Found the Gomes 211-amp in an old l´Audiophile magazine. Reworked it (with ground reference on ct´d output transformer and input-transformer directly to the output tubes) as I want a class-A amp low impedance conversion 2500:8 ohm. As I see it there is no DC in the primary. But will it work in real life?
Interesting circuit, it's a difficult way to implement a PP amplifier.
It also creates problems in the PSU and input transformer. Both HV PSUs need to be floating from DC to at least 20kHz. This will require special transformers or common mode chokes to reduce capacitance to ground. Any capacitance to ground in the PSU will affect high frequency response. The full primary output voltage swing is on the input transformer so the input transformer secondary windings need sufficient insulation. Because the large output signal on the input transformer stray capacitance's in the input transformer may cause unwanted positive feedback causing the amplifier to oscillate at high frequencies.
Did the article in l´Audiophile magazine say what de design philosophy behind this amplifier was?
Corne
It also creates problems in the PSU and input transformer. Both HV PSUs need to be floating from DC to at least 20kHz. This will require special transformers or common mode chokes to reduce capacitance to ground. Any capacitance to ground in the PSU will affect high frequency response. The full primary output voltage swing is on the input transformer so the input transformer secondary windings need sufficient insulation. Because the large output signal on the input transformer stray capacitance's in the input transformer may cause unwanted positive feedback causing the amplifier to oscillate at high frequencies.
Did the article in l´Audiophile magazine say what de design philosophy behind this amplifier was?
Corne
can't see the point of the earth in the middle of the op transformer. the power supplies are floating, no current will flow down there.
seems a very complicated way to get a pp amp working, but hey, what the hell.
as long as the bias resistors are set up accurately to ensure similar currents, then yes these should cancel each other out therefore preventing dc saturation.
I imagine that noise might be something of an issue, the two psu's will probably have different noise signatures, these will add up over the transformer.
just some thoughts.
hope it works out for you.
let's know what happens.
best wishes
bill
seems a very complicated way to get a pp amp working, but hey, what the hell.
as long as the bias resistors are set up accurately to ensure similar currents, then yes these should cancel each other out therefore preventing dc saturation.
I imagine that noise might be something of an issue, the two psu's will probably have different noise signatures, these will add up over the transformer.
just some thoughts.
hope it works out for you.
let's know what happens.
best wishes
bill
The original intention was to bring down the, for 211s optimal primary PP-load (class A 35W), from 19,2k to 4,6k.
They did not ground the CT, instead they had a capacitor of 4,4u to ground.
Also they had an interesting modified SRPP as interstage between input transformer and 211.
They did not ground the CT, instead they had a capacitor of 4,4u to ground.
Also they had an interesting modified SRPP as interstage between input transformer and 211.
This is roughly speaking a transformer coupled variant on the circlotron concept and given the original purpose would work quite well. I would not say that the technical challenges to be overcome make it a great choice for the KT88 however.
The floating power supplies are in the audio path and need to exhibit low series impedance to the audio signal and very high common mode impedance to the outside world. This is not a big issue in low impedance circlotron designs, but getting decent HF response here could be very problematic. Power transformers should have split bobbins to reduce primary to secondary capacitance as much as possible, and high inductance common mode chokes employed after the initial rectification and filtering have been accomplished - with a large very high quality cap across the output side of the choke to provide the high quality series audio current path required. IMO This is more trouble than it is worth for anything other than big DHTs - and even then I think there might be better ways.
With the KT88 it's another whole matter, a conventional interstage transformer driven output stage would get you where you want to be power wise with a single 500V supply and more or less with parts considered to be off the shelf. (Llundahl and MQ likely sources amongst others.)
The floating power supplies are in the audio path and need to exhibit low series impedance to the audio signal and very high common mode impedance to the outside world. This is not a big issue in low impedance circlotron designs, but getting decent HF response here could be very problematic. Power transformers should have split bobbins to reduce primary to secondary capacitance as much as possible, and high inductance common mode chokes employed after the initial rectification and filtering have been accomplished - with a large very high quality cap across the output side of the choke to provide the high quality series audio current path required. IMO This is more trouble than it is worth for anything other than big DHTs - and even then I think there might be better ways.
With the KT88 it's another whole matter, a conventional interstage transformer driven output stage would get you where you want to be power wise with a single 500V supply and more or less with parts considered to be off the shelf. (Llundahl and MQ likely sources amongst others.)
This verifies what I suspected. The original design is working at 1230V with power supplies with one large can CSI capacitor of 32uF.
I guess the differance against Circlotron must be that this one has gain and the circlotron has the gain of 1x.
And do not worry, will not build it. I am at the moment busy with a low-voltage E130L, PP triode-amp.
I guess the differance against Circlotron must be that this one has gain and the circlotron has the gain of 1x.
And do not worry, will not build it. I am at the moment busy with a low-voltage E130L, PP triode-amp.
revintage said:This verifies what I suspected. The original design is working at 1230V with power supplies with one large can CSI capacitor of 32uF.
I guess the differance against Circlotron must be that this one has gain and the circlotron has the gain of 1x.
And do not worry, will not build it. I am at the moment busy with a low-voltage E130L, PP triode-amp.
As a design exercise it was certainly worth the time and effort, and who knows something you learned in the design and analysis of this circuit might prove to more than useful some day.
I designed and built an overly elaborate circlotron amp around 6C33 ten years ago, I wish this forum had existed at the time, and I also lament that spice was not more user friendly (not to mention readily available to hobbyists) - would have saved me considerable effort not to mention money if I had not built the thing in the first place. All I got was a taste of the promise, problems were too numerous and the distraction of a nice sounding SE amplifier design that popped into my head shortly after this debacle meant I never pursued it further..
I am at the moment busy with a low-voltage E130L, PP triode-amp.
Sounds like an interesting idea - care to elaborate, please ? I understand that this tube is very limited in the maximum g2 voltage, so what do you consider low voltage ?
MRupp said:
Sounds like an interesting idea - care to elaborate, please ?
Hi Martin,
Please mail me!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Tubes / Valves
- Gomes amp