Ideas wanted for a medium "Summertime" quiet linear amp

[K1JJ]

The weather will soon be getting hot and I have some extra time like everyone.  I plan to use the big rigs less until fall and build a new smaller linear amp for AM.

I already have a pair of 3-500Zs that I run at 1/2 power at 150 watts out, but I'd like to build something with no blowers or fans, thus no noise...   and about 10-13 DB gain to take my FT-1000D 5-15 watts carrier and give about 100 watts of carrier conservatively.  I want great IMD by running it at 1/2 power capability. I prefer grounded grid for the inherent NFB and stability. I have 2KV and 4KV available, but prefer to use the 2KV supply which is smaller and quieter.

I have all the parts needed.  So far I am considering for 160-40M AM linear service:

A pair of 813s in GG with about 2KV.

A single grid driven, neutralized 833A with 3KV.   (unusual)

Four 4D32s in GG with 1KV.

Eight?  6146Bs would be neat in GG and I could build the 800 V supply into a fully self contained unit. It should do 100 watts AM carrier.

Notice none of them require forced air cooling and get by with just radiation and convection cooling. Quiet.

I will bias the finals into class C for AM use letting the carrier bring it back into B/C mode. This increases efficiency quite a bit.

Any more ideas appreciated and comments critiquing the above suggestions related to strengths and weaknesses appreciated...

T

[WD8BIL]

I'd like to see the 833 idea tried out. As you said, you've done most of the conventional setups before.
Take a li'l walk on the wild side!

[WD5JKO]

Consider a water cooled tube.....

I also remember a 4PR1000 used in a big Varian HV power supply, something like 200kv. I had to change the tube. This was quite a chore since it was upside down in the socket, and cooled with about 25 gallons of oil in a large metal tank. The tube is very buoyant, and submersing it while trying to align the socket pins was really a challenge. That oil would walk up your arms, and before you knew it, you had oil everywhere. Worst thing was that if voltage multiplier diodes failed "open" then a section of the multiplier stayed charged to something like 30KV. One had to be very careful when reaching into that tank. Worst case, your arm is against something sharp or pointed, and then you get nailed. Out comes the arm minus a strip of meat! Oil everywhere.

Jim
Wd5JKO

[Opcom]

Consider a water cooled tube.....

I also remember a 4PR1000 used in a big Varian HV power supply, something like 200kv. I had to change the tube. This was quite a chore since it was upside down in the socket, and cooled with about 25 gallons of oil in a large metal tank. The tube is very buoyant, and submersing it while trying to align the socket pins was really a challenge. That oil would walk up your arms, and before you knew it, you had oil everywhere. Worst thing was that if voltage multiplier diodes failed "open" then a section of the multiplier stayed charged to something like 30KV. One had to be very careful when reaching into that tank. Worst case, your arm is against something sharp or pointed, and then you get nailed. Out comes the arm minus a strip of meat! Oil everywhere.

Jim
Wd5JKO

Hopefully it was not the "disinfectant" type of oil!


Convention cooled single 304T is very quiet, has been done in the Radio Handbook, and points the way to the 833. It is grounded grid, if that is a critical point.

The single 304T amp is on:
page 501 of the 13th ed.
page 135 of the 14th. ed.
page 163 of the 15th ed.

[K1JJ]

http://www.bunkerofdoom.com/lit/Eimac_amateur_service_newsletters1979.pdf

Pat,

I also found some great reading on this subject and other linear tubes in Bill Orr's notes... from your Bunker of Doom.

Right now I'm focusing on IMD 3rd order figures. There isn't much actual test data out there on older tubes, but it seems the more modern tubes designed for linear service have a big IMD cleanliness edge over the older tubes.   It's hard to beat a 4CX-350J, 3-500Z, 8877, etc. in this area. (But notice they all require forced air)

I think Buddly's quad 572Bs are a good choice too. Has anyone seen IMD figures on 572Bs?

T

[w8khk]

I think you should go all-out and run a pair of 849s, or better yet a single 851 in grounded grid.  Those bottles are large enough that radiation and convection work fine with no need for air circulators! 

There are probably folks just sitting on these tubes, never to see the light of thermionic emission again.

[KD6VXI]

572 and the 811 are imd generators.  You have to keep the Pout to about a third of the plate diss to equate to anything clean.

