A NEW KW TUBE Transmitter AM PDM BUILD - the beginning...

[K1JJ]

Well, it's time to end the old thread about building a pissweaker 25w rig.  The end result was a pair of 813s series modulating a single 813. Works great and all finished.
http://amfone.net/Amforum/index.php?topic=46165.0

Time to move on and build something else...


After thinking and talking it over with a few guys, I decided to build a big AM PDM TUBE rig. It is a challenge that scares me a little, so this is good.  I was toying with the idea of doing something like Rick/KHK's big series modulated rig, but I already have one now, though small, but wanted to swing the other way for a high efficiency rig.

Haven't decided on the tube type yet, but it will be at least a pair of 3-500Zs PDM'd by a pair of 4-400As.

Check out the first steps in the pics below.  Shown is the PDM generator, an old 2006 QIX board I have used for both MOSFET and PDM tube rigs in the past. It didn't work when I first fired it up, but after a few hours I have it putting out a nice stable squarewave with good modulation from 1 Hz to about 10 KHz where the artifact LP R/C  filter rolls it off. It also has a very nice low level negative peak limiter (actually clipper) which I feel is one the the best implementations out there. It limits those negative peaks well as an insurance policy. Just a touch at -95% is a good setting.. I use a 10-turn pot to set it.

I added a pot so I can adjust the switching freq from 30 KHz to over 300 KHz.  But 70 KHz is a popular switching freq used by the tube BC PDM pros, so that's where I set it as shown on the freq counter.  

Also, an important part of the rig is the big integration PDM output filter using L-C-L-C-L-C.   The inductors pictured are homemade; wound on 5 gallon plastic cans - and have worked with my old 4-1000A rig PDM'd by a 4X1. They are about 34 mH and the caps are around .001 @ 10 KV.  The filter values will have to be tweaked based on the final plate impedance.

Just cleaned up a very strapping 35" high X  19" wide X 31" deep cabinet and sprayed it Baby Blue. It will have a side window viewing and easy ACCESS of about 2' X 2' using Plexiglas OR sheet aluminum, for testing or permanent use.  It will have three levels inside, one for the PDM deck and one for the PDM filter and PS, fil xfmrs, etc. and the RF deck on top.  At least that's the plan for now.  For safety and stability, I am working hard to fit EVERYTHING into this one cabinet.  This and the tube choice is my next design challenge.


The finals will be in regular class C, maybe 80% efficiency at best.  The PDM tube modulator will probably hit 88% efficiency or even better, based on experience.  I have a pre-driver chip in the PDM generator putting out 10 V p-p with good current. A 11N90 MOSFET or two in the cathode of the PDM modulator tubes will drive it well.  No need for the GFZ driver board this time and no audio transformers.

Have built a few of these rigs in the past using quad 4D32s X 6LF6s, 4-1000As and other big tubes. It is an advanced project and can be dangerous due to the floating supplies, meters, etc.  But the results are really amazing when you key it up and wonder where all the heat has gone...  It will generate audio "shark fins"  on the scope just like the big MOSFET class E rigs. I would compare it very closely to my recently built series modulator rig's performance...  FB triangles and square waves.

In no rush, so it will be several months before completion I would guess.

A journey starts with the first step -  and here we go.

T

Steve did a nice design job on this PDM kit board.  (The 10 turn pot on the front is for precise negative peak limiting adjustments on the fly.)

[K1JJ]

More:

The homebrew filter coils are really big... wound on 5 gallon plastic paint cans.  I tried ferrite cores on the 4D32 rig, but these air coils were superior.  Figure the coils take the place of the mod iron hassle, more or less.

The cabinet weighs about 90 pounds, on casters. It was a bulky bitch getting up on the HV supply enclosure.  It needs to be re-sprayed in areas I see.


The tubes... three 3-500zs PDM'd by two 4-400As?  Still thinking about it.   A bigger lineup wud require a new HV supply.

[K1JJ]

More:

Running out of work space.  But there is a rolling table (out of view) that I use for the main chassis work.


There's the original PissWeaker's certificate from 1996.  During the first net call up in March, 1996, there were 25 check-ins - all having proudly built a homebrew 10 watt or less AM transmitter. Many of the guys went on to build their first big rigs.

[KD6VXI]

So, how many millihenry are those   And what do you use for caps in the low pass?

You get me all good and ready to build a series mod rig and then this.

Though I'm leaning a single 4-1000 pdm a trio of 500Z tubes (if I can find them in the shop, I think I have 5 NOS left).

So many tubes, so little time.  BUT, I'm interested in this.

--Shane
KD6VXI

[K1JJ]

So, how many millihenry are those  And what do you use for caps in the low pass?

You get me all good and ready to build a series mod rig and then this.

Though I'm leaning a single 4-1000 pdm a trio of 500Z tubes (if I can find them in the shop, I think I have 5 NOS left).

So many tubes, so little time.  BUT, I'm interested in this.

--Shane
KD6VXI

Hola Shane,

Yes, it wud be nice to have some company on this project. Hope you build one.  Well, at least a series modulated rig can be easily converted over to a PDM rig. They are so close in layout aside from the switching part.  So you didn't waste too much time I hope...

A tube PDM rig is really a thrilling rig to run.  I remember one day TimTron was over here and I was running a big PDM AM rig at the time. He looked longingly while it was transmitting with the 1000 yard stare in his eyes. He said he liked it and was impressed how much power it put out with so little heat. This was back in the early 2000s I think.  There's just no big homebrew AM PDM rigs on the air... well there's one I know of using 4-400As.

Yep, a trio of 3-500Z class C triodes to eliminate the floating screen supply would be cool  PDM'd  by a 4X1.   I am thinking about that right now.   The 3-500Z is well suited for grid current and class C according to the datasheets.

The filter coils are between 25 to 34 mH. The filter values were based on a design by Frank/ GFZ years ago. But first figger out the plate impedance.  The caps I used were those big micas (Santago?) we talked about in a previous thread.    .001 at 15-20KV is a working value, but again, the filter design comes last.

Don't forget the high speed damper diodes.   Also, an L/C parallel  trap on the filter output to attenuate the PDM switching freq down below ~ -70 DB.  These sidebands can appear at +- 70 KHz up and down the band until the trap is tuned and the last remnants of switching is sucked out.

And, unless you go thru some elaborate work to make the negative peak crossover analog, expect to dial the internal NPL up to limit negative peaks to -95% or so.  But it will come in time. Be patient, refine -  and the rig will fly very well eventually.

I'll draw up a schematic in the near future and post it here.

T

[K1JJ]

BTW, Shane - I wonder what you are using for a 3-500Z fil xfmr?  That's about 44 amps at 5 volts CT with 10 KV insulation.  I wud have to find one or probably wind one.  The 4X1 modulator can use a standard 7.5V @ 21 amp CT fil xfmr.

Also, minimum noise is important.  I remember with Steve's PDM 450TL rig he used a single 4X1 PDM tube and just blew some air past the pins.  The 450TLs needed little air so the overall rig was quiet.

I have also looked at using two or three 833As in the final since they need no forced air.   The bottom line is because there is not a lot of heat, the noise can be reduced with the right tubes.

For safety and stability, I gotta fit EVERYTHING into one 35" high  X   19" wide  X   31" deep cabinet. This is the next design challenge as well as choosing the tubes.
 (Including fitting in the PDM filter and big coils)    

T

[KD6VXI]

Well, this will be one of those things I follow and then get to start in a month or two.  Talks are progressing on a new job, so I may be relocating.  Hopefully with a full gatage / radio shop this time!

My existing pwm setup has a npl built in, pretty much copied from Steve's design.  That then runs into a POS peak stretcher I demo'ed here that allows up to 420ish pct pos peaks, continuously variable from symmetrical to 4x non symmetrical.  Of course anything over 120 to 130 doesn't sound good on most radios, but it's cool wo watch a 1 Watt carrier modulate ip to 16 watts without any baseline!

For filament xformers, the CBers have them in droves.  Dahl stocked on the shelves up to 90A xformers for 6 tube amps. Since a lot of the CB ops are going to ceramic now, these are sometimes parted out. I've also used a pair of sb220 xformers on a previous 3 tube 500Z amp. Just tie everything together in phase and your good to go. As an interesting aside, that 3 tube amp was run off an inverter in the car originally.  With an inverter I put, a si gle sb220 fils xformer was enough to light the 3 tubes.  And did so for maybe 15 plus years???

Switching supplies are also used by some guys.  You have to pick the correct ones, though.  They have to be able to ramp up voltage and not just instantly shut down due to looking at the near shorted cold fils.  I've not used them, but others have.  VE7RF might be able to shed more light on that if he pops into the thread.

Gotta run, time for a job walk. 

--Shane
KD6VXI

[KF7WWW]

I like to use plasma therm filament transformers meant to run a single 5cx1500. They light a trio of 3-500’s well.

