a paper on it and will peruse, though it will not stop present plans for my best (-35dB 3rd order) performance by a Luddite Neanderthal-one step from a rotary spark gap generator…😉.

Also, I have to say that you did all the old linear technology could hope for in that string of class A amps amd -55dB sounds amazing to this ole cave man geezer.
I should have guessed that you had already travelled to the rim of the earth in that journey.
The fact is that power is not the biggest thing for me, but getting 500 clean watts of carrier on AM linear sounds fun. I appreciate your counsel- as always, Tom.
73, Mike.
PS: Hey- what about these defibrillator caps. Any knowledge on them? MB

Hi Mike,

The Flex, Anan and probably some of the "kit" SDR rigs have adaptive pre-distortion software and capability by now. ("Pure Signal," etc)  Just sample some RF off the final amplifier output (using a toroidal transformer) and feed it back to the low level SDR rig.  Commonly done.

The good thang is no matter what linear chain you build now, you can easily turn it into a pristine -55 dB 3rd machine later on using this technology.  Even a nasty 600 watt solid state RF amp can be used as a driver in this case.


For now, you will probably do better using a single 3-500Z as a driver for the two 4X1s in linear.  That's what I am using here now... Last year I built up "Baby Blue,"  a single 3-500Z linear with 1.5 to 3KV on it. (HV Variac controlled)    The driver needs to be cleaner than the final amp and this does the trick. Later on, I will bias them harder for better efficiency and run adaptive pre-distortion.  I admit the 813s are more romantic.  

Remember to include a bias setting into cutoff class C to run "linear AM".   That really works well.  The AM carrier provides the "bias".   It will not work on ssb class C...

* I have no experience with the defibrillator caps. Seems like a repeat of the photo flash caps.  I've  used photoflash caps here for years. I'll bet someone here from industry knows if defibrillator caps will hold up.   10,000 cycles seems a lot, but what if you T/R key the HV supply like I do?   A soft step-start will probably help lifespan.

BTW, with 6KV @ 32 uF, I'm really surprised there are not a lot of defibrillator electrocutions reported in ER rooms or at nursing homes using them.  I mean, even a common ham 2KV @ 20 uF power supply can take you out....  They must be current limited and well insulated.


T

Thanks Tom,
Ok, I will leave off using the 813s, but I may make an overlay of a couple for my front panel…😉.
I actually have a 800 watt 1100-1700 vac tranny that will pit out 400mA in a bridge, so I can get 1500 or 2300volts by just switching the tap and will not need a 10 amp variac…

It came out of some medical equipment, I think. The secondary was 70 ohms I think for the 1700 full winding. I found data somewhere that it was 800va, so figured the winding was rated about 450-500mA.
I have a pair of Eimac 3-500z pulls that a guy told me were full power in a SB220.
Have not checked them yet, but I read his message on the internet, so it has to be true-right? 😬😉😂

Wow, that sample and clean SDR idea sounds great. I do hope that someone figures out how to make my FLEX-5000a do that. I love the open system it has, so maybe some genius can figure that out.
I have always loved panaceas!!🤓😉😂😉
73, Mike

interesting topic.  I enjoyed the GE references esp the old TPL (known officially as Transistorized Progress Line.  Unofficially as toilet paper line) I really liked the Mastr Pro receiver modules for single freq. work

As to the Antek toroids.  I think the assembly techniques are the same across the product line.  We had discussed this several years ago here on the forum when I raised my potential objection for placing hv secondary windings in series.  I still think it is a bad idea since if the secondary windings are bifilar wound with only the winding 'varnish' between them for insulation!

Having no experience with Antek qc they have ok'd the series connection in their drwg J-01255 and have green lighted this practice - so who knows?

Left up to me and having taken one of these toroid xfmrs apart I would stack several lower power hv supplies limited by the 3500 volt spec.

Hi John,
Thanks for the advice on the ANTEK toroids. There seems to be a lot of opinions on it. The guy that published the ER articel on a 2x813 rig hooked them in series and said he had used them for some time.
Reminds me of the lyrics of a song that was written and recorded in 1947, a year before I was born.

