uStomp Maestro FSH - Who's built it? ...need help.

[mydementia]

Howdy y'all.
I got the soldering iron out again last night and tackled the FSH from uStomp (made the PCB a few weeks ago...):
http://www.ustomp.com/index.html?menu=2&section=6&item=48

I don't get ticking in bypass...but I don't get much of an effect when it's engaged either...

I'm curious about a few things:
1)  Stupid me - I committed to a project of this complexity without verifying the schematic - anyone know what it's based on?  Steve?
2)  There's a 0.33uF cap coming off Q4 (along with a 0.047uF cap and a 100k resistor) - this 0.33uF cap doesn't exist in the GGG schematic - I faked the value...but want to be sure I know where it came from.
3) The uStomp layout uses a 1M speed pot where the GGG schematic calls for a 2M... um...
4) What are we trimming with the 10k and 100k trimmers?

I assume I have a solder booger crossing a trace somewhere... so I need to go through a scraping exercise. 

I assumed that the Q's were consistent with the GGG schematic. 
I subbed:
Q1) 2N5088 for the MPS5172
Q2, Q5, Q6) BF245A for the 2N4303's
Q3) 2N5087 for the 2N3906

I used TL072's for the dual op-amps too (instead of the uStomp LM833's)...

Any help will be greatly appreciated.
Thanks in advance.
Mike

[Dan N]

1- Maestro Filter Sample Hold.
2- The collector of Q4 is to -9 so it's connected to everything else at -9. That cap and tons of other stuff.
3- The original use the harder to find 2 meg. There is a parts exchange work-around to use 1 meg. Check the schematic with  the Tonda mods. I don't know whether Steve premodded it or what. Little homework for you!
4- 10K is noise volume. Single turn should be O.K. 100K? Biasing (something). Multiturn if you have one.

I can't remember about the transistors. If Steve used the original values and pinouts, I believe the MPS5172 has a ECB pinout. You better take a hard look at the schematic vs the layout.

Edit- That 0.33 cap is part of the Tonda pot mod. Way to go Steve!

[mydementia]

1) I meant what SCHEMATIC is it based on ... after looking at the GGG schematic with Tonda mods (I thought that link went to a sound clip...not a modded schematic - my bad...), I'm guessing that Steve used some and left others off... glad that's solved!
2) Tonda pot mod - cool
3) Guess I don't have to use my 2MB pot...
4) Okay

The trannies were totally assumed...based on the GGG schem.  Right now I'm sitting with Steve's layout and the GGG schematic and trying to make sense of the trannies and fets.  According to circuit specialist (first site I found with MPS5172 datasheet) both the MPS5172 and 2N5088 have C-B-E pinouts
http://www.circuitspecialists.com/prod.itml/icOid/1243
http://www.circuitspecialists.com/prod.itml/icOid/4569
...but the 2N5088 is only 30mA while the MPS5172 is 100mA...hmmm...

I'm still not sure what the original 2N4303 pinout was... nothing useful on 'google'... I'll keep digging...

Mike

[Dan N]

Hah! I thought #1 was too easy!

About the transistor, I was thinking of a 2N5172. Sorry.

[jrc4558]

There's a transistor there, who's base is disconnected. It is a noise generator. Buy 10-20 pieces of 2n2222, and test them in this position. there will be a variation of the depth of effect.

[ustompsteve]

I have posted the version of the schematic I used.

I used a version of JD Sleep's (GGG) schematic with the Tonda Mods that I wanted on my pedal. I modified it to use specific trannys and dual op-amps.

I just posted the modified schematic I used:

http://www.ustomp.com/edit.html?menu=2&section=6&item=151

Hopefully JD is ok w/ that since it is based off of his drawing.

--steve

[mydementia]

Thanks Steve!!!  That's exactly what I needed!!
I'll have to work through it to be sure I'm consistent...

I went through the de-boogering exercise last night - dug all the flux/goo out from between traces.  Still get a ticking sound on one setting and strange static (kinda like in Silent Hill when you're getting close to the monsters and the transistor radio starts making static?) that changes in pitch when I adjust the 100k trimmer...  Also, the output is VERY low... I'm going to stop by my local surplus store tonight to look for an MPS5172 and 2N3906.  I'll also steal the pair of LM833's I have in one of my Condor builds for the dual opamps...then all the parts will be consistent.

CN: Is there something I can measure on Q4 (2n3904) to figure out which one to use?  ...or just swap until I get the desired effect? 

[ustompsteve]

any dual op amp should work fine. i would just audition a few 2n3904's. you will know when you find the right one.

some of you other symptoms may suggest that there is still a problem w/ the build. does the envelope filter side work ok? That is not dependent on the noise transistor. if it doesn't work then you need to do some debugging first.

--steve

[mydementia]

I picked up the correct trannies and plugged them in - still no go...
Then I heated up and reflowed every solder joint and made sure no solder bridges were honking me up... still no go...
Measuring voltages, Pin 7 on the dual opamps are dramatically different (one is around -7V, the other is around 0.2V) - my problem must be around this area... maybe I didn't get a trace on my PCB?  We're getting laminate flooring over the next two days...so my soldering station/stompbox building desk is out of service . 
I'll get back to this next weekend...
Mike

[ustompsteve]

i'd also check for backwards caps. --steve

[mydementia]

Just placed a snapshot of my circuit in my gallery:
http://aronnelson.com/gallery/album104/MaestroFSH_002
(it's 3MB - so for those of you with slow connections...).
Anyway - Looks to me like the electros are all oriented properly.  I just noticed that one of the 3080's says C3080E and the other says C3080AE - think this matters?
edit: Just looked at the data sheet (http://www.futurlec.com/Linear/CA3080E.shtml) hmm....
edit again: Just swapped out the AE for another E - same result... not the culprit...

If anyone sees anything funky in this picture please write back - you know how it is to spend too much time with a circuit...

Thanks.
Mike

[moosapotamus]

I always got a little bit of bleed-through/ticking in S&H mode on my FSH, too.

For the low output level, the first few things I'd check would be resistor values (for example, ohms vs kohms), transistor pinouts and maybe even try different (but same type) trannies.

I can't say how many times I've gone through the debugging routine, checking power and ground routing/wiring and everything else, only to find that I was plugging into the output or something...

BTW, I'm with Constantin... 2n2222 for the noise trannie.

Good luck!
~ Charlie

[Ry]

I've built two FSHs and tried all kinds of transistors for the noise source.  In the end, I used 2n2222 for both.

[jrc4558]

CN: Is there something I can measure on Q4 (2n3904) to figure out which one to use?  ...or just swap until I get the desired effect? 

As Mark Hammer once explained to me, you need a really noisy transistor here. So the noisier, the better. Thats all.