K5881 Mk1 - baffling problem!

[espidog]

For several days now, I've been trying to discover the cause of a bizarre problem with my left-hand channel. The problem lies somewhere in the relationship between the phase-splitter/driver and the output stage. I've subjected the whole circuit to the most rigorous step-by-step diagnostic routine, but so far all my efforts have failed. I'm baffled!

First of all, let me save time by telling you what ISN'T at fault. Answer: everything! In other words, I have checked very individual component in the circuit - every valve, capacitor, resistor, socket etc., removing each item from circuit if necessary, and found all of them to be healthy. I have also done all the usual checks for possible dry joints, open circuits, short circuits, poor earthing....etc. You name it, I've checked it.

I've also checked the voltage and current conditions at every conceivable point in the circuit, and they all come up within tolerance - with the notable exceptions I'm about to describe...

General Symptoms:
Output from the faulty channel is generally much lower than it should be. When the input signal amplitude reaches a certain level (far lower than would normally cause any overload), the output signal remains small but becomes jagged and unevenly clipped, even showing some crossover distortion (indicating that the output valves are being forced to a bias point way out of class A).

Measurements:
Current at the cathodes of the phase splitter is within tolerance (0.980mA) and is stable. So are all the DC voltages. However, the signals emerging from the phase-splitter anodes are ENORMOUS: 5 or 6 times greater in amplitude than those on the healthy right-hand channel. I have been over the phase splitter stage with a test meter, oscilloscope, soldering iron, fine-toothed comb (and the patience of saint) several times, and can find no apparent cause for this.

When no signal is present, the DC voltage at the cathodes of the output valves reads correctly (34VDC). However, as input signal level increases, the cathode voltages do not maintain their equilibrium, rising dramatically (up to 50VDC and more), throwing the valve way off its correct bias point. Needless to say, the cathode bias resistors and bypass caps test as perfectly healthy.

I'm stumped. Any ideas?

[John Caswell]

Hi Espidog,
Just out of curiousity measure the DC resistances of the OPTxs on the good and bad channels, only the HT side not the speaker sides.

John aka Dr John

[espidog]

Originally Posted by John Caswell

Hi Espidog,
Just out of curiousity measure the DC resistances of the OPTxs on the good and bad channels, only the HT side not the speaker sides.

John aka Dr John

Hi John,
Many thanks for the idea. I've just taken measurements (with the output valves removed and the centre tap disconnected from the HT rail, so the OPT is effectively out of circuit), and there does seem to be an anomaly. I'd very much welcome your opinion on its significance.

Here are the measurements:

RH Channel (healthy one)
Overall resistance: 332.2 Ohms
Centre tap to leg A: 165.7
Centre tap to leg B: 166.6

LH Channel (sick one)
Overall resistance: 299.7 Ohms
Centre tap to leg A: 165.5
Centre tap to leg B: 134.2

Oh my God. Looks like I've got some shorted turns or something on one half of the OPT.

What do you think, John? Do you think that this 31 Ohm discrepancy (about 18% lower than it should be) would be the root cause of my problem? I don't have much experience in judging transformer issues.

One thing's for sure, though: it can't be doing any good!

[John Caswell]

Hi Espidog,
On the face of it yes it looks as though you have shorted turns, I have a spare one that I have just checked and it reads 172R on both legs which is near enough to yours at 166R
In the meantime the simplest confirmation is to swap the good one for the "bad" one and see what happens.

John aka Dr John

[espidog]

Originally Posted by John Caswell

Hi Espidog,
On the face of it yes it looks as though you have shorted turns, I have a spare one that I have just checked and it reads 172R on both legs which is near enough to yours at 166R
In the meantime the simplest confirmation is to swap the good one for the "bad" one and see what happens.

John aka Dr John

Thanks John,
Yes, just to confirm it to myself, I swapped it for a roughly similar transformer (from another 6L6 push/pull amp), and the problem vanished.

I've been in touch with Geoff Budden at SP Wound Components regarding obtaining a brand new replacement. I've also put an ad in the WD Wanted forum just in case anyone out there has a 2nd hand one they're willing to sell.

Many thanks for your decisive input on this one - I was going nuts!
Cheers,
Espidog