Smoke That Cigarette written by Merle Travis and Tex Williams-

Written by Merle Travis and Tex Williams
“Now I'm a feller with a heart of gold
And the ways of a gentleman I've been told
The kind of guy that wouldn't even harm a flea
But if me and a certain character met
The guy that invented the cigarette
I'd murder that son-of-a-gun in the first degree

It ain't cuz I don't smoke myself
And I don't reckon that it'll harm your health
Smoked all my life and I ain't dead yet”

The lyrics tell it all. A man knows that what he sets out to do could go very wrong, but he does it anyway, and if he escapes a while, he justifies it by saying, he has done this ill-advised thing a long time and it hasn’t killed him yet…😉😂😂😂

Luckily for me, I just figured out another way to do this thing.
I remembered that I had a Varian insudtrial high voltage transformer that has two 120vac windings in the primary and a single secondary that is 1850vac with a tap at 1180vac
I measured the windings and the secondary was 68 ohms (42 ohms for the 1180 tap)
The core measurements calculate the iron to be rated at 750VA.
I,also decided to go with a single 3-500z and run 1650vdc on it at 400mA. The datasheets say it will give me over 350watts on SSB full carrier and the 3rd Order IMD is rated at -46dB.
The entire purpose of this amp is to take 20-40 watts from my FLEX5000a and give me enough power to drive a 2x4k gg final running AB1 that will hopefully give me -35dB 3rd order at about 1500 watts out on SSB and 500 watts carrier on AM Linear operation.
The idea is to make fewer enemies while putting out full legal power by making it as clean as possible.
The 1180 winding should give me 450mA out easily at 1650volts dc and that is how I plan to run it as a super clean intermediate power amp and driver.
That us the plan today amyhow…🤪

Yes, the TPL were pure dung in my mind. I used tomtake hookup wire to strap the circuit board traces, which would break and cause intermittent open circuits and angry customers. We were  taking them out of service and junking them when I left GE. The Master Pros were great radios, but then they got “smart” and brought out the Master II series…nice radios except the 100 watt mobile  had 4 discreet bipolars that had to be balanced or they would not work and we were spreading coils to get them going. They had a 25 watt mobile that was a nice reliable mobile,though.
I started designing receivers and transmitters forvwildlife studies after that and lost track…

Hi Mike,

Good decision on using a GG 3-500Z as the driver for the pair of 4X1s in GG linear.  You should not have any issues with getting good IMD results from the 3-500Z itself.

A few suggestions:

Be sure to have more than 1650V available for the 3-500Z in case you need it.  You do not want the tube drawing much grid current for best cleanliness. The grid current demand rises as you try to pull more power out of the tube.  More HV will reduce the grid current requirement and make it cleaner.  I am aware of the -45 dB 3rd order datasheet claim at 1500V, but not sure I believe it. I did not not see a noticable change in IMD when I ran my single one as a driver.  I ended up using 2200V for less grid current as discussed above and it worked out better.  IE, at 2200V, I just had to tickle the 3-500Z to get good power out as a driver for my pair of 4X1s in linear service.

For the 4X1s, use regulated screen voltage with minimum grid current in GG.  (I think that is already your plan)  The 4X1s will be the bottleneck for clean IMD so do everything you can to make them clean.  Expect to have low efficiency and more cooling because of heavily loading the plates and higher idle, thus two tubes for more dissipation to hit your final output requirements.

Now the Flex 5000...   Did you run a two-tone IMD test with it yet to prove out the favorable IMD expection?   If you expect -35 dB 3rd out of the 4X1s, you will probably need at least -38 to 39 dB 3rd IMD out of the Flex 5000 at its required driving power.  Being a solid state final, I don't know if this can be done, considering most solid state ham RF power amplifiers are around -30 db 3rd or so.  Is the Flex 5000 an exception?

As you know, the final result cannot be better than the dirtiest link in the chain.  Now is the time for design mistakes to be made, not later...

And of course, your overall safety net is still using adaptive pre-distortion later on if this does not work out.  

T

Thanks, Tom.
I see what you mean. I had forgotten grid current in my mental calculations:
I want to keep that down to about 120mA or so.
Might as well hook hook up to the top tap, though I will need a 10 amp variac to back it off and I was trying to save space (old problem 😬🤪)

I also appreciate the bump in the  FLEX- ratings.
Just downloaded some QST tests from 2008. 3rd order was measured at -34dB in those lab tests. Figured that was at 100 watts. Need to get more detail. Since I only need about 50 watts max, hoped the FLEX was better at half power, so my -38dB was a “hope so” rather than a fact….😳😬
Actually expect to only need 10 -20 watts from FLEX. Need low level lab tests…

This predistortion thing seems essential. Can it be implemented in a FLEX5000a?
I have older tranceivers that used 12BY7- 2x6146 outputs. Figured that they. Were worse than new rigs, though.
I have KenwoodTwins and a Yaesu FT101ZD and even a KWM2, though all of these rigs need service since they have been dark for 30+ years.

How do I get Predistortion panacaea going in my FLEX?

