The AM Forum
March 28, 2024, 06:36:10 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1] 2   Go Down
  Print  
Author Topic: Single 811A AM Rig  (Read 12056 times)
0 Members and 1 Guest are viewing this topic.
W9ZSL
Contributing
Member
*
Offline Offline

Posts: 769


« on: March 03, 2019, 03:12:48 PM »

Just curious. Has anyone ever built an AM rig using a single or parallel 811A's? I know there is a schematic for a linear amp in the 1964 ARRL Handbook. Now that I'm almost finished with the parallel 807 unit, I think I'm ready for the next step up. I could build an 813 or 4-125A amp but I'm curious about going with a triode or two and not having to mess with a screen supply. Anyone have a schematic?
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #1 on: March 03, 2019, 03:27:26 PM »

You're talking plate modulated, not linear, right?  Unless you're going for redundancy like parking another 807/811A class smaller rig on a different band, I would look to jump up to at least an 813 in the final, preferably a pair..   Look for at least a 4X power increase to see 6dB, a minimum worthwhile project. Look closely at what John / W9JSW is building.

My opinion is triodes work FB, but tetrodes used as  RF finals in plate modulated rigs are worth the screen supply hassle because they modulate very well and are easy to drive. (and very plentiful)  Triodes work great in the modulator. Much will depend on what parts you have access to.

T
 

Transmitter:
https://www.amwindow.org/tech/htm/813/813.htm

Power supply:
https://www.amwindow.org/tech/htm/813/813ps.htm

Building Thread:
http://amfone.net/Amforum/index.php?topic=44558.0


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.
Steve - K4HX
Administrator
Member

Offline Offline

Posts: 2722



« Reply #2 on: March 03, 2019, 09:16:22 PM »

Dick, W2UJR (SK) used to run a rig with 811 finals. Talked to him many times on 160 meters in the early morning hours on that rig.
Logged
W4RFM
Member

Offline Offline

Posts: 397



WWW
« Reply #3 on: March 03, 2019, 09:50:13 PM »

Mike if you change your mind, I can send you a couple of 4-125's, tube sockets are cheap on the Bay.
Logged

BOB / W4RFM  \\\\\\\"I have looked far and wide, (I also checked near and narrow)\\\\\\\"
W8ACR
Contributing
Member
*
Offline Offline

Posts: 729


254W


« Reply #4 on: March 03, 2019, 11:26:59 PM »

A standard push pull triode schematic is published in all of the old Radio Handbooks. That schematic is suitable for 811's or just about any triode you want to plug in there. Here is my version using 254W's. Plug in coils are kind of a pain, (and can be lethal if you forget to turn off the B+)  Wink , but the RF deck is an easy build. Of course, you will also need an exciter, modulator, and power supply. I use a viking II as my exciter, and the power supply and modulator are also standard designs from the Radio Handbook. A pair of 811A's at 1250VDC are capable of about 250 Watts output. You can get more power from a pair of 100TH's, 254W's, or 810's, all still available, but getting expensive. Yes, not having to build a screen supply is a plus, but triodes require more drive power.


* IMG_1233.jpg (1568 KB, 3024x3024 - viewed 484 times.)

* IMG_1237.jpg (1558.28 KB, 3024x3024 - viewed 569 times.)

* IMG_1238.jpg (1428.89 KB, 3024x3024 - viewed 435 times.)
Logged

The life of the nation is secure only while the nation is honest, truthful and virtuous. Frederick Douglass
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #5 on: March 04, 2019, 07:15:15 AM »

Jack, K9ACT did one. Although he later added another supply and changed out the 811 for an 810.

https://schmidling.com/radio.htm

His efforts are documented here in the archives. Search on K9ACT. Here's one.

http://amfone.net/Amforum/index.php?topic=11864.0

I have the PS, some parts and the tubes for a 8000 x 2 811. I was building one until K1JJ made me an offer I could not refuse on building a 2-813 x 2-813 rig. 8000 and 810 are nearly identical. Attached is the schematic that I never got to build which was based on Jack's design. I think I was pretty close to correct on it, but YMMV.

