Marshall JCM 800 2000 250w question

[uneumann]

I'm looking at the schematic of this preamp (below) and I see the lower channel shows V1b output DC coupled to the next stage. The upper channel show a coupling cap between similar stages.

Does anyone knows if this is truly the way they're built or whether a coupling cap is just missing from the bottom channel schematic?

I've been looking for DC coupling between CC stages. There are ways to do it, but this is the only commercial-amp instance I've found so far - and it could just be a drawing mistake.

Your help is appreciated if you happen to know...

[J M Fahey]

Drawing mistake.

[uneumann]

Drawing mistake.

Thanks Juan...

That means I'm still batting zero on finding dc coupled CC stages... if anyone runs into one ... pls point me to it.

[g1]

The upper channel is the same, but has it's coupling cap, C24. So that's the omission.

[J M Fahey]

Here's one, at least the input stage:

www.sowter.co.uk/pdf/Williamson%20Amplifier.pdf

[uneumann]

Here's one, at least the input stage:

www.sowter.co.uk/pdf/Williamson%20Amplifier.pdf

Ahhh - thanks - that makes sense since the second stage is a cathodyne PI. The high anode voltage is actually what you want for the grid of the second stage.

It's interesting, I don't want to read too much into it, but the hifi guys seem to have worked with direct coupled CC circuits, but not the guitar amp designers. The Morgan Jones book has a nice discussion on page 212 about ways to shift the DC component from the anode of the first stage to the grid of the second. It's probably never been needed in guitar amps, but when you have high-gain preamp stages, it seems a good method for controlling bias shifts and blocking. I've looked at all the high-gain circuits I can find and there's no trace of it. I thought the Marshall circuit was an example. Actually, I think their circuit is close to working for DC coupling since they have a level shifter between stages. Just another idea to try sometime...

[mhuss]

Check out the classic early Hiwatt final pre stages, all DC-coupled.
http://hiwatt.org/Schematics/DR_Pre4Input_v1a.pdf
The tube before the PI is actually just a constant voltage reference for the PI.

[uneumann]

Check out the classic early Hiwatt final pre stages, all DC-coupled.
http://hiwatt.org/Schematics/DR_Pre4Input_v1a.pdf
The tube before the PI is actually just a constant voltage reference for the PI.

Thanks - that's interesting too. It's a CC amplifier DC coupled through a level shift to a CF, which is DC coupled to the LTPI.
Nice... the only cap is the 22nf to bypass the DC level shift resistor, and since it feeds the CF there is no blocking problem there.

This is really clean - as in minimal - and it makes me wonder why it's not done more often.
Fewer parts (caps), fewer problems with bias shift/blocking, more reliable (no leaking caps) ... why not dc-couple more often?

[PeanutNore]

more reliable (no leaking caps) ... why not dc-couple more often?

AC coupling between stages is almost always done with film caps which do not leak and will never need to be replaced except in the event of some catastrophic failure. Film caps with the right voltage rating might be the most reliable components in an amp.

[potatofarmer]

This is really clean - as in minimal - and it makes me wonder why it's not done more often.
Fewer parts (caps), fewer problems with bias shift/blocking, more reliable (no leaking caps) ... why not dc-couple more often?

Cynical answer - because 99% of guitar amps are copies of copies of copies of RCA schematics and a lot of the market is catering to bedroom players and bar bands who want to sound like [FAMOUS GUITARIST NAME] on [HIT RECORD], and that guy used a [GUITAR] through an [AMP].

Less cynical answer - because people like tone and volume controls. AC coupling allows for quite a bit of bandwidth shaping, most notably bass attenuation. Also I suspect AC coupling is probably easier, though after a dozen rounds of tweaking a high-gain preamp to break up just right it probably breaks even. There's definitely math involved though; that's not helping either.