73, Mike

Just curious, how are you all determining IMD? I look at things like carrier suppression, 2nd and third harmonics and overall bandwidth but I hear you all throwing IMD numbers around and wonder just how you are determining them? Are you looking at spurious noise around the carrier as a byproduct of modulation? something like occupied bandwidth? Back a thousand years ago when I did annual proofs for AM transmitters we did measurements of THD at different frequencies that was used to determine frequency response but no broadcaster has done that for years now, we only look to confirm that are overall bandwidth fits within the NRSC window.
Sorry for being a bone head but just have to say I have no idea what you are talking about with thees IMD numbers.
Please enlighten.

Ray F/KA3EKH

Hi Ray,
Tom K1JJ is the expert on this. I believe that he uses a spectral analyzer display on an SDR receiver for the purpose, but he is the person to answer.
He may put a two tone test in and observe the near and farther away noise products visible in the narrow band on each side of a signal.
The lower the noise floor in modulation, the lower the IMD. 3rd order products are very close to the frequency an 5th, 7th, 9th order products further away.
Having a very low 3rd order IMD means you have fewer enemies putside of your transmit channel. If you have poor 3rd order IMD, you might sound great to your friends, but guys 5-10kHz away comsider you a LID…😬😉.
73, Mike
As for being a bonehead- we “boneheads” can only be cured by asking those questions and getting counsel from those who know by experience. Tom is a thinker and he always asks for help and it is the reason that he is a great builder and tester of radio equipment and antennas.

Maybe the real question is in these days of the SDR radio is that the new standard? Think maybe I am asking someone to say SDR and DSP are King, the old way of keeping everything well within design specifications for linear operation and spectrum analyzers is over?

No, Ray, I do not think that is true at all.  Of course we all want to have as clean a signal as possible, such that we do not cause QRM to the folks in QSO on adjacent channels with unnecessary splatter.  What is being discussed is no different than maintaining a clean house by monitoring our modulation, with either an oscilloscope or modulation monitor, keeping our modulation percentage and bandwidth under control.

Whether we are using plate-modulated AM, or a lower power AM exciter with a linear amplifier, the signal is only as good as the weakest link.  One of the important things Tom has demonstrated, and is sharing with the masses, is a technique to make the ANALOG linear amplifier as clean as possible.  This involves using the tubes that provide the most linear response within the operating parameters involved, including drive level, bias, gain, output power, as well as plate voltage and current.  Similar parameters are involved with solid-state linear amplifiers.  None of this is new territory, but Tom has taken it to the limit of practicality by spending a great deal of time fine-tuning the parameters for absolute LOWEST IMD possible.  He has even gone to the degree of cascading several linear amplifiers, of increasing power level, to achieve the cleanest signal, as opposed to pushing one or two stages to get the desired output power, albeit more distortion along the way. 

I think a key takeaway here is the fact that the more power one runs, the cleaner the entire system must be, in order to have the same level of spurious energy just outside the needed bandwidth of the transmission. 

The SDR comes into play primarily because the panoramic and waterfall displays provide an excellent, calibrated energy measurement in the frequency domain, obviating the need for the typical ham to purchase an expensive spectrum analyzer.  If one builds an old-school exciter and linear amplifier, or a high-level plate modulated rig, the SDR in this case is simply performing the function of a laboratory instrument to analyze and tune the quality of the resulting transmitted signal.  By the way, the SDR also makes an excellent receiver, able to "hear" much better than some of the best boat-anchor receivers.  The SDR also demonstrates an advantage of receiving full AM signals compared to SSB reception.  The tools at your fingertips, with the mouse, allow you to adjust the passband of each individual sideband, avoiding QRM that affects only one side of the received signal.  It also makes reception under selective fading conditions much more readable.

Mention has been made of the Pure Signal predistortion methods used to clean up artifacts and limitations in the various components of the transmitted signal path.  This is relatively new, as compared to just making the linear amplifiers as clean as possible, but in no way is it required in order to have an AM transmitted signal that is above reporach.

Excellent answer, Rick.
I worked with a Polorad Spectrum analyzer for 20 years and it was $12,000.00 used.
I have a nice HP now, and in my youth, I bought an old 455khz Panadapter and it was a wonder to me.

It is astonishing how fast The new SDR receivers have gotten very cheap and for narrow spectrum in a ham band, they are great.

The same goes for storage scopes. I had a Hitachi that set my boss back 5grand in the 90s, but it is pitiful compared to a $400 new Chinese made scope for both speed and for capture time.

