Real TrainWreck mojo!

[Enzo]

500 printed on a cap CAN mean 500pf. it can also be read 5-0-0, in the format of a color code, so meaning 50pf. If someone draws a diagram from looking in the amp, if they see a cap marked 500, that could explain the two versions. personally in a tone stack, I;d believe 500pf before I'd believe 50pf.

[Silvio55]

This is a correct schematic of Francesca, from the amp garage.

[loudthud]

Most TW clones use a center off mini toggle to select 50pF or 500pF. As I mentioned in an earlier post some clones add a stopper on the second stage. Since there are slight variations in the amps Ken made, and each has a female name given to it by Ken, the name frequently appears on the schematic. "Kelly" is one such amp. The Express I made shows not ripple at all on the clipped output and the sides of the X-Y (speaker V & I) display are very steep meaning the tubes are conducting hard with low voltage drop (and high screen current).

[MarkusBass]

Most TW clones use a center off mini toggle to select 50pF or 500pF.

This is not clear for me. Do you mean TREBLE, or BRIGHT capacitor? Such a switch is on the schematic for BRIGHT cap but it switches between 100pF, 500pF, or off.

I simulated the tone stack and I've got some strange results. I checked 50pF, 250pF and 500pF values for the capacitor and 1M and 250k for the TREBLE pot.
The results can be seen below.
50pF + 1M looks very bad. This is unlikely that this set of components were used.
500pF + 1M looks also bad. Take a look.
500pF + 250k looks much better. Can anyone confirm (or show a photo) whether 1M, or 250k was used as the TREBLE pot?
250pF + 250k also looks good but maybe not that goo as 500pF + 250k.
If I were to build this amp, I'd use 500pF +250k. What do you think?
I haven't checked the 500pF BRIGHT cap but from my experience 220pF is already too much (at least with Stratocaster) - it makes the amp more noisy but does not change the sound of the guitar. Maybe with LesPaul it makes more sense.

Mark



BTW, for the 50/500pF problem I have a different theory. The cap was marked on old Marshall schematics as ".0005". What we have here is almost Marshall tone stack (combined with Fender). When you see ".0005" you have to guess what are the units (uF) and what is the multiplier when there are three zeroes before the digit 5. Someone just couldn't decide whether it is 500pF, or 50pF. I've seen similar mistakes in schematic reading in other amps.

[Silvio55]

I´m pretty sure the treble is 500pf+1M pot, there´s many gut shots of several Express amps, Francesca is the best documented, and all componets values, I assume, were measured. There´s a myth that Ken Fischer tweeked each amp differently, that´s true, but the tweeks were mostly lead dress, tube selection, adding gridstoppers in the second and even the first stage, using shielded cable for input and/or volume pot, made for making the amp stable, the circuit never really changed.

[MarkusBass]

It's funny - I checked few schematics and it seems that the tone stack version with 50pF and 1M TREBLE pot is almost a direct copy of Vox AC30 tone stack. This is strange because the simulation results look bad. The pot acts as a TILT control - it changes both treble and bass contents (most of designers try to avoid such situation). The treble contents changes start already at 200Hz. But, if it works in AC30, it should work here. So it looks like the schematics from Steve are correct. Only in some clones of Trainwreck there was a switch that was switching either 50pF, or 500pF (as a TREBLE cap). The original Trainwreck had only the BRIGHT switch. Silvio, are you sure about the 500pF capacitor?

Mark

[Silvio55]

Yes, absolutely sure (now that I found the info), the cap was measured in Francesca and the actual value is 525pf in that particular amp.

[pdf64]

'50pF + 1M looks very bad.'

What's wrong with it?
Flipping between the FMV stacks in TSC seems to indicate that it's within the norm.
I agree that 500pF - 1M would be too much.

[Tom Phillips]

Re: "...looks very bad" That may be true but as we all know it's what it sound like under real playing conditions and in conjunction with all the other parts of the guitar amp sound system. I can't say which parts values that are under discussion are correct but I do believe that we cannot rely on frequency response curve shape of the tone stack to make a final determination. It is an interesting piece of evidence in the analyses process though. A traditional amp designer would probably be horrified after looking at the waveforms that end up producing both classic and modern guitar sounds.

[MarkusBass]

'50pF + 1M looks very bad.'

What's wrong with it?
Flipping between the FMV stacks in TSC seems to indicate that it's within the norm.
I agree that 500pF - 1M would be too much.

For me what seems to be wrong is that sweeping the TREBLE pot (other pots were not changed during the simulation) changes the frequency response even at @100Hz (by 5dB). Usually, you would expect that changing the TREBLE pot value does not change low frequency response (especially 100Hz). But this is the way Vox tone stack works. You can check it easily with Duncan Tone Stack Calculator. At @50Hz you get 8dB change (by sweeping the TREBLE pot only). That's why I said that it looks bad. But this is the way AC30 is designed and we all know that it sounds quite good.
I think that it would be beneficial to compare TW to AC30. Maybe it's the same .
If you look at various TW clones at ampgarage, you'll notice that the values in the tone stack were tackled and quite often changed to values that were used by other manufacturers. It means that people were not happy with the frequency response of the amp (or they were getting oscillations). You can also see this by looking at the 50/500pF problem. I believe that initially it was 50pF (copied from AC30) and later changed to 500pF. This, together with 500pF BRIGHT cap, causes that the gain of the amp @5kHz is by 30dB greater than the gain at @100Hz. This is huge difference an every designer should be worried about possible oscillations in this case. It is possible that together with missing grid stopper resistors this is a reason of oscillations in this amp. The solution should be quite simple but requires some analysis.

EDIT; Tom is right; the final test is how the amp sounds in real playing conditions - with a band. Simulation is just a basic verification whether the amp works. And what is "good", or "bad" sound can be judged by musicians and listeners of the band that uses the amp.

Mark

[Enzo]

Well that's just TERBLE, innit?

[Chuck H]

Yo, Mark.

The thing to understand is that the tone stack in the Express/Liverpool amps isn't there for EQ control in the classic BMT sense, but rather different tone and response shaping. The amps are designed to be overdriven and the tone stack is placed BEFORE any clipping. A classic, Fender style BF tone stack would sound like, well, a Fender BF being cranked. And that's not what the TW amps are designed to sound like. With the TW amps clipping, any effect of the cut/boost on BMT is lost anyway. So the TW stack is designed for shaping useful frequencies under clipping conditions. The point is simple. The tone stack is being used to setup the signal for clipping and NOT to cut/boost BMT specifically. This is admittedly clumsy as a tone stack if you want to run the amp clean. Since that isn't typically done it doesn't matter much. TW amps are typically cleaned up using the guitar volume control to reduce gain and the amp settings remain constant.

As to the grid stopper... The TW amps DO NOT use an input grid stopper. That isn't to say none were ever built with one, but it's not part of the basic design. For better or worse. Sometimes Ken would use a small value stopper at the grid of the second stage for his own reasons. The power tubes use 1.5k grid stop resistors.

The basic Express/Liverpool bright switch is a three way. 50pf-none-500pf. It's designed to accentuate only very high frequencies for cleaner playing (50pf), nothing (for???) or to boost all the way down into the upper mids for improved overdrive character (500p, most players use this setting). The amps, when tweaked properly, are said to be somewhat dark when clipping. By design or component selection (speakers and OT mostly I think) the very bright nature of the design is brought under control so that harmonics are achieved without too much harshness. What looks to be a design that boosts treble all the way down into the mids is really, in the final picture, boosting mids. While the extra treble in the circuit isn't so much heard as felt. Adding a certain character to the pick attack, accentuating harmonic responsiveness and making the amp feel lively even though it isn't bright. If you've heard clips you no doubt know that they also have a limited and tighter bass than most "vintage" amps. The Express/Liverpool amps could definitely be described as mid heavy. They sound like they sound and if you want to boost any frequency you're out of luck. They just sound like they sound.

Lamenting what the design DOESN'T do is rendered moot because they do one thing VERY well and that is THE reason to own one. It has NOTHING to do with versatility or an amp that should be better at everything than any other. The amp is a one trick pony. But for some players it does that one trick so well that they don't care what it doesn't do.

[Silvio55]

As to the grid stopper... The TW amps DO NOT use an input grid stopper. That isn't to say none were ever built with one, but it's not part of the basic design. For better or worse. Sometimes Ken would use a small value stopper at the grid of the second stage for his own reasons. The power tubes use 1.5k grid stop resistors.

Chuck, Ingrid, a '92 Express, has an input grid stopper and a second stage grid stopper, but is the only TW amp I know that has an input grid stopper.

[Gaz]

Chuck, Ingrid, a '92 Express, has an input grid stopper and a second stage grid stopper, but is the only TW amp I know that has an input grid stopper.
[ATTACH=CONFIG]30984[/ATTACH]

Has anyone considered that perhaps these amps have been carefully laid out to have the LEAST amount of crosstalk/coupling/PFB as possible? I think Ken's layout would suggest that more than the pervasive 'secret parasitic circuit' theory. To me it looks spacious an careful. Maybe the bad sounding clones have a bunch of crosstalk when they should have none!

[Chuck H]

I've always thought the same. Looking at the layout it's always been hard for me to imagine what lead/s or component/s one would move to significantly affect crosstalk. Even the location of the transformers suggests nothing about this design is intended to hint at unstable. But when run as a whole amp it's a fairly high gain design and there ARE unstable examples of original Wrecks with some age or lack of maintenance. There are surely more unstable examples of clones. So there may be some "edge of stability" tweaking that lends to the tone simply because that's the best that was possible and the amps tend to instability if untweaked.