The AM Forum
March 29, 2024, 07:34:07 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: B+ Anode Regulation -- Advantages and Dis-advantages ?  (Read 5265 times)
0 Members and 1 Guest are viewing this topic.
B. Winston - SWL
Member

Offline Offline

Posts: 8


« on: January 20, 2022, 02:03:59 PM »

I've asked this question on a couple of other forums but the small number of responses received have been predominately "inconclusive"  Roll Eyes,  this with the exception of a micro-minority of relatively cogent technical inputs.

My original question has evolved and is for both AM  and also SSB as follows:

 What technical performance advantages and/or disadvantages exist with implementation and usage of  a 3,000 VDC   B+ anode regulator to  AM and SSB  PA's using tubes such as 8295's,  or for that matter,  any more 'modern' tube,  etc. ?

The particular tube, voltage and current values are not the crux of the question....just the idea of having a fast acting B+ regulator which maintains the B+ anode voltage for the fluctuating current due to voice waveform is the main issue.

My current configuration will be for grid driven but my question also applies to future PA's in grounded grid also.

Cost of design and implementation is not a limitation.
 
Components are not a limitation.

Energy usage is not a limitation.

Thanks for your inputs in advance...

Bill
Logged
Detroit47
Member

Offline Offline

Posts: 644



« Reply #1 on: January 20, 2022, 02:32:19 PM »

Proper grid and screen regulation will have a much greater impact on linearity. If you look at most commercial amplifiers you will see straight capacitive input supplies on the B+. The older stuff with tube rectifiers will use L input. So I would say regulated B+ is a waste of time. So for the reality of the situation don't skimp on the B+ capacitor. It will hold the voltage steady avoiding sags on modulation. In sever cases of under capacitance ac component can sneak in when there is a big load on the supply.

Johnathan N8QPC
Logged
W1ITT
Member

Offline Offline

Posts: 573


« Reply #2 on: January 20, 2022, 03:01:03 PM »

I would ask what is the ability of the regulator to follow at syllabic rate, particularly in SSB service.  I don't know the answer to that question, and throw it out to the assembled group.  I'm not sure that we design regulators to follow instantaneous syllabic variations in current draw. 
Thinking that B+ regulation would have at least some effect on IMD, I built the plate supply for a 4-1000A rig 50 years ago with a 240 mfd at 6 Kv  capacitor.  I never tested IMD and, of course, it would be dependent on other factors, but it made me feel smug that I had more C in my plate supply than the other fellows.  I had a long delay step-start, and it took about a minute to bleed down at the end of operation.  By the way, that capacitor would probably be available for purchase as I have acquired others that aren't quite so massive.  It's  a good 2 feet tall...  And darn heavy.
73 de Norm W1ITT
Logged
KA3EKH
Member

Offline Offline

Posts: 772



WWW
« Reply #3 on: January 20, 2022, 03:36:48 PM »

Just curious, how do you regulate a 3-kV power supply? You can always play with resistors to keep the output from rising to some stupid no load level but you would be looking at some big wattage 100K resistors, think the only way I can see it being done is maybe a solid state pwm supply that samples the output and can regulate the pulse with to vary the output but then you end up with a power supply that’s more complicated then the transmitter! Over the years have worked with several Harris MW-5 5 kW AM transmitters and they use PDM Modulators in series with the PA tube with all that mess being feed by a unregulated 12 kV three phase power supply and all they used was simple filtering. Other FM and TV transmitters never had any regulation other then from what the load demanded. On lower power TV transmitters that used tubes, transmitters between 500 to 1 kW they make a big deal out of regulating the G2 supply so much so that Emcee and Acrodog used Sola constant voltage transformers on the primary side of the screen supplies but no one ever attempted to regulate the plate supply and analog TV being AM had a lot more requirements for maintaining linearity the just AM voice.
Logged
K9MB
Member

Offline Offline

Posts: 359


« Reply #4 on: January 20, 2022, 04:07:22 PM »

I do not see the point in precisely regulating plate voltage in a high power transmitter or amplifier.
The use of a choke input filter and a large enough output capacitor coupled with the minimum required bleeder current for the choke inductance will be quite stable and will be limited by winding resistance and transformer power rating.
Most old broadcast transmitters had choke input filters and they had very good regulation for AM. Using a very large output capacitor should limit sagging on SSB peaks.
You will get about 0.9 x the rms voltage of the transformer winding depending in choke and transformer winding resistive losses.

