Obsidian Overdrive Feedback Problem... still

[Joe Davisson]

Can you scan the front & back of the PCB? Which layout did you use? I think you're 90% there. If you have any smaller caps laying around, pick the largest one and hold the leads across the battery and see if it changes anything.

[robbiemcm]

So you don't have to go reading anything else Joe...

I used the schematic located here
http://www.diystompboxes.com/analogalchemy/pedals/obsidian.html
and the only bit I subbed out was the transistor, I used 2N7000 transistors. I reused one (not sure which one) as I only bought 5 (I'm getting a new lot sometime soon, if all else fails I'll try new ones, 50 cents each). I reversed the battery polarity and the electrolytic capacitors aswell.

At the moment I do not have any capacitors in my posession, but I'll pick up some 100uF capacitors on my next trip and try what you said.

The layout can be found here
http://photobucket.com/albums/y13/robbiemcm/?action=view&current=circuit-layout.gif
There are some errors in the top right, but I fixed them all when I was soldering. I was missing a hole for the 33k resistor - 0.1uF capacitor - ground section. I combined the resistor and capacitor into the same hole. Missing a bit of trace, put a lead between them and it's now got a trace of solder between them. And for the tone pot I put a middle hole forgetting that the lead simply feeds off into the volume pot. There is a bit where the trace is almost scratched off at the top, that doesn't matter as it's covered with solder.

Thanks for all help you may be able to give me, from what I can get through the extra noise it's sounding awesome!

[robbiemcm]

Oh yeah, and when I had an audio probe connected to ground a can hear a similar noise even when the probe isn't actually touching anything. It's really faint and my amp has to be turned up fully though.

[lethargytartare]

Following on what vortex said, and to a question you asked -- with those types of pots, try making a loop out of your wire's exposed end, and put that loop around the end of each lug, crimp it down and solder it -- farther away from the chassis.  And you know what -- I've wrecked a pot or two by overheating them -- is it possible that by soldering them so close to the chassis that you may have overheated one of them?  Maybe just recheck their resistances to make sure they're all still ok.

Anyway, LOVED the video!  I watched it laughing saying "Now THAT is an unambiguous way to describe a problem!!" Bravo!

Best of luck!

ltt

[robbiemcm]

Thanks for the suggestion I'll go and recheck the pots, because I get the same when it's not in the enclosure I'm guessing that the problem lies somewhere else and it's not a short or anything.

Can I work out the resistance of a pot using various laws and just the voltages and currents? I've forgotten if my multimeter has something to work out resistance.

[robbiemcm]

I tested all the pots, tone and volume seem to go from 0 to about 94... gain however, about 2 - 86. This a problem?

[lethargytartare]

Probably not -- when I have a pot go bad usually the resistance doesn't change when I move the dial.  For example, the last few I had go bad (or bad ones I tested) all registered infinite resistance no matter where the dial was -- so nothing was getting through them.

And it doesn't sound like your method of connecting to them is a problem.

When I read the schem, there was a lot of confusing language in there, and some things that I might have gotten muddled up:

"Original versions of the Obsidian used N-channel MOSFETs. I found the P-channel counterparts to be much less noisy, on the order of 5-10 times. The circuit now runs on a negative supply. Fortunately, all you have to do is flip the battery around and pretend that + is - and vice-versa. And make sure the electrolytic capacitors are "upside-down" as shown in the schematic."

So the schematic is designed for P-channels or N? And the next few phrases are "the circuit NOW runs on negative supply"  and "make sure the elect caps are upside-down AS SHOWN..."  So I'm guessing that all means the schematic there is for P-channels, you would use the black battery lead to power the board, and the caps are all properly oriented.  So if I were using N-channels, then I'd use the red battery lead to power the board, and would flip the caps.  So at this point I'd ask, since you're using the N-channel, is the red battery lead powering the board, and are your caps oriented OPPOSITE the direction the schematic has them?  And then if it didn't work I'd probably try the other way around since I'm not 100% sure I get the gist of that whole circuit note.

