1000V tubes, 1A supply

[Opcom]

I was asked to rebuild an old amplifier and convert it to 20M mainly for CW but some AM at lower power. It had 8 sweep tubes and is one of those horrible old MaCo amps. The sweep tubes are long gone but the power supply is pretty big and seems good for 800mA continuous at 900-1000V. I can not increase the voltage.

The cabinet is large-ish. I thought to gut it keeping the power supply.
The issue seems to be tube choices. Sweep tubes are too expensive for this junk box project.

I was thinking 811A's but 1KV may not be enough to get the most from them.
In audio service it says 1KV @ 350mA plate current, 240W output, in class AB2 per pair.

How about 6 of them in parallel? We have all see the "KW" amplifiers with 4 of them.
6 might make 740W at 1KV and 1050mA. -assuming 6 would work. It would still seem simple. He has some of them.

The thing may be a waste but the ham has little or no money just a few junk 'base' amps, and I am trying to help the guy out. Any better suggestions than 811's?

[w4bfs]

hi Pat ... 6 should work but will be a bear to tame with a upper band limit of 20 meters or so ... gain compression for an 811 at 1kV is definite unless you can find a way to get the tubes to share the load equally and not overdrive them ....  this is not the best choice javascript:void(0);

Does that sheebee amp have 2 hv transformers in parallel ?  putting them in series would let you run 2kv and put you in Ruskie GI7 territory ...would likely need a new tune cap ....

[W3RSW]

Parallel 4d32's at full scrote. As many as your supply can stand. Reduce loaded plate current somewhat for each to keep Pd in range.

Curious. Why can't voltage be increased.? If possible, change your supply to a voltage doubler, 2000v at 500ma. Then 813's reign supreme.

[N2DTS]

4d32's are interesting tubes.
I run three at 1200 volts and 300 ma, 300 watts out, 1200 watts pep with plate modulation.
They seem to loaf along in that service.

Four would be 400 watts carrier, 1600 watts pep, and 2kw on ssb would seem to be easy to do.
With the 4D32, you can run 600 volts and 200 ma, or 1200 volts and 100 ma, they work fine within the plate dissipation ratings.
They are small tubes (low).
At one time, new 4D32's were $18.00 each.


4x150/4cx250b type tubes are small and work at low voltages and high currents.
813's will work at 1000 volts and 200 or 250 ma a tube, but they are tall.

If you want to use triodes, a quad of 572b's might be ok.

[KD6VXI]

Pat,

I did one last year. . It was a Maco 750.  It had a pile of sweep tubes. . They where sold,  and the funds created a voltage tripler,  new Ctune and a bunch of other parts.

Said Maco now sports a single 500Z tube,  runs 40 through 10 and produces about 800 pep.

I have some power supply boards for sweep tube amps made up. Allows 6A10 diodes and modern day snap in caps.

I am currently doing a 10 tube amp. . Originally it was a golden Falcon 1 x 3 x 6.  Ran the tubes at 1100 volts. I'm keeping Cload and a bandswitch. Lol

You do need a doubler. Or,  some oddball sweep tubes that won't work above 15 meters so the Cbers didn't snap em up.

6550s maybe? . I've used one in place of a 6L6GC before...... . One of the new Svetlana branded ones. That would allow you to keep the power supply factory.

As to the capacitance that will kill you,  run an L-Pi network. . Compute Cout, then multiply by number of tubes. . Compute the L needed to cancel. . Run that between anode and Ctune.

--Shane
KD6VXI

[w1vtp]

I vote for the 4D32's  I've talked to Brett and his 3 x 4d32 and it is a sweet sounding rig.  Your PS would loaf along with that combo with room to spare for a modulator


Al

[W3RSW]

4d32's Push-pull parallel or triellel , Heh, heh, linked coupled might be interesting on 160 through 20 anyway.

[Opcom]

four 4D32's making 400W carrier? I've only ever got 100W per tube on carrier when it was plate modulated. but this is just an amp, no modulation to the plates.

The transformers seem to be already wired as doublers, I'll check again.

A pile of them might be OK, The L input to the PI is a good option for a bunch in parallel recalling the 813 linears for 10M in the handbook. Still, there could be stability issues I don't want to to deal with using 4D32's instead of grounded grid 811's.


Shane, do you have the schematic for the power supply and 3-500Z mods? There are no tubes to sell, kinda limited on converting parts to money on this. There might be a couple of old 3-500's around here. No socket though. It would be interesting to see that, but i don't want to have to change up a bunch of stuff too much to the really high voltage parts.

One thing I noticed reviewing the 811 amps, they all seem to run 1600-2000V. I guess that is what it takes to get a KW input from 4 of them. Wouldn't it be better to use 1KV or so, what there is, and use 4 or 6 of them to make a lower power but more durable amp?

[KD6VXI]

I built it as a straight amp,  scrounging what I could out of other stuff here. Tube and socket out of a Henry 1k.  Ctune from another amp rebuild. Used Cload from the original Maco amp.  It was like the chopper builders call a rat ride. . No schematic,  per se. .

Maco was good about not using a doubler. . Almost all their stuff was grid driven,  so the x former usually has a few voltages out,  instead of just Hv and fils.

I recessed the socket and used the smallest plate cap I could find
. Otherwise,  it sure was iffy,  the spacing from cap to cover.

I did have to shoehorn a filament x former in it. . Did sb220 style cooling,  using a fan. Mounted it on the read cover. I would NOT recommend this in the future. . Having to recess the socket and then not blowing a lot of air through or by it I figured would kill the tube. . It's still going,  though.  I did put 1 inch feet on the rear and inch and a half or three quarter in the front,  and their where a million holes in the chassis cover already...  I'm assuming it's getting enough via convection.

The tripler means it's a decent amp for ssb,  and a decent amp for cw. . As to am use,  hv dropped a bit,  iirc. . It worked,  but I also switched in another 15 volts of bias with the am position. It ran 7 (10 6A10 diodes)  volts in ssb / cw.

Anyway,  that's from memory. . I'll have to find the digital photos,  if they haven't been lost. . I'm great about that lol.

--Shane
KD6VXI