[K1JJ]

I like to use plasma therm filament transformers meant to run a single 5cx1500. They light a trio of 3-500’s well.

Where are these sold?   And Shane, same question.

You both realize that the RF final (3-500Zs) will need a filament xfmr that is insulated for 10 KV between its AC primary, secondary and frame, right?     The 4X1 PDM modulator can be a regular fil xfmr, however.

[KF7WWW]

You might want to get with W7TFO. I helped him kill a Harris pwm rig a few years ago.
The plasma therm stuff came from buying rf gens at the right price and parting them out. Yet you are right. I wouldn’t trust 10kv on those.

[K1JJ]

Yep, I already sent Dennis an email earlier today thinking he may have something.

It's not too hard to wind your own HV insulated fil xfmr. Just get a Variac core and strip off the parts. Remaining is a core with a 120VAC winding.  Just wrap Kapton on the core or use heavy HV wire to make the 30 turns or whatever to make a CT fil winding.  I have one wound for 7.5V that I could take off some turns to get 5V.  The 44 amps is quite a lot when using three 3-500Zs so will take some heavy wire.   The more I think about it, three 3-500Zs in class C neutralized will make a nice RF final.  The 3-500Z is small and only moderate voltage is needed and it will put out the soup.  I am also thinking about two 4-400As as the modulator rather than the 4X1 so to stay with the general voltage class of 3-500Z vs: 4-400As.

Frank/GFZ suggested I could wind the PDM coils as layered with 4", 3", 2", 1"  PVC forms and supported in place inside each other with threaded rod on each end.  This makes a compact coil. I will do this if I cannot find a commercial BC filter.  The Q is higher with wider air spacing compared to a coil layer-wound with close tape spacing.

He also suggested the placement of components in the cabinet:  Three levels - the PDM modulator at the bottom and grounded, the PDM filter in the middle and the RF final at the top.  The fil,  screen, HV iron and general infrastructure can be  where it is needed with shortest leads.   The bottom will be sparse where the HV xfmr could sit in the rear.  I like it... safe and compact all in one cabinet.


BTW, here's an older article by Steve/ QIX about PDM and AM transmitters. It's a good refresher that I read every few years:

http://amfone.net/21stAM/pdm_article.html


T

[KD6VXI]

I had forgotten about the insulated windings for fils.

I do have an amplifier here that does have the necessary filament xformer but I don't want to take anything from it as it's a one off 4-1000 a guy built in the 60s.  There was a magazine article written about it and he also showed it at a convention of some sorts.  It is believed he was trying to take some Collins HV business, he was already firmly entrenched in the FAA as an xmitter / amplifier supplier.

Hammond still winds Dahl stuff.  I'll have to look at my Dahl catalog and see if there are any with the required insulation  and then look for those.

Facebook is where most of those guys hang, Tom.  There are also Cb only forums, but you'll puke if you go read some of their crap 🤔😂

--Shane
KD6VXI

[K1JJ]

Shane and Pat:

What are you guys using for blowers that are relatively quiet but effective?  I thought I saw someone post a Rotron? or maybe something else that was about $100 new.  I wanted something that could be reduced in speed with an AC  Variac or even DC.  I think VE7RF or W1TTT posted a link to one about a month ago, but cannot find it.

T

[KD6VXI]

Shane and Pat:

What are you guys using for blowers that are relatively quiet but effective?  I thought I saw someone post a Rotron? or maybe something else that was about $100 new.  I wanted something that could be reduced in speed with an AC  Variac or even DC.  I think VE7RF or W1TTT posted a link to one about a month ago, but cannot find it.

T

EBM Pabst

That is the reference quiet blower company.

Or find one in the spec book that provides what you need with the lowest wheel rpm.

I did that with the GS35b amp I'm shipping out to Hawaii today and it's not silent but it's pretty damn quiet.  Quieter than the AL1500 on the Ben h now.


--Shane
KD6VXI

[KF7WWW]

I second Rotron, ebm, pabst.
Dayton make some full ball bearing blowers that aren’t bad but don’t seem to be as smooth as the others.
I source mine through eBay most of the time.



Sam

[w8khk]

As requested (in the prior thread linked to this one), here is a summary of my plans for quietly cooling a 3CX2500F3 series modulator tube....

My goal is to design and construct a cooling chamber that reduces the noise to a low enough level that the rig can be near the operating desk and microphone, without compromising the quality of modulation.

The 3CX2500F3 requires a good deal of air flow, and it has been said that the tube radiator fin assembly can cause some quite loud squealing noises if the airflow is sufficient during high dissipation.  It appears necessary to address three different causes of cooling noise - first, the noise caused by the fin assembly as air flows through it, second, the flow of air moving into and out of the the transmitter enclosure, and third, the motor and squirrel cage vibration telegraphed by its mounting to the chassis and cabinet.

In order to provide the required airflow, without exceeding that minimum, I plan to use a variable-speed blower with a DC motor.   My experience with AC motors has not been satisfactory with regard to speed control by a Variac.  In most cases, induction motors, that synchronize with the power line frequency, run warm to hot when slowed down with a Variac, and this is, of course, to be expected.  I chose to try an automotive heat/air conditioner blower unit designed for use with 12 to 15 volts at maximum speed, which can be speed controlled by reducing the voltage either by series resistance, regulator, or PWM power source. This option will allow either manual or automated speed control, in response to heat cycles between receive and transmit periods.
  
The series modulator tube will be dissipating around one kilowatt during periods of unmodulated carrier.  There is no need to run the blower at full speed during receive periods, however it should provide substantial airflow for a period of time after each transmission before slowing down to maintain a safe seal temperature while continuing to remove heat dissipated by the tube filament.

In order to minimize telegraphed vibration, the plan is to take the motor and squirrel cage assembly, and create an enclosure and air flow plenum by cutting and laminating several sheets of one-inch thick styrofoam.  This assembly will then be enclosed in two-inch thick cellular foam rubber, before mounting in the rack cabinet.  

To reduce the intake and exhaust airflow noise, a labyrinth airflow path is planned.  The entire transmitter will be enclosed in a six-foot high rack cabinet, 22 inches wide, and 33 inches deep.  The floor of the cabinet is 80% open, and the cabinet frame is made up of cast and extruded aluminum sections, with removable sheet aluminum skins for the sides.  A sheet of 3/4 inch thick plywood will be used as a base for the plate transformer and other HV power supply components.  Around these components, 3/4 inch holes will be drilled to allow air to enter from beneath the plywood, through a pleated air conditioner filter.  By providing many moderate sized holes around the components, a large volume of air may enter, flowing at a lower speed than the intake or exhaust ports of the blower assembly.
 
The sides of the cabinet, and the front and rear, where possible, will also be lined with two-inch thick cellular foam rubber sheets.
 
Air will flow into the enclosed blower assembly, and then on to a plenum enclosing the 3CX2500F3 modulator, and the 4-400A driver tube.  The exhaust side of the plenum will also be perforated with 1/2 inch to 3/4 inch holes over a large area.  
Two or three layers of fiberglass sheet insulation will be placed above the tube plenum, in a manner creating a labyrinth path whereby the air flow much change direction several times, thus muffling the noise from the motion of the air, as well as the higher frequency noise caused by the tube heat radiator fin assembly.  The airflow will be sufficient that the cellular foam rubber and styrofoam are not exposed to extreme temperatures.

The final amplifier, consisting of push-pull Eimac 304-TL tubes, will reside above the foam labyrinth.  Air will flow around the chassis sides and rear, up to the meter section of the rack.  Behind the meters, and under the top of the rack will reside the high voltage bleeder resistors.  The top of the enclosure has a sheet aluminum skin, which is perforated along each side, from front to rear, allowing sufficient air to escape.  

The intake air filter design, as well as the top cover perforations, were employed by the cooling system when this rack was occupied by six card cages of computer logic, consuming energy from a 5 volt supply at over 200 amperes.  In operation, the computer was extremely quiet due to the air flow design and several small fan assemblies, so it is expected that similar noise reduction can be achieved with this platform.

Further development of this series modulated transmitter will not proceed until the noise abatement development indicates the entire rig will be acceptable residing near the operating desk.

[W1ITT]

Rick...   You mentioned that the bottom of the rack will be about 80% open for air circulation.  If you have or anticipate having small critters like kittens or puppies around, it would be good to use some sort of mesh to exclude them.  I had a 4-1000A in a rack on casters years ago, with a hole in the bottom.  I was in the radio room and could hear my Maine coon kitten, Bullet, making a racket and occasionally meowing.  It took a while for me to figure it out, but he had weaseled his way under the rack, up through the hole and was exploring the circuitry.  Luckily, the power supply and breaker was in a separate enclosure and his environs were cold, so he was safe.  After extracting him, I covered the hole with cane metal.
73 de Norm W1ITT

[w8khk]

Norm, thank you for the safety suggestion.  The 80 percent open area is covered by the plywood transformer support panel, with 3/4 inch holes, large enough for small critters.  (I was attempting to clarify that I did not need to cut the bottom of the enclosure to allow cooling air to enter from the bottom.)  Under this 80 percent open area is the pleated air conditioner filter, and under that is a sheet of perforated aluminum with the entire area covered with 1/4 inch holes, so I think I have rather safely "covered" the large bottom opening.

I certainly hope this project ends up sounding like a light warm summer breeze, and not a Katrina disaster.  Time will tell!

[K1JJ]

Rick:

Interesting on using 2" foam to cover the inside of the rig.  Something like insulation in a car.  I will digest your posting above and see what I can do.

I am still thinking and toying with the idea of making the big rig a screen modulated 4CX-3000 tetrode using the Continental 317B broadcast transmitter circuit of screen modulation.  If I do, I will need to have some quiet air on the tube and use some of your ideas.

Here is a "simplified" schematic of Continental's screen scheme. They used this a lot in their earlier rigs and claim the screen efficiency approached plate modulation once all the modulator tube losses,  fils, iron and everything else were accounted for. Not to mention the fidelity and less cost of screen mod. They are also using a high efficiency linear after the screen mod tube.

I may be using their OB2, 807 and replacing the 4-65s with a single 813 as the screen modulator feeding the 4X3.  Frank is also looking into designing a solid state screen modulator.

QUESTION:  Does anyone have the detailed schematic of this Continental 317B screen modulator?  This "simplified" version off Google does not show parts values or voltages. It wud be a big help.

I'll bet there are a number of guys who would be interested in building a big screen modulated rig once all the design (and even a PC board) is ironed out.

Thanks.

T

[w8khk]

The Continental 316B has circuitry very similar to what you are looking for:

https://www.steampoweredradio.com/pdf/continental%20electronics/Continental%20Electronics%20316B%20AM%20Transmitter%20Circa%201959.pdf

It is a big manual, you need to dig for the schematics, but they are there in full detail.

Three 4CX5000s modulated by a trio of 4-65As.  The OB2 and 807 bias regulation circuit is clearly defined in the detailed schematic in the above manual.  I believe the 813 would be a good substitute for the 4-65s, as it can tolerate a much higher screen voltage.  The 813 does have a somewhat lower maximum plate voltage rating, but we all know that limit is not cast in concrete!  The simplified circuit showed the 4-65A drivers "triode connected", but the detailed schematic of this transmitter indicates a standard tetrode configuration.

I also found the book for the Continental 317C:

https://worldradiohistory.com/Archive-Catalogs/Continental/Continental-317C-1-50-kw-AM-Transmitter-1975.pdf

This one uses a 4-400A class-A audio driver for the parallel 3CX3000A1 screen modulators.  The 4-400A input is all discrete transistor circuitry, almost exactly what I want to do with my series modulated 3CX2500F3.  The class-A 4-400A works into a 40K series plate load resistor at around 5 kV, capacitor coupled to the 3CX3000 screen modulator, working as a cathode follower, exactly what I envision doing to place the modulator between the positive plate supply and the final amplifier.

A wealth of good information can be gleaned by studying these schematics!

By the way, this thread has been "PROMOTED" from QSO all the way up to TECHNICAL FORUM!   WAHOO!

[KD6VXI]

Years ago, when I built my pwm, I bought some 2kv fets for it.  I had planned in using them to scream modulate a 4cx5k (and then a 4cx7500 someone dropped in my lap).  Never got around to it and moving has most of the stuff still in boxes.

**sigh**

--Shane
KD6VXI

[K8DI]

Regarding noise control --

You don't want to use "foam" to absorb sound, you want to use acoustical materials that are flame retardant, high temperature capable, and non-friable.  Stuff like rockwool panels are not expensive, won't burn, won't melt, won't out-gas, and so forth. Some other things:  a bunch of little holes are noisier than one big one. High velocity is noisier than low velocity. Sharp edges are noisier than rounded. Think smooth curves, radiused corners/edges, low speed, high volume, passing along absorptive materials.  Finally, sound will travel through the rack walls themselves wonderfully.  Consider lining it all first with a sound barrier type product (look up Dynamat), then put panels on the large areas to absorb sound.

My day job is pro audio; when I read stuff like building a box of styrofoam with a fan in it, my eyes can only see the aftermath of the Station fire in Rhode Island 17 years ago...  Please don't make a literal flamethrower out of your "flamethrower" QRO rigs.....

