Troubleshooting a Harbinger One build... before I do something drastic

[taeagan]

So I have a Harbinger One build on my hands (I got one of the thin boards on sale).  I wired it up for 18v power supply and I had to sub in 2n5089s for a couple of the 2n5088s (I ran out, but I'm going to order some more soon... all transistors are socketed). 

I'm getting the audio path through the effect, but the lamp is not pulsing.  I can adjust its brightness with the trimmers just fine, but the brightness never varies.  The speed control does nothing and the on board LED never lights up.  I did some research on the forum and by all accounts this seems to mean that I'm not getting voltage to the 15v side of the equation. 

So I took a look at the schematic and started troubleshooting... and here's where it gets a bit weird... or maybe not, someone who knows more than I do can tell me. 

I started taking voltage readings and found that I was getting 15v coming out of the LM7815 where it's supposed to and next on the one leg of the lamp.  Good.  But then as I followed the trace, I found that I wasn't getting 15v at Q11, Q12, Q15, R38 and R55.  I checked solder joints, etc...  Then I disconnected it from power and decided to test resistance.  Resistance between the LM7815 and the lamp is 0 as expected.  However, it turns out that I'm measuring about 36kohms between the lamp and Q11.  I've isolated it to this part of the trace.  Between Q11 and Q12, Q15, R38, and R55 I get no resistance.  It's just between the lamp and Q11.  The meter quickly gives me a low reading and then slowly starts climbing over the course of a couple of seconds up to about 36kohms. 

I can't figure a good reason for this.  According to the layout it's a straight run from the lamp to Q11 so I don't see why I should get resistance.  Unless there's a flaw in the board?  My thought was to run a jumper along this path on the solder side of the board, but I thought I'd check here first.  It feels like a hassle and something that I'd rather not do unless absolutely necessary.  Let me know what you guys think and thanks in advance. 

[midwayfair]

Post your voltages on all transistors and photos of both sides of the board.

[taeagan]

Thanks.  Some notes about my build:

1) The power supply that I'm using actually kicks out 19vdc
2) I rely on a stock of 1/2W resistors for a lot of my pedals, so in the pictures you'll see that I have a number of them that are "standing up" on one side.  I'm very careful about making sure that the exposed leads don't touch anywhere and I've looked that over a number of times on this build.  I have six others that I did this weekend using the same process and they all checked out.  But it's obviously a bit crowded in spots. 
3) Q11, Q12, and Q13 are subbed out with 2n5089 instead of 88.  From the voltage readings below, I don't think that this is the source of any problems.  At least not right now anyway. 
4) All of the transistors and the indicator LED are socketed - yes, I know that I put the indicator LED on the wrong side of the board.  I'm a dummy.  They're all standing tall right now because I don't want to trim anything to length until it checks out. 
5) There's no 1054 chip (even though I wasted a socket on it) or any of the other parts for the charge pump as per the build instructions. 

Voltages - I think that they're all within reason, except of course for Q11, Q12, and Q13 not getting any voltage (and hence the lamp not pulsing):

LM7815: 18.8, 0, 14.8
Q1: 2.2, 1.5, 1.1
Q2: 4.7, 2.2, 1.6
Q3: 14.5, 4.7, 4.1
Q4: 18.8, 5.9, 5.6
Q5: 13.8, 5.6, 5.0
Q6: 18.8, 5.9, 5.6
Q7: 13.8, 5.6, 5.0
Q8: 18.8, 5.9, 5.6
Q9: 13.8, 5.6, 5.0
Q10: 18.8, 7.5, 7.0
Q11: basically 0s
Q12: basically 0s
Q13: basically 0s
Q14: 18.8, 0.5, 0

Pictures to follow

[taeagan]

Crossing fingers that this works.  Let me know if you can't see the pictures:

[taeagan]

How about just this instead:

https://drive.google.com/file/d/0B-VSm6ER7FEzVTlMcDFCRlVtcEE/view?usp=sharing

https://drive.google.com/file/d/0B-VSm6ER7FEzT0piZG9ydFZNTFk/view?usp=sharing

https://drive.google.com/file/d/0B-VSm6ER7FEzdDg5N2ZVMzB2VEk/view?usp=sharing

[madbean]

A bit strange sounding. You are getting 15v from the regulator and on one pin of the lamp. But, the pin on the lamp connected to the 15v supply is also connected to the collectors of Q11 and Q12 where you seem to be getting no voltage.