I find predistortion both fascinating and a bit disheartening.
I want the cleanest signal I can get, but it makes me think of those commercials for air freshener spray. A woman walks ina room with a can and the place is,full,of dirty clothes and whatever,and,she walks,over and sprays her jeans ans says, “now I can wear them a couple more days before I need to wash them….😳😬🥺😝😝😝
Just think- a LID with a CB linear can sound as good as a skilled RF engineer who takes great pride in running his equipment properly and monitoring it constantly….
Another simile might be liposuction for LIDs. 😉

Not you Tom. K1JJ is thinking a 3rd order IMD of -60 to -80dB. Tom makes the point rightly- The more power you run- the cleaner it has to be to not cause others to notice and feel disgust. That- is what motivates my own efforts.

Mike, you make some very good points.  Brings to mind one of Tom's transmitters that was designed NOT to be a shack heater - he called it "Summer Breeze".

Iff'n I ever set up a rig with predistortion, it will likely be called "Febreze"  Some things are better washed and rinsed than hidden with overspray!

😂😂😂🤣🤣
Rick, you have wonderful taste in humor, but no sense of smell…😉

Funny story about Fabrese, the toxic stuff that is featured on my commercial story.
I bought some little 2 gallon trash bags to put in our dually when we went in our fifth wheel to South Texas two years ago , so trash did not build up in the floor.
I did not read the label and when I opened the box, it was like getting anhydrous ammonia sprayed in your face…😬. Yep- they featured the fresh Fabrese abomination…
 I washed my hands 8 times because my wife loathes these cheap perfumed products… I threw them in the corner of my garage amd forgot about them until this Summer. I wondered if they had faded any, but they were like the energizer bunny or an old Twinkie- still putrid, so I put them in the trash.

I am definitely going to remember your name for a predistortion rig if I get one to push my mid- 30s IMD up 30 points and may plagiarize the name  it myself, if I can stand the smell…😉.
More likely, though, I will offer to pay a copyright fee.
73, MB

Mike, thank you for the kind compliment. It's just a site of wierd old stuff! I hate to see unusual circuit ideas lost over time, as well as some magnificent ones. "Everything old is new again".

FB on all the comments.  I was away from the computer for a few days and missed all the dialogue.  Mike and Rick pretty well explained it all.  

There is not a lot of "practical" and simple IMD amplifier info available to the average ham. I mean the formulas and complex testing equipment and procedures usually scares away the average guy who passed a multiple choice test to get a ticket. He wants to keep it simple and cheap. Pay the money and blend in.

I've tried to keep transmitter testing simple and in plain language grouped into two main areas:  1) A simple objective two-tone test for 3rd/5th order IMD (and harmonic dstortion) to give a general baseline for changes and improvements.   2) a touchy-feely test involving a common SDR spectrum analyzer using actual microphone voice audio inputs to get a look at how the overall dynamic bandwidth looks in the real world.  It's amazing what an "ssss" "ch"  "yallo" and normal voice speech can tell you about your signal, once you develop an eye for it.  For example, I can tune in any AM or ssb signal using my SDR spectrum analyzer and tell you about his voice power band ranges, sideband suppression, bandwidth using a mask, ratio of blower/background noise to peak signal, carrier suppression, and approximate IMD by freezing a frame and counting the grid dB squares.  Easy to do once time is spent observing good and bad signals.  The voice tests are relative, though the two-tone tests can be quite precise.

It's something akin to setting up our audio on AM.  There is some technical expertise and practical voice tests required, though they do not have to be overcomplicated.

The first objective to all this testing and improving is to be a good neighbor.  Being a good neighbor means being able to operate close to another QSO and simply blend in with the crowd. If you run more power than the average ham, then you will need to work harder than the average ham who just buys a rig and plugs it in.    The second objective is to have a generally clean, pleasant to the ear - and good sounding signal with little background noise (hum and blower noise, etc).  We want a sharp roll off of our sidebands (after the desired audio power band)  with cleanliness that we can actually hear.  

Yes, I think adaptive predistortion IS the new standard. I can instantly see who is running it by the sharp skirts. It is quite dramatic.  If the overall active amateur radio population were increasing like in the distant past, the crowded bands would encourage it.  But because there is getting to be plenty of room to operate, we will probably see no changes in rules in the future.  I mean wideband AM is no problem at this point.   The driving force will simply be manufacturers wanting to sell more radios by telling us to switch to the ultimate mode....  "Yaesu Thin Body" "Like a good neighbor..."    or whatever.... :-)

One last point:  Once a rig is "tricked out" and running FB, it must be checked periodically. How many times I have not used a rig for a year and then fired it up with problems?  Be sure to check the rig's spectrum every time you operate... or better yet, have an outgoing and incoming look at the spectrum.  We want to be able to see our own signal as well as help out others who may have developed an issue.  Rigs love to break.... :-)

T

Patrick, I live the sight because just when I think a thing is not obtainable,
I look there and you have it waiting.
Just like all us ole geezers, this art snd knowledge is becoming more and more scarce.
Luckily, there are younger guys who like to explore the old arts and parts.
Weird?- Nah, I like to think of it as rare and esoteric. 😉😁 73, Mike

Tom, Sorry stepped on your toes because I hit the button s few seconds after you did.
I agree with everything you said and I credit you for moving me from getting the highest output and setting mic gain so the plate meter hovers at about half the tuneup current to actually thinking about the region that my amp will occupy in the transfer curve of the amp. My dumb old school way did prevent me from splattering and causing disgust on all directions, but your ideas on improvement in IMD by better matching to the cathode and operating my big tetrodes with  with fully stabilized screen supplies and monitoring the spectrum at least +-10kHz for Intermodulation distortion that puts fuzz on the sideband(s).
We have all observed a guy with a bucketmouth in your face signal and everybody giving him 40 over and (good audio) reports. The guys 5khz away see this LID differently, however and many are too polite or timid to point it out. When one hears truly great audio and a vertical wall around the channel, it becomes very important to be that guy.
I am taking my time andcspplyingbthevdtuffbyiu have shared from hundreds of hours of testing and experiment so I have equipment that is capable of excellence.
Unless you just want to talk only- the pride of making your own stuff is the real joy if this art.
I also agree in principle to your philosophy of KISS, but my brain always wants to know more, so I love to run numbers snd hope to get closer to a good design before I begin to understand how to run it.
I admit it- I am nutz and afflicted with brain worms and questions.
😉😂
73, Mike

Mike,

It's always nice to see someone make good use of information and improve their lot with it... :-)


I'd like to do a shout-out for Don, K4KYV....

The other night I was tuning around and came across Don's AM signal on 3875.  Don runs a big, old school, plate modulated transmitter.  I would consider it in the big rig class. He has been known for clean and easy to copy audio for decades.  So I gave his signal a close going-over on my SDR spectrum analyzer.  

I think he may have the cleanest analog big rig signal on the band.  His audio power bandwidth was about +- 4.5 KHz max. His skirts after that dropped like a rock, like down 70 dB from full carrier.  It actually looked like an adapted predistortion rig.

He is proof that old school analog plate modulated rigs CAN be very clean and sharp with the right amount of work.  We tend to get lulled into thinking that a big plate modulated rig needs some slack, but he obviously knows different.

I sent him an email to give him a pat on the back and find out how he was limiting the highs at 4- 4.5 KHz so well. He said (on that particular evening) he was using a 3.5 KHz audio transformer low pass filter. The filter appears to roll off sharply after 4 KHz.   He also has a VERY clean rig IMD overall to show skirts that drop into the noise rapidly after 4 KHz. It really does resemble an SDR adaptive predistortion rig.  He matched the band activity to his narrower filter.

I could hear the 4 KHz limitation to my ear, but it was very slight and Don still gets very favorable audio reports - some saying he has the best audio on AM for overall communication and listening.   This is really doing something at 4 KHz cutoff and using generally pre-1940 tubes and parts.

Congrats to Don for a nice job building, testing and optimizing that classic rig.


Anyone looking to add a clean AM signal to their mental library of viewing experience, tune in Don next time you hear him on.

T

Tom,
It is funny that you talk about Don Chester. When I started out as a short wave listener in 1962, K4KYV had the strongest signal on 75 meters. I can still hear him- “K4KYV in Woodlawn Tennessee”.
He was always talking to an Old Timer from Paducah KY. Don was an inspiration to me to want to get on 75 meter phone. His audio was clear and crisp and powerful and natural. He very often expressed contempt for SSB and considered most of them LIDs. I remember one time, he modified his transmitter to a kind of control carrier setup to prove that AM could sound good even if the carrier varied. He put it back to his regular setup after making his point.
He was always more like a Dutch Uncle than Mr Rogers and never suffered LIDs or fools. The fact is that he was right and his respect for the Art was first to him.
He always was an AM guy and OWL was his religion.
I have read and reread his articles on AMfone archives and his Q&A for a great deal of wisdom and knowledge of the classic AM Art. Seems like he ran a pair of 4xKs back in the 60s.
He was not well loved by a lot of appliance operators which he lectured on putting out splatter, but I always had a deep respect for his knowledge and the quality of his AM and still do. Don has said many times that classic iron and tubes should be cared for and preserved because they cannot ever be replaced. I have been influenced by that reverence for the classic equipment from the golden age of AM radio. Don’s legacy is one that I hope will be appreciated.

The defibrillator cap I'm using in my 4x4-125A amp is 56 uF@4.2KV. The amp is nearly 20 years old (the beginning of the build can be seen in my profile pic) and I don't use it very much,  but it's still going strong  8)

If using the large stack of diodes, which I recommend, to set bias.....  Also throw a large value electrolytic across them.

Helps with dynamic regulation oodles and gobs.


--Shane
WP2ASS / ex KD6VXI

  Hi Shane, I see what you are saying, and I hear you.

The fact that it makes a difference suggests that the dynamic regulation of the series diode based bias is poor, where a large swing in cathode current could bring the diode forward bias from around 0.6v at idle, to .9v or so at full boogie. This is for each diode! That is a 50% swing!

There are many circuits floating around with a small zener diode, a power transistor, and a few passive parts. Some versions of them might be better. I like the adjustable Zener diode, the TL-431 as the basis for such a circuit; the zener knee is text book perfect whereas those big stud mounted zeners, around 5V have a poor zener avalanche knee.

Back to the series diode method, it is very simple, and rugged. The big bypass trick you mention is a good idea, but there is a downside. When you key the amplifier, that bypass capacitor must charge up from zero volts, and for an instant, the bias is zero. As a result the cathode current will surge, perhaps as long as a second, all depending on the circuit conditions.
A remedy for this, is an external bias source to trickle charge the capacitor, and could have a series diode to make sure when keyed up, there is no tube bias feeding some other circuit in reverse.

Jim,
Wd5JKO

Thanks for the tips, Shane and Jim.

Just a couple of questions:

1- Shane, what value electrolytic cap would you recommend?
2- By dynamic regulation, do you mean that it is supposed to integrate the voltage drop variations (0.9-0.6)*(number of diodes) to some average DC value?
As an example- let’s say I want +7.5 volts on the cathode:
10 rectifiers can give anything from +6 to +9 volts, depending on the current and the temperature of the diodes.

Below is a graph showing both current and temperature variations.
Given this variation, does the cap need to be 100uF, 1000uF, what?
Seems like an ugly way to generate bias…

Jim, Your point about the delay caused by charging this large cap through a string of “high” resistance diodes sounds scary for say- a 4- 1000a with screen voltage applied….
If I understand your suggestion, you suggest putting a positive DC voltage on series with a diode and resistor to feed the cathode to support a minimum sustained bias in the large cap. The diode would back bias during operation then…

All this makes this method look ugly, though ugly may be acceptable. Comments?

I find the TL431 idea interesting… using a precision regulator to stabilize the bias. It obviously needs a low impedance pass transistor or FET. Anybody have a suggested circuit.
73, Mike

I link to the Triode board PDF download. Also see the image of the TL-431 circuit.

They use a TL-431 with a PNP power transistor to set the PA cathode bias anywhere between 3-27V. They have a Mod to change that range to 27-45v by adding a fixed zener diode.

The PDF discusses the IMD impact of bias variation with SSB transmission. With AM linear operation, the cathode current remains relatively constant, so this is of less concern on AM.

http://www.f1uvn.com/f1cxx/gs35b/triode-manual.pdf

Jim
Wd5JKO

Hi Jim,
Thanks- that is a great resource. I will study the manual.
I wonder how far one might go to increase the range of the citcuit?
For this 3-500z, the 3-27 volt range will suffice for my purposes, but my plans for a 4x1Kx2 amp will need to go up to about 120 volts or more, since I will be putting a regulated 400-450volt screen voltage on the tubes.
For now, the original range looks fine.
73, Mike

I've been using the TL431 circuit for well over a decade as well.  Great circuit.

Make sure you bypass the heck out of the input and output of the variable supply.

To increase voltage you just add a series zener.  The only downside to the TL431 is you only get xxx amount of volts of swing.  You can increase the bias level 100 volts with a 100 volt zener, but you will still only have 27 or so volts of adjustable bias.

