Dead Channel on BeePre

[boulos]

Hi,
Last night I finished installing the BeeQuiet on Gerry's BeePre.  I turned the preamp on, and everything worked fine.  Today, Gerry and I turned on again to listen, and for about 15 minutes it sounded great, then all of a sudden one channel started developing noise and died out into static noise (like TV "snow" noise).  The sound from that channel is like the sound that comes out of the BeePre when it hasn't warmed up (the first few seconds of turn on).  We did all the necessary tube and cable swapping, but the bad channel didn't budge.

Voltages on both channels checked out except for OB on the C4S boards. The good channel measured ~130, while the bad one measured ~143.  I checked the potential difference between the collector and emitter of the MJE5731A transistors on both boards. On the good channel it is 15V, while on the bad channel, it is 0V.  The difference between base and emitter on both channels is 0.6-0.7V. Note that on the bad channel, the two LEDs on the B-side of the board are dim; while the 6 other LEDs (the ones on the A-side and the four on the other board) are bright. On the filament board Kreg measures 4.0V on the bad channel and about 4.9V on the good channel while +reg measures about the same on both boards.

Next, I took out the EL84 tubes from both channels and turned on the preamp.  While the good channel worked almost immediately, the bad channel slowly went from static noise, to muffled sound to low sound then to full sound equal to the good channel.  With the EL84s out, the B-side LEDs on both C4S boards are bright, while the A-side LEDs on both are dim.  Kreg and +reg on both filament boards measure equally with the EL84s out.

Thanks so much for your help.
Boulos

[Paul Birkeland]

Did you swap 300B's between channels?

Your OB voltage being that high means the 300B isn't drawing any high voltage current.

[boulos]

That makes sense. I think the dimmed LEDs corroborate that.  Two questions:
-- Why does it work with the EL84s out?
-- How can I fix that?

[boulos]

Sorry, I didn't answer your question.  Yes, I swapped the 300Bs.

[Paul Birkeland]

Sorry, I didn't answer your question.  Yes, I swapped the 300Bs.

What were the results?  Did the high voltage at OB followed the tube?

If it did, you have a problematic tube.

If it didn't you have a loose connection somewhere in the offending channel. 

-PB

[boulos]

Hi Paul,
I hadn't re-tested voltages when I switched the tubes last night, but when I did, the bad channel didn't switch. I tried it again today, and OB did not change when I switched the tubes remaining at 143V in the bad channel. But... OB in the "good" channel now went from 130 to 143V!  So I exchanged both tubes for a different pair, and guess what happened, OB went to 110V on one channel and 105V on the other.  So BOTH tubes were bad!  What are the chances of that? Perhaps it just happened that one side was going over the edge enough yesterday to work and was throwing me completely off.
Thank you for your help!  Gladly, I changed the pair of tubes before I started melting solder joints.
Boulos

[Doc B.]

The tubes that aren't working may just have a little corrosion on the pins. Try cleaning them with a scotch brite pad or a little fine sandpaper.

[boulos]

I'll try cleaning the pins. These are 10 year old EH tubes, but they weren't being used during all this time, so it is strange that both would fail.

For reference, all LEDs on the C4S boards are now bright.

The BeeQuiet was a fun build. It was relatively easy, but that is mostly thanks to the excellent instructions.  Attaching a photo

[Doc B.]

Could also be the contacts in the sockets have stretched a bit, creating intermittent contact. You can usually tighten them up a bit by working on them with a small screwdriver tip.

[Paul Birkeland]

So BOTH tubes were bad!  What are the chances of that?

Very, very remote, that's very odd.

These are 10 year old EH tubes

Oh, never mind, that's not super surprising.

If you grab a new pair of EH 300B's, you'll notice that the filament supports in the top mica got a redesign, as the old ones tended to physically fail after a certain amount of time. 

-PB

[Chris65]

The BeeQuiet was a fun build. It was relatively easy, but that is mostly thanks to the excellent instructions.

Thanks for the photo, just what I was looking for as I start my build. What are the caps there? Some vintage Russian?

[boulos]

Thanks for the photo, just what I was looking for as I start my build. What are the caps there? Some vintage Russian?

They are Russian MBGO 10uF 300V caps.  If you install them, be careful that they're heavy. They needed some RTV Silicone to support the base of the ties. I actually thought the caps were the problem yesterday and measured them. They both measured 10.5uF and their ESR, dissipation factor, and Q were matched and very good.

[Chris65]

Thanks for the info. Something to consider!

[mozarti]

You may want to remove the o-ring under the tube on the offending channel.  My Beepre had occasional white noise and the problem was solved by taking the o-ring out. Apparently, the tube's pins did not have good contact with their socket on account of the o-ring.

[boulos]

You may want to remove the o-ring under the tube on the offending channel.  My Beepre had occasional white noise and the problem was solved by taking the o-ring out. Apparently, the tube's pins did not have good contact with their socket on account of the o-ring.

Thanks for the tip!  I hadn't thought about that, but since the BeePre was upside down during testing, I wasn't using the O-rings to lesson the chance of the tubes falling out. So the measurements were all taken without the O-rings.