SOLVED Octanaut - please help!

[ftumch]

So I'm completely stumped by this

I've got an Octanaut build and everything is working except Q4

I've been through the circuit with an audio probe and there's no gain on Q4 at all. I can remove Q4 from the socket and the sound doesn't change. There is gain on Q1, that's fine. It sounds great at T1, but the volume drops after T1 and then Q4 does not bring it back up

I've reflowed the joints, double checked component placement and orientation etc

I've checked the resistance from C100 to R2 and R10 and both read 5K1

I measured~-8v on the drain for Q1 and basically 0 on the drain of Q4

I measure -.37A on the SGD for Q1 and -.37A on the drain of Q4 but 0 on the source and gate

The resistance to ground is much higher from Q4 than it is from Q1 which is probably the biggest clue?

Since the photos were taken I've changed the 10uF caps for ones that fit. The trimmer on the Q4 side was a bit stiff compared to the others so I replaced it, no change. Tried three different transistors now in Q4. I'm completely lost now

Any suggestions welcome

 

[Feral Feline]

You said you've "... reflowed the joints,", and did that include the Q4 socket itself?
Maybe replace the socket.
Aside: I noticed you've soldered the rest of the transistors into their sockets, but not Q4... curious as to why you're soldering them into sockets instead of soldering right into the board...

You've tried 3 different transistors there at Q4, but perhaps your whole remaining batch of transistors got zapped by ESD? Did you try a KNOWN-WORKING transistor in Q4?
Breadboard a MOSFET boost, make sure you've got a working BS170 and plug that one into Q4.

If you've got VCC going into R2, then so too should it be flowing into R10; looks like you've got the correct values for R8 R9 & R10 — still, it might be worth posting new photos since you've changed out a few things.


Also, you say "gain" when I think you mean "signal"?



One thing unrelated to your not having any juice getting to Q4's Drain:

Incorrect diode arrangement of the two sets:
SET1) D1 D2 D3 D4
SET2) D6 D7 D8 D9

Both the build doc ref-des layout and component-value layout pic of the board show both sets of diodes pointing:

north south south north.​

Looks like you've got them pointing in alternating directions:

north south north south.​

I don't think it matters which orientation they're pointing so long as both sets of diodes have matching inner and outer:

north south south north ... north south south north​

OR

south north north south ... south north north south​

Since you've got them socketed, once you've got the circuit working properly, it'd be interesting to see/hear if there's any difference between having one set differ from the other, ie:
south north north south
and the other set
north south south north

 

[ftumch]

@Feral Feline they're Russian diodes so the stripe is backwards. I've probed the audio with them both ways and I didn't hear any change. Will definitely have a play around with them once it's all going

I completely replaced the socket for Q4 while troubleshooting and left it unsoldered for now

I tested the Q4 BS170s in my transistor tester but I can breadboard a SHO to be extra sure

Also, you say "gain" when I think you mean "signal"?

Actually this might be a clue. I have the same signal at the gate and the drain of Q4. I assumed it was passing through the transistor without gain being applied, but would a signal pass through it at all if it wasn't functioning? Maybe something else is bridged somewhere

Here's the current state of it, I hope it is something easy to spot like a misplaced resistor

 

[Feral Feline]

Yeah, I hope it's something simple.

Get the strongest magnifier you can and look for solder-bridges, I'd even take a dental-pick or similar and drag it between pads — a solder-bridge can be a thin-blonde-straight wisp of solder, that's all it takes.



I was hoping someone more experienced than myself would jump in and help. As it is, I've got to go run some errands. I'll take a look at the new pic later tonight.





Before I go, though:
RE striped end of diodes...

Yes, I saw the diodes were Russians, that's not the issue.

As far as I can tell, you've got them alternating:



While both the build-doc and PCB clearly show they should be:



Doesn't matter if they're Russian, turned around they'd still be matching directions on the outsides and same again for the insides:

 

[ftumch]

RE striped end of diodes...

Yes, I saw the diodes were Russians, that's not the issue.

As far as I can tell, you've got them alternating:



While both the build-doc and PCB clearly show they should be:



Doesn't matter if they're Russian, turned around they'd still be matching directions on the outsides and same again for the insides:

Oops, you know what I might have f--ked that up lol

 

[ftumch]

@Feral Feline well, as the saying goes. You can lead a horse to water, but you sometimes have to lead it there a second time with pictures

So that's working now, and I am a wally, thank you

On a positive note, I've learnt a lot about troubleshooting today, I've re-learnt what I already knew about working while tired, and most importantly it sounds glorious

 

[ftumch]

This mostly boils down to my being so neurotic about germanium diodes, the prices people want for them these days make me cry

I bought a bag of 100 D9J's for $20 from an eBay seller few years ago and after the sale they sent me a link to their webstore: https://nextfive.pro/products/d9

I found the link in my email earlier today. Looks like the prices there have not changed so much. I might give it a punt

 

[ftumch]

Well I spoke too soon actually, after all that the problem with Q4 getting no voltage persisted

I measured the resistance from C100 to R10 and it was all over the place, going from a few hundred k to a few meg and back again, so I jumpered the connection with a wire and then it worked perfectly



My only guess is that something got in the hole and contaminated it? I think I'll desolder that section and give it a good clean and start again

I'm absolutely sure that the solder connection is sound, I can see it through to the other side so it's gone down there. I'm baffled by this I've never had it before

Here's the little fella:





I think I'll leave this marked as resolved and I'll desolder that area and give it a good clean and start again. I'll measure the resistance from end to end of that trace too

BUT IF ANYONE CAN TELL ME WTF IS HAPPENING HERE I'D BE VERY HAPPY TO KNOW KTHX

 

[Feral Feline]

I did say the diodes-issue was an unrelated to the Q4 conundrum...

One problem at a time...


Solder can flow through a PCB pad and still not make a great connection with the associated component leg, just heat up the pad and not the leg and you've still got a cold solder joint even though there's a mess of solder spoojed all over the PCB.


Maybe make a list of possible causes, then start knocking down the list.


Here's a Chuck D Bone's post worth repeating here, checking things off you've done:
1. Start a new thread for troubleshooting your pedal.
2. Post detailed pictures of both sides of the board. [some photos at odd angles make it difficult to see components/solder joints]
3. Recheck all of your component values and orientations against the build docs. We're here to help but you need to do your due diligence.
4. Explain all of your mods or parts subs.
5. Be prepared to make and report voltage and continuity measurements.
6. Be methodical. Do not start swapping parts at random unless you are 100% sure the part is the wrong one. Do not assume that an opamp or delay chip is bad.



So you jumpered power to Q4 and it worked — so definitely there's a blockage somewhere ie cold-joint or something.
Now that you've confirmed Q4 CAN work, maybe try working in reverse AWAY from Q4 off each of its legs one by one, confirming electrons are flowing as they should.

Key thought of mine being that final step, #6: BE METHODICAL (I'm quite crap at doing so myself).




PS: If you leave it marked "SOLVED", nobody will come to help...

 

[ftumch]

Thanks for your help @Feral Feline. I sucked some solder out and flowed some fresh stuff in but no change. I'll mark this solved until I've had a chance to remove the components and clean the area up since I know now exactly where the issue lies

 

[ftumch]

I did measure 5.1k from C100 and across R10 at one point. Mad

 

[giovanni]

Thanks for your help @Feral Feline. I sucked some solder out and flowed some fresh stuff in but no change. I'll mark this solved until I've had a chance to remove the components and clean the area up since I know now exactly where the issue lies

Worst case scenario the trace might have gotten damaged by too much heat, in which case the jumper is the only possible fix. Good luck!

 

[ftumch]

Worst case scenario the trace might have gotten damaged by too much heat, in which case the jumper is the only possible fix. Good luck!

Yeah might have f---ed that faffing around with it when my diodes were wrong. Half tempted hot glue the jumper down and call it done lol