5D8 Twin input hi/low switching circuit

[octal]

I just got a Johnson Monarch M-4 (a Canadian built amp) off of e-bay. After some blind alleys, I eventually figured out that it is all but identical to a Fender Twin 5D8, even the board layout. (Different volume pot values, different presence circuit values.... that's about it.)

Can someone explain the theory of how the high/low input jacks work? It's quite different from the usual Fender arrangement... the plates of the input tubes are all tied to B+ with 100K resistors, then there are 100K mixer resistors joined together. If you use one of the inputs, the grid of one of the two triode halves will be grounded, so I guess the signal will be attenuated by the mixer resistor plus the ra of tube with no signal going into it. Using the other input is what has me confused. The summing point of the two triode halves is AC coupled back into the grid... it almost seems like a positive feedback circuit.

Please explain!

Nathan

[octal]

This is the arrangement I'm talking about.

[loudthud]

It looks to me like when the lower input is used, the upper triode is connected like and inverting opamp with a gain of minus 1 and the signal is picked off at the summing junction. Since the open loop gain isn't very high, there is still some signal at the summing junction, but not much. No increase in headroom like with the Blackface attenuator. When the upper input is used the feedback loop is broken and much more signal gets through. Not Leo's best work, that's probably why it wasn't used more than one year IIRC.

[octal]

It looks to me like when the lower input is used, the upper triode is connected like and inverting opamp with a gain of minus 1 and the signal is picked off at the summing junction. Since the open loop gain isn't very high, there is still some signal at the summing junction, but not much. No increase in headroom like with the Blackface attenuator. When the upper input is used the feedback loop is broken and much more signal gets through. Not Leo's best work, that's probably why it wasn't used more than one year IIRC.

Thanks! I can only imagine the kind of havoc a dirty jack switching contact could wreak on this arrangement. I'm torn on what to do with this amp. It's super original... not a single part appears to have been changed since it was built, though someone did clip out all the input jacks. It kind of seems like a shame to mod it, but it's not a real Fender, it's a clone....but it IS from the '50s. The 5E8 twin has the same mixer circuit, but switched to running the triodes in parallel for the sensitive input and grounding the input of one of the triodes for the low sensitivity input... that makes more sense and doesn't have the screwy feedback loop deal.

[tubeswell]

So its grid-leak biased as well huh? (If so, that's probably another reason why you might not get too precious about not modding it)

[octal]

Nope, the preamp tubes are cathode biased. There's that 470 ohm resistor shared by all the input tubes.

[tubeswell]

Oh yes - missed that.

What do you think of the sound of it?

[octal]

It's got a lot of flexibility with the tone controls- much more so than the typical FMV tone stack. I think it's a Baxandall type arrangement? Cleans are nice, not overly sterile, the overdrive seems good, but it's a pretty loud amp, so I haven't had the chance to push it hard. I'm not really a guitar player- I just noodle- I'm waiting for my guitarist friend to put it through its paces. I posted a schematic under "schematic requests" (see the version at the bottom of the thread-my first version has errors)- the circuit topology is exactly the same as a 5D8 Twin, with some different values in the tone stack. I had to add a pretty substantial sag resistor on the CT of the power transformer as the HV was out of control... really high with our 125V mains voltage here.

[tubeswell]

If you were (heaven forbid) wanting an idea for a simple modification without changing the exterior cosmetics or the tube count, there is always this idea of Merlin's (which I have implemented in a couple of amps), that you could try on one of the input tubes, seeing as how the amp basically has two input channels that are (in my humble view needlessly) doing the same thing:



YMMV

[octal]

That's funny. All the input jacks were missing for channel one, so I wired it up like a 5e8 Twin, which is similar to Merlin's circuit... the triode sections are in parallel with the high sensitivity input and just a single section is used for the low sensitivity input, but the other unused triode is still in parallel which reduces the gain. I kept channel two stock. There is a _much_ bigger sensitivity difference between the hi/lo inputs on channel two than on channel 1.
I do agree with you though- having so many redundant inputs is wasteful. They're even all EQed the same on this amp. No "bright" channel here. I guess it may have made more sense back when it was more in vogue for multiple musicians to share a single amp on stage?