Bass attenuation with bypass caps vs coupling caps in cascading gain stages

[Merlinb]

I was just thinking it would be less confusing if all these capacitors had unique names.

I think so too. That's why I refer to the power supply caps as the reservoir cap (first one) and the smoothing caps. Then you have coupling caps (obvious), and the cathode bypass cap (it has the word 'cathode' in it see?). For guitar amps I also reserve the name 'bright cap' for one placed across a volume pot. The words 'decoupling' and 'filter' refer to general principles and can be applied in many situations, including all of the above (although I generally think of 'filter' as referring to as a signal-frequency shaping circuit like a tone stack, rather than power supply caps *shrugs*).

[Chuck H]

Yup. Most of the time we (the collective, but reading posts as individuals) need to take certain terms in context. As in "bypass cap". If the thread is about a tone shaping high pass "filter" then bypass cap could be (and probably is) a capacitor parallel to a series resistance in the signal chain. If the thread is about a cathode circuit... etc. I'm guilty of using half terms that I expect to be taken in context. It's not usually a problem with "old timers" but I have seen some cases where newer members or less tech posters do become confused because their vocabulary IS term specific. As in, to them "bypass cap" means high pass filter OR a preamp cathode bypass circuit, but not either.

The trouble here would be coming to some agreements on naming components for specific application that are also electronically accurate and exclusive. And then chastising each other repeatedly until we're all on the same page. And then chastising walk in and occasional posters for not knowing "our" language.?.

Probably easier to just continue trying to take terms in context and ask when unsure.

[elipsey]

Thanks for the plots, that makes total sense (and I think unless I'm mistaken is pretty much what I described).

[elipsey]

Probably easier to just continue trying to take terms in context and ask when unsure.

Seems like that worked out ok here. As a beginner, I might as well ask and be wrong because I'll probably be wrong pretty soon anyway.

[elipsey]

Why I asked...








EDIT: I guess I should have explained better. I got interested in this because I tried two different ways of added a channel switch, and I thought it was interesting that they made the high-gain mode sound a little different.

[pdf64]

I suggest the 1st version, with the cut from the coupling cap; though neither may cut much low end.
Resistance Frequency Capacitance Calculator

The standby implementation is nasty, as it 'hot switches' a capacitive load on to the 5U4; better to put it after the reservoir, leaving the reservoir permanently connected to the rectifier, to avoid stress to the rectifier from high charging current surges. Though its voltage rating may need to be increased.
But traditional standby on the B+ line is pretty poor anyway, eg switches are used well above their rating.
With a fixed bias amp and the smooth B+ ramp up provided by a tube rectifier, it offers no technical benefit; some users like the mute function but there are much better ways to achieve that, the most simple being to pull the input jack plug out a notch, allowing the switch contact to short the input to ground.

With high dissipation cathode bias, I think there's a argument for a standby which cuts the power tubes off; that (and a near mute) can be achieved by lifting their cathodes.
Merlin has some other options The Valve Wizard

[olddawg]

I would suggest considering just getting a smaller amp. A lot of good options exist. Those old Marshall amps sound great for what they do. Unfortunately it not 1985 anymore. If you are clubbing you will drive the crowd out of the room nowadays. IMHO.. There is nothing that you can do that will make it sound as good at a much lower volume. Get a smaller amp, run it through a single 12" speaker, and save your back and hearing.

[elipsey]

Ok, I guess I should have explained this better. I wanted to discuss, in general, the relative merits of these two different ways of attenuating bass in the first two gain stages, not try to fix an amp.

I wanted to replace the traditional low gain input in an 800 style preamp with a virtual “channel” switch that would switch an extra gain stage in and out of the preamp. I thought the low gain “channel” would sound better with a cathode bypass cap and normal first stage biasing. I guess this kind of thing is pretty common; some makers use a “decade” switch, etc.

It seemed to me that the obvious way to design the switch was with a DPDT switch that dedicated one set of poles to switching the signal path, and the other set to switching the full cathode bypass cap in and out of the first stage. I tried it, and it worked pretty well (except for popping a little). The low gain “channel” sounded way better to me than the traditional 800 style low gain input that I had before.

Then I noticed that if I added a second coupling cap to attenuate bass between stages and used a fixed full cathode bypass cap, I could do the same thing with an SPDT switch, which is neat. But when I tried it, it didn't sound quite as good in high gain mode. So that got me interested in the theoretical questions that I asked in this thread.

I didn't post any details about my amp to begin with because I wanted to have a discussion about theory, and I though it would make things more confusing. And it kind of did.

But if we're going talk about actually fixing stuff, I would like to know how to add de-popping resistors to the switches, and how to make the high-gain mode a little less noisy

[guitician]

I thought that the 250uf bypass cap on the Bassman/JTM45 was to shunt cathode hum to ground, if hum is the noise your referring to.

[elipsey]

I thought that the 250uf bypass cap on the Bassman/JTM45 was to shunt cathode hum to ground, if hum is the noise your referring to.

Good point, I never thought of that. I guess I thought that was in the Bassman to allow very low frequencies for bass players, and Marshall just left it in, but maybe it was there for noise too. They put a way smaller first stage cathode bypass cap in the JCM800, which is way more prone to noise, so if noise was the reason for using 250uf, I guess they changed their minds later. Anyway, I'll try it and see if it helps.

[guitician]

I think Fender originally used Western Electric circuits and that is were it came from. I remember reading a very old tube textbook that explained using large bypass caps for hum reduction due to the filament supply 60hz noise. A DC filament supply would eliminate the 60hz too.