John W9JSW

* 8000 Transmitter.pdf (938.56 KB - downloaded 144 times.)
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #6 on: March 04, 2019, 08:51:34 AM »

"Back in the days of 1kW power input" (for highest efficiency), I used a pair of 450TL triodes (in parallel - PI section output).  Efficiency was VERY high - 90+ percent.

A PI section output coupling network is easy for band changing, so I went with it.

With such an arrangement, you provide a balanced grid circuit, and neutralization is achieved by using a neutralizing capacitor off the plates of the parallel triodes, and running back to the non-grid side of the balanced input circuit.  Works slick.

Have used 812s, 250THs, 833As and other triodes in this same arrangement with a balanced grid circuit.  In fact, the first PWM transmitter I ever built was a pair of 250THs modulated by a single 833A (PWM).
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
W9ZSL
Contributing
Member
*
Offline Offline

Posts: 769


« Reply #7 on: March 04, 2019, 11:28:36 AM »

I'm thinking in terms of an experiment more than anything else.  I have most of the parts needed to build an amp in the 300 to 400 watt input range including  plate iron that will provide 2KV. I also have another with dual outputs.  Plenty of choices, 2KV, 1750, 1500 and 1250 VDC. Got the tubes also. This is back-burner for now since my modulator is restricted by the ART-13 iron. Was just thinking I could put together a classic 813 or 4-65A OR parallel 811A's and juice it with 1250 or 1500 to push everything a bit. I have exciters including that Heathkit AT-1, a home-brew, a DX-20 and a DX-60. I can also crank the power down on my Kenwood 440 to about 25 watts.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #8 on: March 04, 2019, 11:43:51 AM »

Yes, lest we forget, plate efficiency is certainly an advantage of triodes in the final. I have seen 90%+ using certain triodes. Biased deep into class C with heavy drive with high plate voltage - triodes really sing.

I'm guessing from experience the efficiency difference between optimized triodes and tetrodes runs about 10% in favor of the triodes.  But no free lunches being harder to drive and the mandatory requirement for neutralization. Does the increased drive power requirement wash out the increased plate efficiency?   And the cost of some bigger triodes these days (if you can find them) is possibly $ triple+  that of tetrodes.

Though, there is something sexy about 833As, 450TLs, 750TLs or 2000Ts class C in the RF final.

T
Logged

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

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

There's nothing like an old dog.
W8ACR
Contributing
Member
*
Offline Offline

Posts: 729


254W


« Reply #9 on: March 04, 2019, 01:00:55 PM »

My next HB rig will have 304TL's in the final. 304TL's are also electroerotic.  

Yes Steve, my 810 RF deck has a pi network tank and neutralization back to the balanced grid circuit. Works well for me also. Covers 160/75/40 meters.

Ron
Logged

The life of the nation is secure only while the nation is honest, truthful and virtuous. Frederick Douglass
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #10 on: March 04, 2019, 01:18:25 PM »

I am helping another ham get his Wilcox 36D on the air. This monster uses a pair of 450TL's with grid neutralization to get 2500 watts carrier output on AM! The RF output at 600 ohm balanced into open wire feeders will keep out any FCC person carrying a Bird watt meter!

I attach a partial schematic. B+ meant to be 4200v.

Jim
Wd5JKO



* WilCox 36D.jpg (127.83 KB, 1228x258 - viewed 1253 times.)
Logged
W9ZSL
Contributing
Member
*
Offline Offline

Posts: 769


« Reply #11 on: March 04, 2019, 02:37:24 PM »

Yeah, I remember the BC-1T we had at WOBT. Four 833's. That's why I was thinking about the triodes.
Logged
Steve - K4HX
Administrator
Member

Offline Offline

Posts: 2722



« Reply #12 on: March 04, 2019, 02:55:29 PM »

And very high plate voltage - the higher the better, especially the Eimac tubes. Not sure if this applies to the 811 of RCA heritage.


Yes, lest we forget, plate efficiency is certainly an advantage of triodes in the final. I have seen 90%+ using certain triodes. Biased deep into class C with heavy drive with high plate voltage - triodes really sing.

I'm guessing from experience the efficiency difference between optimized triodes and tetrodes runs about 10% in favor of the triodes.  But no free lunches being harder to drive and the mandatory requirement for neutralization. Does the increased drive power requirement wash out the increased plate efficiency?   And the cost of some bigger triodes these days (if you can find them) is possibly $ triple+  that of tetrodes.

Though, there is something sexy about 833As, 450TLs, 750TLs or 2000Ts class C in the RF final.

T
Logged
KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #13 on: March 04, 2019, 10:49:35 PM »

One of the old classic amateur AM transmitter designs was a pair of 812's modulated by a pair of 811's. Why didn't they just use four 812's? Because 811's make better class B modulator tubes than the 812. Why didn't they use four 811's? Because 812's make better class C plate modulated tubes than 811's. The most significant difference between 811's and 812's is the amplification factor. 811's are high mu (mu=160), 812's are low/medium mu (mu=29). It almost makes me think that RCA had this in mind when they introduced these tubes to the market.  Grin

It's my understanding that to achieve linear modulation, the RF tube must be able operate in a square-law manner. As such, when the modulating voltage is double that of the plate voltage (at 100% positive modulation), the plate current must also double to achieve the 4 times instantaneous power required for linear modulation. For reasons that I, admittedly, don't understand, low/medium mu tubes, like the 812, are better able to do this than are high mu tubes, like the 811.

So, an 811 can be plate modulated but it may not be the best tube for the job. A low/medium mu tube, such as the 812, would be a more natural choice. If it were me, I'd use a V70D (mu=28). The Handbook says a single V70D will do 185 watts output on AM. Not too shabby. And how many V70D transmitters have you ever heard on the air? Anyone, anyone?


Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
w8khk
Member

Offline Offline

Posts: 1200


This ham got his ticket the old fashioned way.


WWW
« Reply #14 on: March 04, 2019, 11:25:44 PM »


Why didn't they use four 811's? Because 812's make better class C plate modulated tubes than 811's. The most significant difference between 811's and 812's is the amplification factor. 811's are high mu (mu=160), 812's are low/medium mu (mu=29). It almost makes me think that RCA had this in mind when they introduced these tubes to the market.  Grin

It's my understanding that to achieve linear modulation, the RF tube must be able operate in a square-law manner. As such, when the modulating voltage is double that of the plate voltage (at 100% positive modulation), the plate current must also double to achieve the 4 times instantaneous power required for linear modulation. For reasons that I, admittedly, don't understand, low/medium mu tubes, like the 812, are better able to do this than are high mu tubes, like the 811.


See RCA HAM TIPS October 1939

All ham tips:
https://www.americanradiohistory.com/RCA_Ham_Tips.htm

October 1939 tips:
https://www.americanradiohistory.com/ARCHIVE-RCA/RCA-Ham-TIps/RCA-Ham-Tips-39-10.pdf


NOTE: High mu tubes require considerably more driving power than medium mu tubes.  This is especially important if the Class-C final will be plate modulated.....  See snip image of the article......
 


* 811812.JPG (78.93 KB, 320x537 - viewed 313 times.)
Logged

Rick / W8KHK  ex WB2HKX, WB4GNR
"Both politicians and diapers need to be changed often and for the same reason.”   Ronald Reagan

My smart?phone voicetext screws up homophones, but they are crystal clear from my 75 meter plate-modulated AM transmitter
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #15 on: March 04, 2019, 11:35:02 PM »

I suppose any tube (that can handle the grid current) can be set up to operate "reasonably" well in either class C  RF service or class B audio/linear service.

Tetrodes as modulators got a bad rap over the years because of the terrible slave service in Valiants, and other 100w ham rigs (6146s) using poor parameters and saggy regulation. Whereas, I run my 4-1000A tetrode modulators as they should be w/ NFB and the performance approaches a PDM rig.  4X1s as tetrodes are used in BC rigs.  813s are clean both as triode-connected or as tetrodes.  The transfer curves can tell us a lot too.

