Left channel problem [solved]

[pranabindu]

Here's my story:

Finished build; resistance and voltage both tested fine.  At first, the sound had its definite positives, but they were made irrelevant by the ringing of the tubes - very fatiguing sound, bordering on headache inducing.  The ringing was way worse on the left channel.  Sorbothane under the stock feet helped the ringing some, and the vinyl covered lead rings helped immensely.  I'm not sure if it was the lead rings or break-in, but by then (about 70 hours of play) the harshness was gone and the sound was lovely (no fatigue), with the ringing only somewhat discernible in the silences between tracks - especially really dynamic or bass-heavy music.

Then, as the unit warmed up (over the course of an hour or so, gradually decreasing in occurrence but not in volume), the unit would make tearing, popping, humming noises through the speakers, whether playing music or not.  The noise would go away after the unit warmed up for a while.  This happened every day for a couple of weeks.  Then, one day, the left channel cut out, except for a very low-level signal (either coming only from the tweeter, or too low to discern in the woofer).  30 minutes later, the left channel went out altogether.  Tapping on the top plate caused the same popping, tearing sounds, and momentarily (split second) would bring the left channel back.

Switching the 300B tubes to the other respective channel did not change things (left channel still out).

I put it aside for a few days.  Tonight, the unit tested normal for resistance.  It tested normal for voltage, too, except for the following: C4/D4 tested at 6.8V each (supposed to test "N/A"), C5 tested at 0V, and C6 tested at 6.8V (both supposed to test "N/A").  Resistance and voltage tests done with tubes installed.

Help me Obi-Wan Kenobi.  You're my only hope.

[Paul Birkeland]

Tapping on the top plate caused the same popping, tearing sounds, and momentarily (split second) would bring the left channel back.

This is a very sure sign of a loose connection or cold solder joint. 

-PB

[pranabindu]

I went through each test point, attaching the multimeter with a clip at each so as to be able to wiggle every wire and lead soldered to it.  The test points associated with the left channel transformer mounting screws (all but one of them) were not stable at 0, and the resistance went all over the place when I lightly twisted the board on the offsets.  I had had the transformer caps powder coated, so I went through and scraped the powder coat off each side of the cap so that each mounting screw and washer would have a clean contact (both sides of the transformer cap).  That brought the resistance reading at each test point to 0, with no variance when twisted.  I thought that would be it, but the problem didn't change at all.  The left channel would still pop and scratch and tear, especially when tapping on the top plate.

I went through and tested resistance at each test point again.  Every test point tested within normal for resistance, and no amount of wiggling of wire or lead (I did it with a wooden chopstick) would change the resistance reading.  I did all this with all tubes installed.  The problem remains.  I also wiggled the tubes while testing their test points, and this did nothing.

I don't know what to do next.  Switching the tubes (either of the two pairs) doesn't change the location of the problem (left channel).  I don't know how to test the solder joints at the locations that are not test points.

Thanks for any suggestions you might have.

[2wo]

You have a poor connection/solder joint. Re-flow all of your  connections paying extra care to the left channel, you may have a case were 2 of 3 wires  are fine and 1 is flaky.

People seem to resist doing this but it works almost every time...John       

[pranabindu]

Re-flowing the solder joints on the left channel did the trick.  2wo's observation about folks' resistance to doing this certainly applied to me.  I kept thinking about how long it took to assemble the whole unit.  If you're reading this and have a similar problem, don't despair of the task.  It is so much easier than assembly, and you don't have to follow the order of assembly (I progressed from the back of the unit to the front).  A flux pen touched to each joint before it is re-flowed gets the solder flowing a lot faster, so I would definitely recommend it.  Depending on how much solder each joint had, I would treat each joint differently: too little solder, and I would heat to flow and then add a touch of solder; just the right amount, and I would just re-flow; too much, and I would use a solder wick to remove most of it and then follow up with fresh solder (flux applied to the wick made it really absorbent).  I also took care to make sure each wire or lead in the joint had a mechanical connection to the joint before releasing the iron.  I kept the mechanical connection with pressure (finger, chopstick) until I was sure the solder had hardened.

The preamp now sounds completely different.  It's actually a quiet preamp!  The ringing is 95% gone (can't discern what causes it when it happens, and it's very low level now).  Microphony in general is still threatening, but the current setup makes it unnoticeable - heavy maple shelf resting on racquetball halves, BeePree's stock feet on little sorbothane hemispheres, vinyl covered lead rings on the 300B's.  The long-term plan is to build a birch ply knock-off of the MusicWorks Revo shelf, with diy roller bearings and diy Symposium shelves under the components.  Until then, I'm a happy camper and eager to play around with my new Rothwell attenuators and various mods to other gear.

Thank you 2wo and Caucasian Blackplate for your help.  You made the work I put into the whole project totally worthwhile.  Despite not having a real opportunity to do critical listening, the improvements to the system are obvious (from VTL IT-85 to BeePree/VTL ST-85).  More texture (detail?), clarity, microdynamic pep, wider (or more fine-grained, I can't tell which yet) tonal palette.

Happy listening to all!

[Paul Birkeland]

If I had to write a set of instructions regarding how I perform repairs that arrive at our facility, your post would about sum it up.  (Though I use a sprung vacuum pump to pull off the excess solder)

Your observation about how the sonic performance of the preamp changed with better mechanical connections is something I have noticed with frequency.

I'm glad you got it sorted!

-PB