Marshall 2205 failure question

[stephenl]

I repaired/updated a 2205 a couple of months ago, discussed in a previous thread - and it failed again...

After talking with the owner, I found out that he had (2) 8 ohm cabs plugged in (parallel) and set the impedence selector for 16 ohms...

While not as destructive as an open output circuit, a 4x mismatch is pretty hard on the amp - but would it cause failure? It failed pretty quickly.

He said that he had the volume knobs @ 8 for reference. At that volume combined with the low output impedence - I assume that the amp was running at the current limit of the powersupply.

He said it sounded great at first, then got thinner, quieter, gone...

[Stevem]

Please tell us what failed again, as I left my crystal ball home today, lol!

Too low of a load on the OT can burn it up if played loud enough for certain, and it will also chew thru output tubes pretty fast, but not make them short out all on there own if one did fail on you in that way.

As the output tubes get weaker the OT has less current passing thru it and so is less likely to burn up if it has not already!

Has your amp blown the HT fuse?

[Roe]

check fuses, output tubes, and both transformers

[xtian]

Also check screen and grid resistors.

[stephenl]

Socket.jpg

When he originally brought the amp to me it had evidence of power tube socket arcing. The impedence selector was set @ 16 ohms - at the time I didn't know he was running (2) 8 ohms cabs in parallel...

I put in new sockets, cap cans, bias cap, and new KT77's (burned in by Tube Depot) and all new preamp tubes. I set the bias @ 35mA, played it at fairly high volume for over an hour, and left it powered on for a couple of hours, rechecked bias, played a little more - all good.

He killed it in less than an hour.

[stephenl]

Also check screen and grid resistors.

I did check both the screen and grid resistors, when I had it, and all were good.

[martin manning]

Running at 1/4 the nominal load would result in excessive current draw, leading to red plating and cooking the OT primary. Was this the configuration (selector set to 16Ω, actual load 4Ω) that caused it to land on your bench originally? In other words, Oops I Did It Again?

[pdf64]

I suspect an intermittent connection in the load, eg bad speaker lead, impedance selector.

[stephenl]

Running at 1/4 the nominal load would result in excessive current draw, leading to red plating and cooking the OT primary. Was this the configuration (selector set to 16Ω, actual load 4Ω) that caused it to land on your bench originally? In other words, Oops I Did It Again?

I suspect that this was the original cause of failure - it did come to me with the selector set to 16Ω and I assume was connected to the same speaker setup.

[Stevem]

A intermittent load on the OT will not make the outputs red plate and burn themselves up though!

[pdf64]

A intermittent load on the OT will not make the outputs red plate and burn themselves up though!

So far, redplating is just supposition, it hasn’t been observed.
And we don’t truly know what load impedance has been used with it.
Whereas tracking has occurred, and an intermittent open circuit on the secondary is likely to cause that.

[stephenl]

So far, redplating is just supposition, it hasn’t been observed.
And we don’t truly know what load impedance has been used with it.
Whereas tracking has occurred, and an intermittent open circuit on the secondary is likely to cause that.

Just to clarify:
The first failure mode is speculative, but likely. It did come to me with the selector on 16Ω.
Relative to the 2nd failure, it is factual that the impedence selector was set to 16Ω with a 4Ω speaker load connected.

[bepone]

While not as destructive as an open output circuit, a 4x mismatch is pretty hard on the amp - but would it cause failure? It failed pretty quickly.
He said it sounded great at first, then got thinner, quieter, gone...

when off and caps discharged, repair again, remeasure all, primaries of power transformers resistance, continuity primar - secondaries ..plus charge more ofcourse this is the fastest school to your friend to start to pay attention

[johnnyreece]

Do you have a way to test the tubes? If they've been run ragged enough to redplate, surely that would show on a tester. Not saying that's the problem, but trying to sort out what's going on when there's another unknown variable complicates things.

[pdf64]

…
Relative to the 2nd failure, it is factual that the impedence selector was set to 16Ω with a 4Ω speaker load connected.

If you’re sure of that then fine, I was thinking the load info may have been from a non trustworthy source, eg a lot of players tend to be a bit flaky regarding that type of detail.

Do you have a way to test the tubes? …

Plugging them into a good amp seems a suitable test.
If there’s appropriate voltage on the relevant electrodes yet no anode current flows, then the valve is dead.