Real TrainWreck mojo!

[jazbo8]

Chuck H - please allow me to go OT a bit, hope you don't mind.

Alan - first, let me apologize for starting this "pissing match" with you, it got a bit personal, for that I am sorry. But my whole point was when discussing feedback theory, it is generally accepted to use phase margin, by putting phase shift in bold letters as you did was both confusing and un-neccessary. Since you are constantly waving your EE background about, it's quite surprising to see commonly-accepted terms mis-applied or skipped altogether. I guess the last one just pushed me over the edge.

And as it has been pointed out elsewhere, the tone you use for posting are often condescending and patronizing, do we all really need to be lectured/schooled by you on a regular basis? You have used the excuses of "wanting to learn" to post on various threads, and many of them simply turned into long lecture sessions so you can show off your stuff, such as they are. If that's how you enjoy yourself, then more power to you, but just expect some push-back from time to time when people get fed up. I'm not sure if any of this will sink in for you, but at least I tried...

Nothing to see here, folks, move on...

[Alan0354]

Chuck H - please allow me to go OT a bit, hope you don't mind.

Alan - first, let me apologize for starting this "pissing match" with you, it got a bit personal, for that I am sorry. But my whole point was when discussing feedback theory, it is generally accepted to use phase margin, by putting phase shift in bold letters as you did was both confusing and un-neccessary. Since you are constantly waving your EE background about, it's quite surprising to see commonly-accepted terms mis-applied or skipped altogether. I guess the last one just pushed me over the edge.

And as it has been pointed out elsewhere, the tone you use for posting are often condescending and patronizing, do we all really need to be lectured/schooled by you on a regular basis? You have used the excuses of "wanting to learn" to post on various threads, and many of them simply turned into long lecture sessions so you can show off your stuff, such as they are. If that's how you enjoy yourself, then more power to you, but just expect some push-back from time to time when people get fed up. I'm not sure if any of this will sink in for you, but at least I tried...

Nothing to see here, folks, move on...

I have to say first that I came here at the beginning with bad attitude, have a know it all attitude. But since, I started to see the intricate side of guitar amp, I really try to tone down and be humble to learn. But I think the damage is done, I can just tone it down and talk.

This post and in amp garage took me by surprise, I look at post #92 in this thread, I truly don't see where I am condescending or trying to lecture. I was actually excited to join in about the instability and offer some of my experience in taming an instability loop ( which is positive feedback). Doesn't everybody use their experience and give opinion? I am by no means putting people down here.

regarding to phase shift, I guess the terms we use are different from in guitar amps. To me, a positive feedback is when the signal feedback is in phase with the input signal so the feed back signal is added to the original input signal and become bigger and bigger and result in oscillation. In phase is either 0 or 360 deg. So we always talk in terms of 360 deg to get +ve feedback.

[jazbo8]

Alan - thank you for your understanding and again I am sorry that I went over the line - it was a childish thing to do. Now to get back on topic, PFB does occur in guitar amps, but more often than not, they are not done by design, and I don't believe it was the case with the TWE. It's hard to see why KF would build the Express to be prone to oscillation, more likely explainations might be that when the users change the tubes (with different gain than the original tubes) or using hotter pickups that resulted in un-wanted feedback, etc. If the amps were somehow inherently unstable, then I doubt TWE would have gotten the reputation that it deserved, or else KF would be doing after-sales service all the time.

For me, it is more interesting to look at the circuit design and how it differs from the Fenders and the Marshalls, as done by Gingertube and katopan et al. While the component selection and wiring/layout no doubt also contributed to the "mojo", I still view them as secondary contributors, rightly or wrongly...

[Steve A.]

90 degrees is a right angle. And it is also quite warm.

Especially when its 90 degrees celsius!

I hadn't looked at this thread previously and I figured I'd just jump to the last page. Without reading anything other than Chuck H.'s original post here are a few observations.

