[resolved] Crack + SB Woes. Left Channel dies after 5 minutes

[Blue]

So far it has been 10 days using the stock crack with the small board. No problems so far.
I'd like to attempt to reinstall the big board in the crack.

I haven't done any soldering rework on the big board yet. Any suggestions on what I should do before I put it back in?
I'm tempted to resolder all the components for the left channel on the big board, however I am waiting on a new set of LEDs before I do that (I read that one shouldn't reflow the LEDs due a high possibility of them cracking on a reflow).
How tight does the transistor need to be against the heat-sink? I can't tighten it anymore that I originally had it.
However, for the right channel, all i used was a flimsy screw driver and a pair of pliers, and yet the right channel has non of the problems I have in this thread.

Could this problem occur from any other places? Perhaps poor solder join somewhere else?

 

[Paul Birkeland]

It might not be a bad idea to replace the TIP50 and 2N2222 where you originally had the mounting insulation issue (A side).

 

[Blue]

I did replace both the TIP50 and 2N2222 once already. When I replaced the two, the crack worked without problem for two weeks.
Once the problem occurred again, it would fail after 5 minutes of operation like originally. I have another 2N2222 and a few TIP50s remaining.
Once the LEDs come in, I'll replace all of them at once.

 

[Paul Birkeland]

It would also help to know what voltages change once the board stops working.

 

[Blue]

So I decided to try just replacing the 2N resistor and TIP50 in the meantime.
Before I removed the 3kohm ceramic resistors, I did a full resistance and voltage check, all good.

Now I attempted to put back the big board. This time the B+ voltage on the bigboard fails. It falls to 112 volts. I only quickly checked OA/OB (they were about 79 volts), but stopped measuring because I started to feel an intense amount of heat coming from the Crack. I didn't want to destroy any other components  Did I destroy my big board? I wouldn't be surprised as I have reworked it many times now.

Therefore I put back the 3kohm resistors and left the small board in. I did a full resistance and voltage check, and everything seems to fit, other than T2 and T4 being a bit higher than the tolerances., but I am guessing that is due to the small board

Here are my voltage checks with stock crack and small board:

T1: 80.8
T2: 197
T3: 0
T4: 197.2
T5: 83.9

T6: 0
T7: 111.1
T8: 0
T9: 117
T10: 0V

Small board:

OA: 80.5
IA: 196.6
B-A/B: 0
IB: 197
OB: 83.8

Should I just give up on the big board and order a new one? I'm afraid that eventually I'll destroy my transformer.

 

[Paul Birkeland]

If the B+ voltage drops like that, the TIP50 transistors are not properly insulated from the heatsinks.

 

[Blue]

I’ll get back to it again on the weekend.
As for mounting the tip 50, it goes screw, Fiber washer on top with protruding end facing towards the insertion side, transistor, thermal insulator tab, heat sink, metal washer, lock washer, nut.

I’m sure my picture shows that too. I find it perplexing that isolating the transistor is so complicated .
I’ll take a picture of each step of mounting the transistor against the heat sink.
I’m also tempted to try one of your tests in an older thread: pulling back the transistor from the mounting kit, and measuring the voltages then, running the amp for no longer than 30 seconds

 

[diynewbie]

I  would think that you could check continuity or resistance between the heat sink and metal tap on the transistor to verify proper installation. Correct?

If the shoulder on the fiber washer did not seat properly it could result in a short between the transistor and heat sink even if order and orientation of all parts are correct.

 

[Tom-s]

This might help.

It's essential that the screw with fiber washer are centered in the transistor.

https://www.youtube.com/watch?v=UWNVMkNwkUw

 

[Blue]

Yea I followed the exact same video when I originally built the speedball, and probably watched it another 10 times trying to solve this problem.

I also checked for contact between the transistor metal tab and the heat sink.
My Multimeter was not beeping. The transistor is currently disassembled from the heat sink. I’ll be trying it again on the weekend.

 

[Blue]

So I pulled both TIP50 transistors away from their heat sinks to measure the voltages, the amp was on for no longer than 25 seconds.
B+: Starts at 185 volts, and then drops to 112 volts after a few seconds just like before.
What should I check for next?

 

[Paul Birkeland]

Honestly I would just replace the big PC board.  There's only so much debugging and rework that you can do on one of those PC boards before they are destroyed, and my guess is that you're getting close.

If the B+ drops like that with the TIP50 transistors not touching the heatsink, then you almost certainly have a shorted transistor on there. 

 

[Blue]

Yea, I’m going going to order an entire new big board.

I’ve already replaced the offending transistors twice, and I see no visible bridges between the transistor legs. I am not the best at soldering, but I really don’t see any bad joints.
Last week I checked for shorts in the transistors with my multimeter, but it showed it was fine.
Perhaps reinstalling the big board today caused a short.

However I’m not interested in damaging the rest of the components, and I’m slowly going crazy checking the same things over and over. I need to email replacementparts@bottlehead.com correct?

 

[Paul Birkeland]

Yes.

 

[Doc B]

I just did a repair of a different kit, but with similar style of upgrade PC bards. The issue with that kit was that lots of the wires connecting the board to the circuit below had been nicked by the wire strippers used and the wire had fractured just inside the insulation where it wasn't visible. lt's worth checking the connecting wires themselves for continuity.

 

[Blue]

Thanks for your reply Doc.
I went ahead and checked for continuity in the wires, they seemed fine. However, I replaced the all the wires on the speeball big board anyway.
Still have the same problem.

I decided to check the 2N2222 Transistor and the LEDs on the board. The top side  of the big board were measuring properly.

The bottom side measured wrong though. The TIP50 measured correctly on the bottom side of the big board as before, but the
2N2222 transistor was measuring 1.025 volts from Emitter to Base when it should show over limit.  Also the the LED closest to the TIP50 transistor was not showing any reading at all.
It's the third time I had to replace the 2N2222 for similar reasons. However this is the first time that the LED failed to measure.

I'm still waiting on my delivery for the LEDs. If installing all new LEDs and transistors for the left channel don't work, I am going to purchase a new big board, as Paul said, it's probably better to do just that.
Thanks for your help everyone.