BeePree Silent and amazing - for the first 200 hours - loud hum in lt channel!

[ewingfox]

Hello Bottleheads!

After nearly 200 hours of runtime with my amazing BeePree, the left channel has developed a very disturbing hum that is present after a few moments of power up and is audible at listening volume.  It does not change with volume knob, and is present with or without sources connected, is unchanged between source selector positions and does not follow any of the tubes when I flip them from channel to channel.

I flipped it over and there are no signs of vented caps, burned wires etc, and a resistance and volt check shows that everything is still within spec.  Thinking it could be power or amp related, I brought it to a friend's house and the hum did not change.  Other than the tiniest of voltage present (few mv) on startup, there is no evidence of voltage across any of the inputs or outputs. The resistances for terminals marked "greater than 100Kohm" (OL) do not nor have they ever climbed above 40kohm but I think that may be my meter, and it is is very consistent across the whole build (IE both channels)

I'm at a total loss for what to try next and hope that a guru (or pack of gurus!) has a good suggestion or two!!

Thank you in advance,

Ewing

[ewingfox]

So... You may notice that I diligently, carefully listed every step I did in my hairbrained attempt to NOT REFLOW EVERY VISABLE SOLDIER JOINT. If I had a quarter for every time I see PB or another guru ask if a frustrated bottleheader had done this... I would have many, many quarters.

It took perhaps 30 minutes to carefully flow out every joint and suprise - bee pree is back to dead quiet and lovely.

Here is an interesting question - what would cause this to start happening after 200 hours?  It wasn't moved or bumped or jostled or launched from a trebuchet. I have had cold solder joints be an issue in the past back when I first started doing this kind of stuff... But they always reveal themselves during checks or early on during burn in. This is my only (weak) excuse for not jumping to reflowing from the start.

Thoughts anyone?

Ewing

[Jay]

Perhaps repeated heat cool cycles could compromise a solder joint over time if it was closer to some threshold.  That or the cat did it.
Jay

[ewingfox]

Jay,

I was wondering about the hot/cold.  This thing is like a space heater!!!  Paired up with a Parasound A21 and PSB Imagine T3's this thing is just sublime.  The room is "warm" so pairing it with a SS amp is a great combo. 

I did end up replacing my cheapo amazon special soldering station with a Haikko about halfway through this build.  Wonder if that lead to a few cold joints as the garbage unit was starting to die.

E

[Paul Birkeland]

Can you measure the voltages at POS IN and POS OUT on each channel.  Also check the 10,000uF caps under the power supply PCB.

[ewingfox]

I will - I was feeling really  good about things and was doing some listening when the sound returned.  Tapping the caps under the transformer for the LT channel changed the sound.  It then went away, and listening was great.  20 minutes later, really bad sound then two of the diodes went out, and the tube is now not lit.  I immediately killed power, rolled the tube to the RT channel /no amp connected and the  tube lit, so I will need to go back and run voltages and see what broke. I feel that the cap is probably bad  - your intuition is pretty awesome to have made that suggestion!

[Jay]

Any follow up?

[ewingfox]

Hello Bottleheads!

I have had a bit of a personal setback that makes working on HV electronics a bit nasty - a pretty wicked tremor that has proven difficult to tame enough to work inside the amplifier and feel safe about doing so.  The issue also makes it a bit tough for me to organize my thoughts, so I apologize if I'm not being very clear in my update. The loud hum on the left channel is gone, it has completely stopped working. I haven't had the 'pre in my system, it has been benched since my last post as I was finding it too challenging to steady myself to work on it.

I fired up the 'pre today in advance of a call to support regarding getting this fixed by the in house team and decided to try and get a few measurements.  What is interesting is that from viewing the board from the bottom, the diodes on the left side of the chassis (Left channel) are off on the B side of the board and blinking on the A side. The blinks correspond with some of the voltages that I found.  Everything that deviates has been indicated here. Where indicated by (Variable) they are jumping between the two voltages indicated at a similar cadence to the blinking diodes.

11 - 0v-146v (variable)
15 - 0-190v (Variable)
16. 2.8v
17 1-190v (Variable)
19 .23v
30 -7.3v
31. 0-3v (Variable)
37 .05v
40 227-230V (Variable)
43 227-230V (Variable)
46 227-230V (Variable)
B2 - 220v

[Paul Birkeland]

Every time that I've had blinking LED's, it has indicated a loose or broken wire.

[Doc B.]

Yeah, that's the first thing I would look for too.

[ewingfox]

based on the voltages I am reporting, any place to begin my search?

[ewingfox]

Thanks to those who posted and to Doc who suffered through 23 minutes of listening to me stutter -

I had diligently re-flowed all of the numbered connections but had failed to notice that the 80SQ045N diode running from terminal 28 (which had been soldered, re-soldered and flowed again and again) was running through terminal 5 of PT-7 A but was NOT soldered - the diode tail running to terminal 31 had been wrapped around PT-7 terminal 5 and soldered, but I had left the other tail loose so I could carefully square things up before locking it down (I'm a bit of a nut for right angle intersects)  I had somehow not considered re-flowing them as I had worked my way around the kit from 1 to 57 in numerical fashion.

The LEDs immediately re-lit and things look very good, all the voltages had stabilized across the entire channel and my B+ is within 2v channel to channel heading into the C4S.  All I can assume is that after a number of hot/cold cycles, the wire had lifted away from what was basically a bare mechanical connection.  In my prior troubleshooting, wiggling the cap between terminal 28 and 29 was actually bringing the diode tail running from 28 to terminal 5 PT7 into contact.

More than a bit embarrassing that the dumbest mistake I have made with assembling electronics gets discovered while on the phone with one of my greatest heroes, but I appreciate everyone's feedback and follow up even as I let this thread lie dormant for so long!

Now back to doing what I'm better at these days - drinking seltzer water and listening to music with my feet up!

[Paul Birkeland]

There's no shame in asking for help.  The only way that we are able to provide it with any proficiency is by making the same mistakes ourselves, over and over again!

[Doc B.]

Ewing's leaving out the part where I mentioned that all of us here have plenty of stories about missing a solder joint. Happens to everyone. Speaking of stuttering - which I didn't really notice from Ewing because we were having a great conversation - here's a little story I saw yesterday:

(https://scontent-sea1-1.xx.fbcdn.net/v/t1.0-9/20431265_10211741897117682_2260447918648826258_n.jpg?oh=163efabd527029d79ff163bbd726ba80&oe=5A3181C2)