The Kelly '90 schematic shows 21 caps and resistors in the preamp. Ken built his amps on perfboard and reportedly spent up to 40 hours fine-tuning the amp. What the heck was he doing for those 40 hours? My guess is that he was trying out different components and moving them around on perfboard until he got the sound he was looking for. He worked in his basement where I doubt that he was able to crank his amps up to stage volumes, yet all of his amps apparently worked very well on stage. So I think that there were certain characteristics or properties that he was looking for before he considered an amp ready to roll.

I always suspected that he was working with parasitic effects from some of the components. Film and foil caps have an inner lead and an outer lead. You would normally connect the outer lead to the copper pad closest to ground so that it would act as shielding to some extent. Switch them around and that capacitor might start picking up some of the signal from other components- or start broadcasting a signal that might be picked up my other components. One experiment I tried at the suggestion of an amp tech here was to wrap wire around an 0.022uF Orange Drop cap and connect that lead to other parts of the circuit (like the plates of other stages.)

Parasitic effects can act as positive feedback to make a wilder sound closer to the edge or as negative feedback to calm things down a bit. So much for my theory...

One thing I heard from someone very familiar with the amps was that part of the magic had something to do with the phase of the power supply and the phase of the audio section. (Way over my head!)

Steve Ahola

[Alan0354]

Alan - thank you for your understanding and again I am sorry that I went over the line - it was a childish thing to do. Now to get back on topic, PFB does occur in guitar amps, but more often than not, they are not done by design, and I don't believe it was the case with the TWE. It's hard to see why KF would build the Express to be prone to oscillation, more likely explainations might be that when the users change the tubes (with different gain than the original tubes) or using hotter pickups that resulted in un-wanted feedback, etc. If the amps were somehow inherently unstable, then I doubt TWE would have gotten the reputation that it deserved, or else KF would be doing after-sales service all the time.

For me, it is more interesting to look at the circuit design and how it differs from the Fenders and the Marshalls, as done by Gingertube and katopan et al. While the component selection and wiring/layout no doubt also contributed to the "mojo", I still view them as secondary contributors, rightly or wrongly...

Right now, I am more disturbed by Mark's comments than anything else. I was absolutely surprised that I came across condescending and lecturing. I am really thinking about what am I doing wrong because this is absolutely not my intention. I read my post #92. Is it because I go over and over explaining, label in 1) and 2) that sounds like I am talking down to people? I always worry I don't convey what I try to say, so I always try to explain what I want to convey in more than one angle to make sure I am clear. I try to separate in paragraphs so I have one point at a time so the whole thing is not mumble jumble together. I don't know, if you see other things, please say it.

I get use to environment that we through out an idea, people agree or disagree, then we pull out reasons to support our theory and nail it out. I throw out an idea in post #92 and I fully expect people to come back either to agree or disagree and stated why so we can talk more. I thought this is normal.

Lastly, Chuck is one I am very grateful to because he really helped me when I was struggling at the beginning. When I read this post that talked about instability, I did worked on stability issue, I really try to jump in and join the discussion hoping I can contribute something as I did learn a lot here.

[Chuck H]

Actually jazbo8, many TW amps have been known to suffer some instability within the range of the controls. One solution is to simply not set the controls that way. The other is to take it to a tech and have it worked out. As I said before:

I doubt they left Ken's bench that way, BUT...

Ken voiced the amps with NOS tubes from his personal stash AND...

Ken was said to voice some amps to the edge of stability, SO...

It's entirely possible that component drift and tube replacement can cause instability, AND...

It was common for Ken, or one of his trusted cohorts to service TW amps. There were still always a few TW amps around that had developed problems and never got back to TW for service. This has always been considered a failure of the owners part. Not Ken's. Just to add my $.02 on this, I've met a great many techs that can "repair" an amp and have a firm grasp on electronics but are truly clueless about the nature of guitar amps overall. Results from these techs WRT amps like vintage Fender's, Marshall's, Vox and other amps that may spend much or all of their time cranked, are often disappointing in the extreme. This, IMHO, is part of where the "guru" moniker comes from. Likewise, I've met many techs that have a flimsy grasp on electronics but know guitar amps better than most. Results from techs like this are spotty at best. I would estimate that 95% of the techs out there fall into one of these two categories. That's bad news for vintage amp owners to be sure. I'd like to add that the ratio of competent techs that actually speak "guitar amp" is very high on this forum for obvious reasons.

Basically, what I'm saying is that Ken had a formula that worked. Techs competent in both electronics and guitar amps should do fine with them, but this is more rare than is supposed. Without such support any TW amp will deteriorate into a glitchy, problematic and even bad sounding amp. Truth.

[Enzo]

I don't know if it pertains to TW amps or not, but having an amplifier circuit on the edge of feedback or instability can make it peaky and have some interesting gain effects.

There is a type of radio receiver called regenerative. They set up the RF circuits so it can oscillate. But as you turn up the gain, there is a point where the thing gets ultra sensitive, just on the verge of oscillating, the RF gain goes way up.. You tune the RF gain for this effect, then the radio is very sensitive to weak radio station signals. Think of a microphone turned up as far as possible so the slightest little thing will break it out into feedback. It gets real sensitive. Same sort of thing. Or maybe driving a sports car with the tail out. Just in the edge of spinning out, but not quite.

perhaps the TW takes advantage of this sort of thing to some extent.

[Justin Thomas]

"Instability within the range of controls..."

Between Chuck's observation and Enzo's technical explanation, I think we have the answer. Nearly all of my favorite amps, guitars, and pedals to play have this type of problem. A particular "undesirable" trait of a piece of gear(oscillation or instability in this case) is not always a bad thing. And it is my firm belief that if I can NOT "get a bad tone out of it," I will probably find it boring. I would bet that most of our favorite players played at the borders of sonic chaos, and for myself, much of my appreciation for certain players is how well they do on those borders. My personal favorite is probably Mike Pritzl of the Violet Burning.

Being so close to the edge of control adds an element of surprise and adventure. The people who play these amps are not afraid of walking a razor's edge, and they take the time to learn their gear. It's almost a relationship. Guys like Roy Buchanan, David Gilmour, Rory Gallagher - they KNOW their instruments. I cannot pick up any of their guitars and sound like them. I chose thbecause they have been identified with a single instrument most of their carreer. Oh - Malcolm Young.

I probably could not play a TW to its potential without actually owning it. And there is a fine line between utterly useless noise and pleasingly unpredictable. Sure, if I yurn it on and it screeches like a banshee with no input and the volume half up, I probably wouldn't use it. But if it starts to lose control when I crank the volume & treble, hmmm... sounds like a good time to me. And I disfavor master volumes except on bass amps. Too much control involved. Too loud? Crank a smaller amp through a bigger efficient speaker.

Justin

[Chuck H]

Justin raises a good point. Enzo's post has me thinking too. It occurs to me that any random instabilities would be too unpredictable to create consistent results. I would guess then that the TW amps properties of touch sensitivity, having that elusive quality of "grabbing the note" and usually feeding back with even order harmonics on any note are the result of tweaks to a specific interaction. As inferred by Alan. I've never built a clone of a TW so I've never had the opportunity to experience for myself just which part of the amp and layout might be responsible. I would think that after building a clone it wouldn't be too hard to identify if one were looking for it. But if it takes forty hours to dial it in (as posted by Steve A) I could be wrong. I still have to wonder if anyone other than those Ken tried to "train" has ever looked into it with these considerations specifically. This has been as good a thread on this particular aspect of the TW amps as I have read so far. Not surprisingly, there are no definitive answers as yet But perspectives have been taken that are more plausible than others and some that I haven't seen discussed before.

Steve, you mentioned the phase of the amp in relation to the power supply. I could buy into this somewhat. I've also read that the filters are one of the most important aspects of getting the sound "right" on this design. It was a thread including Glen Kuykendall. Perhaps the strongest proponent for TW amps that offers sound clips. He built a couple of clones to take on the road so he wouldn't have to risk his originals.

I figure, if Fender HR series amps can oscillate due to filter failure on a regular basis where other amps don't share this reputation there may be some validity to the idea.

I have a feeling the "secret formula" is probably subject to a couple of areas being tweaked in an untypical manor. With a circuit like a guitar amp this could make identifying the secret sauce almost impossible. But I take heart in the reality that there are A LOT of people looking for it and it WILL become known.

[Steve A.]

Steve, you mentioned the phase of the amp in relation to the power supply. I could buy into this somewhat.

I never could grasp that concept- can you explain it to me?

As for being on the edge of stability when I was experimenting with Trainwreck inspired circuits in 2000* there was one Orange Drop coupling cap** that when wired up correctly (outer lead to circuit point closest to ground) resulted in a rather boring sound but when I reversed the leads the amp really came alive and was right there on the edge of stability. To double check that I switched it back and it was boring again. Proof positive that something strange was going on!

One thing I like about the 225P Orange Drop caps is that you can determine which leg is the connected to the outside of the foil just by looking at it. At least the ones commonly used in guitar amps. Other than that I think that they sound a lot like Mallory 150's which are cheaper and much easier to work with because of their size.

* See attached PDF file of various incarnations of my Blues Express amp...
blxprschdup.pdf
** As I recall it was the .002uF coupling cap between the 2nd and 3rd preamp stages.

Steve Ahola

P.S. I just ran across this MP3 file of sound samples from 2000 using my Blues Express amp.
bluesxpr.mp3

[Tom Phillips]

Where can I get schematics of Train Wreck?

Covered in posts 29, 30, 31 and 35 of this thread.

[Chuck H]

I never could grasp that concept- can you explain it to me?

There is some coupling in the power supply. That means, depending on how things are arranged, like phase and out of phase signals are coupled there for better or worse. And indeed the difference can be heard in a high gain amp. Especially if you're using old school filters that typically have a higher impedance than modern filters. Perhaps much higher if they are old. Nothing more to it than that. I need to look over the TW schems and see if there might be any positive feedback at the preamp filter nodes. If it's arranged "properly" there won't be. "properly" being subjective of course. If there is a good result and no risk of damage these things can, and should, be fudged.

[teemuk]

At least all "Express" model schematics seem to follow sound power supply rail decoupling rules:

- No gain stages working in same phase share the same power supply node
- Nodes are separated largely stage by stage, and by magnitude of current draw, e.g. B+, screen, PI, 3th gain stage, 1st and 2nd gain stage
- Only very low current draw stages share same power supply node (this means supply voltage variation caused by signal current draw remains absolutely minimal)
- No high current draw stages sharing nodes with sensitive and lower current draw stages
- Adequate RC coupling and filtering per each node

I don't see, if built as is depicted by the schematic, why the power supply should cause any instability.

[Chuck H]

Agreed. I looked into it too, this morning. The Express power supplies are done exactly right and are therefor "unremarkable" WRT tonal additions or detriments.

[MarkusBass]

I'd like to see the schematic of the amp. I have some doubts about the correctness of the schematics presented so far in this thread. In some cases the treble cap is 50pF while in other cases it is 500pF. 50pF seems too low but with 500pF the bright cap 500pF seems too high. Sometimes the BASS pot is 250k and sometimes 1M. It seems to me that 1M is the correct value but I'm not sure. The same applies to missing/or not missing grid stopper resistor on the input tube (missing resistor may lead to oscillations). In the beginning of this thread someone talks about missing grid stopper resistors on the power tubes and the schematic (layout) in just previous post shows that these resistors are present. So it's hard to refer to any of the schematics since most of them contain some possible mistakes.
I built similar amps (but with much higher gain) and never had any problems with oscillations. Maybe I was lucky, or maybe I just followed basic rules (like e.g. using grid stopper resistors). I'd like to check the amp but for this reason I need a correct schematic. It looks like the schematics from Steve are quite good but the 50pF treble cap doesn't seem to be correct.

Mark