The only other thing I'd suggest is posting the voltages for all pins on all three trannys and see if anyone here has readings they can compare.  I'm amazed how guys here can look at the readings and figure out, with some precision, what's probably going on...

Any pics of the soldered side of the board (as it is now)?

Cheers!

ltt

[robbiemcm]

Battery leads are the right way round, capacitors I was sure were the right way round, not so sure now. I'll take them out and reverse them and see what affect that has on my circuit, nothing better to do. Would that cause this sort of problem? Is there a way I can tell which way round they need to go once I've cut the leads off?

And in terms of the transistors...

Q1
D: 3.45
G: 1.40
S: 1.38

Q2
D: 3.54
G: 1.38
S: 1.38

Q3
D: 3.57
G: 1.39
S: 1.34

[robbiemcm]

Electrolytic caps are in the right way, all caps & resistors are right values, battery is connected with +9v supply as should be done with a 2N7000 transistor in there.

Can anyone see any problems with the transistor values in the previous post?

[Joe Davisson]

I read them wrong before, the voltages should be closer to:

drain - 4.5
gate - 2.25
source - .75

On your pedal the gate and source are nearly the same voltage. I think the transistors could be damaged from static, I don't know what other conditions would cause that.

[robbiemcm]

I wouldn't be surprised. I did a schocking job soldering them. No heatsink the second time round after I screwed up the first lot. Also, considering one of them was pulled out... which was hard... it got really hot while I was pulling it out and tugging. Maybe this one bad one is screwint he others up? It doesn't matter either way, I'll just replace all of them. I'll get a batch of 25 tomorrow, and they're only 45 cents each, 33 cents if you get 25. If that doesn't fix the problem I will be back.

[robbiemcm]

Alright I'm just finding this annoying now... I changed all the transistors, and it was good for a bit, no crappy high pitched squeals. The volume knob wasn't doing much, and tapping the side of the box was enough to make the pedal operate at no, half or full volume. I decided this was because of the crappy job I'd done on the pots. So I pulled it out of the enclosure again, did my pots really cleanly and now that's all good, volume is working, knocking it doesn't half or kill my output

Then I plug it back in again... what's this... just wait a second... it's back again. I was so pissed off. Why is it doing this? I looked at the voltages of te transistors down before and after this happened. It didn't seem to change (from memory of the voltages when it worked). Although the voltages are better, it's still not what you seem to suggest it should be Joe. Q1's D is actually 3.5 though... anyway most are around...
D = 3.6
G = 1.3
S = 1.2

[robbiemcm]

Gahhh... still nothing... nobody knows why my pedal is screwing up? Am I meant to take voltages while the jacks are in?

[Joe Davisson]

Ok, you drew your layout as if the transistors would go in from the copper side. Usually layouts are flipped before being transferred to the board. Just flip the transistors around and bend the middle lead the other way so it fits.

I couldn't get any voltages that come close to yours unless the transistor was flipped, so that could be it.

[robbiemcm]

Sorry Joe, but that was the problem in version 1.1 I've fixed that, when that was happening it was effectively killing the signal. The thing that is really odd is right now I get a huge amount of distortion, I love it! If I could just get rid of the squeal I would be 100% satisfied with this. When the guitar is played and it's overpowering the squeal, it's awesome. So well done Joe But the squeal can overtake if a note is not played loudly within a couple of seconds...

Developments of my build of the Obsidian Overdrive

Code Select Expand

| Version | Problem                           | Fix                                   |
+---------+-----------------------------------+---------------------------------------+
| 1.0     | Static noise, guitar barely heard | Jacks had to be grounded              |
| 1.1     | Signal killer, almost no output   | Transistor orientation, D + S flipped |
| 1.2     | Extreme squealing noise           | Still unknown                         |

[robbiemcm]

Gah still no luck. Is there a way I could solder in a transistor mount or something... so I could push them in and reuse them if I need?

Or if you happen to know what the problem is I'll take that too