Sunn Coliseum Lead railitis

[Enzo]

It is direct coupled, so it requires different thinking when troubleshooting. But as we just saw, once we look at the circuit and take proper voltage reading, the problem leaps into view. AHA! those two parts have different voltages but are supposed to be wired together. WOuld that not be just the same as say the wire between pins 3 and 8 being broken on a 12AX7 with a shared cathode in a Fender? It wouldn't pass signal, you'd find the plate voltage on one side really high, no cathode voltage, but the other side has cathode voltage - which is now wrong voltage - but they are supposed to be wired together - AHA!

And while it might take a while on this forum to go back and forth a post or two a day across the world, but on the bench right in front of me or you we systematically do what this long trail of posts boils down to. The whole thing is offset, we look for the bad parts and replace them in the output. Without a load, it can make DC all it wants. If the outputs are not shorted together, then DC results from the thing being turned on hard one way. So we look back a stage and find the relation between the + and - drives is right, just shoved over to one end. What drives that stack? The voltage amp xstr. Was it OK? then move back to what drive it - the diffy pair. Explore them and find an errant voltage. Either then we had a bad diffy xstr or a broken circuit path. and then AHA!

We could add a whole lot of complexity to make any failure localized - like keeping all the compartment doors sealed on a ship. But when we look analytically at it, either the output is shorted, or both sides turning on at once, or it is just massively offset. Three troubleshooting paths to take there. If there are shorted output xstrs we know right away and we may have to move back through shorted drivers. If both sides are turning on, then the bias string is open, and for offset, well we just went through it.

What if we made tube amps all transformer coupled? Then we would never have to deal with a leaky coupling cap.

Tube amps are pretty intuitive, until you encounter for example parasitic oscillations. Little chirps on peaks. ONly thing intuitive about solving those is if you heard about it from somewhere else. Every technology has its strengths and weaknesses.

[Euthymia]

....once we look at the circuit and take proper voltage reading, the problem leaps into view.

Yeah, but it took me 3 months to find the right area of the circuit to read. If you hadn't told me to poke the differential pair (including me having to ask "what's the differential pair?"), it would still be broken.

And I have to confess that I STILL don't know what it was about floating that emitter that made it drag everything to one rail. When I can stand to look at the schematic again, I'm going to try to figure it out....

Also, the thing first came to me with something else wrong with it, which apparently got fixed somehow along the way. I broke it again by sticking the transistor back in its socket with one pin 0.005" off to one side. So to add insult to injury, we don't even know what was originally wrong with it! It was broken, I started working on it, broke it further, fixed the original problem, then fixed the thing I broke.

I don't think I'm lazy when it comes to trying to learn how to troubleshoot SS circuitry. I own half a dozen books on just that topic and have read twice that many. You know I don't give up easily.

The way my business is now, I do not need to accept SS amps to survive. I choose to work on them because I like to fix things, because people need someone to fix their SS amps, and because some of the old SS amps are cool. In my area, I am actually a "go-to" guy for old SS Sunns; this one wound up on my bench due to that reputation.

AHA! those two parts have different voltages but are supposed to be wired together. WOuld that not be just the same as say the wire between pins 3 and 8 being broken on a 12AX7 with a shared cathode in a Fender? It wouldn't pass signal, you'd find the plate voltage on one side really high, no cathode voltage, but the other side has cathode voltage - which is now wrong voltage - but they are supposed to be wired together - AHA!

If that had that been a Fender Twin with no sound coming through the Normal channel, but sound coming through the Vibrato channel, which is, I think, the scenario you're thinking of, it would have taken me maybe 15 minutes to find it, including pulling the chassis.

I would have first swapped in a new tube in V1, then poked my scope probe at the plate of the Normal channel's tone recovery stage, then the grid (working backwards from where we know we have signal). Okay, we have signal at the grid but not the plate. How are the DC voltages?

Whoops, pin 8 seems kinda off. There we go.

A total of poking a test instrument at a pin 5 different times (maybe 6 if I started scoping at the Normal channel's first stage plate).

Such an issue would NOT wipe out every output device, as happened with the Sunn. The failure would be confined entirely to one stage, and would have resulted in replacement of maybe one resistor.

I wouldn't have had to take all my measurements 30 seconds at a time with the unit hooked up to a lightbulb.

And while it might take a while on this forum to go back and forth a post or two a day across the world, but on the bench right in front of me or you we systematically do what this long trail of posts boils down to. The whole thing is offset, we look for the bad parts and replace them in the output. Without a load, it can make DC all it wants.

It was roasting components even with the output transistors disconnected.

