LOUD cupcake and some distortion issues -- any suggestions?

[toneboner]

Kind of funny.  All my madbean builds so far - Sunking, Pork Barrel, DeadRinger, Silverfox, Neutrino, Pointdexter (except for reversing all the transistors, doh!)  - have gone off without a hitch and worked perfect from 1st plug in.  Cupcake is easily the simplest build and I'm having some problems hopefully the hive mind here will help me solve! 

1st: I can bias the circuit ok with the trimmer, get some good compression or back off for a bit less - but it seems at any setting i get a clipping like distortion mostly on lower notes.  sounds almost like my cupcake is trying to overdrive!  This also occurs at any level setting as well, which brings me to...

2nd: The level control (yes, a 10k B) starts getting really LOUD to the point where its boosting way past original bypassed levels when turned up, even half way up. 

maybe the two are related?
thanks for helping,
sasha

[joelvanginkel]

Did you get this resolved? I'm having exactly the same issue with my Pork Barrel. What's weird is that it worked perfectly and then I boxed it (redoing virtually all the hookup wire) and then this loud/distorted problem arose.

Joel

[Jurassic]

I'm having the same problem with my cupcake build.. there is some really nasty buzzing/distortion especially with lower notes/chords. My pedal is also really really loud, I can turn the volume pot up only around 2cm and after that the effect is way beyond unity gain. Usable range of the volume pot is really small.

Did you guys got this distortion-problem solved? I changed the transistors and the op-amp, but that didn't make no difference.

I also have some problems with the bias-trimmer. I turned the trimmer clockwise and suddenly there was a small flash under the trimmer and it started to smell burned..?! I dialed the trimmer quickly back down and everything seems to be working ok. Just wondering why this happened? I didn't short circuit the trimmer with my screwdriver or anything and I had the distortion issue already before this happened.

If anybody has any tips for solving this problem, please help me out!

-Juri

[jkokura]

I first suggest that you read the sticky thread "rules for getting tech help" and use the tips there to do some debugging on your own first, then make a new tech help thread for any help you with anything that you can't seem to do.

Jacob

[Jurassic]

Ok. Thanks! I will check that topic out and start debugging today.

[oldhousescott]

Replace the 10kB pot with a 10kA (log/audio) to get a more useful taper for the level control.

For very strong signals, the feedforward portion of the circuit (C3 and R3) does effectively clip half the input signal, at least until C7 charges up enough to be the dominant controller of gain reduction (after the time constant set by R10 and C7, or about 45mS). So you might get a tick or click on the leading edge of strong signals. You can try to reduce this by increasing R3 to 1M or 2M2 (but you'll get more volume peaks coming through on the leading edge of percussive signals).

R10 (in conjunction with C7) sets the attack time, and R11 (also in conjunction with C7) sets the release time. The stock values yield time constants of 45mS and 3S respectively. You might socket those two resistors and try different values to get the attack and release characteristics you prefer.

Q2 is set up as a current source. If you got a flash under the trim pot, it suggests you've got the wrong pinout for Q2. Both Q1 and Q2 are D-S-G on the layout. Make sure the FETs you used have that pinout. If you have D-G-S for Q2, then turning up the trim pot all the way would force all the current through a small portion of the trim pot resistance track, exceeding its power limit.

As a side note, it's interesting they didn't use that extra opamp stage to create a full-wave rectifier instead of just a half-wave rectifier with D1. The sidechain control would have been faster and smoother, and it only would have cost a couple extra diodes and resistors.

[Jurassic]

Wow.. oldhousescott, thanks a lot for the information! That really helped and now I actually understand the circuit better! You were right, the problem was that the Q2 transitor was reversed and that damaged the trimpot. After replacing the trimpot and putting the transistor in correctly, the pedal seems to work! 
Sometimes it's really hard to spot a simple mistake.. I checked everything over and over again, and still couldn't see the reversed transistor. 

Anyway, problem solved. Thanks!

[toneboner]

Haven't checked in here for a while... VERY THANKFUL to see some replies (and to see I'm not the only screwup!)  Will do some more debugging as per the tech help sticky tomorrow...

[9Lives]

it's crazy how you can build the most complex stuff and the smalls can work you to death. I managed to nail the cupcake but when I first started I couldn't get the fkn thunderpuss to work.. I mean I cussed and screamed and everything between. Prob ended up being my mosfet. I used the 2n7000.. Can anyone guess? I didn't reverse it.. That stumped me for days.

[toneboner]

Well, finally, metered voltages on the IC and it tested within spec (as per this thread: http://www.madbeanpedals.com/forum/index.php?topic=7467.0)   SO, I upped r3 to 1M2 as per scotts suggestions above - haven't fooled with attack or release changes yet. And so far so GOOD. No more clipping!!!  Only sweet squish now...

[Flaconsius]

I´d got the same problems, except for the Burned out Trimpot and I reduced the Clipping replacing R1 (4M7) for 1M. That makes a difference in the Input Impedance.
I´ll try replacing R3 for 1M too, and see what hapen.
Regards and thanks for the info.