Can somebody please help me debug the PHUNK?!

[doc_drop]

So, this is the first time I have gone through the "what to do when it doesn't work" routine, but I had a frustrating weekend trying to build an envelope follower that worked, so I could start getting all Phunked up!

I decided to build 2 of them, just because I like variety, and I figured that way I should wind up with at least one cool circuit. I used layouts from the layouts gallery, and etched the PCB's myself.

I built the Q-Tripper based on these documents:



And I built the Phuncgnosis based on these documents:



Neither works. So, here is the rest of the debugging info. for anyone kind enough to take a look:

Q-Tripper:
It just passes a very faint non-effected sound.
I used 2SK30A silicone because I don't have any 2SK30A-GA. I also tried J201's and MPF102's with the same results. For kicks I tried them all in both orientations in case there was a pinout problem.

Voltages:

Filtered/regulated wall wart: 9.28v

IC1
Pin 1 4.35v
Pin 2 4.43v
Pin 3 4.43v
Pin 4 0v
Pin 5 4.42v
Pin 6 4.43v
Pin 7 4.43v
Pin 8 9.29v

Q1
D 9.29v
G 4.02v
S 6.18v

Q2
D 4.4v
G 2.49v
S 4.43v

D1
A 0v
K 9.3v

D2
A 0v
K .88v

D3
A .87v
K .83v

Phuncgnosis

In the down position it passes a slightly compressed unity level guitar signal - no effect. The loudness is changes with the intensity knob. In the Up position it passes a faint guitar signal with an awful pulsing monotone noise if I turn up the intensity knob and/or strum hard.

I had to double up a couple of caps and resistors to get the values called for. R6 measures 49.2K Ohm not 50K. R9 measures 326K Ohm not 330K. R8 mesures almost exactly 5M Ohm. C7 + C8 are (2) 330pF which measure right arounf 680pF.

I tried different color LED's, with similar results. I tried a TL072 and a 4558 before going back to the TL062.

Voltages:

IC1
Pin 1 467
Pin 2 4.68v
Pin 3 2.11v
Pin 4 0v
Pin 5 4.63v
Pin 6 4.55v
Pin 7 4.66v
Pin 8 9.29v

Here are pics of the Q-Tripper. I know my soldering isn't great, but I did continuity checks, and it appears everything is connected, but of course "famous last words". I etched both PCB's on Radio Shack boards. I had trouble getting the solder to stick which explains some of the blobs. You can see I did some scraping to make sure there weren't any whiskers, etc.



And here is the Phuncgnosis, including the blank PCB. :




Any help would be greatly appreciated!

[flo]

Q-Tripper: Try to measure the envelope voltage om R16 while plucking notes, see if its ok.

Use an audio-probe and listen to the signal through the circuit. Find where you "lose" it.
See: http://www.diystompboxes.com/smfforum/index.php?topic=29816.0

[doc_drop]

Thanks Flo. I will try the audio probe. (I am aware of the debugging post, and everything it mentions. I was hoping pics and voltages might reveal something.) 

But I'm not sure I understand what you mean by

measure the envelope voltage

What am I looking for to determine if it is o.k.?

[CynicalMan]

The PHUNC  on the PCB looks like it might be bridging two traces.
      GNOSIS

[doc_drop]

I'll check that again...I did a continuity test to make sure the traces weren't bridged before I populated 'em, and again during unsuccessful debug. I think they are o.k., but there is no reason why I can't check again.

Thanks!

[Mark Hammer]

I recommend using a toner transfer approach next time.  I used to make them just like that, so I won't get all snooty on you, but both those boards present serious challenges to troubleshooting.

[bigandtall]

Both envelopes look pretty interesting. Can anyone tell me how they sound in comparison to say a 440? Any thoughts on how they'd perform with a bass?

[doc_drop]

Mark,

I'm not sure what you mean. I tested them, and there was very clear continuity from any point on any trace to any other point. They are conducting as well as wires. What would make them so hard to debug?

I did have a solder whisker or two. But that was my sloppy soldering on lousy Radio Shack PCB stock.

Thanks!

[Mark Hammer]

I just meant that when one has traces that are crisp and cleanly defined such that you can visually confirm continuity, then you don't become obliged to do all the continuity testing by hand.  Lord knows there is enough of a checklist when troubleshooting a non-working effect that it's a blessing to be able to say "Well I know it can't be THAT, because I can see it".