Low-mu triode modulators like 833As have been popular with the BC industry for years. Though hi-mu 3-500Z triodes as modulators are as pristine as it gets.    Even a hi-mu (200) 8877 makes a good modulator, though in RF service is limited by grid power in class C.

In class C  an RF  tube operates as a switch so I am told linearity is not as important as in say a linear amplifier or audio modulator.  Though the  X2 voltage = X4 power relationship holds to achieve a clean signal.   I mean, we can use some nasty old triodes that would exhibit < -20 dB 3rd order IMD in RF linear service but in class C  AM service will produce -35 dB 3rd IMD or better.

The problem arises when using older design triodes and other tubes that were not designed for linear RF service. At full power, they cannot be made to do better than -20 to -25dB 3rd IMD no matter what you do. (exception: using SDR Pure Signal or heavy NFB techniques)  But put in an optimized zero bias 811A, 8877, 3-500Z, 3CX-3000A7 or other tubes designed for linear service and they will do -40 DB 3rd without a problem.  Try this with a 750-TL in linear RF service and you'll be tarred and feathered off the band...  Don't axe me how I know...  Wink

T
Logged

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

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

There's nothing like an old dog.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #16 on: March 05, 2019, 01:04:43 PM »

I was hoping someone would challenge a specific idea I  mentioned above. No takers, so I will challenge it myself because I am wondering the answer.... Wink

Question: As far as a tube is concerned, linear service is linear service whether in RF linear or modulator linear service, right?  (10 KHz vs: 3.8 MHz)  So why should older low mu triodes  (like 833As, 450TLs, etc) do so well in BC linear modulator service but fall apart in RF linear service?  Or do they?

If this is true, shouldn't "modern" tubes designed for linear service (3CX-1500s, 8877s, 3-500Zs, etc., totally blow away these old triodes when used as modulators?  Why do the classic older BC rigs sound so good?

Maybe they really don't -  and the AM BC performance is hidden in the distortion created by modulation transformers. Maybe the better IMD figure is lost when using a better linear designed tube with -40DB IMD specs because of the mod iron?


IE, we are VERY sensitive to IMD performance when listening to a tube in RF linear service on the band. Even an excellent signal -40DB 3rd down can cause us QRM.  But this same "excellent" signal in modulator service gets lost in the mod iron and the difference is not enough to be noticed.

A tube that is running at -30DB 3rd has a distortion of only 0.1% ?  (not sure unless I look this up) so how can we even hear a cleaner tube in modulator service at -40DB 3rd?

To summarize, why does an old triode like an 833A appear to run so well in modulator service but falls apart when measured critically in RF service?    Perhaps the mod transformer answer is the key here.  

Maybe the answer is, yes, the modern linear-designed  tubes really are 10-15 DB cleaner than older tubes BUT the mod transformers (and tube audio drivers) are masking this difference -   And BC rigs would not improve even with more modern tubes due to the mod xfmr limitation.

Ideas on this?

T
Logged

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

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

There's nothing like an old dog.
KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #17 on: March 05, 2019, 07:44:15 PM »

Rick,

Thanks for the tips, Ham Tips that is. From the data sheets, the 811A needs about 30% more drive than the 812A. How this effects the square-law transfer function is still a mystery to me, but suffice to say, a low mu tube is a better choice for class C RF.

Those tubes sold for $3.50 in 1939. But before we jump in a time machine, that's the equivalent to $64 today. Not too bad, really. With a quick look at eBay, there are a few 812's listed there, most selling for less than $64. With some patience, I'd bet a good deal can be had.



Tom,

Consider that modulators are typically set-up in push-pull and RF linears are single ended (at least every one I've ever seen). A P-P modulator, working at audio frequencies, cancels the "in-band" harmonics. But, even if an RF linear was set-up in P-P it wouldn't cancel "on-channel" trash like a P-P audio amplifier does at AF.

Almost every tube-type audio amplifier goes through a transformer of one kind or another, so I'm not sure the full answer is there.

