Saturday, September 30, 2017

Help, Part 9. Character building opportunities.

I strongly dislike the obsession with "simplicity."  I think this is very freedom limiting, and has no essential basis in what people can actually hear.  Sure, have a 3 way system is more complicated than a 1 way system.  And so on.  But done properly, a somewhat more complicated system can have better performance in the ways that are most meaningful to a particular audiophile.  Ways that I'm interested in, such as having bandwidth from the infrasonic to the ultrasonic.

However, there is absolutely no doubt, that a more complicated system is more difficult to set up properly, AND to keep working.  So, it's a challenge, and that means that a complicated system that works, let alone works well, is a testament to an audiophile's talent and perseverance.  And so it's a character building exercise.

Sadly, the "keep working" part has been my primary focus since I apparently damaged something in May of 2017, and started having amplifier shutdowns.  Tracking down the true problem or problems here has certain been a test of my perseverance.

After trying to stop the problem with various tweaks, such as new less capacitive speaker cables (made of hand twisted 12 gauge silver plated PTFE coated wire) for several weeks, I noticed that the problem had occurred when I was playing through the right speaker only, but not the left speaker only, so did a further series of tests pairing up each speaker in each amplifier channel, figuring it might be one speaker, or one channel, but hopefully not both.  I figured the problem could be either a shorting transformer or other part in one of the speakers, or a malfunction in one amplifier channel.

I determined that either speaker playing on the right channel would cause a shutdown, but either speaker playing on the left channel would not.  So, this suggested that the amplifier, which had recently had capacitor service costing $1800, had developed a faulty channel.

I sent it in for repair (as always, itself a fairly challenging exercise) and Krell found a problem and fixed it and sent it back for free.

Well, after that repair, now BOTH channels could shut down with either speaker.  I could have thrown in the towel at this point, sworn never again to use a Krell amplifier, and so on.  But actually, it dawned on me that shutting down is exactly what the Krell is supposed to do when there is a speaker short.  The the final Krell repair was actually to restore this correct behavior to both channels.

And so back to what had been highest on my list after the tweaks didn't help: the speaker transformers.  (In fact, it took several weeks of testing before it occurred to me it even might be the amplifier.)

I had in fact tested the high frequency (HF) transformer in the most suspect right speaker back toward the end of June.  I disconnected the HF transformer, and, playing with no highs and a very recessed sound, and there was still a shutdown.

I hadn't believe the low frequency (LF) transformer would have a short for several reasons, the first being that I highpass the speaker signal at 100 Hz, taking a big chunk out of the low frequency current in the  transformer, and second that it's a big honking transformer that looks like it could handle anything, whereas the HF transformer is this little tiny thing.

It was right about then that I narrowed the problem to an apparently malfunctioning amplifier channel (actually, it was the channel that was operating correctly, and the other channel which was not shutting down was the one that was not operating correctly).

So I never got around to testing the LF transformers.  Until I got the amplifier back and it then seemed the most logical thing to do, again.

But first I decided to tackle something which had long bugged me, and would be less expensive to fix, I cleaned the HF power resistor connection which had turned green in both units.  I took a day to thoroughly clean away all the green and make sure the contacts were making good solid contacts (which, in one channel, it might not have been).

Sadly that did not stop the shutdown.  Or as usual it worked for one day, and I was elated.  Then, the next morning, I got a shutdown again.

By this time, I had told my story in the DIYAudio AcoustatAnswerMan blog, and the AcousatAnswerMan himself, Andy Szabo, said he had not seen the HF transformers go bad.  He strongly believed the LF transformers were at fault, is that had always been at the root of intermittant speaker shorting.

Well, I then had a very busy weekend, so the hopefully final really big deal would occur.

Disconnecting the LF transformers was easier than I expected.  They simply unplug with a slide connector.  But then you have to insulate the loose end.  I figured out how to first wrap the connector in non-sticky plastic (to keep it from getting sticky), then electrical tape, attached very thoroughly to the wire insulation.  Then the whole thing is wrapped around the now essentially disconnected 1 ohm resistor.  One one side, I further protected it from coming loose with a nylon tie wrap.

And so, I started to play on Saturday September 30.  The most difficult source: my F-26 tuner digitized through Lavry AD 10 at 96kHz.

It was playing great for over 4 hours.  By this time I was giving myself high fives again.  This was it, I am sure, the LF transformers have an intermittant short.

This was also an opportunity to do some other things.  For one, I was hearing the HF transformers more directly, and now I think I might have been able to hear the old electrolytic in one speaker, vs the new large polypropylene in the other.

I turned the balance all the way to the left channel, which now has the speaker with the polypropylene cap.  It was sounding cleaner, purer, I thought.

To be sure, I turned the Tact to Mono mode, so I was getting the whole signal.

It continued to sound clean.

Then, 5 minutes later, playing only through the PP capped speaker and the left channel, it did the impossible again, for the 200th time.

It shut down.

Maybe I just can't play the FM that way...which I had never in fact done for very long before May.  I vaguely recall getting shutdowns playing FM that way, so I went back to using a more bandwidth limited Sonos connection to the FM tuner.

So that will be the test tomorrow.



No comments:

Post a Comment