As been mentioned, if you are using a tetrode wth grounded catbode circuit, the screen voltage is very critical and must be regulated for best linearity.
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #5 on: January 20, 2022, 06:48:10 PM »

If you're that worried about regulation then use choke input for easy.

That said, they do have xiayers that will work at 2kv now.  IGBJT I believe is the nomenclature.

I have used transistors to regulate my screen supply in my 4cx5000 amp.

Rich Measures had a lm723 based screen regulator that would run up to 1.4kv.  That can also be scaled up.

Use a 3-500Z or other robust tube as the pass element.  Look at Rich's circuit for ideas on how to scale the voltage levels.

In the end, you'll eek out a few dB better IMD.

IMHO you'd be better off running a choke input LCLC or a cap input on the B+ and choke in the negative.  Use a FW rectumfrier.

Or, if IMD is what you are chasing then look at running tubes on lower voltage.  A 3-500Z with 2kv on the plate and a resistor in series with the drive can do better than - 40 dB down.

Tom, K1JJ built an amplifier chain using great tubes that achieve the dame IMD ratings that you can get with predistortion.

And then that is the crux.  With presistortion you can see - 50 or better IMD.  Without all this regulation nonsense.

I saw your post on Facebook and semi ignored it thinking it was a troll attempt.  But this is my honest answer.

Also, running a big enough transformer goes miles.  My Dahl will drop about 20 volts at legal limit.  It's capable of 12kva CCS and 24KVA intermittent.  So if you wanna build a couple hundred lb supply, you can drop the regulation too.

Hope this helps.

--Shane
WP2ASS / ex KD6VXI
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #6 on: January 20, 2022, 06:51:07 PM »

Also, cost.  Having 3 or 4 500z tubes running as regulators for a 2 or 3 tube amp is pretty expensive, since they are consumable.

A 4 or 5kva supply looks pretty inexpensive, since it is a one time cost.

If I was still stateside I'd go e you a choke for the cost of shipping.  Alas, here in the Carribean I don't have access to my storage right now lol.

--Shane
WP2ASS / ex KD6VXI
Logged
W1ITT
Member

Offline Offline

Posts: 573


« Reply #7 on: January 20, 2022, 09:35:35 PM »

I think it was sometime back in the 1970s that 73 Magazine had a construction article on a regulated multi-kilovolt supply.   The design sampled the B+ down a ways on a voltage divider, compared it to a reference, and then used op amps to control the gates of triacs or SCRs , pulsing them with the 60 hz sync to control the primary volts duty cycle on the plate transformer.
As with my previous concern, I doubt how well this setup would be able to follow syllabic rate load variations, given the time constant of the big transformer and whatever filter capacitors might be on the far side of the rectifiers.  I saved the article....but finding it would be yet another matter. 
73 de Norm W1ITT
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #8 on: January 20, 2022, 10:22:07 PM »

Very good question and logical answers....

I have to agree that regulating the plate supply in a linear system (and class C system) is way overkill assuming that the HV supply uses  a large filter cap.  A filter choke is not as important as a large output capacitor, however.

As Shane mentioned, I built an HF RF linear amplifier chain using the following:  100 mW from the low level FT-1000D >  1 watt ZHL-1 SS lab amp (class A) > 4CX-350J  (class A) > 8877  (class A) >   BIG mofo triode (MRI pull, class B).   Only the  lab amp had a regulated VCC.  All screens and grids were regulated.  The plate HV power supplies were all unregulated for the 4CX-350J, 8877 and big triode. All stages were using super linear tubes. The 4CX-350J "J" and MRI pull were designed for super clean linear use and were rather expensive.  The  5KV HV supply used a 10 KV @ 180 uF filter capacitor, so it was pretty stiff.

Using no RF negative feedback, I was able to achieve  -55 dB 3rd order IMD at full power using this lashup.  I built it in a time before predistortion was widely used by hams, back a decade ago. The fact that I was able to get this performance without regulated HV (plate) supplies shows the futility of going thru the effort to go any further - especially with the difficulty and serious power wasted by regulating large HV supplies.  I realize that the driver amplifiers were class A, thus they had no dynamic draw down on their supply regulation... however the big triode was in class B, thus its HV load varied from idle to full power input.  No IMD problem when using a large HV filter cap.