I had to run it through a lightbulb limiter, and even then, I could only turn it on long enough to take voltage readings or it would fry more resistors.

If the outputs are not shorted together, then DC results from the thing being turned on hard one way. So we look back a stage and find the relation between the + and - drives is right, just shoved over to one end. What drives that stack? The voltage amp xstr. Was it OK? then move back to what drive it - the diffy pair. Explore them and find an errant voltage. Either then we had a bad diffy xstr or a broken circuit path. and then AHA!

Well in this case, that "AHA" took a hell of a long time. You yourself were convinced that the problem was somehow in the bias string, remember?

We could add a whole lot of complexity to make any failure localized - like keeping all the compartment doors sealed on a ship. But when we look analytically at it, either the output is shorted, or both sides turning on at once, or it is just massively offset. Three troubleshooting paths to take there. If there are shorted output xstrs we know right away and we may have to move back through shorted drivers. If both sides are turning on, then the bias string is open, and for offset, well we just went through it.

I really do hope you write this down in book form some day. If there's anything I can do to help with that project....

What if we made tube amps all transformer coupled? Then we would never have to deal with a leaky coupling cap.

Well, the leaky coupling cap doesn't take out $75 in components. It probably just makes one of your controls sound scratchy when you turn it.

Also, when you're looking for the leaky coupling cap, you can have the amp powered up and with a signal running through it without worrying about whether you're going to kill all your output devices, peel traces off the circuit board, and roast your bench speaker.

Tube amps are pretty intuitive, until you encounter for example parasitic oscillations. Little chirps on peaks. ONly thing intuitive about solving those is if you heard about it from somewhere else. Every technology has its strengths and weaknesses.

Well, it's my belief that one of the weaknesses of SS technology for musical instrument amplification applications is its fragility and subsequent difficulty of servicing.

Doesn't mean I think the technology is "bad," or that I won't work on it or use it, but I do think that it is one of the things to factor in when making the decision whether to use it.

[Euthymia]

That was at least 2 transistor amps that burst back into life yesterday. How does the Sunn sound?

Congratulations on your victory, Liam. Remember how I said to pray to the Enzo!

They were both suffering from railitis, too, which is the most fearsome of SS amp symptoms.

One of the reasons I like working on these old Sunns is that they do sound good after you get them working. This one is no exception. Well over 100W before it clips the sine into 8 ohms.

The thing is built like the proverbial tank, with cooling fan, sockets for everything (doh!)....

[Liam]

I feel like I learnt quite a lot, and I take Enzo's point about direct coupling meaning you can just chase the offsets through the amp. Very eloquently put. If I come across another injured transistor amp I'll happily wade into it.

Yeah, but it took me 3 months to find the right area of the circuit to read.

I take it from your long post you feel like you've got time on your hands now. Funnily enough I think the Fender was here for 7 weeks. I only mend amps for fun (therapy?;-)), so it may take me a while to figure out what to do with my spare time.

Glad the Sunn sounds good. And now the Fender is back with its owner I'm quite relieved. Is it me, or do all these modern "channel switching" transistor amps sound very much alike? I'm actually starting to doubt whether the transistor guitar amp will ever truly catch on.

Liam

[Euthymia]

I feel like I learnt quite a lot, and I take Enzo's point about direct coupling meaning you can just chase the offsets through the amp. Very eloquently put. If I come across another injured transistor amp I'll happily wade into it.

I'm still trying to digest what I learned in this episode. That's a good way to put it: "chase the offsets."

A big part of my frustration is my own lack of ability to look at SS circuit building blocks and understand what they do and what voltages I should expect to find.

I take it from your long post you feel like you've got time on your hands now. Funnily enough I think the Fender was here for 7 weeks. I only mend amps for fun (therapy?;-)), so it may take me a while to figure out what to do with my spare time.

Things just slowed down enough here that I could get back to the tough doggie. While I had that Sunn on and off the bench, I repaired dozens of other amps and got them merrily on their way.

I'm also working on the designs for my own production amplifiers (tube, natch). Having a hard time finding someone to do my sheet metal for a reasonable cost.

I mend and build amps to pay the mortgage.

Is it me, or do all these modern "channel switching" transistor amps sound very much alike? I'm actually starting to doubt whether the transistor guitar amp will ever truly catch on.
Liam

What I think is that the modern transistor amps to which you refer are indeed designed to sound like what novices want. The area of the market that SS amps seem to have settled into is that of "starter" amp. You get an SS amp for a hundy, then when it's time for a "pro" amp, go tube.

The sound of the starter amp is probably going to be something that sounds as much like what's on the radio as possible. No character of its own, because the target buyer is not looking for an amp with an individual character. Stick an op-amp diode clipper in the preamp for distortion and be done with it.