[doc_drop]

Mark,

I see what you mean. I just bought some better PCB stock off E-Bay, so I'll see if the magazine paper works better, but if not I will have to try the toner transfer thing. I did complete a Magnus Modulus recently which was a much more complicated etch, but it worked perfectly first thing...

Thanks

[flo]

But I'm not sure I understand what you mean by

measure the envelope voltage

What am I looking for to determine if it is o.k.?

Use the digital multimeter and set it to measure DC voltages (set it to 20V DC range or something).
Set the envelope decay to "slow", measure the voltage on R16 and pluck a note. You should be able to see the voltage change from the envelope detector circuit of the Q-tripper. This way you can verify if that circuit part works.

[doc_drop]

Flo,

Thanks for the explaination. I will try that. Luckily I have a nice autorange DMM, so I don't have to worry about that.

Mark,

O.K. I had to see if I could do a better job with the transfer using my new and improved PCB stock and magazine paper. I think it came out pretty well. (This is Frequency Central's Phase-Revolution layout. I just made the traces wider and enlarged the ground plane.)

[arma61]

And here is the Phuncgnosis, including the blank PCB. :

I'm sure you already checked all the traces, anyway I just noticed this, though it may well be the quality of pictures, but aren't the N of "Phunc" and the 2nd S of "gnosis" shorting respectively on top and of the left, hence shorting the 2 traces?



Edit: oops! Didn't notice Alex has already pointed it out 

[doc_drop]

Arma61. I better check that again. I see what you mean. God knows I've thought I checked things only to find out I missed something in the past...

Thanks for looking it over!

[doc_drop]

Arma61, you were right. There was a "Phuncgnosis" short. I fixed it, but had the same results sound-wise. I've kind of given up on the Phuncgnosis, and decided to focus on Q-Tripper.

I audio probed the Q-Tripper, which only produced bizarre results. Basically the sound stops at the input jack... I test the guitar plug and cable unplugged from the circuit, and it works fine. But when I connect it to the circuit, I lose all sound. I assume there is something stupid I am missing?...What is a common cause for something like that? (It isn't like this is the first time I built a circuit, but you never know!)

Thanks

[arma61]

Arma61, you were right. There was a "Phuncgnosis" short. I fixed it, but had the same results sound-wise. I've kind of given up on the Phuncgnosis, and decided to focus on Q-Tripper.

Great, one problem solved

I audio probed the Q-Tripper, which only produced bizarre results. Basically the sound stops at the input jack... I test the guitar plug and cable unplugged from the circuit, and it works fine. But when I connect it to the circuit, I lose all sound. I assume there is something stupid I am missing?...What is a common cause for something like that? (It isn't like this is the first time I built a circuit, but you never know!)

Thanks

I know nothing about these circuit but I had the same problem (sound stop at input jack) in another circuit, then testing with a DMM I've foud (again) a short on the PCB, it was shorting the signal to GND.

Armando

[doc_drop]

then testing with a DMM I've foud (again) a short on the PCB, it was shorting the signal to GND

Yeah, that's what I'm thinking. I guess I better check for that one more freakin' time!

Thanks!

[Mark Hammer]

Since the initial semiconductor in the Q-tripper is a JFET, my reflex is to suggest you check and doublecheck the pinout on the device.  Honestly, you'd think that with only three pins, there would be fewer ways to mix up the pinouts....but somehow JFETs can defy the laws of probability and present what can seem like an infinity of possibilities sometimes!

[doc_drop]

That's a good idea, Mark. I will check the pinouts. I did try turning them around, but maybe the middle pin is different. I will triple check the data sheets just in case...

And by the way, that last board that I posted a picture of etched very well. I didn't have to test it at all. Plus the board is translucent, so I can look through it in a strong light and see exactly what the traces are doing. So, thanks for pushing me to get better at the transfer process.  (And I hope anyone reading this refuses to buy PCB stock from Radio Shack. The stuff I got on E-Bay is 1,000 better for what we do! The copper is thicker, and accepts solder much better and the board is stiffer and of much better quality.) 

[Mark Hammer]

I think we're well on the way to getting you a system that works consistently.  Pleased to hear that a better etch saves you a big step in getting a circuit up off the ground.

While it does not happen very often, different companies CAN make the same numbered part and have different pinout from their competitors.  So, make sure you have a datasheet from the manufacturer whose tranny YOU are using, and not just any datasheet for the same part number.  Plus, make sure you have at least one with a decent picture that clearly identifies for you which pin is 1, 2, and 3.  I find a lot of them are pretty ambiguous