Q1 transistor sending all the power to 'R' on Thunderbird build

[hot_rod]

Hi - I'm doing a 1776 Thunderbird build. Got power working at least on the DC connector and my LED works! However, my latest issue at present is that power is not making it all the way through the circuit, in fact it gets stopped fairly early by the 2N4403 transistor in 'Q1' slot.

The build notes say "Use a stereo jack for the input and wire the "R" pad to the ring terminal. The purpose of the 2N4403 transistor is to avoid the return current from the charge pump IC through the same ground wire that carries the return of the input signal. The charge pump ICs draw current peaks at the switching frequency which will otherwise make its way through the signal path."

However, I am not getting any power through to my charge pump IC.
I tried 2 different 2N4403 with same result, even tried swapping my socket too. In all cases I measure ~9V going in the 'E' pin, but 9V going to 'B' and nothing reading at 'C' which looks on the specs like it then goes to the LT1054. However, I read a full 9V on the 'R' terminal that goes to the ring terminal of my stereo jack. I checked my transistor to make sure I was using the right EBC order too.

I posted pictures below. Thanks!

[TGP39]

Try to reflow the solder on D10.  A cold solder joint will shunt pin 1 to ground. 
-Steve.

[hot_rod]

Hi - thanks that fixed Pin 1. However, still no sound. The rest of the pins on LT1054 are roughly in line with specs except pin 2 has no voltage (!?!) Based on the results from fixing pin 1, I reflowed every solder joint on the board - twice - but no luck. I tried changing out the LT1054 as well but same issue on the other one. I am getting 8.42V on Pin 1+8, but just microvolts on Pin 2 which is far less than the specs where it looks like it should be about half pin 1.
All my electro capacitors and diodes look to be installed in proper polarity And the parts involved were ordered using the Mouser BOM link in the Thunderbird build doc so I think are correct? I am baffled but still wanting to hear this thing work!

Pin 1

Pin 2

Overview top

Overview reverse

side angle

[TGP39]

Ok.  Now check that C35, C36, C37 and C38 are properly oriented.   
Remember Tantalum capacitors are polarized (i.e. they have a positive and negative pole).
These caps can be tricky because usually the Positive pole is marked. (With aluminum electrolytics the negative terminal is usually marked).  I can't see the orientation of your Tantalum (yellow) capacitors from the photos. 
If you're unsure, snap some photos of those caps and I'll tell you if they are properly oriented.   If they are in the wrong way, I would get new caps because the dielectric in these caps break down quick when electric flow is reversed. 
-Steve. 

[hot_rod]

Hi Steve - thanks for the help so far. Here is what I am using for C35-C38. The positive pole is long leg so on right side if looking at text side, and left side if looking at blank side. I think I inserted them all correctly with long leg going into the +/square pad.  I almost wish I had them backwards since that would have been easier to troubleshoot. 

[TGP39]

Hmmmm. The Tantalum caps look correct.   You reflowed the solder for cap35 and cap37?
Something is pulling most of the voltage away from pin2 on the charge pump. 
-Steve. 

[hot_rod]

Yes -  I reflowed the solder on everything, and then after you highlighted them I even did the tantalum caps again. Maybe one is bad? I would have to order another batch from mouser as I only have 2 spares...

[midwayfair]

If you pull the charge pump, does the circuit work?

What's the jack you used? There are two types of switched jacks.

Are we sure pin 2 should have voltage in this setup? It's hooked up to the -ve side of a capacitor even though it's the cap+ pin. (The tripler is not the usual setup so I'm unfamiliar with the details.)

[hot_rod]

unfortunately I took an unexplained step backwards --- today went out to work on it and I was back to the first issue as before, all power flowing to R and nothing on Pin 1 of the charge pump. I changed *literally nothing* from last night when I took all the photos to troubleshoot the C35-38 power for pin 2, and at that time Pin 1 was humming along fine.

Since I was building 2 of these at once, I swapped the second assembled board in with same jacks etc. Same issue. (!!!)  And I checked the solder on D10 again.

So, at this point I am wondering if I have some dud 2N4403 units, used in the Q1 slot. I thought maybe it was my socketing of Q1 that made a difference, but I desoldered and then resoldered the 2N4403 directly and still it was sending all the power out through the middle leg to R. By the end of it, one of them only tests as a diode in my tester so I think it is fried, but the other one tests fine so it should work but doesn't.

I have no spare 2N4403 so I am going to have to order more to evaluate this further.  For now, I am off to go work on a build that I have made before. I don't have enough hair to afford pulling out more over this! 

ps - @midwayfair I am using a stereo in, mono out (if that is what you are asking for jacks), with the ring wired to R per instructions. And Pin 2 on the LT1054 is supposed to have voltage about half of Pin 1, per spec sheet pg 3 if I am reading it right (https://cdn.shopify.com/s/files/1/0022/3952/9069/files/Thunderbird-1-18-15.pdf)

[TGP39]

John,   Would he need to have a cable plugged into the stereo jack while testing so the ring of the jack isn't isolated?   I also questioned pin 2 of the LT1054 being connected to the - side of the Tantalum cap. 

Hot_rod,   I'm pretty sure the ring and 2n4403 part of the circuit are necessary if you plan on using a battery.   The build docs should say something about what to omit if you are not planning on battery power. 

-Steve.

[hot_rod]

Good point - I think my initial testing was done with power but nothing plugged into the input or output jacks. I didn't think if that would affect the 'R' going to stereo as I haven't had a build use something going back to the jack before!

For my testing yesterday, I had things plugged into the I/O jacks, maybe that is what fixed the Pin 1 issue. Today I didn't have the jacks plugged in and Pin 1 had its issue. Anyhow, I will try some more tomorrow and see if I can get some voltage to Pin 2. FWIW I am not using a battery, just trying to get it to work with my normal 9V input.

[midwayfair]

Yeah, you need to be plugged in. I didn't say anything about it before because you were actually plugged in on the test set.

This circuit will run without the LT1054 in. If you want to know where the problem is you can isolate any problems with the charge pump from the rest of the circuit.