I don't see why you couldn't make an integrated mono, but functionally I don't like those. Having two individual volume controls that need to be adjusted correctly to keep things in balance I find to be very annoying. I'd much rather have a stereo preamp with one volume knob and a balance if I felt the need.
CA2010
Member
In Mr gillespe's schematic "the KT66 can still be used". I will try these as well as the 7027A. I don't currently have a 5AR4/GZ34 would keeping the original 5R4GY cause any damage? I suspect less headroom but they seem similar am I missing something important to the life of my power transformer?
CA2010
Member
In learning about bypass capacitors I read a response from Kegger:
Advice on B+ bypass caps
Discussion in 'Tube Audio' started by tsbentley, Sep 24, 2015.
https://audiokarma.org/forums/index.php?threads/advice-on-b-bypass-caps.681041/
"Don't listen to the naysayers.
Bypass caps are perfectly fine and warranted in better designs..
Basically what your trying to do with a bypass cap is make a large electrolytic a "better" capacitor.
If I'm going the extra mile I'll bypass with a 3-40uf poly cap then bypass that with a small Film/PIO cap.
Electrolytic capacitors are not perfect, they don't filter high frequencies as well, and tend to have high esr.
(bypassing them helps turn them into a better all around cap)
I'm not debating anyone, I hear what I hear and know I've learned over the years, works for me so I do it." - Kegger
Would this be where you might add the Kegger "extra mile"?
Advice on B+ bypass caps
Discussion in 'Tube Audio' started by tsbentley, Sep 24, 2015.
https://audiokarma.org/forums/index.php?threads/advice-on-b-bypass-caps.681041/
"Don't listen to the naysayers.
Bypass caps are perfectly fine and warranted in better designs..
Basically what your trying to do with a bypass cap is make a large electrolytic a "better" capacitor.
If I'm going the extra mile I'll bypass with a 3-40uf poly cap then bypass that with a small Film/PIO cap.
Electrolytic capacitors are not perfect, they don't filter high frequencies as well, and tend to have high esr.
(bypassing them helps turn them into a better all around cap)
I'm not debating anyone, I hear what I hear and know I've learned over the years, works for me so I do it." - Kegger
Would this be where you might add the Kegger "extra mile"?
There's much debate about this, I have no doubt. I have thought about it and Kegger's logic does make sense although I have no idea if it can be supported in testing.
I was testing some 40uf @ 450v caps last night and was really surprised to see how much the capacitance changed with frequency and how differently two different brands behaved. A Sprague Atom tested about 38.6uf @ 120Hz, but only 28.1uf @ 1KHz. I then tested a "Tung" branded cap of identical ratings, but physically less than half the size, and it measured 39.6uf @ 120Hz and 37.8uf @ 1KHz. In their intended application of filtering 120Hz B+ ripple, it seems likely they will perform the same, but if these two caps were in an audio circuit (directly passing the audio signal), I don't see how they could not perform differently and therefore also "sound" different. I didn't test for it, but I would be willing to bet that the inductance and series resistance measures significantly different between these two brands at different frequencies as well.
The argument against bypassing is that when you split the audio path between the two mismatched capacitors and then rejoin them, there's a possibility of "smearing" the signal. I guess that would be a timing/phase mismatch issue, it that's possible.
I'd say just try it and see if you can hear or measure any difference. It can't hurt anything!
I was testing some 40uf @ 450v caps last night and was really surprised to see how much the capacitance changed with frequency and how differently two different brands behaved. A Sprague Atom tested about 38.6uf @ 120Hz, but only 28.1uf @ 1KHz. I then tested a "Tung" branded cap of identical ratings, but physically less than half the size, and it measured 39.6uf @ 120Hz and 37.8uf @ 1KHz. In their intended application of filtering 120Hz B+ ripple, it seems likely they will perform the same, but if these two caps were in an audio circuit (directly passing the audio signal), I don't see how they could not perform differently and therefore also "sound" different. I didn't test for it, but I would be willing to bet that the inductance and series resistance measures significantly different between these two brands at different frequencies as well.
The argument against bypassing is that when you split the audio path between the two mismatched capacitors and then rejoin them, there's a possibility of "smearing" the signal. I guess that would be a timing/phase mismatch issue, it that's possible.
I'd say just try it and see if you can hear or measure any difference. It can't hurt anything!