Ed

[w8khk]

Ed, thank you so much for the thoughtful constructive criticism.  I do have many rock wool panels that were used in a small recording studio, and they will be perfect for the enclosure and duct work.  I was not aware that the foam would magnify the acoustical noise, that is also very helpful information.  I was aware of the smooth curves vs sharp corners, and that is part of the plan.  High frequency sound does not like to find its way around corners, hence the labyrinth path.  Slow air flow a large volume is much quieter than fast air flow, using multiple parallel paths. 

[steve_qix]

Hi Tom, good project !!

Important question: Were any of your previous tube PWM transmitters capable of modulating a triangle wave, without distortion, to 99% negative (or even 97% negative), and 150% positive?

I ask this question because, as far as I can tell, there is no analog compensation in your design (you didn't mention it).  There is a very good reason why Harris and other manufacturers use analog compensation in their tube PWM transmitters - tubes are BAD switches - AND - the PWM filter cannot be fully emptied as the current approaches 0 (high negative modulation), causing the modulating waveform to "flatten out".

So, with analog compensation, the modulator is essentially operating in analog mode for the last 15% of modulation of so (from 85% to 100% negative).

-AND-, since the tube is a bad switch, analog compensation drives the tube harder as more current is required (positive modulation).  The design referenced elsewhere in this thread (my PWM article) has analog compensation, and it is very important if you want a low distortion transmitter that will actually modulate a triangle wave from 99% neg to 150% pos.   If the rig will modulate a triangle, it will modulate anything else, too !

Footnote:  Why isn't analog compensation part of the solid state rig designs (class E rigs)?  Two reasons.  1) We have the class E "gift".  MOSFETs, when driven into saturation at RF frequencies, will pull current with 0V across the device (0 volts on the drain), and will pull the DC drain voltage negative.  This is perfect for discharging the PWM filter completely at 100% negative modulation.  Tubes don't do this.    2) MOSFETs are nearly perfect switches, so no compensation is needed to linearize them.

If this is a new design, analog compensation should really be part of it.  It's not that hard to implement. 

Anyway, thought I'd mention it.

[K1JJ]

Ed: Very interesting on using acoustic, fire-proof soundproofing material.  I'll be looking into that.  This is why it pays to drag our feet and come up with several designs before proceeding.

Rick: The quietest BIG blower system I ever had was when I mounted a big 220V blower just outside the shack pulling in outside air. It was in a waterproof enclosure. Man, was it quiet in the shack, being just outlet air noise and no motor, bearings, etc.  The problem was the moisture and cold air in the winter that made things rough in the shack. But it was the quietest amplifier I ever ran, especially considering the amount of air it could quietly pass.  To eliminate the intake air noise, bearings and motor vibration, using fiberglass ducting with smooth insides worked well.

So maybe I can do it with a muffler type routing like you described.  And thanks for the Continental manuals. The full blown version screen modulator is more involved than I thought. Lots of power resistors and tubes..... 

Steve:  No, I never installed any analog compensation on the tube PDM rigs.  (4X1s  or quad 4D32s rigs)  I was about to but when I saw the complexity of the circuit you sent me, I held off. Or maybe it was something that Harris had designed or could it be I was cathode switching and you were grid switching the modulator?  I seem to remember the Harris version looking like a rat's nest.  But my un-compensated rig did pass nice triangles up to about 93% negative and I just masked the other 7% by setting the negative limiter tight. I thought the rig worked well up to 150%, but don't remember how far it would go without the comp.

But tell ya what:  If you have an analog comp circuit that has a good chance of working well, send it over and I will give it a try. I planned to cathode switch the modulator(s) with an 11N90 as before. I think you were talking grid driving it, so that may be an issue?

BTW, what do you think would be the best lineup of tubes in the 4-400A / 3-500Z class?   Would three 3-500Zs in the RF final PDM'd by a pair of 4-400As be best -  OR  three 4-400As in the final and two 3-500Zs in the modulator, or some other combination of these tubes?  Would a single 4-1000A as a modulator for three 3-500Zs be a good choice?    I will be starting at about 4.5 KV (low) at first and may have the option to bring it up to 7KV later on.

I'm still thinking of designs and talking to my "radio-lawyers" about the various choices for a big rig.  I like the tube PDM project and the possibility of a big screen modulated rig. Also, a big linear with heavy bias on AM could be a possibility.  I think my true heart lies with a tube PDM rig, being the most difficult, least blower noise  -  and most heat efficient for both winter and summer use.


T