Real TrainWreck mojo!

[Chuck H]

It was Kens design after all. As much time and energy as the guys at the Amp Garage have spent on that circuit, Ken spent more! So that makes sense.

FWIW I never got around to posting my final considerations on the matter. I did find my "Express" file. Looking at amp porn of a couple of Express's there and then a couple of others on line I've determined that layout manipulation is NOT part of the final voicing. I believe the final voicing is done with minute changes to component values (within their tolerance in fact) and, most importantly, tube selection and biasing. The right amount of mismatch (which is probably not a lot) and the bias adjustment being the most important. I've read that Ken took great pains to tube the amps. Most used a combination of different brand NOS tubes. And we've all heard what ear swapping a pile of preamp tubes into different positions in an amp can do for the tone. So there you go. If Ken DID spend any time on moving lead dress around it wasn't much because all the amps I looked at are so similar and it looks to be a pretty stable layout anyway. Plus there are minute differences in some of the amps board layouts or a half inch give or take on the PI tube hole. That sort of thing. That tells me that the layout and lead dress are working well enough that he considered those things non critical factors to the final tone as long as they were very close. So, I don't know it that seems convincing to anyone else but I'm putting the lead dress possibility to rest.

[Leo_Gnardo]

Referring to D1-D6 connecting plates to ground is very strange and have not seen that before. Was thinking about this today and just could not understand what is the purpose of these diodes. So I got researching and found that they were used on older Ampeg amps, hence the early years of KF.

At some point early 70's all of a sudden SVT, V4, VT22, V2, VT40 all sprouted flyback suppressor diodes. Whether it was Ken's doing, I don't know. But he does recommend them in The Trainwreck Pages, and you can see the flyback spikes on a scope if you slam enough signal thru an amp that doesn't have 'em. Curiously some amps don't develop those spikes but that's a question for another day. When the speaker cone is moving, always trying to but not quite keeping up with the amp output waveform, there are times when the cone and voice coil are being pulled thru the magnetic gap by spring force from the surround and spider. At these moments, the speaker is acting as a generator, not a linear motor, and the current it develops passes thru the output transformer secondary, is multiplied by the OT winding ratio, and ends up on the plates of the output tubes. These spikes can hit 3000 volts, and that can shorten the life of your output tubes. Also can punch holes thru the insulation in an output transformer. After installing them, I hear a very small decrease in overall distortion, and it's an obnoxious component that goes away. Still plenty of "the right kind of" distortion when you overdrive the amp. Note that you will not see flyback spikes when driving load resistors.

Yes as Jazz P Bass mentioned, a shorted flyback diode can wreck an OT. I've had it happen exactly twice. Once it wrecked the OT, another time just popped the fuse. So there's that drawback . . . use 'em or not as you please. FWIW there's a single 3000 or 5000V diode in each MusicMan amp, instead of a string of 3 1000V as Ken recommends. Also Peavey's been using 'em for years, and Fender too. Hmm, might be something to it. When I'm dealing with an amp with extra-high B+ I string 4 1N4007's in a row, and for low voltage amps say 350V or less, 2 are sufficient.

[Alan0354]

Where can I get schematics of Train Wreck?

[Enzo]

A couple are here:

Post 70's Amp Heaven Schematics - Acoustic Bogner Kettner HiWatt Matchless Orange Peavey Rickenbacker Rivera Soldano

[Chuck H]

See post 37 for what is considered the most accurate representation of the Express circuit. The links Enzo posted access speculated circuits for the Express. The Liverpool schem drawn by Mark Abbot is supposed to be his witness of an actual Liverpool, but there are some inaccuracies so I'm not sure what the real story is. The power amp section of his schematic is accurate except that the NFB loop is not shown to connect at the LTP tail via a 100k resistor and the input cap to the PI should be a .022uf. For a Liverpool just use the Express preamp instead of Mark Abbot's rendition.

[loudthud]

The biggest collection of Trainwreck schematics is at Amp Garage. There is a files section that you can't see unless you are a logged in member. Link: The Amp Garage :: Index

[echoride]

You're going on about all the differences from model to model and the evolution of the circuit during it's production. None of this happened. I'm not sure where you could have read any of this.



There are a few people and companies that have stepped forward since Kens death claiming to have been an integral part of the amps. Some even claiming to have built amps that were sold as KF built amps. I can't speak to the truth of it but I do think that it seems awful convenient now that Ken is dead. If Dave Funk could build an amp that performed just like a KF serviced Express Thunderfunk would be notorious for being the only real deal Wreck clone going. Period. Players CAN tell the difference. There is some hype, but it's not JUST hype. Who is "Mark"? Mark Abbott? Kendrick never claimed to make any TW amps. They had a partnership on a couple of Kendrick models. Similar to TW's thing with Dr. Z only less fruitful.