You are correct in your assumptions in the newer tubes.

LDMOS is promising.  With your drive level, you could drive a quad to 6k plus. Assuming you don't mind not staying hollow state.

--Shane
KD6VXI

[K1JJ]

LDMOS is promising.  With your drive level, you could drive a quad to 6k plus. Assuming you don't mind not staying hollow state.

--Shane
KD6VXI

You are right Shane... this LDMOS is amazing technology.   

https://www.youtube.com/watch?v=M1wmYoQ2Uko

[WD5JKO]

Then there is the 861, made for Class B Linear AM.

http://www.rfcafe.com/references/qst/rca-victor-radiotron-uv-861-qst-february-1931.htm

Jim
Wd5JKO

[w8khk]

Then there is the 861, made for Class B Linear AM.

http://www.rfcafe.com/references/qst/rca-victor-radiotron-uv-861-qst-february-1931.htm

Jim
Wd5JKO

If famous AM strapper Tom Vu make big linear with 851 or 861, word get out, everybody need one, demand cause Chinese to make more new "old school" toobs.  Audio Fools also love supply of big Class-A toobs for SE overpriced snake oil amplifiers, make big money.  World economy restored in months!
 Hope that not tooo political???   Cawmaaannnn

[K1JJ]

I think the 861 was back in that brief period when broadcash was still amplifying AM telephony with RF linears.  Then came Heising, plate modulation, etc.  Interesting that the tube had an amplification factor of 300. And used a screen grid.

T

[K1JJ]

Here's a guy who built a single, grid driven, class AB2 linear 833A, neutralized. He claims it's doing 70% efficiency on 20M ssb.  Very cool build putting out 800 watts on dead key.

https://www.youtube.com/watch?v=mbaYUkZlMsw
https://www.youtube.com/watch?v=ChHPL84f22Q

I am just toying with the idea of a linear pair of AB2 833As in push-pull, grid driven with cross neutralization.  

Also looking closely at a pair of 813s in GG...

Just thinnkking and making my misstakkees on paapper for now.

T

** As a fitting epitaph, the next video shows him converting the 833A over to a 3-1000Z a year later -  because the 833A is hard to drive... :-)

https://www.youtube.com/watch?v=CTcOg4fy_Uw

[K1JJ]

After looking at a lot of data and thinking about it, I decided on a single 3-500Z with only 1500 volts on the plate. This produces a 3rd order IMD freak at -46 DB.... super cleanest tube parameters out there.  Look at the link below showing the 3-500Z typical operating numbers on the 2nd page down.

I have run these same 3-500Z parameters before when I was using the class A linear chain and it actually tested out to those IMD numbers.  I have a Variac on the HV and will be able to dial in more voltage if desired, but at 1500V I will be able to get almost 90 watts AM carrier according to the data.  The good thing is I will need only about 5 watts of drive from the modified FT-1000D which is its cleanest output power according to tests I've run.  I will experiment with moving some air thru the 3-500Z socket using a Variac controlled blower. In the past I have been able to slow it enuff to make virtually no noise and keep it cool when running at such a low power...  This lash-up will make a really nice overall exciter that with the DSP audio filter and heavy amplifier loading should approach SDR-type sharp bandwidth, though not as good as pre-distortion..

Notice at 2KV the 3-500Z will do -38 DB IMD = 125 watts carrier (500 W pep.)    Perfect for my application!  I have a few 3-500Zs, so in luck.  The sweet spot for IMD vs: power is about 1800 VDC.  Like running Johnson Valiant power.

Basically I am taking my old "Mr. Clean" 4CX-350J linear and replacing the grid driven 350J with the  GG 3-500Z and associated circuitry. I don't use the 350J anymore, so might as well convert it over.   Mr. Clean II, I guess.

Here's the data sheet - look at the 2nd page.  see Single tube -  "Minimum distortion products at 1500V."   (and 2000V.)

https://frank.pocnet.net/sheets/088/3/3-500Z.pdf

I'll post some construction pictures of the project soon.

T

[Opcom]

The 1980 data sheet has also some interestingly low IMD at higher outputs. Strange there are that many differences between the 1968 and 1980 data. The tubes look slightly different in the plate. Are they the same as in interchangeable?