BTW, I felt pretty stupid after going thru so much effort to get great IMD and then seeing predistortion and Pure Signal get optimized a few years later.  A super clean linear amplifier chain had been my goal for many years, but when I finally got there, everyone was already leaving the party with SDRs under their arms...  Grin

BTW, here's some pictures of the IMD testing of this linear system:

http://amfone.net/Amforum/index.php?topic=33505.0


Tom, K1JJ
Logged

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

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

There's nothing like an old dog.
W7TFO
WTF-OVER in 7 land Dennis
Contributing
Member
*
Offline Offline

Posts: 2521


IN A TRIODE NO ONE CAN HEAR YOUR SCREEN


WWW
« Reply #9 on: January 21, 2022, 12:29:12 PM »

The whole game with power supplies is the source impedance.

The lower, the better in terms of regulation.

The current (no pun) scheme is to use iron with low resistance windings and soft start, silicon rectifiers, and a fat cap to knock down the ripple and 'fill in the gaps' of power via T-R/C.  This yields a very good reserve of syllabic DC power.

Older circuits using tubes introduced more resistance, with vacuum rectifiers much worse than Mercury vapors with their constant voltage drop. 

With the variable load presented via plate modulated finals, swinging chokes were employed as dynamic regulators.  That worked well enough for a long time, but the ability to source all the right parts nowadays has led to the previous paragraph.

This is most important in single-phase supplies, as when the waveform goes thru zero it means there are lots of 'holes' in the flow of electrons.

3-phase supplies, if available, afford a much more robust energy source due to there never being a zero crossing.  With full-wave rectification, you get easily filtered ripple @ 360Hz.

One can then get by with a little as 2uF & 2Hy for passable regulation.

73DG

Logged

Just pacing the Farady cage...
B. Winston - SWL
Member

Offline Offline

Posts: 8


« Reply #10 on: January 21, 2022, 09:05:16 PM »

Thanks to everyone who took the time to respond. 

I appreciate the time and consideration given in the various posts.


With cheap $ 40  IGBT's  that can easily handle 1200 V and 75 Amps,   ....a bit more in cost can handle 4500 V and 30 Amps   ( .70 inch wide, .20 inch thick package) , the cost to me is not prohibitive if there is an advantage in the performance aspects of the PA.

I 'get' the high capacitance  and bleeder resistor 'regulation' alternative of traditional PS's. 

My concern is the possible B+ 'sag' that occurs with the voice peaks and any other performance based parameters. 



Bill




 

Logged
B. Winston - SWL
Member

Offline Offline

Posts: 8


« Reply #11 on: January 21, 2022, 09:16:00 PM »

I would ask what is the ability of the regulator to follow at syllabic rate, particularly in SSB service.  I don't know the answer to that question, and throw it out to the assembled group.  I'm not sure that we design regulators to follow instantaneous syllabic variations in current draw. 
Thinking that B+ regulation would have at least some effect on IMD, I built the plate supply for a 4-1000A rig 50 years ago with a 240 mfd at 6 Kv  capacitor.  I never tested IMD and, of course, it would be dependent on other factors, but it made me feel smug that I had more C in my plate supply than the other fellows.  I had a long delay step-start, and it took about a minute to bleed down at the end of operation.  By the way, that capacitor would probably be available for purchase as I have acquired others that aren't quite so massive.  It's  a good 2 feet tall...  And darn heavy.
73 de Norm W1ITT


Agree that the time response is important.

Those IGBT's  have pretty fast response times ...around 1000 nS for the 4500 Volt version.  Couple this with a fast  7 nS opamp (Analog Devices)...the cumulative time stack response should have a minimal response time in relation to a 3 KHz voice bandwidth.

Logged
B. Winston - SWL
Member

Offline Offline

Posts: 8


« Reply #12 on: January 21, 2022, 09:31:26 PM »

Very good question and logical answers....

I have to agree that regulating the plate supply in a linear system (and class C system) is way overkill assuming that the HV supply uses  a large filter cap.  A filter choke is not as important as a large output capacitor, however.

