Author Topic: Modulator... where is best?  (Read 1360 times)

0 Members and 1 Guest are viewing this topic.

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Modulator... where is best?
« on: April 08, 2014, 01:55:52 PM »
I got the idea for my homebrew rig after watching Brian's video on the TMC T-368 exciter. IIRC, he ran it into a 6146 buffer/amp, then onto the final. I promptly found a T-368 exciter & plan to use it & a 6146 to drive PP 250TH. ( Haven't worked out all the details yet ) Over on the TMC list, there is a discussion about running a TMC SSB rig in "AM" mode.
( which, their particular model wasn't really designed to do.)
Someone posted the following to the TMC mail list recently;

"... An old buddy of mine devised still another solution many years ago.  He added a small plate/screen modulator to the 6146 output stage of an SBE-3 exciter.  A small modulation transformer which he improvised was mounted inside on the rear apron of the exciter, and provided an 8 ohm input impedance for a small external audio amplifier...  In retrospect, he considers the mod a debauchery, but it worked great."

Modulating the exciter seems to me to be counter-intuitive. Inserting audio, at that point, with the 6146 used as a buffer, may be no problem.  Reading up on different modulating schemes today,  starting  with the ones from Brian's  old site, I see that low level grid modulation is possible. But, there's a trade off... efficiency and / or the ability to reach 100% modulation.
 I have an old audio amp capable of +/- 400 watts, my best guess, depending on how I run it... ( 8 X 211A PPP ) and a NOS, RCA modulation xfmr fit for the job of high level plate modulation. So the question is, modulate at a low level or high level? Pros & cons?
TNX, Mike 

BTW: I have separate B+ supplies for the audio and RF sections.

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #1 on: April 08, 2014, 01:56:06 PM »
I'll let someone else come in here and answer your questions, but there is one point that needs to be highlighted.

250th's are triodes and the control grids should be modulated in some way when used in the final RF amplifier. It's the same thing as modulating the screens with tetrodes. I'm assuming that's what your plan is and by modulating the 6146 you will achieve that goal? Many think that just because they have triodes with no screen grids that it eliminates the need for grid modulation which is not so. The amount of control grid modulation needed will vary with different tubes, but it is still needed regardless even when using triodes.

P.S. I would go with high-level plate mod with some added control grid mod combined. Control grid mod by itself (if that's really what you meant) is doable, but you have the RCA iron, so why not use it?

73’s

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Re: Modulator... where is best?
« Reply #2 on: April 08, 2014, 01:56:18 PM »
Brian:
I'm not seeing the need for grid modulation on PP triodes looking back at old 1940's & 50's handbooks.
They're all driving the primary of the mod xfmr with the secondary in series with the B+ to the final plates. This was my original idea. I was thinking of using shunt-fed plates to keep all that
current out of the mod xfmr secondary.... I only have one, and if I burn it up, I'm screwed.
Lots of fuses and meters.

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #3 on: April 08, 2014, 01:56:34 PM »
You are not allowed to view links. Register or Login
Brian:
I'm not seeing the need for grid modulation on PP triodes looking back at old 1940's & 50's handbooks.
They're all driving the primary of the mod xfmr with the secondary in series with the B+ to the final plates. This was my original idea. I was thinking of using shunt-fed plates to keep all that
current out of the mod xfmr secondary.... I only have one, and if I burn it up, I'm screwed.
Lots of fuses and meters.

You are correct, but the control grids did self modulate in those old circuits. They either had grid leak bias resistors which would automatically self modulate or they had a fixed grid supply that was not well regulated so it would move around a little during modulation. Those old handbooks didn't always go into all of the details. They often just only scratched the surface.

Using grid leak resistors usually works best and will help produce the highest positive modulation peaks, but you need a safety bias there in case the RF grid drive disappears. It can get a little tricky though.

You could also use a stiffer fixed grid supply then modulate the RF driver tube which has been done before and I thought that's where you were originally going with this until I realized you meant a grid modulated rig only without any plate modulation.