A few cents worth:

1) I never met Ken Fischer but I had a couple phone conversations with him at around the time Steve (Angela) M. put his work in his catalog. This was when he was still selling his amps new for reasonable money. He seemed nice enough, but he really did not have a thorough grasp of real electronic engineering principles and he seemed a little Aspergery. Or that was my impression then.

2) Shortly thereafter I met a man that claimed to know Ken really well, had hung out at his house apparently and said that Ken had made him, and at least one other person a deal, if they would build Trainwrecks at their home shops on a piece work basis he would supply all the parts and pay them by having them build N number of amps they could keep one, to keep or sell as they saw fit. He had photographs of what he said was Ken's workspace and bench and of Ken himself. I no longer have these photos but it did appear that the man in the photos was Mr. Fischer although I couldn't swear to that.

3. He stated that Ken had a large number of unusual practices, including that he would drill a random hole on the chassis somewhere to drive future cloners nuts. He was convinced everyone would copy these amps well before they became really valuable. he also stated that the light gauge aluminum chassis was a concern of his.

4. Apparently Fischer allowed him to watch him build but he came to the conclusion between visits that there were steps that Fischer would take that he would not allow to be observed. He suspected that Fischer was using lead dress and wire in a sort of tuned fashion and had evolved some way of doing it in a hurry. Fischer named his amps with female names and he had seen the same amp at different times and it would be wholly rewired. This observer said that his father had been a foundryman and that was how the old molders and melters kept their apprentices from really learning the trade so as not to be real competition for their jobs. He felt Fishcher had had some issues that caused him to be on powerful psych meds and that this would cause issues from time to time and was probably the root cause of his illness.

5.) I had thought about buying a Wreck myself, but never found it quite convenient to come up with the funds-they were not very expensive by comparison to today's prices, but at the time I had a young infant and had just bought a project car that would turn into a high-five-figure sinkhole. Live and learn.

[Shadrock2]

So I have been considering a Trainwreck style build and have been reading the posts on this thead and over at the amp garage and...

A column that KF wrote for some gear magazine. A file from Ampage called "Trainwreck Thinking". I saw where Ken was talking about swapping film/foil caps and changing the polarity to figure out which way sounds best. He was talking about the tone cap inside the guitar but I was thinking that there is no reason this can't apply to amps.
Do any of you guys ever do that? It would be a PITA but...

I had decided that a spiral wrapped cap like this would have to have an inside foil and an outside foil but capacitance is capacitance... then from another thread here... I started looking at Aiken Amps white papers and I came across this

Where To Connect the Outside Foil on Capacitors

Any comments on this line of thinking? Maybe they are not all oriented to minimize noise.

BTW... I have a guitar playing friend who buys and sells boutique amps all the time and he had a Dr. Z Z-Wreck a few years back and that amp definitely had the "chirp" on the string attack. We both commented on it at the time. Can't speak to the "bloom and swirl". That chirp was definitely unique to the amp compared with the others he had around at the time.

[silverfox]

Don't think it matters

My thoughts on this would be the difference, if there is any at all, would be so slight as to be nearly imperceptible. The most noticeable change would be in the listeners mind and I don't mean that in a derogatory fashion.

I don't think there will be any difference or the generally accepted practice would be to polarize the caps. Everyone would know the difference.

Silverfox.

[Chuck H]

I built two identical amps. That is, at the same time with components from the same lots and chassis and boards made side by side. Some component testing showed that whatever small % off spec (5% tolerance) was the same for resistor lots, though I didn't stringently test every component. I used OD caps for these amps. These are outer foil caps. I intentionally oriented the caps differently for each amp. There was no difference in noise. One amp did seem to have a slightly different tone. Not better or worse, just different. Such a small difference though. If you were to blindfold someone and AB test, anyone would soon become confused. That close. I doubt it was the cap orientation but I never swapped the caps around to find out because it clearly wasn't worth it.

[Gingertube]

FWIW - The Gingertube corruption of the Trainwreck Express:



Well I did build an Express Clone - the "Junkbox Trainwreck"
Parasitic Oscillation was a real pain to cure and so I can believe in "lead dress" stories, final trim of components such as grid stop resistors which are often left off the schematics, and various "edge of stability" stories.

Grid stop on the 2nd gain stage was essential, so I just put them on 3rd gain stage too.
The final problem which took ages to track down is that the Phase Inverter was subject to parasitics too and it required grid stops as well. I did'nt try a lot of values just put 22K on both sides of the PI and on the 2 gain stages (the input stage already had a grid stop) - Note the schematics above does not show the grid stops on the PI.

I did copy the ceramic + parallel polyester coupling cap that is talked about on the Blue Guitar Schematics (the 2n2 after stage 2) but personally believe that if anything it probably degraded the sound. Ceramics have very nasty dielectric absorbsion and any "cork sniffer" worth his title will tell you firmly that ceramics are for bypassing power supplies on RF Amplifiers and have no place in anything designed to produce audio. They can also be nastily microphonic.