Works, works well, just remember that one caveat

--Shane
WP2ASS / ex KD6VXI

Hi Shane,
Thank you. Builds confidence to know that the circuit has worked well for so long.
It seems to me that the small range is not s problem as long as one may place a reliable two pole rotary switch in the circuit to allow the switching of several different Zeners in as needed. Not necessary for the 3-500z, but nice for 2x 4k amp.
A make before break switch would prevent anything floating or one might just use some small dpdt relays with proper sequencing to assure make before break occurs.
How about that idea?
Of course, the TIP147 would not do for over 100volts, but a 250volt pnp like a MJH11021 would work…
Are there any P-Channel FETs that might be adapted? The offset would be higher, but have not looked at that… probably the PNP BPJT would be easier as long as it handles voltage.
How about varisters? Need to consider higher values for 100+ volts- right? Would need another wafer on the switch it that was added, though one does not need a universal circuit, just custom designs for each tube type profile…
Ideas?? 73, Mike

I made sure I put a 100k resistor across the whole shebang. Then you don't have a floating cathode even when switching between bias ranges.

I don't have the part number here (actual parts in California in storage) but I do know that they make devices rated to hundreds of volts.  I used an LR8 and pass transistor to make voltage regulators for screen supplies and grid using them.

Many ways to skin this cat.

--Shane
WP2ASS / ex KD6VXI

Thanks Shane.
I like the LR8 for screen regulator. I found this circuit attached below.
I would use a 900volt FET in place of the 400volt one they have in there.
How does it look to you?
A good heat sink essential, too. 73, Mike

In my experience, if you are using a pass regulator (like a typical Astron, etc) then you'll need to dump have the exoect d peak screen current in a resistor, depending on the tube.  Some tubes go negative on screen current actually dumping electrons and causing screen voltage to increase under some tuning conditions.

A shunt regulator eliminates that problem.

Keep this in mind.

I used a bipolar device for my shunt regulator out of a stereo.  Iirc they where good to 900 or 1200 volts.  I also had some mosfets I played with but wasn't happy with them and went back to bipolar.  For the life of me, I can't remember why though.


If your tube(s) won't go negative on the screen, a series regulator with a small amount of bleed R will be fine.  250B style tubes will need more bleed current.


--Shane
WP2ASS / ex KD6VXI

Thanks for the advice, Shane.
I actually have some high voltage NPN power transistors coming in.
They are ON FJP5027 high voltage devices in SO220 packs.
They are good for 1100v C-B and 800v C-E
Data sheet below:
The SOA graph is also there. I will be using about 300-350volts on my 4-1000A tubes. The DC graph shows about 40-50mA at 300 volts and maybe 30-40mA for 350 volts.
I figure that I can put 3-4 in parallel for my screen supply and ballast the emitters with (10ohm?) resistors to balance things and stabilize them. Three of them would give at least 100mA at close to 400 volts, I think, which is more than the 80-90mA I will see in Grounded Grid operation at about 4500 volts and about 350 screen volts.
Comments on this device?

Yes, I am very familiar with the 4CX250B, 4CX300A and 4CX350A tubes.
Screen currents would almost always go several mA negative in AB1 for SSB. I used 0xx cold cathode tubes as shunt regulators and they would brighten with modulation on a pair in AB1. I still have a box full of 0B3, OA2, etc…
Nice blue glow like 866 rectifiers. I do not use them- opting for the souless but compact silicon stacks….😉
Have you parallel bipolar pass transistors- like this with emitter ballast for higher currents?
A pair of 4-1000A tubes will not go negative, I think, but if I decided to push them, the screens could pull 0ver 200mA, though that is not my intent, beimg focused exclusively on low IMD, not high output.
73, Mike

Hi Tom,
Since you mentioned Adaptive Predistortion in this discussion, I have casually tried to find a way to add adaptive Predistortion to my Flex 5000 transceiver.
I found some comments in forums that there were operators who had adapted it to the Flex5000a with excellent results.
I can find no real application information or interfaces that can sample the amplifier output and a board that feeds the signal back into the transceiver and software to make instantaneous changes to correct non-linearities to improve 3rd order IMD of the amplifier output.

I am proceeding with the construction of my single hole 3-500z and my 2x4K or 2x3K amplifier.
The single hole design is complete and I am ready to mount components.
I will be doing everything possible to improve the inherent IMD of the single hole amp, but the ceiling is in the high -30s for this amp and that is better than my transceiver.
I could lay out $4500+ for an ANAN with pure audio built in to push it up to -60 or so, but the question is- can even these pricey radios accept sampler input from a linear amplifier and where can one buy or build that interface?
I wish that there was more information available in a how to guide, but I cannot find it.
Can anyone give me some clues on this?
Thanks in advance for any advice?

Hi Mike,

I recently did some reading trying to figger out if the older Flex radios will do predistortion.  (You mentioned that some clever guys have figgered it out) Otherwise, the Anan line (and associated kits)  are the only ones with the software, and hardware available for hams.  The Anan 200 seems to be a great used buy thes days, if you can find one.

The HPSDR "kit" radios will do it too, (HPSDR, SmartSDR) being closely associated with Anan's technology.