I think that it is possible for a well-designed, well-constructed, reliable, serviceable good-sounding solid state amplifier to eventually be successful. There are some barriers to get past before that happens.

First is the reputation. SS ain't the "cool" thing for guitar players to use. Even Dimebag was working with a tube amp designer before he died.

Next is that innovation in the high-end amp marketplace tends to be driven by labor-of-love tinkerers. This goes back at least as far as Fender and Marshall. Their innovations trickle down or up into the mass-marketed stuff. So we'd need someone who really loved and understood solid-state guitar amps to design and build them.

Where does this person come from? People fall in love with tube MI amplifier technology every day, there are many websites devoted to messing with it, books, etc. How many people have you ever encountered who are really in love with SS MI amplifier technology? How many books are there to study?

Is the marketplace clamoring for great solid-state guitar amps? If so, I don't hear it.

Tube technology is still around, it's fun to work with, rugged, and can readily produce a sound that we hear as musical. The only advantages SS has over it are weight and component cost (which is something that means much less to our theoretical tinkerer than it does to Crate, Maker of Fine Practice Amps). The disadvantages are that it doesn't as readily produce the sound that we have heard as musical for the past 60 years, it's fragile (susceptible to cascade failure), and more difficult to service (whether that difficulty is inherent or due to a lack of skilled technicians is moot: it's harder for Joe or Jane Musician to get the stuff fixed).

[52 Bill]

Euthymia:
I've been following this thread since the beginning and I'm glad to hear that it's finally fixed. I've gone back to the start and re-read all of the posts and have a couple of questions regarding this amp.

1- What was the original problem? Blown fuse/outputs?

2- When did you pull and mis-insert the input differential transistor?

Just trying to learn from your experience.
Thanks!

[Euthymia]

1- What was the original problem? Blown fuse/outputs?

The original complaint was that it was making a big hum and popping the breaker. No mere fuse; this old Sunn has a breaker.

When I put it on the bench, it had, as I figured it would, a handful of shorted output transistors.

2- When did you pull and mis-insert the input differential transistor?

I'm not sure exactly when. I had the thing on and off the bench so many times.

I replaced the output transistors that had blown, then started testing everything on that driver board. I think it still had DC on the output. I have a nice transistor tester/analyzer, so I pulled them all one by one and checked 'em with the analyzer. Must have been during one of those operations.

Along the way, I found and replaced burned resistors and old electrolytic caps; one or more of those may have been the original problem.

Just trying to learn from your experience.

Me, too!!

[Enzo]

A big part of my frustration is my own lack of ability to look at SS circuit building blocks and understand what they do and what voltages I should expect to find.

But that is just a matter of experience. DOn't you think if you encounter another one of these sometime soon, you will not take nearly as many false paths? The power stage of most any push pull tube amp is pretty much identical to any other. I mean really, when was the last time you saw one that was different in any meaningful way? COuld you not draw one on a napkin and have it be right? I see SS power amps the same way. I can draw one on a napkin. They are all pretty much the same. Output, driver, maybe predriver, some sort of bias string to keep the bases spaced properly. A voltage amp to move the bias string up and down, and an input diffy stage to drive the VA. Doesn't look that way to you now, but it would if you did it enough times. Remember the first time you saw a split load phase splitter? THAT is pretty intuitive. Remember the first time you saw some sort of two triode PI? Did you have to try to figuer out how the second side really got a signal? I did.

I have to disagree about the frailty of SS amps. They CAN have cascade failures, but that doesn't mean they usually DO. Most times, I replace failed outputs and that is that. If the drivers fail, then I have to check for burnt resistors. But mostly they just work. Sure they can do mean things, but tubes often short and take out a screen resistor. You know - but only from experience I bet - that when the new tube doesn't work, check for an open screen resistor. Maybe it is only one step, but it is still a tube failure causing a further failure. Or a tube shorts internally and burns up the 100 ohm false ground resistors on the heaters. Happens enough. SS amps never have shorted OTs. Rarely get microphonic. DOn't care if the load is absent. EVery technology has its weak points. And its strengths.

[Enzo]

How it works?

The input xstr has the base at ground (DC). Q201 and friend Q202 share their emmiter resistors R203,204. The current through them biases the xstr. Just like a 12AX7 with shared cathodes.