I think the "newer" tubes that you mention are optimized for low IMD in RF service, where the older ones weren't. I suspect that tube design, like most everything else, is a compromise of different factors, and tweaking one parameter is done at the sacrifice of some others.

These are good questions that you ask! I hope my comments are of some help.


Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #18 on: March 05, 2019, 09:14:29 PM »

Tom,
Consider that modulators are typically set-up in push-pull and RF linears are single ended (at least every one I've ever seen). A P-P modulator, working at audio frequencies, cancels the "in-band" harmonics. But, even if an RF linear was set-up in P-P it wouldn't cancel "on-channel" trash like a P-P audio amplifier does at AF.
Don


That's an interesting point, Don.

So you're saying that most audio modulators are in p-p and the harmonics that are useful and heard like X2, etc., are canceled within the audio band and this makes the tubes look good.

Whereas,  the same tubes in p-p RF service still cancel harmonics, but these harmonics are too far removed to even notice the improvement near channel. So we are essentially stuck with single-ended performance in RF service whether in p-p or single-ended.

I must give this some more thought.

Thanks for the info, OM.

T
Logged

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

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

There's nothing like an old dog.
Steve - K4HX
Administrator
Member

Offline Offline

Posts: 2722



« Reply #19 on: March 05, 2019, 09:18:02 PM »

Also most (almost all) tube audio amps had/have negative feedback. Almost no tube linear amps have negative feedback.
Logged
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #20 on: March 05, 2019, 09:27:24 PM »


Grounded Grid is a gain stage with built in negative feedback

http://amfone.net/Amforum/index.php?topic=41504.0
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #21 on: March 05, 2019, 09:33:25 PM »

Also most (almost all) tube audio amps had/have negative feedback. Almost no tube linear amps have negative feedback.


Vely true.  Heavy NFB can make a pair of old triodes behave like rented mules.


Now apples to apples:

What happens if we were to replace a pair of 833As with no NFB  (modulator) with a pair of 3-500Zs with no NFB - in a BC1T?   Would you see and hear the superior linearity of the "modern" hi-mu 3-500Zs?


* Yes, GG with its inherent NFB is a blessing. The problem is with old low-mu triodes, GG RF becomes impossibly hard to drive. I once ran a pair of 750TLs in RF GG. These have an amplification factor of 12.  It took a pair of 813s to drive it. What a tree stump that was.

T
Logged

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

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

There's nothing like an old dog.
w8khk
Member

Offline Offline

Posts: 1200


This ham got his ticket the old fashioned way.


WWW
« Reply #22 on: March 05, 2019, 11:48:47 PM »

I suppose any tube (that can handle the grid current) can be set up to operate "reasonably" well in either class C  RF service or class B audio/linear service.

Tetrodes as modulators got a bad rap over the years because of the terrible slave service in Valiants, and other 100w ham rigs (6146s) using poor parameters and saggy regulation. Whereas, I run my 4-1000A tetrode modulators as they should be w/ NFB and the performance approaches a PDM rig.  4X1s as tetrodes are used in BC rigs.  813s are clean both as triode-connected or as tetrodes.  The transfer curves can tell us a lot too.


Tom, I did not consider challenging your statements (as mentioned in a later post in this thread).  On the contrary, I wholeheartedly agree with your comment about the tetrodes, and specifically the 6146, getting a bad rap as an audio (modulator) tube, while it seems perfectly suited to linear amplification - - - (how many SSB transceivers use 6146 tubes as finals - too many to count?)

As you mentioned, poor regulation of screen and grid bias, and operating the tube outside the linear portion of the transfer characteristic curves leads to this erroneous conclusion that the tube is no good for audio.  If this were true, then it would never have been selected for high-end audio amplifiers by Dyna, Hafler, RCA, UTC, etc.  United Transformer Corporation provided design and schematic data for use with their high-end Linear Standard output transformers, as did Hafler and Keroes for the Acrosound TO-350 in their UltraLinear circuits, etc. etc. etc.  With good regulation, proper bias adjustment, and careful impedance matching, minimal feedback was employed and performance was stellar.  They have the same potential (no pun intended) as modulators in AM equipment, with attention to the aforementioned details.  Question:  How much time is spent in finding the "sweet spot" for minimal IMD and maximum performance and efficiency with any new linear amplifier designed and constructed by the competent amateur?  Does the modulator for an AM rig deserve the same TLC?

It is true that most of the RF linear amplifiers in use today for SSB (and AM) probably fall into the grounded-grid triode category.  This of course provides some negative feedback, by default.  Other configurations can be employed as well, with or without NFB.  Either single-ended or push-pull.  It has been my experience that the most important consideration in achieving acceptable IMD is operating the amplifier within the most linear portion of the transfer characteristic curve.  

Case in point:  Before I restored my grandfather's 1937-vintage push-pull 250-TH rig to its original classic AM configuration with push-pull 810 modulators, my dad (W2DU) reconfigured it to function as a linear amplifier for his Heathkit HW-101 transceiver.  He ran 3500 volts on the plates, and had a very tightly regulated, adjustable bias supply.  He added "grid stoppers" to tame parasitics, but none were required in the plate circuit.  Standard split stator plate tuning capacitor, center tapped plate inductor with swinging link, feeding a link-coupled tuner to ladder line-fed dipoles and lazy-h antennas.  He operated on 75, 40, 20, and 15, with no need to re-neutralize when changing plug-in coils.  Neutralization was immaculate.  He participated regularly in the RCA International SideBand Nets, often on 3999 KHz LSB.  Most of the folks were extremely critical of signal quality, and his was always above reproach.  He continued to operate that rig until I donated my dual 4-400A rig to his shack.  He preferred that only because it included the Illumitronics Pi-Dux 195-2 with a BC-375 bandswitch, eliminating the need to swap grid and plate coil assemblies.

Before he put that amplifier on anything but a dummy load, he ran two-tone tests and fine-tuned the operating point to ensure a clean signal with minimal IMD.   Of course in the '60s there were no SDRs with waterfall displays, but it was still common practice to make sure there was no monkey-chatter up and down the band from improperly-operated "linear" amplifiers.  Listen on 75 today, I am not sure that is still the case!

I believe the answer to the question is that tubes designed and manufactured for lowest IMD at high power excel at providing a wide margin in the linear portion of the curve, but most any tube could probably be made to perform acceptably at a reasonable power level in linear service without negative feedback.  The user will likely have to limit power output to a level considerably less than the tube's dissipation capability, but, come to think of it, I seem to recall some of the folks on this forum have recommended that the linear amplifier be "over-built" so that it can be operated at far-less than its maximum capability to result in a pristine signal for all to enjoy.  Sound familiar?

I embarked on a project over a year ago to build the ultimate AM transmitter, with a clean, distortion free, "DC to daylight" modulator, without a trace of iron in the audio path.  I am still in the process of collecting the required components, and I believe that in the near future I will have a low-powered prototype, then a legal limit final configuration will be assembled.  This will be the most "INEFFICIENT" rig known to man, sneered at by the Class-E PDM crowd, but there will be no cleaner modulation from any rig heard on HF.  The idea is to use a single-ended, one-tube class-A series modulator, running the legal limit, plate modulating (NOT HEISING) a single Class-C triode final.  The power supply is complete, but I am awaiting tubes and a few other components for the lower-power prototype.  The modulator will look much like the vacuum-tube series regulator, providing plate voltage for the final.  The regulator will allow adjustment of the operating point, and therefore the plate voltage applied to the Class-C final from the 5000 volt 1 amp CCS supply.  From there we determine the most linear portion of the transfer curve of the series tube (modulator) and apply exquisite audio to the final, with no inductors or transformers to corrupt the signal.  Almost one kilowatt will be expended collectively by the filaments for the series modulator and final, and the series modulator will dissipate in excess of one kilowatt with no modulation.  All these watts are 100 percent efficient in heating the shack on a night like tonight when it is 28 degrees f in HotLanta in March.  (A few "Old Buzzard" transmissions may be required to stabilize the shack temperature when the mercury falls below 15 f in January, because the plate dissipation is only a shack heater when in transmit mode.)  But the pair of 3CX3000 tubes will be coasting at far less than their rated dissipation.  With this rig, adjustable bias will allow us to demonstrate the fidelity and cleanliness of the proper operating point on the straight portion of the curve.  I can't wait until the rest of the components arrive to proceed with final assembly!
Logged