However, I would contact Rob/ W1AEX to get his latest opinion.  Probably the cheapest method would be to get a Hermes, (or latest equivalent) wind your own toroidal RF sampler for your 4X1 amp and feed it into the Hermes predistortion set up. Any linear amp will work.  There are likely pre-made samplers available.

I may be doing something similar in the future too. Predistortion is certainly the path to take for the cleanest signal.  

BTW, the dual GG 3-1000Z amp driven by the single GG 3-500Z sounds like your best bet for old school technology.



Flex's positions on predistortion:

Old:
https://community.flexradio.com/discussion/6512026/official-position-on-adaptive-predistortion

Newer:
https://www.systems4silicon.com/ip-cores/digital-predistortion


Does it mean only the latest Flex rigs apply?


Since you're doing the legwork, let me know if you find a complete plan for using a Flex 5000 to do the job...  


** And here is an excellent post about setting up a complete station using a Flex 6700:
http://www.nn4zz.com/FLEX6700.htm


T

Many thanks, Tom.
I found a number of articles for build your own couplers all based in an article by W5QN.
They look simple to build.
That is the easy part.
I have also looked ar the Hermes 14bit board and it is attractive and cheap. It has inherently low 3rd order products at -53dB.
The only rub is the 500mW output and it should be run at half that.
This might be a problem without the Adaptive Predistortion system in place, but given that- an amp that is not too expemsive could be used to bring it up to a constant 25-30 watts carrier (100 watt amp) and then that could be used to drive my single hole 3-500z to 150watts carrier and then the big jugs.
I agree that the 3-1000z X 2 amp is the attractive option and I now have 4 “good” ones that cosmetically look new.  I wanted to get a set and some spares before completing the build.
I will-however- make plenty of headroom and elbow room and air flow for changing to the 2x 4K setup if needed. I have 6 good 4xKs to work with and if Adaptibe Predistortion works, those tetrodes run in AB1 with 350 screen voltage and appropriate cathode bias to clean them up.
I will likely run the 3-1000z x2 at about 4kv but my variac can go up to 4500 if I run the 4XKs.

I will certainly share anything I discover regarding the adaption of my FLEX radios.

By the way, I noticed thatyou were back on the audio improvement track.
I will send you some info I got from working with Jim Tonne in 2021 that you may find useful. He focused on antialiasing in these sharp multistage filters and has some nice analog answers using discreet op amps that are configured for easy cutoff changing based on changing only the resistor values. Will semd that along you to look at.
Thanks again for your counsel.
73, Mike



https://www.collinsradio.org/wp-content/uploads/2015/05/Build-a-Quality-RF-Power-Sampler-Jackson.pdf

http://www.n4ga.com/50db-hf-rf-sampler/

https://www.qsl.net/on7eq/projects/rf_sampler.htm

http://www.lyonscomputer.com.au/Test-Equipment/50-Ohm-HF-RF-Sampler/50-Ohm-HF-RF-Sampler.html

Very good on all, Mike. Sounds like  a good plan.  Starting with a clean amplifier and then adding predistortion will only improve things more, vs: starting with a rat to clean up... ;D.  

As far as audio improvements, Rick/ W8KHK rebuilt my old Switched-Capacitor Filter (SCF) based around two Maxim MAX295 8th-order Butterworth filter ICs in cascade.  I was using the homebrew L/C analog filter but didn't feel it was sharp enuff for the big 4X1 plate modulated rig. (Fabio 2) Fabio is pretty clean for IMD as is.   But the audio skirts rolled off at 5-6 KHz, but still had some slight fuzz out to 7-8 KHz.   The Maxin 295 chips rolled the audio off much sharper. It is also variable, so has an advantage to adjust on the fly.  

Rick plans to make these filters available in the Max audio Processor or as a stand-alone filter wired / or as a kit.  I think the filter alone could be a very valuable asset in the shack to tame big rigs that are inherently clean, but need some additional skirt management.  For example, using the MAXIM filter, the spectrum analyzer pics I did of the 4X1 rig look almost like predistortion skirts.

I think Rick's processor and filter will become popular once guys see what it can do.

T

Pictured below are three waterfall bandwidth snapshots of about +- 4.5 KHz,  +- 6.5KHz  and  +- 7.5KHz  audio using two Maxim 295 chips in cascade.   The 4X1 modulated by 4X1s rig is flat out to about +- 12 KHz, but is reigned in by the filter.  The filter is placed just before the limiter near the end of the audio chain.  The top line-waveform shows the +- 7.5KHz response using a real voice into the mic.