Confession of a rank amateur tube amp repairman

[loudthud]

If the signal started getting bad at the phase inverter then how was the OT bad?

Negative feedback from the OT back to the PI makes crazy looking waveforms in an attempt to correct what's going on with the OT.

[nsubulysses]

Thank you for sharing how the signal could look bad at the PI if the OT is bad becasue I have never seen that. I usually just see a great signal all the way up to the power tube grid, and a terrible signal at the plate. I always clip in an OT just to confirm 100% that it is the OT and not something else that I've missed.

[xtian]

Negative feedback from the OT back to the PI makes crazy looking waveforms in an attempt to correct what's going on with the OT.

Thanks for the clarification! Great point.

[drewl]

The "The Twin" I recently worked on had an odd phase inverter problem which could have been mistaken for a bad OT.

I always return the bad parts with the repaired amp.
Some people want them, others could care less.
Mainly old vintage stuff people want the parts when they go to sell them.

[mhuss]

I always return the bad parts with the repaired amp.

That keeps the e-waste disposal costs down too.

[Steve A.]

That pretty much describes my approach to concluding "bad transformer." Considering the replacement cost, I want to eliminate any other possibility. Then it doesn't take a whole lot of time to clip in a known working transformer - if the amp improves at that point you know what you have to do.

Good point! It is much easier than "clipping in" a PT...

Steve A.

[Steve A.]

I always return the bad parts with the repaired amp.
Some people want them, others could care less...

It is usually the only way that a customer can be sure that the parts were actually replaced (assuming that they are not given parts from another customer!)

One exception is warranty parts which often must be returned to the mfg or "scrapped" which I learned the hard way meant holding on to them for 90 days in case the regional service wanted to inspect them... at least according to Amana back then. But I would still show them to the customer...

Steve A.

[Jazz P Bass]

I do not give back bad tubes.

I put them where they will never get tried: in the bin.

[Mick Bailey]

An OT is easy to test, but A resistance check will 99.99% give a misleading reading - the exception being if there's a dead short (highly unlikely) or open.

Say you get 34 ohms and 42 ohms from centre of the primary. Is that good or bad? It would most likely be good - transformers are wound to a turns count, not resistance value and and the physical nature of the construction leads to asymmetric resistance. So let's say its good. Except it may not be; it could have a shorted turn. This is the most common failure mode and a resistance check cannot detect it because one turn less out of hundreds or thousands won't make a bit of difference to your meter reading. A shorted transformer usually reads good on a DMM, unless there's a serious meltdown.

Reading the secondary is also misleading; they read near to shorted on a DMM. is 1.8 ohms good or bad?

The check I make is a quick resistance check to see if the primary is fully open or shorted on either side. Then I do a ring test. I have a dedicated test unit for OTs that I use, but RG's neon trick is just as good in giving a pass/fail.

[Justin Thomas]

I just fire it up, crank it on ten, and then play it til the smoke comes out... confirmed: dead OT!

Just kidding... But isn't it best to replicate the conditions which led to the failure in the first place?

Justin

[Mick Bailey]

The amps I saw with bad OTs had the same symptom! Because of the "unhealthy" load of the bad OT, it was causing the signal after the PI to look really ugly, making me think it was a PI issue.

I recently had a strange problem with a Marshall 100W amp. The owner complained of low volume and distorted output. I scoped the PI and the waveform was equally bad on both halves. So I'm thinking something is wrong there, but it was the output tubes causing the PI to distort. A new set of tubes fixed the problem. The odd thing is that those tubes worked fine in another test amp.

[nosaj]

I recently had a strange problem with a Marshall 100W amp. The owner complained of low volume and distorted output. I scoped the PI and the waveform was equally bad on both halves. So I'm thinking something is wrong there, but it was the output tubes causing the PI to distort. A new set of tubes fixed the problem. The odd thing is that those tubes worked fine in another test amp.

Was it because the output tubes were pulling more current? Curious on the answer for this one.

nosaj

[Mick Bailey]

I never concluded what the problem was, and that's a situation I don't like. When I pulled the tubes the waveform was fine. Installing any two out of the four tubes caused the waveform to distort badly. Current draw was fine. I wondered at the time if there was some oscillation going on, but couldn't see anything on the scope. So whatever was wrong with those tubes was wrong with all four of them. I still have them (Marshall branded Chinese 5881s) and have set them aside for further investigation at some point.

[Ted]

I've got an old 2V Heater transformer, mounted in a box, with jack and terminal post outputs. It takes about 30 seconds to plug in to the transformer secondary and measure voltages on the primary. If voltages aren't symmetrical about the CT (to within 5% or so) then the transformers gets replaced. To be sure, I calculate the load impedance that the plates see to make sure its reasonable for the speaker impedance.

Like Leo says, you sometimes get to a bad transformer diagnosis by elimination, but sometimes (unexplained asymmetrical distortion, low output when all else seems well) this approach really speeds things up. I hate clipping temporary wiring to 500V.

[drewl]

I hate clipping temporary wiring to 500V.

...Aw, C'mon....ya gotta' live dangerously!