In a tube if you disconnected the second plate, only one triode would conduct curren tthrough the tube, and the cathode voltage would fall. This would affect the bias. In the xstr circuit, if you disconnect one - which is what happened to you - then the current through that xstr no longer flows. The voltage dropped across R203,204 then drops. That means the grounded base is mroe positive than the emitter compared to normal. More base current then flows, so the xstr conducts harder. ANd that in turn means more current dragged from the base of the VA, xstr Q203 up there. That increased base current turns on Q203 harder. That means it pulls its collector closer to its emitter, and the whole bias string is thus pulled up towards the + rail.

[Enzo]

Yeah, but it took me 3 months to find the right area of the circuit to read. If you hadn't told me to poke the differential pair (including me having to ask "what's the differential pair?"), it would still be broken.

DOn't be unfair to yourself. it didn't take you 3 months to do it. Three months elapsed while we all exchange tiny little incremental steps one at a time each day or two. How long would it take you to build a 5E3 if you made one solder joint a day five days a week? And while doing that we had to discuss how you determine which resistor was the 100k and which the 1.5k. And how to determine which pins were which numbers on the sockets, and so on. WOuldn't take you three months again.

And I suggested the diffy pair because I was never getting satisfying information about Q203. If the xstr itself is OK, SOMETHING is controlling it. In this case that diffy pair.

I don't think you are lazy either. You might fix that exampled open cathode on the Fender in 15 minutes, but that is because you know what you are doing. I bet if that was your first encounter with a tube amp, it would not have gone that swiftly. Now, you not only are familiar with the tube circuits, you also know the implications of funny readings on the meter. If I see 350 instead of 200v on a 12AX7 plate, it would not be a surprise to find zero on the cathode. That is because I know how tubes work. A noobie might not put that together and would wonder why do I have to fix my B+.

(EDIT: DAMMIT - it blanked out some paragraphs again. I'll try to fix.)

CAn you imagine telling someone how to build a 5E3 who had never done it, using this daily post method. Not easy. Not because you or he can't do it, it is just not simple to communicate all the information in a coherent manner.

[Enzo]

Is the marketplace clamoring for great solid-state guitar amps? If so, I don't hear it.

The tweakos like us that congregate here, maybe not, but if you look, SS amps sell by the boatload.

Entire boards are devoted to SS effects units and guys who like to mod them and build them. Why even here on the AMpage over in the other room there are such folks. Preamp circuits agonizong over which chip to use, whar resistors to change, etc. And if you go over to someplace like www.diyaudio.com and follow their forums you will find planty guys building SS stuff. Oh it is mainly hifi, but still.

[Euthymia]

The tweakos like us that congregate here, maybe not, but if you look, SS amps sell by the boatload.

I know they sell a ton of them, what I meant is that the market for SS guitar amps seems stuck in a lower tier. There's plenty of demand for adequate ones, but I don't see much demand for great ones.

Entire boards are devoted to SS effects units and guys who like to mod them and build them. Why even here on the AMpage over in the other room there are such folks. Preamp circuits agonizong over which chip to use, whar resistors to change, etc.

Dunno if you've ever checked my web page, but my other revenue stream is fuzzboxes. I have 3 that I build myself and another that I designed for a client.

Of those products, 3 of them are actually direct-coupled transistor amplifier circuits (variations of the evergreen Fuzz Face circuit), and the ones sold by me each have their bias set by resistor selection.

The other one is an op-amp/diode clipper. In that product, there are only certain brands and vintages of op-amps that sound truly great in it, so I have all these tubes of old op-amps around.

Never really thought I'd be caring about vintage freaking op-amps.

And if you go over to someplace like www.diyaudio.com and follow their forums you will find planty guys building SS stuff. Oh it is mainly hifi, but still.

I chose my words carefully when I said "guitar amps." I think the market and hobbyist scene is way different for hi-fi and effects, and a lot of the things I've said about the SS guitar amp marketplace do not even apply to bass amps.

[Liam]

Euthymia, my head went down the same path as yours over the last few days. In fact I resuscitated the thread on "good sounding SS amps", and somehow missed out Andy Summers' Roland JC120.

Someone said SS amps are more flexible, whereas tube amps have a better sound. This might be why I've got quite a few tube amps, but "flexibility" has never figured high in my list of what I want from an amp. I want a proper ear to ear grin...

I'm mad about SS effects pedals, and have a big collection, but for some reason they only do anything for me when they are driving an ECC83 or somesuch. Have you ever heard a Tubescreamer driving a Sessionette SS amp? The result is just pitiful IMHO.

Here's another tack on the issue. In the area I last lived were two really good techs. One of them designed a solid state amp that sounded good. The other designed a tube amp that also sounded good. Guess who's making a living out of it... (and no, neither of them is me) The solid state amp never had a nice clean sound to my knowledge, but it sounded way cool dirty.

Liam