Express Output Waveform Shots

[jazbo8]

No, it's only the duty cycle shift created in the PI that would work for any LTP PI driven with enough signal. It's the PI headroom reduction from the 1K screen power chain resistor sag that allows it to sneak under the power grid clipping on the wider side. Any other amp that doesn't have that screen resistor sag can still have the asymmetry (I believe this is happening in a Marshall as well, but won't say that for sure until I test one, but I've seen the output of one on my CRO before and it wouldn't surprise me given it's basically the same PI & power stage) but will stay power grid clipping dominant on both sides because there's no reduction in PI headroom. The asymmetry at the power grids will still cause movement there, but with no reduction in PI headroom it stays above the power grid clipping threshold. So it won't have the change in tone and reduction in clipped rms output that the mixed mode distortion gives in the Express. My friend's JCM800 that I've had my CRO connected to (set for clean preamp and higher master to overdrive the power section) had some asymmetry, but not as much as the Express, and the height on each side of the waveform was consistent with power grid clipping and flat topped on both sides. It certainly didn't reach a peak and then back off on the wider side like the Express does. And even if you add the 1K screen supply resistor, it won't have the same dynamic unless it's got that slowed response to the PI supply node from the PI supply chain 18K2 resistance and cap.

Ah, even more things to think about... I guess that's why KF's designs were well "tuned" machines. Ok, slightly off topic, why is it then, when the Express are cloned, they still do not sound like the real ones? People have gone to great length to duplicate the layout, the parts value, OPT, PT... yet still few (if any) could claim that they got the tone. If all the intricate interplay between the supply voltages could be duplicated, what else could be missing?

[katopan]

Good questions. Of course if a real one is different to a clone we might as well throw all that testing out the window. All these waveforms are of my clone which although follows the schematic, has my own transformers, new production EL34s and a layout that is different to the one published on this forum. Also keep in mind that different models of Express have been found with circuit differences and different tones. If originals don't sound the same as each other what hope have clone amps got!

Now if only I could get a real Express on my test bench in Australia for a week to compare.....

By the way, I'd like to personally thank you for following along Jazbo8. The initial interest was OK but if I didn't have someone (and that one and only person was you) taking an interest I would've found it hard to keep posting my test results. I did promise to reveal all and hope that I have done that. I'm sure (hopefully!) that there have been people reading with interest that haven't posted as well. But thanks again.

I'll get to testing different tail resistor values and confirming the B+ sag hopefully soon. If there's any other tests of interest to anyone please let me know.

[martin manning]

I think clones built to the original circuit and using similar parts will have something like the tone of an original example, maybe even very close to an original. The last few percent is part tolerances and lead dress. Glen K made a copy of his original and got close, and then after acquiring a second original found that the second one did not sound exactly the same as the first either. No surprise there, though.

Another question is after all if this detailed analysis of waveforms, how does one know which ones are important to producing the TW tone? In my mind it will be very difficult to identify which of the many are responsible, and perhaps even more difficult to prove the hypothesis.

[katopan]

Hi Martin. I set out to find out the mechanisms that were responsible for what I saw in the output waveforms, primarily the asymmetry and the weird looking clipping on the bottom half. I could see that combined this had a reduced rms output compared to a purely power grid clipped signal, and that it would help make the 'mean' perceived volume comparable to the 'clean'.

Tone is something completely different. I did do some plots of the whole amp frequency response with both a resistive load and reactive load. They were indicative but not accurate enough for any detailed information and so I didn't post them here. They were measured clean and basically confirmed the expected shape from overlaying the tonestack response from the Duncan Tonestack Calc with the 530Hz 2nF/150K high pass filter between the 2nd and 3rd stage. Distorted just gets way too complex with baseline tone shaping, harmonics and intermodulation distortion, so I agree that identifying and explaining the tone of an Express would be quite an impossible task.

I've certainly found this all very interesting. It's also a good resource to have a collection of waveforms posted here for everyone to see.

[jazbo8]

Tone is something completely different... so I agree that identifying and explaining the tone of an Express would be quite an impossible task.

I've certainly found this all very interesting. It's also a good resource to have a collection of waveforms posted here for everyone to see.

I'm not sure why there have not been more participants in the discussion after you updated the CRO shots, perhaps they have already figured it all out I think you have made a great contribution to the community, and there is a wealth of data to explore and interpret, perhaps not now but later, we will get to the bottom of it...

When I said "tone" earlier, I wasn't referring to the frequency response in particular, but rather how the amp behave overall, with C2M, touch-ability, feel, etc. If GK's experience is anything to go by, then perhaps "different" is the best we can aim for.

[katopan]

Thanks. It does astound me that you and Martin are the only ones participating. There are pages of discussion here (I've read all I can find) on ideas to explain the behaviour of the Express. We now have something to actually base that discussion on and it all goes pretty quiet. Some forums can be funny like that. Maybe more will surface after people have had time to digest and mull it all over. It is a lot to take in at once.

Anyway, I'll post the final tail resistor change tests and B+ sag figures when I next get the change to open it up again.

[Colossal]

I've been reading this thread with great interest since the beginning. Excellent work. I'm also interested in the next iteration of your efforts Craig, changing the PI tail resistor.

[brewdude]

I can't participate in the discussion because it all goes over my head.

I am eagerly awaiting some sort of synopsis of what this all means. Is a conclusion eminent?

[katopan]

Thanks guys. Good to know some people have been reading with interest.

Synopsis.... mmm. Once I've done the tail resistor and B+ sag tests, plus anything else anyone thinks might be relevant, I'll have to put some thought into how to summarise my observations. There's a lot to condense down.

[matt h]

(deleted)

[jazbo8]

I don't have a wreck, so haven't really pitched in. What I imagine is that by changing the tail resistor, you'll see "the same thing." Mostly because I think you're already feeding it an asymmetric signal.

I thought the input signals of the PI are relatively symmentrical. Could you please point me to the scope shots that showed otherwise, I could have missed it....

It'll be interesting to see how it affects the insertion of the global NFB, as in that position it's doing double duty as a PI-balance-tail, but also a voltage divider for the "third" PI input and better distributing the gNFB from the "presence" circuit.

I don't imagine it'll change the nature of the amp a tremendous amount, and you'll see the same trends you've seen otherwise... but it is conceivable that it'll slightly change the relationship between the "clean-to-mean" point against the amount of compression displayed as the gNFB fades in/out.

Well, we did briefly touch on the gNFB a few pages back, the conclusion was that once the volume is over 2, with the output already clipped, the NFB collapses, so there is no longer interacting with the PI. And Craig also verified this by dis-connecting the NFB loop altogether - it did not make the C2M effect more or less prominent.

[matt h]

(deleted)

[katopan]

The NFB has the effect of linearising things more before power stage clipping than without it (just as you would expect). So it does have some influence in the clean to mean transition, but none in the symmetrical power grid clipped to the asymmetrical mixed mode distortion transition (also just as you would expect).

[Tillydog]

the conclusion was that once the volume is over 2, with the output already clipped, the NFB collapses, so there is no longer interacting with the PI. And Craig also verified this by dis-connecting the NFB loop altogether - it did not make the C2M effect more or less prominent.

That's about when I ducked out, because in my limited experience (theoretical and practical) collapsing NFB *is* the key to the clean to mean. Disconnecting NFB loses the abrupt clean-to-mean transition on my Express, so either my amp is a-typical or Craig's is (or both). I can appreciate the effort that Craig has put into this, and don't wish to de-rail his efforts.

If/when I get time, I'll try to make a comparison and come back with some info (but it's not likely to be anytime soon!).

Andy

[jazbo8]

That's about when I ducked out, because in my limited experience (theoretical and practical) collapsing NFB *is* the key to the clean to mean.

Andy

Perhaps I mis-spoke, what I meant was that the NFB can only be responsible for a part of the C2M - once the volume passes "2", the NFB collapses, which should no longer have an effect on the output, however as the waveforms show, the asymmetry continues to shift depending on the volume setting - perhaps that no longer qualifies as C2M but mean-2-meaner!

And from what I have read, most people leave the volume setting pretty high, perhaps 5-8, in any case well pass 2, with C2M controlled via the guitar's volume control. That's also a reason to rule out the NFB as the key ingredient for C2M.