curious about the "devil's channel"

[marco08]

did a bit of tightning of the socket on the left channel 300b, as i had intermittent dropouts that stopped when flicking a fingernail on the glass enveloppe. Got rid of it, but it started me thinking : dont you guys and gals find it weird that roughly 80% of the electrical (or mechanical in my case) bugs and bad soldering reported on the beepre are on the left channel I did check the math before posting !

[marco08]

as a matter of fact, the left channel curse is not over.. il still get the drop outs. Same issue someone reported a while back : volume drops by 95% but their's a little bit coming throuh.

Doc suggested an issue with the voltage regulator. I checked, it's snug against the heatsink. I tried reflowing all the joints, except the ones on the PCB.

strange, as the volume comes back with a small whack... I'll try reflowing again (PCB included)

Unless someone has a nice vodoo suggestion, relevant to the Halloween theme...

[Doc B.]

Are you monitoring the cathode and signal voltages to help to understand what is happening when the sound drops out? Does the tube stop glowing?

[marco08]

tube is still glowing, and it's not cross feed (i took out the other input RCA).

Difficult to monitor the voltage, it only happens after a few hours of up time. Maybe I could connect an aligator plug to the pins and "leave it out" on the meter next to it (I would have to be carefull no one is near).

I'll try next weekend, which points should I try to monitor ?

thanks

[Doc B.]

If it only happens after a few hours it's a thermal issue. So it could be inside the tube (have you swapped tubes?), or maybe a pin connection is expanding or warping a wee bit and losing contact with a tube pin (hence my question about the filament glow). Or maybe a solder joint on the socket is getting flaky with heat. You could monitor the plate voltage at pin A2.

[marco08]

thanks doc. Not the tubes, I swapped them and tried another pair. I wiggled the socket and cleaned it too.

had a quick look this morning before going on a three day trip to egypt.

the wires from the regulator PCB where dodgy and sheared of when I lifted the board to measure voltage. must have hurt them when stripping the insulator (or they got brittle with the heat ?). Rewired, checked the voltage on B3, got 108v... Beepre worked  for an hour, then muted again (the filament still glows well). Tapped the base plate, and re-started...

I'll get this bug !! but pyramids first for a few days.

By the way, it still does not tell us why the majority of fails is on the left channel !! smile

[marco08]

fourth reflow since day one... seems to do the trick !! (like usual).

Does not tell us why the left channel is more prone to human misshaps when soldering .. anyone have a guess ?

I did remake the base using a nice zebrano wood, and I left a few millimeters between the baseplate and the sides. This, I think, has the double benefit of airflow to improve cooling the underside, and isolating the microphonics by having the baseplate rest on silicone mats fixed on the sides.

looks and sounds good !!

(http://)

[Rocketman248]

I suppose one theory could be that for most BH kits, you start with the right channel, then move over to the left.  I know I've made mistakes because I don't read the steps as thoroughly the second time.  Either I assume its exactly like the right channel, or I just plain rush through it.  Messed that up on the Kaiju the other day.

[RPMac]

Interesting that my only problem has been a socket, the left one.

[Doc B.]

I blame aliens and the Loch Ness monster.

[Natural Sound]

I blame aliens and the Loch Ness monster.

You left out Sasquatch. How can a self respecting Pacific Northwest resident omit Sasquatch?