Rick / W8KHK  ex WB2HKX, WB4GNR
"Both politicians and diapers need to be changed often and for the same reason.”   Ronald Reagan

My smart?phone voicetext screws up homophones, but they are crystal clear from my 75 meter plate-modulated AM transmitter
KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #23 on: March 06, 2019, 08:05:13 AM »

Tom,
Consider that modulators are typically set-up in push-pull and RF linears are single ended (at least every one I've ever seen). A P-P modulator, working at audio frequencies, cancels the "in-band" harmonics. But, even if an RF linear was set-up in P-P it wouldn't cancel "on-channel" trash like a P-P audio amplifier does at AF.
Don


That's an interesting point, Don.

So you're saying that most audio modulators are in p-p and the harmonics that are useful and heard like X2, etc., are canceled within the audio band and this makes the tubes look good.

Whereas,  the same tubes in p-p RF service still cancel harmonics, but these harmonics are too far removed to even notice the improvement near channel. So we are essentially stuck with single-ended performance in RF service whether in p-p or single-ended.

I must give this some more thought.

Thanks for the info, OM.

T

Tom,

Exactly! I just wish I could have written that thought as succinctly as you have.


Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #24 on: March 06, 2019, 11:41:58 AM »

This will be the most "INEFFICIENT" rig known to man, sneered at by the Class-E PDM crowd, but there will be no cleaner modulation from any rig heard on HF.  The idea is to use a single-ended, one-tube class-A series modulator, running the legal limit, plate modulating (NOT HEISING) a single Class-C triode final.  The power supply is complete, but I am awaiting tubes and a few other components for the lower-power prototype.  The modulator will look much like the vacuum-tube series regulator, providing plate voltage for the final.  The regulator will allow adjustment of the operating point, and therefore the plate voltage applied to the Class-C final from the 5000 volt 1 amp CCS supply.  

Hi Rick -

That was one of the best, most informative posts of the year.   I read it several times and plan to read it again.  Your posts usually show the mentorship and wisdom you've had for years from your father. You were a lucky guy to have a coach like him.

I'll make some more comments on the tube material later.

I am very interested in your new class A series modulated class C rig.  I had thought of doing that with my 4-1000A rig early on, using a 3CX-5000 as the series modulator, but settled for the conventional p-p modulator.  The heat generated would be incredible and prohibative in the summer, but a nice heater in the winter as you say.  But what exquisite audio!  There's nothing getting in the way except the transfer curves of the tube and a 3CX-3000A7 is already one of the very best tubes to use.

What do you plan to use to drive the 3CX-3000A7?    Will you float a fil xfmr / CT and use a series MOSFET for the 3CX ?  Or maybe a fiber optics grid feed or some other clever method?  Tell me more why you chose to modulate the HV plate circuit  rather than in the cathode.


I imagine your design would be along the lines of my little PW series modulated rig, but instead modulated in the plate circuit:
 
https://www.amwindow.org/tech/htm/series.htm

I was very pleased with series modulation for audio performance. The only trouble was crossover anomalies to work out... but the beautiful linear audio and huge, unlimited audio peaks were amazing.   I found that  three class A modulator tubes were needed for one class C final, so you are correct using the big 3CX-3000A7 series tube.

What RF final(s) are you using?  5KV  peak will give you about 2KV when dead carrier. (using ~130% peak mod)  There are so many ways you could go here.  

Now that I think of it, the reason I didn't go with the series modulator for the 4-1000A rig is because I needed an 8 KV DC supply to equal the peak  voltage I could get when using a p-p modulator system. Like a big PDM tube rig, it got a little hairy...  Wink

T
Logged

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

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

There's nothing like an old dog.
Pages: [1] 2   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.147 seconds with 19 queries.