Last edited:
You might be over my head with that, but by stiffening, do you mean increasing the capacitance of the stages so that high current transients could be serviced by that stages capacitance and not have to pull from the surrounding stages and the overall B+ sag as it deals with the demand? I haven't been able to test full power yet, but Dave said he was quite sure that this amp would have very little drop in B+ from quiescence to full power. I had asked him whether it would benefit from his EFB(tm) circuit and he said that it would not due to the stability of the B+.
Pio1980
AK Member
The elyticaps on the rails would do that acting as reservoirs, but may not keep the stages on the rail from affecting each other at higher freqs. The small film shunt caps are supposed to reduce the tendency for midrange/ hf crosstalk thru the rails, in theory. O'scoping the rails during bench testing would show practical effectiveness.
I've seen bypass caps on large electrolytics in power supply filtering. The reasoning as explained is to improve the noise filtering above the electrolytic's unaided frequency response. I can see how that could make sense in a power supply. But alternatively in a power supply, if HF hash noise is an issue, you could use a traditional snubber circuit. In the signal path, I rarely use electrolytic bypass caps, but when I do, I have never bypassed them with film caps, and the amps have worked and sounded fine without them. Honestly didn't give it much thought in using bypass caps on 'lytics in the signal paths.
I can see the potential advantage, but whether its actually audible or measurable is hard to say. Bypassing a cathode cap with a film cap is certainly not going to hurt it though.
I have used film caps across power supply caps though, especially in amps where there is a step network at the voltage amp across the plate resistor rather than to ground. That relies on AC grounding through the electrolytic, and if the electrolytic begins to degrade your filtering degrades with it. In cases where the amp can become unstable without the network, I figure the film cap is just some extra insurance. They're cheap enough.
I have used film caps across power supply caps though, especially in amps where there is a step network at the voltage amp across the plate resistor rather than to ground. That relies on AC grounding through the electrolytic, and if the electrolytic begins to degrade your filtering degrades with it. In cases where the amp can become unstable without the network, I figure the film cap is just some extra insurance. They're cheap enough.
What value cap would you use to bypass a 40uf power supply filter cap, for example? Are the high frequencies that we're trying to filter a product of rectification, and if so, wouldn't a single bypass at the first filter section take care of any high frequencies for all the remaining sections downstream?
Well, I just tried this on my Musician's Amplifiers and I like it. I had a bunch of 600vdc CD 940C series caps from an abandoned project so I applied them in "decades"-- .1 across the 10uF input cap, .47 and .22 across the others accordingly. Does that make sense? The amps certainly sound crisper--better attack on the highs. I think they'll stay.
I've noticed this approach a lot on the old Japanese single-ended designs. They'll use 40uF electro, then 4uF and .47uF bypasses, for instance. I don't think I have that much room--or money at the moment. ;-)
I've noticed this approach a lot on the old Japanese single-ended designs. They'll use 40uF electro, then 4uF and .47uF bypasses, for instance. I don't think I have that much room--or money at the moment. ;-)
Probably whatever I had surplus of in the 0.01 - 0.1uf range. I'm not sure the value is terribly critical but I'm sure thats up for debate. I know some people advocate "decades" of caps, so 1, 0.1, 0.01, 0.001. I really don't feel that to be necessary.
HF trash from rectification should be sorted nicely with a bypass at the first cap. If you're using step networks across the plate resistors like some amps use, then you'd want one at that stage of the supply too.
HF trash from rectification should be sorted nicely with a bypass at the first cap. If you're using step networks across the plate resistors like some amps use, then you'd want one at that stage of the supply too.
I tend to like a choke, to deal with HF rectifier hash. That sort of inherently takes care of that.
As for bypassing power supply stages connected to circuit stages with step (RC) networks- that COULD help, or it could also possibly change the HF phase/gain relationship to a degree that might degrade stability. It's all about the desired transfer function- it's not a one-size-fits-all situation...
Regards,
Gordon.
As for bypassing power supply stages connected to circuit stages with step (RC) networks- that COULD help, or it could also possibly change the HF phase/gain relationship to a degree that might degrade stability. It's all about the desired transfer function- it's not a one-size-fits-all situation...
Regards,
Gordon.
Well, I could 'scope the Musician's Amplifiers post-tweak and compare the stability to see if the bypasses affect them.
Are you talking about scoping the audio output, or looking at the B+ supply for any high frequency component riding on top of it?