|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
96.255.150.2
In Reply to: RE: SRPP question posted by cpotl on March 01, 2015 at 10:53:51
Years ago there was an article in Glass Audio that made a strong argument for NOT bypassing Rk1 (Rk on the lower valve section), having to do with equalizing current flow through both top and bottom halves of the SRPP. I cannot recreate it chapter and verse, but here is another reference that does discuss the SRPP in detail and does also end up with unbypassed Rk1 as optimal. This author also talks about how to select Rk. In reading Broskie on SRPP, it seemed he never looked at the bypass issue; all his schematics show Rk1 bypassed by a capacitor, if I recall correctly.
Follow Ups:
"Years ago there was an article in Glass Audio that made a strong argument for NOT bypassing Rk1 (Rk on the lower valve section), having to do with equalizing current flow through both top and bottom halves of the SRPP."
Bingo! I've just tried it in the simulation, and that makes a huge difference, in the case when the load is tuned to the value that balances the antiphase currents in the two tubes. It looks like the 2nd harmonic can now become highly suppressed (80 or 90 dB down on the fundamental) if the load is tuned carefully enough. Of course this may be unrealistically good, since the simulation is a bit idealistic, but nonetheless it does show that there is a cancellation phenomenon now.
But it does, however, depend crucially on have the load impedance exactly right. I'll play around a bit more...
Chris
For what it's worth, back then I built an 807 PP amplifier with an SRPP input stage using a 6SN7. I listened to it with vs without Rk1 bypass, and it sounded much better without, which became the permanent topology.
Chris, I'm not seeing it. Removing the cap seems to improve the entire spectrum in line with the NFB it provides. Also when you increase the input signal to compensate for the lower gain, 2ndH is not much better.
What I do see is a vast improvement in high order distortion by unloading the OP as expected.
Naz
"Chris, I'm not seeing it. Removing the cap seems to improve the entire spectrum in line with the NFB it provides. Also when you increase the input signal to compensate for the lower gain, 2ndH is not much better."
I find that with the bypass capacitor removed, it is possible to make a very fine tuning of the load resistor that gives a very great reduction in the 2nd harmonic. A bit either side, and the 2nd harmonic distortion is much higher. 3rd harmonic shows no particular dip around that load impedance, of course. In fact 3rd harmonic and above seems to get better as the load impedance becomes large.
In my case, even after increasing the input signal to compensate for the reduced gain, I get figures like:
24K load: 2nd -95dB, 3rd -61dB, 4th -96dB
Infinite load: 2nd -52dB, 3rd -86dB, 4th --
But actually, that 24K critical load has to be very closely fine tuned to get the very low 2nd harmonic distortion. (In my case, it optimised at about 23.96K.)
I'm sure such low 2nd harmonic distortion is unrealistic, since the two tubes in the simulation are identical, etc. But it does illustrate the point that a fine balancing is possible. (But no evidence of such a balancing if the bypass cap is there.)
Chris
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: