Might have wiggled something loose... right channel static [resolved]

[Colonl_Charisma]

Hi everyone,

My BH Crack has always had a little bit of a buzz in the right channel. It's always been faint, and it does not change with increasing or decreasing the volume. Well, tonight, I decided to change my input tube after not removing it for close to two years. I tube rolled a new 12au7 I obtained, so there's no fear of using an incorrect tube.

Upon turning on the amp my right channel buzz became so LOUD that it was noticeable across my room. I confirmed that it was only coming from the right channel in my headphone, and the sound eventually went back to its normal faint sound after about 10 seconds. I allowed my BH Crack to warm up for 20 minutes, and then I tested to see if anything sounded different (e.g., no sound from the right channel). Thankfully, everything sounded great except for the static.

Since this occurred, do you think I may have wiggled loose a poorly soldered joint somewhere near the input tube socket (or on the socket)?

Thanks for all the help!

EDIT: The sound from the right channel is specifically a constant, low buzz.

[Doc B.]

Why don't you reflow the connections on the socket and see what happens? Probably best not to reflow the LEDs, just the other connections.

[Colonl_Charisma]

Thanks, Doc! I'll definitely look under the hood once I get home later today. I'll be sure to not reflow the LEDs on the socket. Even though I gently switched input tubes, there was a little bit of audible noise inside the amp. This might be normal since my input tube sockets are fairly tight, and I had to "rock" the tube a bit to get it out.

[Colonl_Charisma]

For the sake of thoroughness, here are photos of what my input tube socket soldered joints look like. I believe I see some areas of improvement in terms of soldering.

[Doc B.]

One thing that is not well understood in this hobby is that, while you may be able to easily spot a bad solder joint like a crystalized cold joint or one that was completely missed when soldering things up, you cannot look at a solder joint and know that it is good. Even if the solder looks shiny with a smooth fillet and covers the entire joint you can't know for sure that it is good underneath that unless you measure resistance at two points that are on either side of the joint, with the test lead tips explicitly not touching the solder joint itself. That way you are measuring the conductivity thru to joint from one component to the other. This is not always easy to do and oftimes it is a whole lot easier and faster to just reheat each joint until it flows well and cools to a shiny smooth joint - adding a tiny bit of solder when necessary. Then test to see if the issue has been resolved.

PB does all repair work these days. But when I did it - on a simple circuit like a Crack amp that was exhibiting any issue that seemed potentially related to connections I would take 5 to 10 minutes and reflow every joint on the amp. And the majority of the time this would fix the issue. This doesn't apply to our kits alone. I was having trouble this winter with cold starts with R-Ambo, my race hauler ambulance. It's a common issue and a dealer will happliy charge you $600 to replace the fuel injection control module to solve the problem. It turns out that the cause is exactly what we are talking about, a couple of bad solder joints that look perfectly good dropping the FICM voltage due to their increased resistance. The problem appears when the air temperature gets cold and the cause is heat cycle cracks in the solder that develop over time. Half an hour spent pulling the FICM out of the truck and reflowing those two resistor connections on a PC board had the cold start FICM voltage right back in the sweet spot. Even if a bad solder joint doesn't turn out to be the problem it should have been  checked and crossed off the list of suspects during troubleshooting.

[Colonl_Charisma]

Thanks Doc for the shared wisdom. It is much appreciated.

I finally found some time to look under the hood. First, I turned on my amp to do a chopstick test, and I noticed that my input socket LEDs are not lighting at all. I've included a photo to show. This is quite odd, because all I've done at this point is flip my amp over lol. It was working just fine except for the right channel buzz.

My chopstick test was negative.

What is the opinion on the input socket LEDs? The others light up just fine. I will say that since I installed the speedball, those LEDs were always dimmer, but never completely unlit.

EDIT: I also know the drill haha. Here are my voltages from the Speedball board:

OA = 188 V
OB = 187 V
G = 0 V
B+ = 194 V

[Paul Birkeland]

Is the 12AU7 tube still glowing?

[Colonl_Charisma]

Hi, PB! Yeah, the tube is still glowing. I flipped the amp back over, let it warm up, and then I played music with sound coming from both channels.

The buzz is still present in the right channel, but I have not done anything in terms of re-soldering.

[Paul Birkeland]

I would not use this amp until the issue is fixed.  The LEDs going out and the Speedball voltages going up like that means that the 12AU7 is no longer conducting.  100% there is a broken wire, unsoldered connection, or loose connection in there doing that.  You should be able to poke around with a wooden chopstick in order to figure out what needs to be poked to get those socket LEDs to come back on.

Whatever you do, don't just obsessively resolder the LEDs on the sockets themselves, as that is just going to damage the LEDs and it's most likely that your issue is elsewhere.

[Colonl_Charisma]

Thanks, PB! I'll be sure to not use the amp until this issue is resolved. What's crazy is that I have no idea how long the LEDs have been out for, because it's been a couple years since I last flipped my amp over haha.

I'll perform a more thorough chopstick test. I will state that I only poked around a few joints the other night.

Is there an idea as to specific joints to check?

[Paul Birkeland]

The black wires in the amp are common to both channels and you have an issue common to both channels.

[Colonl_Charisma]

Hi, PB! I did two extensive chopstick tests with headphones plugged in, one yesterday and one today. No luck when pushing and even tugging a bit on cables.

I did some re-soldering today, but that did not get the two input socket LEDs to turn on.

One of my posts from 2021 had my speedball voltages in it, and I agree that mine are way off from what is normal.

Two questions:

1) Is it possible that my LEDs may have burnt out? The reason I mention this is because at one time a couple years ago they were both very dim.
2) Could I use my multi-meter to help track down the issue?

[Paul Birkeland]

They will operate relatively dimly in that position.  If you can carefully measure the DC voltages at A3 and A8, that would be helpful. 

For what it's worth, I have really old kits here in my basement that are 20+ years old stuffed full of LEDs and they are all working totally fine.  The problem you can run into is that they absorb moisture over time, so if you try to resolder them after sitting for a few years, the moisture will flash boil out of them and destroy them.

[Colonl_Charisma]

Thanks, for the timely reply, PB! That's reassuring about having kits that old, and them working without issue.

I just measured the DC voltages at A3 and A8 on the input socket:

A3 = 182.6 V
A8 = 167 V

I also remeasured voltages on the speedball board, and they remain unchanged.

[Paul Birkeland]

Are you certain that's A3 and A8?  That would be extremely odd and would make me pump the brakes full stop on using that amp again.  What DC voltage do you get on the center post of the 9 pin socket?