LS.
I’ve been happily building this and it sounds great, but it has a little bit of hum and I can’t get rid of it, it must be something I don’t know/understand yet.
The basic idea is a self-splitting output stage with plate to plate feedback from the driver.
At first I just directly grounded the grid of the bottom 6550, only resistor the gridstopper. Hummed. Added the 220K ground leak. Same result.
Added the feedback resistor and cap as in the schematic. Still hum.
Spent lots of time looking for ground loops, no result.
Key thing is: the amp is dead quiet with the feedback resistor disconnected.
What am I missing?
Thanks in advance.
I’ve been happily building this and it sounds great, but it has a little bit of hum and I can’t get rid of it, it must be something I don’t know/understand yet.
The basic idea is a self-splitting output stage with plate to plate feedback from the driver.
At first I just directly grounded the grid of the bottom 6550, only resistor the gridstopper. Hummed. Added the 220K ground leak. Same result.
Added the feedback resistor and cap as in the schematic. Still hum.
Spent lots of time looking for ground loops, no result.
Key thing is: the amp is dead quiet with the feedback resistor disconnected.
What am I missing?
Thanks in advance.
I did find a small AC voltage measured between the cathodes, I should check my notes.
The power supply cap is 390 µF, seemed plenty to me, I'll measure ripple tomorrow.
The power supply cap is 390 µF, seemed plenty to me, I'll measure ripple tomorrow.
I don't think it does, the bottom part of the output stage is essentially a very expensive resistor as far as I can tell. Without the cathode decoupling in the output stage, it would be a (sort of) long-tailed pair.
Last edited:
That was the reason for my question, I can't understand how it works, and at the moment don't have good Internet access to check. I'm familiar with LTP-style output stages, or using the screen signal of one tube to feed the second one, also auto transformers as phase splitter, but can't make this circuit work as a phase sitter in my head.
Last edited:
I have seen attempts using push-pull transformers in a single-ended setup, where the second tube has no signal, and it's only there to compensate the idle DC component flowing through the primary.
Can someone explain to me the advantages/disadvantages of this topology?
I tried to figure this out https://dalmura.com.au/static/RH April 1942.pdf but couldn't.
I tried to figure this out https://dalmura.com.au/static/RH April 1942.pdf but couldn't.
I see it as the output stage looks like jlh69 or burning amp but i can't understand the process.
I think that's what it is doing. I think it could also largely cancel out the transfer from the supplies to the output when that 33 kohm from post #9 is added.
Then you must remove the common cathode capacitor 120 uF. Otherwise the self splitting does not take place and the other 6550 is just idling.
At present your amp is single ended.
Just can't see a reason to do something like this. What is the purported advantage to be gained? The disadvantages are obvious. Interesting though!
All good fortune,
Chris
All good fortune,
Chris
Theoretically speaking for best splitting performance the cathode resistor should be as large as possible, in practice that would be met with a CCS, say a self-biased FET.
The idea is that the shared cathode current is fixed so when one tube is sucking up some amount of current, the other tube gets starved with the exact same amount, hence the signal becomes split.
Those two 10 Ohm cathode resistors are not really needed here, or could be decreased.
Make sure at least the input tube have a clean voltage supply.
The idea is that the shared cathode current is fixed so when one tube is sucking up some amount of current, the other tube gets starved with the exact same amount, hence the signal becomes split.
Those two 10 Ohm cathode resistors are not really needed here, or could be decreased.
Make sure at least the input tube have a clean voltage supply.
Time for a quick update, very busy this afternoon.
First: the ripple on the B+ is 4V p-p.
Second: the bottom tube does do something, the (practically) identical amplitude of the top tube in opposite phase of course.
The feedback on the bottom tube is baaaad for balance, it alsmost halves the amplitude on the bottom tube. Out with that.
Thanks for the input everyone, I'll try some of your suggestions next.
First: the ripple on the B+ is 4V p-p.
Second: the bottom tube does do something, the (practically) identical amplitude of the top tube in opposite phase of course.
The feedback on the bottom tube is baaaad for balance, it alsmost halves the amplitude on the bottom tube. Out with that.
Thanks for the input everyone, I'll try some of your suggestions next.
There is signal on the anode of the bottom valve because the primary winding of the output transformer works as an autotransformer.
Without feedback to the lower valve, my suggestion won't work, so you can skip it.
Without feedback to the lower valve, my suggestion won't work, so you can skip it.