I was downloading some of the Radiotronics AU magazines, and paid particular attention to an article on the 7027 tube. Included in the article was this schematic for a PP amp using 7027s. I'll make the bold assumption this was an RCA design as the company behind the Radiotronics magazines seemed to be an RCA importer/rebrander for Australia. I have seen other, complex, RCA PP designs but never one with this form of cathode bias, nor do I recall ever seeing it elsewhere. Can anyone, far more knowledgable than myself (which means most of you), explain the purpose of the voltage tap off the screen supply onto the output cathodes?
In the Marantz C8 adds B+ ripple into the signal path. There is no DC component.
Here we have R28 and R29 adding significantly to the cathode bias. It looks to me like an attempt to add some "fixed bias" to the cathode-biased tubes. I don't know what C11 and C14 do.
Since cathode and anode are 180 degrees out of phase C11 and C14 will bring a vary stable operating point. It comes at the expense of over 20mA per channel though.
How exactly are C11 and C14 bringing a "very stable operating point"? They are injecting the same ripple in both cathodes, which cancels in the OPT anyway. How does that make anything better? And what does that have to do with the added DC bias?
Well, I was thinking about the charging of the cathode decoupling cap at loud program, shifting bias.
Fixed bias AND/OR cathode bias work the same way. They make G1 more negative than the cathode. You can do that by making G1 more negative or K more positive.
The difference between fixed and K bias is that... well... one is fixed and one is not.
The difference between fixed and K bias is that... well... one is fixed and one is not.
Some super-similar plans were published in the RCA Tube Manuals of that decade.
They are frankly not much like the designs of any of RCA's commercial products; nor those of major tube amplifier companies. They are somewhat like the enthusiastic amateur designs published in AUDIO et al, but not exactly. I have always assumed these were home projects of an RCA employee, not company designs, but published to fill-out the Tube Manual.
There is about 67mA in the idle tube and 23mA in the voltage divider. We could say the bias is 3/4 self-bias, 1/4 fix-bias. This does not significantly improve the bias stability. It may marginally suck-down the idle B+, though there is no good reason to do so. I suspect it is left-overs from prototype experimentation. Try this, try that, try the other thing, and some of these changes were never tested and removed.
The capacitors are correct. The cathode gets a heavy bypass to B-. The screens should generally be bypassed to *cathode*. With the cathode already heavily bypassed, the screen return may not be critical. But the way shown is correct.
> say, just one 90uF to ground.
From screen? From cathode? Both want bypass. Two caps.
The 40/50 difference of value is an artifact of the vast difference of voltage rating and commercial catalogs.
From screen? From cathode? Both want bypass. Two caps.
The 40/50 difference of value is an artifact of the vast difference of voltage rating and commercial catalogs.
PRR: I'm not sure what part of my post you don't understand.
Leave the screen alone. I didn't mention it and it's not bypassed. If you argue that it should be bypassed, fine, but it has nothing to do with this thread.
Each cathode has 2 bypasses, i.e. C11 and C12. One to ground and one to B+. What does this accomplish and why is this "correct" as you put it?
Leave the screen alone. I didn't mention it and it's not bypassed. If you argue that it should be bypassed, fine, but it has nothing to do with this thread.
Each cathode has 2 bypasses, i.e. C11 and C12. One to ground and one to B+. What does this accomplish and why is this "correct" as you put it?
The AC current in the PP OT primary is largely balanced by the AC current in the secondary while the DC current is largely balanced by being in opposite directions in the two legs. PS ripple intermodulation does not cancel in the same way that ripple itself does.
Has IM impact on class AB tube crossover distortion and switching noise resulting from OT leakage?
Has IM impact on class AB tube crossover distortion and switching noise resulting from OT leakage?
I'm not sure what the term 'by-passing' exactly means in this discussion, but I suppose the same as 'decoupling'. I have no answer to the question of TS. I just want to mention C5 (40 uF) in the supply rail. This capacitor is (already/also) by-passing/decoupling the screens to ground.
That's a novel definition of decoupling. I have not seen that before. By the normal definition, C5 is on the wrong side of the screen resistors. So you claim that C5 decouples all the screens of both finals and driver tubes?
Ok, suppose I agree to that definition. Now how would I have to change that circuit so that the final tube screens are NOT decoupled?
Maybe I use a wrong term/definition.
The two resistors on the screen resistors of the finals are grid stoppers (100 Ohm) so I would think there's hardly a wrong side of the screen resistors in the context of decoupling.
The screens of the drivers are being fed over R12 (200K). I would not call the screens of the drivers decoupled, just like I would not call the anodes of the drivers decoupled.
It was and is not my intention to steer this thread into discussions about definitions. Like TS, I just like to know how the circuit around the cathodes of the finals works.
The two resistors on the screen resistors of the finals are grid stoppers (100 Ohm) so I would think there's hardly a wrong side of the screen resistors in the context of decoupling.
The screens of the drivers are being fed over R12 (200K). I would not call the screens of the drivers decoupled, just like I would not call the anodes of the drivers decoupled.
It was and is not my intention to steer this thread into discussions about definitions. Like TS, I just like to know how the circuit around the cathodes of the finals works.
Last edited: