Selmer Treble and Bass 50W MkII

[yeeshkul]

Hi guys, i am new to tube amps and i am trying to analyze the famous Selmer Treble and Bass 50W MkII (Syd barrett, paul McCartney, ...).

Here is the schematics:


It has separate preamps for each input mixed to the power amp via two 470k resistors.

I thought the distortion is created in the preamp, but the LT Spice simulation shows otherwise ...
I simulate the guitar signal by 300mV, 440Hz sinus.
The first stage boosts it to 13V, then the filter stage takes it down to 1V and the last stage up to 60V amplitude.
The anode of the last stage is sitting on 160VDC and the clean signal swings around.




Can you please tell me where is the distortion created?


Here is the sound

[J M Fahey]

Hint: Volume= 10 (TEN)

Pure power amp distortion glory

Drive your Guitar input , the high one, with your 300mV, 440Hz sinus signal, set all tone (and volume) controls to 10.
You checked preamp only; go 1 step further and check what happens on PI plates, then EL34 plates ... and on speaker, of course.

[yeeshkul]

Thank you. Actually, what is the pair of ECC83 before the EL34 pair for? I have seen this topology on other tube amps ...

[J M Fahey]

You need out of phase signals feeding the power tube grids.
That is called a "Phase Inverter" and this particular one does double duty: has two outputs, out of phase, and also adds some 20X gain.

Its output is very large, as needed to drive EL34 grids with some 30 or 40V RMS which drives *them* into heavy distortion once you go beyond a certain level.
It also clips/overdrives itself because preamp signal is LOUD if pushed.
And output signal also gets clipped at EL34 grids.

So hidden in plain sight you have *3* distortion makers

Mind you, they were not intended to by original designers and in fact up to clipping they are reasonably clean, but these crazy Rockers love to overdrive them to Hell.
Only problem is that by then amp is really LOUD, not a problem in the old days when Musicians tried to fill large halls with just the backline, PA was primitive at best.

[yeeshkul]

Thank you very much Juan, i get it now

[J M Fahey]

You´re welcome.
Now please simulate end to end, meaning including PI, power tubes, and speaker ... and post results

Not sure there are simulation models for Guitar speakers; if not, somebody should write them, their crazy impedance curves influence big time how those power pentodes work.

I fear standard simulators "think" speakers are 4/8/16 ohm resistors ... nothing further from the truth.

Just a free sample, this is what a Marshall 20W, in principle very similar to yours, looks like when overdriven:



here´s the distortion you couldn´t find earlier

[Helmholtz]

This thread :
https://music-electronics-forum.com/...ive+attenuator

should give some idea how to simulate speaker impedance in LTSpice.