thunderchief bias problems .. Q4 drain shifts with gain pot

[chemman]

I have been been debuggin the ROG circuit for a while now and am stuck at Q4. The Q4 drain drastically changes bias voltage depending on the position of the gain and volume pot. From 4.5 with no gain to 7.1 with full gain. All traces look good and no shorts. Wondering what it could be? All the other trannies stay at 4.5 though Q3 is still a little off at the source and gate.
The pedal sounds ok but def has a bias issue that is audible.

Made the EC with no probs and it sounds killer . Thunderchief is being a big pain in the ass.

THanks
matt

[puretube]

leaky caps?
(or a hidden tiny solder-bridge?)

[MartyMart]

Make absolutely sure that you have NO solder bridges or bad connections around the 22n from
Q3 and 100k from the 4k7's/100n junction

Make sure that 220k is NOT a 220 ohm !!

MM.

[chemman]

I was messing around resoldering connections and then Q3 and Q4 responded to the gain control (voltage up as gain up). Messing around some more it and almost givin up it fixed somehow. Have no idea but why complain. probably a bad solder connection (gotta clean my iron) but i triple checked all connections before my first post. First of 20 pedals to give so much trouble. Thanks for the replies. Sounds awesome without the farty transistor buzz i was getting before.

Mods:
added 1k in place of 4.7k as described by ROG to help Q3 get lower on the source and gate. (still high on source and gate but sounds very good)
Lifted one of the 2.2n caps at the output filters. (no more muddy bass in my strat or SG but still can dish out lots of bass with treble rolled down)

Awesome sounding circuit that puts my marshall combo gain to shame. 

-matt

[chemman]

OMG. next day this thing is doing the same shit again. Ha Ha. I also notices that if my finger is on the pedal the Q4 biases differently.

Q 4 moves from 4.5 to 7.6 V with gain pot turned but is also responsive to Volume.

I have checked solder connections lots and even recleaned between connections and think its something else.

I have a cable in the input to activate the battery, no output cable connected, and only Q4 seems to be reactive to the gain and volume knobs.
Why would the voltage change as volume is turned I have no idea because it is the last thing in the circuit.

Any ideas before I rip it apart and try again? Someone mentioned leaky caps but it seems there are stable voltages across them except for the 0.0022 filter to ground after 15k.
The 22n cap before Q4 has a voltage that changes from 4.5 to 6.8 as the pots are turned. the others coupling caps before the trannies are lower than the Vref provided by the 100k trims

Any other things to try to fix this beast ?

thanks,
matt

[puretube]

HF oscillations?
(scope!)

[snyder80]

HI!

I have the same problem in my JCM800 emu  which is a ripoff of the thunderchief (for me it´s Q5, http://www.diystompboxes.com/smfforum/index.php?topic=54990.0)!!!
Could u solve your problems right now? I figuered out, that biasing with cables connected to the box i/o helps to bias Q5 as its like "in use". And i set the gain pot to middle (or slightly lower) when biasing Q5, thats making the sound better even on full gain and Q5 D= more than 4,5 V.
But i also have some slightly fizz in sound if i dial middle and highs to 10 as well as gain and Mastervolume (NOT overallvolume).

So im realy interested in how u solved ur problems...

[Ben Lyman]

I just breadboarded the Thunderchief and had the same issue.
Volts go up and down on Q3 and Q4 drains when I turn the gain pot up and down.
Also, when measuring the drains I get different readings when I touch my hand to the breadboard or ground.

If I adjust the drains to 4.5v with the gain pot on full, it sounds fine but if I turn the gain down, the drains drop to about 3v and the "clean" sound is awful.

My solution so far:
Set drains to 4.5v with the gain pot almost all the way down, just to get that nice clean/light o.d. tone.
Now if I turn the gain all the way up, the drains rise to about 6v but the distortion sounds are all fine (to my ear)

Anyone think that could cause a problem? Q1 and Q2 always hang at 4.5v no matter what.

[TejfolvonDanone]

Can you post schematics of which you are working? It makes everyone's life easier: other know what you are talking about and you can get more help.

[Ben Lyman]

Oh ya, sorry. I had been reading every thread with the word "Thunderchief" and I thought this was one of them that had the link in the first post. I've also never been 100% sure about the protocol for linking vs. directly posting someone's image but I'm just gonna go ahead and throw the pic up right here anyway... I hope nobody minds, ROG seems like cool dudes anyway.

[duck_arse]

measure your resistances from gates to grounds. Q2 shouldn't de/rebias with pot change, hard to see why or where Q3 and 4 would get change of pot information from.

[Ben Lyman]

OK Duck, thanks, I'll do that later.
Q2 stays the same all the time, 4.5v with the trimmer set just right.
It's only Q3 and Q4 that go up and down with the gain pot.

[Kipper4]

Might be an issue with the 15nf? Is it shorted?

[Frank_NH]

Try putting a 1M resistor from the gate of Q2 to ground.  A properly biased gate (grounded in this case) is needed for the common source gain stage to work.

(edit.  Actually, if just Q3 and Q4 are affected, that may be a quirk of the circuit, given that you have a feedback connection affecting the bias of Q3 through the 2.2nF cap and 100K resistor.)

[TejfolvonDanone]

I love that there is no actual numbering on the schematic and everybody talks about the Q1 to Q4... I assume that everybody assumed that Q1 is the leftmost and Q4 is the rightmost JFET.

Try putting a 1M resistor from the gate of Q2 to ground.  A properly biased gate (grounded in this case) is needed for the common source gain stage to work.

(edit.  Actually, if just Q3 and Q4 are affected, that may be a quirk of the circuit, given that you have a feedback connection affecting the bias of Q3 through the 2.2nF cap and 100K resistor.)

1) The 1M resistor would just introduce another damping with the 470k resistor. Also the gate of Q2 should be grounded via the Gain pot.
2) the 2n2 cap and the 100k shouldn't affect because there is no DC current path to either of the stages.

Did you check the pinout?
Did you check the gate and source voltages?
Also you can try to measure all the voltages from power supply ground to other ground points near the component leg. Eg.: the 1k at the Q4 just around the leg of the resistor. If you messed up something with the ground wiring you can get crazy things.  (as Duck stated.)

[Frank_NH]

1) The 1M resistor would just introduce another damping with the 470k resistor. Also the gate of Q2 should be grounded via the Gain pot.

Good point.  You could use a 10M resistor instead.  A little attenuation wouldn't harm the output too much.

2) the 2n2 cap and the 100k shouldn't affect because there is no DC current path to either of the stages.

Also a good point.  Of course I would expect that the DC bias should not be impacted by the gain at all as long as the blocking caps are OK.

[Ben Lyman]

Never mind, I forgot to check the I.D. 10t and sure enough, there it was... user error 

I guess I had my guitar plugged in and the volume up when I was having the bias changes. Turn the guitar off or unplug it and the bias remains constant.
Here's another thing: when I touched the (bare metal) gain pot to turn it, the bias goes up BUT only if I'm bare foot.. so the fact that sometimes I'm shoeless and other times not... well you get the idea.

So, I think it's all sorted. Breadboard connections all look good, circuit sounds great.
I've added a tone control after the so called "cab sim" at the end of the circuit (2x 15k & 2x 2n2).
I tried putting it before but it didn't sound as good. I also tried taking the cab sim out and that didn't sound good to me either.
I measured the trimmers and replaced with fixed r's, still sounds good.
Only other change at the moment is a 2k to stop the gain pot from going all the way off and an 18nF instead of 15nF.

I love that there is no actual numbering on the schematic and everybody talks about the Q1 to Q4... I assume that everybody assumed that Q1 is the leftmost and Q4 is the rightmost JFET.

Yes! You're so right, I hadn't even noticed. Maybe I'll get bored and draw up my own version of this with part numbers and the tone control

[Ben Lyman]

Here's my version with SWTC2 tone control from AMZ. 
Trimmers replaced with fixed resistors for my personal J201 set, values shown in parentheses.

edit: took a couple parts out, simpler and seems to sound the same
edit edit: simpler tone control from plexi drive. still sounds fine