Your transistors are socketed, and that's good. What voltage do you read on the collector socket if you remove them? What I'm wondering is if those transistors are misbehaving they might be sinking current. I have not used 5089 in a Univibe LFO, IIRC, but I don't know if they would have a negative effect or not.

Lastly, the resistance you are measuring is a non-starter. It's not useful because there are still resistors loading the 15v supply which means the measurement is not indicative of anything amiss. If anything, it shows continuity on the 15v rail, which is a good thing.

[taeagan]

I also have the lamp socketed.  I removed the lamp and Q11, Q12, Q13, and Q15 from the sockets and now I get 0v on all of them.  I still get 15v on the LM7815 and the positive side of the C34 cap.  Stumped.

[taeagan]

Since I had the lamp un-socketed, I took a jumper wire and hooked it into the socket and then connected it directly to the 15v leg on the 7815.  So I measured 15v on the socket of the lamp, but still nothing on the transistors.

This isn't making sense to me.  I should be getting the same voltage on everything connected to the 15v bus here shouldn't I?  Unless there's some continuity issue somewhere? 

[mjg]

Possibly a continuity issue between the right hand pin of the lamp, and the components on the left side of the board (if you are looking at the board with IC2 in the top right corner - like in the build doc picture of the board). 

With the lamp and Q11 etc removed from their sockets, are you still getting the 36k resistance between the right hand pin of the lamp, and the bottom pin of Q11 or Q12?  If that trace is broken, I would expect you'd get infinite resistance with the lamp removed, but maybe some resistance with the lamp in, as there is a path through the other side of the lamp, through all the SPD and OFFSET circuit, and back to the pin of Q11.  It would probably change from 36k to other values if you twiddled the SPD1, SPD2, OFFSET and INT if that was the case. 

I'd double check the soldering on the socket of the lamp - and try a jumper from the right hand pin of the lamp across to Q11, just as an experiment to see if that fixes the missing 15V over that side. 

[midwayfair]

Put your meter on continuity. First check for continuity from the regulator's far right pin to the left pin of the lamp. Next check the collectors of Q11 and Q12. R38, R55, and Q15 are also directly connected to the regulator.

If you don't have continuity at any point, just reflow your solder joints (actually, you should have done that already -- if you haven't, do so before you do any further debugging), test again, and if you still don't have continuity to one or more parts you can either run a wire to fix a presumably broken trace, or you can start scraping away solder mask to figure out where the trace breaks.

If you HAVE continuity to the power rail but no voltage, start checking for shorts to other components, but 0V on Q12 would require a dead short to ground, and you'd get 0V out of the regulator in that case, so a broken trace is much more likely.

[taeagan]

So it looks like it was something with a broken trace. I had no continuity past the regulator and the cap, so I ran jumpers to the lamp and the next transistor. What's weird is that I had continuity to the lamp at one point and then it vanished. Either way I got it up and running, lamp pulsing (and with a full compliment of 2n5088s - I found a few) and its doing the vibe thing.

I've got a few other problems though. Theres a very low level whine or squeal in the background, but it's the kind I'm going to hope disappears after I trim the leads on the transistors and box it up.

Also, I get no sound when the DPDT is in the other position. But when the board flexes a bit as I switch it the sound cuts in. So I have a feeling it's another trace but I could easily be wrong. Interesting that the only other board I had these kinds of problems with was another thin board. I've done at least 10 other madbean boards that went together without a hitch (normal board thickness).

Thanks all for the help.