As Shane mentioned, I built an HF RF linear amplifier chain using the following:  100 mW from the low level FT-1000D >  1 watt ZHL-1 SS lab amp (class A) > 4CX-350J  (class A) > 8877  (class A) >   BIG mofo triode (MRI pull, class B).   Only the  lab amp had a regulated VCC.  All screens and grids were regulated.  The plate HV power supplies were all unregulated for the 4CX-350J, 8877 and big triode. All stages were using super linear tubes. The 4CX-350J "J" and MRI pull were designed for super clean linear use and were rather expensive.  The  5KV HV supply used a 10 KV @ 180 uF filter capacitor, so it was pretty stiff.

Using no RF negative feedback, I was able to achieve  -55 dB 3rd order IMD at full power using this lashup.  I built it in a time before predistortion was widely used by hams, back a decade ago. The fact that I was able to get this performance without regulated HV (plate) supplies shows the futility of going thru the effort to go any further - especially with the difficulty and serious power wasted by regulating large HV supplies.  I realize that the driver amplifiers were class A, thus they had no dynamic draw down on their supply regulation... however the big triode was in class B, thus its HV load varied from idle to full power input.  No IMD problem when using a large HV filter cap.

BTW, I felt pretty stupid after going thru so much effort to get great IMD and then seeing predistortion and Pure Signal get optimized a few years later.  A super clean linear amplifier chain had been my goal for many years, but when I finally got there, everyone was already leaving the party with SDRs under their arms...  Grin

BTW, here's some pictures of the IMD testing of this linear system:

http://amfone.net/Amforum/index.php?topic=33505.0


Tom, K1JJ


This PA is one of my 'Bucket List' projects   Wink.....

As far as SDR....worked that stuff 20 years ago on an Army comm DARPA program.

My receive set-up uses RFSpace SDR's since almost 2 decades already....excellent dynamic range. 

 For kicks I bought a couple of those RTL's and there they sit in a cardboard box.   

And the SDRPlay when it first came out....returned the SDRPlay to HRO a couple days later.





Logged
B. Winston - SWL
Member

Offline Offline

Posts: 8


« Reply #13 on: January 21, 2022, 09:35:57 PM »

Regulating the bias, screen and suppressor is already in the 'Plan'.
Logged
W1RKW
Contributing
Member
*
Offline Offline

Posts: 4410



« Reply #14 on: January 22, 2022, 07:24:49 AM »

I ran into to this a while back and bookmarked it for future reading.  Forgot I had it until I saw this thread.  It may be of interest to one or more.

http://w5jgv.com/hv-ps1/
Logged

Bob
W1RKW
Home of GORT. A buddy of mine named the 813 rig GORT.
His fear was when I turned it on for the first time life on earth would come to a stand still.
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #15 on: January 22, 2022, 09:22:42 AM »

Very good question and logical answers....

I have to agree that regulating the plate supply in a linear system (and class C system) is way overkill assuming that the HV supply uses  a large filter cap.  A filter choke is not as important as a large output capacitor, however.

As Shane mentioned, I built an HF RF linear amplifier chain using the following:  100 mW from the low level FT-1000D >  1 watt ZHL-1 SS lab amp (class A) > 4CX-350J  (class A) > 8877  (class A) >   BIG mofo triode (MRI pull, class B).   Only the  lab amp had a regulated VCC.  All screens and grids were regulated.  The plate HV power supplies were all unregulated for the 4CX-350J, 8877 and big triode. All stages were using super linear tubes. The 4CX-350J "J" and MRI pull were designed for super clean linear use and were rather expensive.  The  5KV HV supply used a 10 KV @ 180 uF filter capacitor, so it was pretty stiff.

Using no RF negative feedback, I was able to achieve  -55 dB 3rd order IMD at full power using this lashup.  I built it in a time before predistortion was widely used by hams, back a decade ago. The fact that I was able to get this performance without regulated HV (plate) supplies shows the futility of going thru the effort to go any further - especially with the difficulty and serious power wasted by regulating large HV supplies.  I realize that the driver amplifiers were class A, thus they had no dynamic draw down on their supply regulation... however the big triode was in class B, thus its HV load varied from idle to full power input.  No IMD problem when using a large HV filter cap.

BTW, I felt pretty stupid after going thru so much effort to get great IMD and then seeing predistortion and Pure Signal get optimized a few years later.  A super clean linear amplifier chain had been my goal for many years, but when I finally got there, everyone was already leaving the party with SDRs under their arms...  Grin

BTW, here's some pictures of the IMD testing of this linear system:

http://amfone.net/Amforum/index.php?topic=33505.0


Tom, K1JJ


This PA is one of my 'Bucket List' projects   Wink.....