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Re: Modulator... where is best?
« Reply #4 on: April 08, 2014, 01:56:47 PM »
Yes, looking at Terman (1947) it is quite clear about the bias, etc. What I didn't understand about the low level modulation being introduced at the buffer stage is...  they obviously could not reach 100% mod with that scheme. I have only seen modulators run at high level plate mod, so introducing  it at a lower level at the driver / buffer didn't seem the best way to do it.  I have several HB rigs, and I will be building my own, so maybe I'll try it both ways and see what I get... :confused:
Thanks, Mike

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #5 on: April 08, 2014, 01:57:01 PM »
You are not allowed to view links. Register or Login
Yes, looking at Terman (1947) it is quite clear about the bias, etc. What I didn't understand about the low level modulation being introduced at the buffer stage is...  they obviously could not reach 100% mod with that scheme. I have only seen modulators run at high level plate mod, so introducing  it at a lower level at the driver / buffer didn't seem the best way to do it.  I have several HB rigs, and I will be building my own, so maybe I'll try it both ways and see what I get... :confused:
Thanks, Mike

Yeah that's the thing, most of the old handbooks didn't really elaborate on the fine details and if you built it just like they had it in the drawings you were often lucky to get about 90% max modulation. Those circuits never considered transformer losses hams would face in the real world and other details like getting optimum tube modulation using correct biasing techniques and they never came out and said that you should always over-build the modulator to compensate for a lot of it. At the time most hams probably didn't really care and the circuits worked good enough, but many of us today want it to be better if not perfect :ugeek:

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #6 on: April 08, 2014, 01:57:21 PM »
Since no one else has come in here so far and made any comment I guess I will.

If it were me and I had that 1:1 RCA transformer I would not even waste time messing around with low level grid modulation. To me that would be pointless. Most people only do that when they don't have any iron and if your transformer is in fact new and never used then it shouldn't be a problem. The only thing I can think of is the varnish between the laminations may be a little dried up due to its age after sitting around for 60 years or more, but if you don't run it with more than a +2000Vdc plate supply it should work fine and never die. Find a pair of 810's for class B push-pull to modulate the 250th's in pull-pull for that transformers 1:1 ratio because that’s what it was designed for. Just run the center-taps on both sides of the transformer with +2000Vdc. Then you only need a -50V grid supply for the 810's.

As far as shunt-feeding the transformer is concerned that would be a bit difficult. Shunt-feeding is not really designed for push-pull RF amps. Not saying you couldn't make it work, but it would require two big matched chokes in series instead of one. You would have to feed them with the DC in the middle where they connect together. You would have a problem with the ratio if you tried to use one shunt choke with the 250th’s in parallel and would probably need to increase the plate voltage way up on the 810's to even make it work. That would cause other problems.

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Re: Modulator... where is best?
« Reply #7 on: April 08, 2014, 01:57:33 PM »
Brian:
Thanks for your input. It has caused me to dust off some old books and wipe away some of the cobwebs in my old brain. :eek:

Decisions, decisions... I was trying to pull the old  trick of using what I had laying around. The Norton audio amp has the advantage of plenty of big iron in the power supply and was originally built to run class "A". It is in it's own rack cabinet and it's complete with 60A filament supply & HV, separate driver chassis ( 2 X EL34 ) etc... The big thing I need to worry about is getting the impedance right on both sides of the RCA xfmr. With eight audio output tubes, I could do some swapping around and most likely come pretty close. (  single ended 211s like a 10K load ) The HV supply for the RF final has, at it's core, a 3750V 1A xfmr which, after rectification & using a swinging choke, will get me in the ballpark for HV. I can use a variac on the primary of the HV xfmr to give me a little flexibility. If I can't get the Norton amp to fit this scheme, then the 810's would be the way to go.
( I now have a pair)
That comment over on the TMC list, about modulating the 6146, just threw me off track...
TNX, Mike

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #8 on: April 08, 2014, 01:57:48 PM »
The problem you are going to have with that audio amp is the ratio, mainly the modulator's voltage. That RCA transformer is setup for a 1:1 ratio which means you need about the same plate voltage on each side of the transformer. Even if the 211's or EL34's produce enough power their plate voltage is too low to modulate the 250TH's in push-pull on the RF side unless you run the 250th's at the same lower plate voltage. Then you probably wouldn't have hardly any RF power out of them.

What I'm trying to say is think in terms of voltage and not power. A pair of 810's or another pair of 250TH's on the modulator side will work best.

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Re: Modulator... where is best?
« Reply #9 on: April 08, 2014, 01:58:01 PM »
The 211 is rated at 1250V. The 250TH will be running 2KV or more...
the 810 is a better fit. ( I may have an extra pair of 250TH also )

I think I'm finally getting the picture... If I have 2KV on the final (secondary) , by modulating the  primary, sitting @1KV, it just won't "swing" enough, even though enough power may be available. I have been thinking in terms of driving power in watts and completely missed what should have been obvious. ::)

Well, that just bums me out. I thought that big old Norton amp could be put back in service.
I guess a future project would be to design an RF final to match the Norton "modulator".   

Today, I start on the "Radio Handbook" (10th ed) by Editors & Engineers  1946
TNX, MD

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #10 on: April 08, 2014, 01:58:15 PM »
Yep, its the peak-to-peak voltage swing on the primary side that needs to be at least as high or greater than what's on the RF secondary side. Push-pull to push pull, a 1:1 ratio.

Another thing is don't use the transformer with the 250th's in parallel. In push-pull only one tube conducts at a time so the current (duty cycle) is less. You will still see the full current on the meter, but if the tubes are in parallel it might be too much for the secondary winding to handle. That's why the secondary is rated lower via the number on the label. Push-pull is much easier on it.

To be honest you really have the perfect transformer. 1:1 transformers are great because the impedance on both sides is not as critical and doesn't need to be exact. With the typical push-push high-Z step down type to a single tube or tubes in parallel like most people have the impedance match is much more critical and if it’s off by the slightest bit it does make a difference. It's because you’re taking a high-Z output down to a much lower-Z load. With push-pull to push-pull its still pretty much a high-Z output connected to a high-Z load.

Push-pull RF amps were always better anyway, but when the military dumped all of the unbalanced coax on the surplus market after WWII it changed everything and they slowly went away as the pi-network took over. It’s a shame too because push-pull amps solved most of the matching/coupling problems with the modulator.

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #11 on: April 08, 2014, 01:58:30 PM »
There are times where it is practical to grid modulate RF tubes like when you don't have or want to find a modulation transformer. I'm planing to build a compact transmitter just for the higher HF bands like from 20 or 17-meters up to 10-meter. It depends on whether I can make my big rig work on 20-meters properly or not. It currently loads on 20, but it needs some more tweaking to make it fully efficient. It might even work on 17-meters if I do it right because the T-368 exciter does go up that high.

I have some rare 828 tubes that work well up high and can be grid modulated, but I'm thinking a single screen modulated 4D32 might work best. I could use a 5W or so 600V peak-to-peak screen modulator with a negative peak limiter on it to obtain really high positive modulation. At 600V on the plate it would give me 100W out and if I used the 828's the power would be less and I'd need around a 1250V plate transformer. The 4D32 is small too and it wouldn't take a very big cabinet. I could build the whole transmitter on one rack.

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #12 on: April 08, 2014, 01:59:06 PM »
As a matter of fact here's a really simple design using a 4D32 for an example which will produce a perfect modulation envelope. You would only need one plate transformer that yields +600V and one negative -100V bias supply. The 4D32's screen can handle a pretty high modulation voltage, but I think forcing a 600V peak-to-peak AC source like I mentioned in my last post would be overkill. The below circuit would be really easy to do and the neat thing is no modulation transformer would be required and it would sound just as good as any plate modulation rig.

Offline KM1H

  • Full Member
  • Posts: 39
Re: Modulator... where is best?
« Reply #13 on: April 14, 2014, 03:45:24 PM »
I agree about PP being a bit easier on the transformer especially if it is one of the economy "ham" models.
OTOH a BCB transformer is designed for 24/7 music which puts the maximum strain on it. With reasonable ham audio 200-4000 Hz or so and the typical locked mike with plenty of ahhs, ehs, and pauses of a typical transmission there should be no added strain with a parallel pair. My old parallel pair of 250TH's with 810 audio ran fine for years with a RCA transformer at around a 800W carrier and some modest processing.

Back in the late 50's when in HS, I had access to all sorts of free WW2 parts thanks to MARS and built a PP 250TH's with 810's using BC-610 parts. The plate and mod transformers never complained about the double load either and I knew little about impedance matching! BUT that was the last time I used plug in coils in an amp. A pi network is the only way to go if you like to move around the bands....and use coax feedlines.

Carl

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Re: Modulator... where is best?
« Reply #14 on: April 17, 2014, 06:40:21 AM »
You are not allowed to view links. Register or Login
....OTOH a BCB transformer is designed for 24/7 music which puts the maximum strain on it. With reasonable ham audio 200-4000 Hz or so and the typical locked mike with plenty of ahhs, ehs, and pauses of a typical transmission there should be no added strain with a parallel pair...  Carl

Back in 1963-64 I worked as an engineering intern at a 1KW daytime AM station. One of the experiences I had was replacing the modulation xfmr. The RCA xmitter was made post war, so it was probably 15 years old,+/-. Only running 8-12 hours per day, depending on the sunset, it seems a little premature that this mod iron would have opened up.  Can't remember what the modulators were off-hand.
The one I plan to use, see photo, is new-in-crate.

