Crack + Speedball build voltage issue

[aarond]

Hi from New Zealand!

I recently completed the Crack build and after an initial issue with a 180

[Paul Birkeland]

Hello Aaron,

Welcome to the forum. 

My best initial guess is that you have an LED in backwards, I'd double check that first.  Otherwise:

Do both sets of B+ pads read the same voltage with the amp on?

Do both sets of G pads read the same resistance to ground with the amp off? (zero)

Since the rest of the LED's light up, the transistors are most likely in the correct places.

-PB

[aarond]

Thanks for the quick reply, PB!

All LEDs are in the correct way, the B+ pads read the same (196V cf nominal 170 V) and the resistance of the G pads is the same (about 1 ohm).

I wonder if one of the LEDs is faulty, since they are in series if one doesn't work then the circuit would be broken - or is it more likely for the failure to result in an open circuit? I can't read any continuity across the LEDs.

[Paul Birkeland]

The LED's can be checked with the diode function of your meter.  You should get a reading using the probes one way, but not the other.  Do note that different meters behave differently, I have one meter that will light up both LED's in the C4S circuit when doing a diode check, another meter that only lights up one LED when doing the same check, and another that lights none of them, so go by the numbers.

The 1.5V on the O terminal/terminal 9 looks like the C4S just passing DC from one of the LED's, which perhaps could indicate a short in the 2N2222.  You can set your meter to resistance and measure between each pair of legs on the 2N222 transistor (helpful if you can set your meter to beep for continuity).  See if there is a short between any pairs of legs.  Sometimes this can happen if the body of the transistor is sitting against the board.

-PB

[aarond]

Ok, so I've checked the transistors for shorts and there is no continuity between any of the pairs of legs on either one (and the bodies are well clear of the board).

With regards the LEDs, there's something strange going on. I get a reading on the diode setting of my meter, with black lead to stripe end of LED. But this is only for the outer most LED of both pairs on the main board. I have tried and tried to get a reading over the inner one of both pairs, but I just get an above range reading, in both directions.

To summarise, the outer two LEDs on both sides give me a reading with black lead to striped end, but no reading the other way. The inner most LEDs from both pairs give me no reading in both directions. Both LEDs on the right side of the main board light up, both LEDs on the left do not.

[Paul Birkeland]

I'd unscrew the board, flip it over, and resolder the joints on the transistors in that area of the PCB.  Specifically, give some extra heat to the center solder joint on the TIP-50C, as this connects to the tab of the transistor, which is heatsinked, so it will sink quite some heat from your iron.

-PB

[aarond]

I reheated all the solder joints for the 3rd time, still no change. I also checked the continuity of each transistor leg with the next component on the board, all solder connections were good.

I guess the next step is to get some new parts to systematically replace the transistors and LEDs on that side?

Thanks for the advice so far, it seems so close, yet no cigar at this stage.

[Paul Birkeland]

Yeah, I'd contact queen(at)bottlehead.com, and ask for a new TIP50C, 2N2222, and a pair of LED's.

If you don't have any, grab some solder wick to help you get them out. 

[BNAL]

That was what I did after my crazy x through my Crack with speedball against the wall. I visually was able to see the TIP50 was damaged, so I replaced it, but I still did not have one channel. I decided to revert the Crack and took the speedball out to make sure that everything else was OK.  I finally got around to installing the speedball after re-soldering everything on the board. Still one channel was out. I ended up replacing the 2N2222A and LED

[Paul Birkeland]

Yeah, sometimes it's best to just replace all those solid state parts to be sure everything is A-OK.

Looking towards the future, we will try to include transistors with the kits that all have major appearance differences so we can minimize these errors, and we switched to flipping the LED's over to ensure that the stripe is always visible, which has had an impact on instances of backwards LED's.

[aarond]

I finally got the replacement parts, many thanks to Eileen. Had some issues with NZ post sending them back to the US for some reason, but the second time sent they arrived within a week.

I replaced the LEDs on the side where they weren't lighting, but when I went to put the board back, I noted that I'd installed the 'top' towards the back, so re-inserted correctly. Now the new LEDs are at the right, which was the side which was lighting up before hand. The LEDs which were lighting are now on the left, and they refuse to light now - so it's pretty clear that there is a issue with that side of circuit, if the LED + transistors which work on the right side now do not work on the left.

Unfortunately, I'm at a complete loss as I've rechecked all of the connections, re soldered all the wires to that main board 10 times at least now.

Any ideas where I should be focusing my attentions now?

One more thing, all of the LEDs on the small boards light up initially, but over a period of about 2 minutes they gradually dim until they are fully dark.

Aaron

[Paul Birkeland]

One more thing, all of the LEDs on the small boards light up initially, but over a period of about 2 minutes they gradually dim until they are fully dark.

Run the amp without the 6080 and let us know if this still happens.

Also, monitor the voltages on 1/5 and let us know what those terminals look like during this process.

If those voltages are off, that directly impacts the C4S board on the 6080.

-PB

[aarond]

With the 6080 removed none of the LEDs on the main big board light, and the ones on the small board light and remain on at steady brightness for at least the 10 mins I left them on.

This is compared to the initial illumination followed by decrease over about 2 mins to darkness, with the 6080 in place. I also noticed that the LED on the small tube socket, the one connecting A8 to the centre pin, also illuminates initially then gradually goes dark, just like the ones on the small boards, when the 6080 tube is in place.

The voltages at 1 and 5 are fine with the 6080 removed - at a steady 71 and 67 volts. With the 6080 in place these are poor - 1 = constant 6.5 V and 2 = around 65 volts initially and steadily dropping

[Paul Birkeland]

OK, this eliminates half of the circuit as the problem. 

There has to be a wiring error on the larger PC board, either with the feeds going to and from the board, or with the transistors populating the board.

For now, you can pop the 3K resistors back in where they were and remove the big PC board, leaving the smaller boards in place.

-PB

[aarond]

Thank you so much for your patience, Paul. Now I've put the 3K resistors back in and I've completely lost the left channel.

I suspect that this is the cause of my issues - some how in the speedball installation I've done something to lose that channel, which could explain why one side isn't lighting up. The only places it could be are the joints where the 3K and 22K resistors came out, so i'll try re flowing these joints.