As far as SDR....worked that stuff 20 years ago on an Army comm DARPA program.

My receive set-up uses RFSpace SDR's since almost 2 decades already....excellent dynamic range. 

 For kicks I bought a couple of those RTL's and there they sit in a cardboard box.   

And the SDRPlay when it first came out....returned the SDRPlay to HRO a couple days later.







None of those offer TX.  And for predistortion to work, you need to TX.

20 years ago hams where using sdr as well.  Things have moved along A LOT.

It appears you asked your question about regulated b+ supplies more in a quest for reaffirmation on something you already wanted to do.  I get it.

However, as pointed out, it's just not necessary.  And time, money and energy can be spent elsewhere.

Regulating your bias supplies, yes.  Needs to be done.

Regating B+.....  Other than to say been there done that, I just don't see a need.

But keep us updated if you do it.  If anything, it's going to be a learning experience.

--Shane
WP2ASS / ex KD6VXI
Logged
B. Winston - SWL
Member

Offline Offline

Posts: 8


« Reply #16 on: January 22, 2022, 06:45:10 PM »

I ran into to this a while back and bookmarked it for future reading.  Forgot I had it until I saw this thread.  It may be of interest to one or more.

http://w5jgv.com/hv-ps1/

I took a look at this today.

Impressive engineering...including the regulation responses.

Thanks for sharing.

Bill
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #17 on: January 23, 2022, 10:09:37 PM »

Overregulation with too much feedback could result in attenuating the actual waveform's peaks and valleys. This would result in a carrier for example, with little modulation.

In broadcast transmitter power supplies we only use enough regulation to keep us within the stated power output limits we guarantee for FCC regulations.

Power supplies with low output impedance are the best approach.  


Phil - AC0OB  
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #18 on: January 24, 2022, 05:18:23 PM »


I link to some old sites related to audio effected by power supplies. This is in regard to tube type audio amplifiers. A similar if not exactly the same issue exists with the power supplies feeding an audio modulator, or for an RF linear amplifier used on SSB. In both cases, the power draw varies at a syllabic rate. The power supply for a class C plate modulated amplifier is also important since the modulator is supposed to modulate the RF PA tube, and not modulate the power supply! This last point is important with vintage gear that might be low on power supply output capacitance, and then some Ham comes along and opens up the bass response in the modulator....the result is that the lower audio frequencies get muddied up and reduced in amplitude as the power supply varies with the lower frequencies in the speech. Those old oil Caps at 8 uf, or so need to be beefed up to 20uf or more...

This link from Mr Curcio is decades old, and more of a marketing Glossie for a service he provided. He provided an LM317 based floating regulator for tube power amplifiers. They might work OK, or even great, but virtually all circuitry like this using silicon will fail the screwdriver short circuit test...

http://www.curcioaudio.com/Audibility_of_PwrSupply.htm

Here is an 18 year old post from me concerning a project I completed. I added several regulators to a Sherwood S8000 tube type Stereo, and FM receiver. It is about 62 years old now, and still going strong. As usual with the rec.audio.tube group, the back and forth can be quite brash, and in ones face. Some of those folks now are long gone. Still, the conversation offers differing viewpoints, pretty rare these days.
https://rec.audio.tubes.narkive.com/u3WKwjmq/the-audibility-of-power-supplies#post25

Jim
Wd5JKO
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #19 on: January 25, 2022, 01:09:00 AM »

I appreciate the hobby aspects of perfection-seeking.

-worked-out HV regulator circuits from STC available,
  https://bunkerofdoom.com/tube/man/STCx.pdf

military-preferred circuits of 300V 0.1% regulation circuits that could be adapted.
  https://www.navy-radio.com/manuals/NAVAER16-1-519-1.pdf

HV regulator need only concern itself with the top 300-500V or so of the B+ if the rectifier and filter are competent. Not going to do that here, it's just something to study because it's interesting. Predistortion would have the same effect on the output signal? A non-regulating workaround but the improvement goal is there.

It is true that plate power supply voltage fluctuations and other noises modulate the RF amplifier but will anyone ever hear or notice it over the air? Only test equipment will pick that out.

Something like a 2x-oversized choke input supply and critical bleed current can give <3% regulation and <0.5% ripple. No active regulation required. But no one will notice the effort, just saying.

Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.119 seconds with 18 queries.