The guy I built it for is the local guitar "god". He lived and played in the US for a few years and tells me he has played a "real" trainwreck and that what I built was as good as he recalls of the original. That was a bit of a surprise because I did two other things which are definitely not on the originals and would have changed the sound.
The power amp ran cathode biased 6CA7 (EL34) and I changed bypass cap values to be the same as the "Bad Cat" (200uF on each cathode).
The unused triode was put to use as parallel of the input stage but biased at a different point, that is the parallel triodes on the input stage ran at dissimilar operating points for some harmonic emphasis.

It runs a 12" speaker (Strauss) in a 3/4 closed back box. It was a half closed back but nailing on another board was a final "set to work" tweak to get the sound just right (in lieu of adding more feedback in the amp, for a slightly tighter sound).

Cheers,
Ian

[Chuck H]

Wow! That's a lot of NFB you've got there. A 7:1 divider ratio off the 16 ohm tap seems like a huge amount to me.

All seems like sensible differences otherwise.

My own version of the TW circuit, well, lemme splain... I have a circuit that I designed for a friend around 1999 before anyone knew much about the TW circuit. In fact I wasn't even interested in those mythical amps at the time. Later I found that the biggest difference between my circuit and the TW circuit was the cathode resistor value on the third stage. I tried the colder biased value there and liked some of what was happening. I settled on a 6.8k (rather than 10k) for that circuit. That one change making my amp similar enough to a TW that anyone looking at the two would think it was MY interpretation rather than an original design. Bummer. but NBD. Anyway...

My own circuit is designed very carefully WRT layout and grounding. The nature of the circuit is very high sensitivity due to lack of HF attenuation and the use of the power stage as part of the clipping circuit. The TW doesn't have a bad layout, it's just a super sensitive circuit. There are a lot of reports of instability even in KF built TW's. Though I doubt they left his care that way. Still, this may be what gave rise to the "edge of stability" banter. It's hooey of course.

[Shadrock2]

Once More Into the Breach

"I built two identical amps. That is, at the same time with components from the same lots and chassis and boards made side by side. Some component testing showed that whatever small % off spec (5% tolerance) was the same for resistor lots, though I didn't stringently test every component. I used OD caps for these amps. These are outer foil caps. I intentionally oriented the caps differently for each amp. There was no difference in noise. One amp did seem to have a slightly different tone. Not better or worse, just different. Such a small difference though. If you were to blindfold someone and AB test, anyone would soon become confused. That close. I doubt it was the cap orientation but I never swapped the caps around to find out because it clearly wasn't worth it. "

Just feel I need to beat on this cap polarity thing one more time.
I know you guys got far more experience with these amps than me, but I was a research chemist in my former life and I know a little about scientific method.
So, Chuck, what bothers me is that you say you changed the polarity of ALL of the caps between the two otherwise identical amps and detected a slight change in sound/tone. Well, for a proper experiment, you should build them both identical and change one cap at a time. A - B the two amps and decide which you like better, change the 1rst amp to that polarity configuration and then move on to the next cap. In this way you "might" get stacked changes that lead to a significant difference/improvement. To me there are only 4 - 6 caps that matter in this kind of experiment. The ones after the PI would be the last I changed because they are inside the NFB loop.
It seems I read or heard somewhere that KF said that you have to change one thing at a time and listen for the difference. Other people on this forum have already said KF was always swapping parts around in his amps.

[silverfox]

To satisfy the scientist within, why not build the amp with an easy to change method for the caps in question. AB the amp after it is running and let us know the outcome. You could even change select caps to see if there is a particular section of the amp that responds to the test criteria. (spell ck is off)

Ginger Tube, I am interested in building a train wreck in the future. If I were to use this schematic and a layout from say Amp Garage is there anything I need to know beyond what you have stated in the notes and schematics?

Fox.

[Gingertube]

Ginger Tube, I am interested in building a train wreck in the future. If I were to use this schematic and a layout from say Amp Garage is there anything I need to know beyond what you have stated in the notes and schematics?

Fox.

Add the 22K Grid stops to the triodes on the Phase Inverter (not shown on my schematic) - also a WAG of mine, I believe that some of the lovely sound can be attributed to running the 6CA7 (EL34) at 380V to give 32Watts at just below clipping. My own opinion is that they sound better when not being push to 50-55W with 480V B+ supplies. As Chuck says I'm running a fair amount of feedback, you could experiment with that 33K, try values up to 100K. This may have just been to suit the "loose" speaker box, you will see in my post above that I commented on converting the box from 1/2 closed back to 3/4 closed back in order to tighten the sound without adding even more feedback.
32 Watts is plenty anyway since "Trainwrecks" sound best when turned up - their whole design philosophy in fact was "non master volume/pushed power amp".

Cheers,
Ian