Author Topic: Thunderchief debug from hell  (Read 546 times)

HeavyFog

Thunderchief debug from hell
« on: August 29, 2019, 01:04:40 PM »
First post in a long while here but i'm desperate for a solution at this point because this debug is testing my sanity.

I'm working on a modified ROG thunderchief with a tonestack for a friend of mine and i've got a few issues with it:

1- When bypassed it produces a noticeable high pitch whine that changes pitch depending on how the knobs are set. This is making me think it might be a grounding or power supply issue.

2- It passes audio fine and sounds great until i turn the treble up to full, then the bias of the last two JFETS gets thrown off a bit, plus there ends up being a negative voltage of around -0.70v to -0.35v on the gate of the last JFET. When this happens the circuit starts to gate a bit, and there's a bit of a jump in gain.

For the life of me i can't seem to figure out whats up with this thing. I've poured over the stripboard layout i designed and the board itself and i can't seem to find any mistakes. I used the same layout to build a very similar less modified thunderchief a while back and i had absolutely no problems with it, so i'm suspecting it's either a power supply thing or something to do with the modifications i made.

Full list of mods made (compared to the tonestack version on rog's website):

5 siliconix J201s used
Q1 source resistor changed to 820 ohms
4.7nf bright cap added to gain pot
mix resistor changed to 270k
Q2 source resistor changed to 820 ohms
Q2 source bypass cap removed
Tonestack changed 500pf and 33k to 250pf and 56k
4k7 resistor on source of Q4 changed to 5kb pot (100nf resistor on lug 2) to make a presence control
22nf on drain of Q4 changed to 0.1uf
5k1 grid stopped added to gate of Q5
27k negative feedback taked directly from drain of Q5

Probably would have been a good idea to breadboard it before i boxed it up, but i figured since i already did one before that was a bit closer to the ROG schem i probably wouldn't need to (there's my first mistake). Iv'e audio probed the whole thing and the problem seen to be isolated mostly to Q5 (specifically right after the o.1 coupling cap), as the shifting bias doesn't seem to be affecting Q4 audio wise. I looked over all the external wiring with a meter and it all seems to check out fine, so i'm completely stumped as to why it's doing this. IL try and post a complete set of voltages when i get home if it helps as well.

If anyone could offer any advice it would be a massive help!



HeavyFog

Re: Thunderchief debug from hell
« Reply #1 on: September 02, 2019, 07:35:50 PM »
Posting some voltages for reference
(all knobs at full)

Q1
d 4.51v
s 0.26v
g 0v

Q2
d 4.38v
s 0.26v
g 0v
 
Q3
d 9.38v
s 4.5v
g 4.38v

Q4
d 4.63v
s 2.37v
g 2.51v

(for whatever reason the source voltage measured 4.02v when i measured from the top of the vero board but 2.37v when measured from the bottom)

Q5
d 4.38v
s 0.28v
g 0.28v

Judging from things its looking like Q4 might be the issue giver the abnormally high voltage on the source and gate but who knows at this point. It also looks like the negative voltage problem i was having before is gone but its still gating a bit with treble on full. The weird thing is that sometimes i plug it in and the treble knob makes it gate, and other times it doesn't. Regardless it still makes a loud whine when bypassed. I found that if i disconnect the cable on the input with the pedal on it still makes the whining noise but louder.


HeavyFog

Re: Thunderchief debug from hell
« Reply #2 on: September 03, 2019, 01:32:33 PM »
Just remembered I drew up a schematic a while back. Probably should have posted this earlier. It's a bit different from the one I actually built, with the changes I listed above being the main difference


PRR

Re: Thunderchief debug from hell
« Reply #3 on: September 03, 2019, 03:52:42 PM »
You aren't getting anywhere.

I can't read your schematic, even cleaned/rotated.


"Whine" -- this is a HIGH gain circuit. Proximity of output to input WILL induce oscillation. You apparently have a second (perhaps supersonic) oscillation when you turn the treble up. Leaving the input open makes more whine is another hint. LAYOUT!!

I do not understand the 4.7k in Q4 source.

Taking the 47k NFB from Q5 drain introduces some DC shift which I don't want to study today.

> source voltage measured 4.02v when i measured from the top of the vero board but 2.37v when measured from the bottom

What is "top" or "bottom"? What are they? What is the difference? We can't see your build from here.


HeavyFog

Re: Thunderchief debug from hell
« Reply #4 on: September 03, 2019, 04:40:43 PM »
I've managed to solve the whine issue by grounding the input with the stomp switch, so now when I bypass it it doesnt make a noise.

The 4k7 is what was used on the original thunderchief, I saw no reason to change it. I've also verified the layout I designed, after having a successful build with it as I mentioned in the first post.

For the neg feedback i probably will add a dc blocking cap to it, but again, in the last build i did this was not an issue at all in the final build. Il add the cap anyways to see if it helps.

The source voltage on Q4 was different depending on if I was measuring from the underside of the board on the solder joint, or the leg of the component from the side where the components are mounted. Why it was different I still dont know.

I re biased it again today and so far its acting as It should have in the beginning with no issues, though I dont know if it's going to start giving me trouble again like it has before

DIY Bass

Re: Thunderchief debug from hell
« Reply #5 on: September 04, 2019, 07:07:18 AM »

The source voltage on Q4 was different depending on if I was measuring from the underside of the board on the solder joint, or the leg of the component from the side where the components are mounted. Why it was different I still dont know.



The only times I have ever seen a difference in voltage between the solder and the leg of a component it was a bad solder joint or a lifted track on the PCB.

HeavyFog

Re: Thunderchief debug from hell
« Reply #6 on: September 04, 2019, 08:30:43 AM »
Update

Added a 220k resistor from the input to ground and it did the trick in getting rid of the oscillation. Also saves room on the stomp switch. It was going good but I noticed the problem with the treble eq started happening again but only when I put a treble booster in front the pedal. It seems that when the gain is high the problem returns.

I re biased it and now it's just farting out on me and the negative voltage on Q5 gate is back again even without the boost in front. This is making me think I might have some sort of loose connection or bad joint. Not sure if the voltages could point to anything. I did go over the whole board and fixed up a handful of bad solder joints but still nothing. Gotta check over the off board wiring, especially the tonestack