Offline KM1H

  • Full Member
  • Posts: 39
Re: Modulator... where is best?
« Reply #15 on: April 17, 2014, 11:24:57 AM »
Im not familiar with that transformer but with only 200 ma of DC allowed it certainly isnt a hi power one, most likely a 250W TX, with a pair of 813's and 813's audio which are RCA designed tubes and would meet the mod iron specs. That was the lineup in a RCA BTA-250M TX at WGBB where I worked part time weekends in the late 50's as the evening shift injuneer with a fresh First Class ticket.
The transformer part # is later than in the BTA-250L manual and older than the BTA-250M so is likely a replacement or earlier than the 250M manual Im looking at; that model was in use for decades.

Carl

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Re: Modulator... where is best?
« Reply #16 on: April 17, 2014, 12:01:57 PM »
RCA Type 901769-501, 500 watt Modulation Transformer
General

The primary is center-tapped for push-pull modulators. The tranformer has two secondaries. Secondary #2 will carry 80 mils to modulate screens of beam power tubes or screen grid tubes. It is equipped with adjustable protective spark gaps on both primary and secondary. Primary will match any class B tubes up to 10,000 ohms plate-to-plate such as 810's, 75T's, 805's, 8005's, ZB120's, 203's, HY51Z's, 211's, 813's, 828's, 203Z's, etc.


Maximum Audio Power: 550 watts

PS: Brian, this info seems to contradict the info you sent me on the old message board.

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #17 on: April 17, 2014, 01:05:51 PM »
The datasheet for your transformer is attached.

It's a 1:1 BC transformer designed for a full 1KW.

The label on the secondary is just misleading for someone who doesn't know what they are looking at because it's for push-pull AF to push-pull RF. If you do the math with the way it was designed to be used then divide it in half for push-pull it comes out to 200mA. The 198mA in your pic is just what RCA got during their setup with the tubes being used or with their test jig setup.

Post the other pic showing all the secondary terminals and the spark gap.

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Re: Modulator... where is best?
« Reply #18 on: April 17, 2014, 04:16:05 PM »
Brian: I hear you about the push-pull & ratings... makes sense to me. :icon_thumbup:

I got that info from the other AM list...

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #19 on: April 17, 2014, 04:54:09 PM »
You are not allowed to view links. Register or Login
I got that info from the other AM list...

Actually, that information in you last post was correct. The data sheet says the same thing so I'm not sure where the confusion is.

Offline KM1H

  • Full Member
  • Posts: 39
Re: Modulator... where is best?
« Reply #20 on: April 17, 2014, 07:00:09 PM »
That printed data is what Id expect and what Im used to reading and there is no argument it was for a KW TX.

Carl

Offline K4TQF

  • Full Member
  • Posts: 77
  • SWL 1959, FCC 3rd 1964, WA4ZYL 1967, FCC 1st phone
Re: Modulator... where is best?
« Reply #21 on: April 17, 2014, 07:17:11 PM »
Here are the full spec:
RCA Type 901769-501, 500 watt Modulation Transformer    General

The primary is center-tapped for push-pull modulators. The tranformer has two secondaries. Secondary #2 will carry 80 mils to modulate screens of beam power tubes or screen grid tubes. It is equipped with adjustable protective spark gaps on both primary and secondary. Primary will match any class B tubes up to 10,000 ohms plate-to-plate such as 810's, 75T's, 805's, 8005's, ZB120's, 203's, HY51Z's, 211's, 813's, 828's, 203Z's, etc.

Maximum Audio Power: 550 watts

Maximum Current

Secondary #1: 450 mA

Secondary #2: 80 mA

Turns Ratio

Pri to Sec #1: 1:1

Pri to Sec #2: 5:1

Pri to Sec #2 tap 25:1

Impedance Ratio

Pri to Sec #1: 1:1

Pri to Sec #2: 25:1

Pri to Sec #2 tap: 625:1

DC Resistance

Primary 135 Ohms

Sec. #1: 112 Ohms

Sec #2: 99 Ohms

Transformer Insulation

Primary: 8000V

Sec. #1: 11000 V

Sec.#2: 2000V to the rest of the coils and the core

Dimensions

9" wide, 7 " deep, 7 " high

Mounting Centers: 5" x 8 ", with heavy channel iron mounting brackets

Weight: approx. 40 Lbs.

Offline w5hro

  • Administrator
  • Sr. Member
  • Posts: 331
Re: Modulator... where is best?
« Reply #22 on: April 17, 2014, 07:20:47 PM »
You are not allowed to view links. Register or Login
That printed data is what Id expect and what I'm used to reading and there is no argument it was for a KW TX.Carl

Yeah, we had already gone thru this via the original topic, but of course that's gone now :'(

Anyway, I hope you guys like the new board software, its very stable and robust. This is what the other AM board uses, but its the new version. I've been watching the CPU usage on the server and the usage is almost nonexistent. I just have to upload all of the magazine files and etc. to the new download section. That also solved the other problem :icon_thumbup: