MI Audio Crunch Box - gated sounding, then dies.

[J0K3RX]

Not a big fan of vero board... can get confusing! Ever etch a board?

[jymaze]

Hey Jok3r, good to see you.

Mikestahlme,

I would desolder C6 and check it has infinite resistance in DC with a meter; then if it is not C6, desolder R5 out of the board and check it is really 1 Meg with a meter,. Then change the IC if R5 and C6 you are 100% sure that R5 and C6 are not damaged. It is quite common for ceramic caps to fail by short-cutting while resistors would more likely fail by becoming non-conductive at all. Because of that I would check C6 in priority.

If you have eliminated that and want to swap the IC, a TL072 will be fine, but any double opamp will do for the purpose of testing. I hope your IC is socketed. It is better to socket them, less risk of frying them when soldering, and easy swap if they fry for any other reason.

[mikestahlme]

Jok3r, I haven't etched my own board yet, but I have a friend who has and it doesn't seem to be too hard. 

Jymaze,
Thanks for all your help.
I checked both components and they tested fine. I ordered some sockets a couple of weeks ago when I realized how much I'd be messing around with this one. (lesson learned!) Unfortunately they won't be here until ~ December 26. I'll have to look around and see where I can find one.

[jymaze]

If you live in North America, they have decent DIP-8 sockets at Radioshack.

[mikestahlme]

Ok, I replaced the IC with a TL072 and a JRC4580D and got the same gated fuzz sound. I think I may have fried the LM833 earlier because when I put it back in it is considerably quieter than the others.

Any tips from here? I was thinking maybe I should start replacing caps that I haven't already replaced.

[slacker]

Are your voltages still wrong for pins 6 and 7? What voltages do you get if you measure those pins with the opamp removed and the circuit powered?

[jymaze]

I really don't understand what is wrong with your circuit.

Can you do one more round of voltages at all pins while powered, and measure resistances between 5-6 and 6-7 while unpowered, all of the previous with the IC on and the IC off. With that much info, we should be able to fully characterize the problem hopefully.

Changing other caps is probably not needed now as long as you are sure that C3 and C7 are fine. I remember you checked C3, so you can check C7. Value of C7 is not critical as long as it is at least 2.2u so you can change it with anything you have that is at least that value.

[mikestahlme]

OK, here we go

Powered, IC OFF:
Note: The longer it is powered, the LEDs go from being very dim to moderately bright. They start out flashing and eventually stop.
1. 0.80
2. 0.98
3. 1.08
4. 0
5. 5.5
6. 2.2  (not steady, fluctuating between 2.1 & 3.0)
7. (fluctuating between 0.1 & 0.9)
8. 12.1

Resistance between...with IC OFF
Pins 5 & 6: -1.5 (negative?)
Pins 6 & 7: (fluctuating between 125 & 128)


Powered, IC ON:
Note: At this point the LEDs are moderately bright and steady as soon as I plug it in.
1. 1.4
2. 1.2
3. 1.07
4. 0
5. 4.0
6. 2.5
7. 2.7
8. 11.9

Resistances...with IC ON:
Pins 5 & 6: 88 and climbing into the 90s. Checked again, this time reading infinite on the 200k range and 777 when set to the 2000k range.
Pins 6 & 7: 97

I have replaced the 2.2u once already.

I hope this helps, thank you for your time.

[jymaze]

Bad news, your readings are totally messed up. Besides the power supply readings, not even one of them makes sense! Not even one!

There is just so much that is wrong that I don't even know where to start. For example, the LEDs should not light up in any case without signal, IC being on or not! The rest is in the same vein, with discrepancies in every kind of ways.

At this point, I think the only thing left is to take pictures of both sides of you board (put your IC on too), as clear a possible, and post them here.

I am confident it will help.

Hold on!

[J0K3RX]

Bad news, your readings are totally messed up. Besides the power supply readings, not even one of them makes sense! Not even one!

There is just so much that is wrong that I don't even know where to start. For example, the LEDs should not light up in any case without signal, IC being on or not! The rest is in the same vein, with discrepancies in every kind of ways.

At this point, I think the only thing left is to take pictures of both sides of you board (put your IC on too), as clear a possible, and post them here.

I am confident it will help.

Hold on!

+1  What he said! 

[mikestahlme]

Here are pics of the front and back.
The top of the back side picture corresponds with the top of the front side.
https://dl.dropbox.com/u/54382528/DSC_1973.JPG
https://dl.dropbox.com/u/54382528/DSC_1972.JPG

In case you can't tell from the picture: Because of the size of the electroyltics, the 10k coming off the top track has its lead bent around the side of the 100uf elect cap. That same cap's negative lead is in fact going to the ground rail right next to D1.

[Ronan]

Are you using lead-free solder by any chance? It does look like the solder may not have wetted onto some of the component leads on the pcb. The red wire soldered to the LED is similar in the way that the solder has not flowed enough to properly bond the LED leg and the red wire. Just a thought. It's going to be a nice looking build once/when you get it going.

[slacker]

I see a couple of places where you might have a short between traces, I've shown them below. Either go over those with your iron to remove the unwanted solder or use a sharp knife to scrape it away.



If it still doesn't work after doing that, please could you do the measurements with the opamp removed like you did before.

[J0K3RX]

Probably not the problem but you never know?

Why don't you leave the stomp/foot switch off until you are done trouble shooting in case you may have wired it wrong? Just wire it straight without a switch/ always on...

Also, I don't know if you checked this and I am not going back through and re-read the past posts but make sure you have the polarity wired right...

[jymaze]

Thanks for the pictures. I cannot find anything suspicious right now, but I'll look better.

Some more measurements request:

1) The resistance you measure between pins is with the circuit being unplugged right? I feel like you measured it with the power being on, it is not how it should be (it would explain the wacky values, like a negative resistance). Also, check that your multimeter is reading correctly on some spare resistors in your stock and/or check the battery: Sometimes it can get crazy when the batteries are getting low. Please remeasure if any of the above was causing a problem, I really would like to have accurate measurements.

2) Can you measure the DC across the LEDs, with power being on this time: It should be 0 volt. Actually you should check that DC across all pots is about 0 volt.

3) The bias should be 4.5V both on pins 3 and 5. I have no idea how it could be anything different. What is th resistance from the bias point (pin5) to V+ and then to ground (power being off)?

[mikestahlme]

Hey, while double checking my polarity I noticed I'm not even getting infinite resistance between the positive and negative wires at the power supply! Instead it reads 145k what's with that?

Slacker, the spots you circled are just longer leads that appear to be crossing over because of the angle. No shorts there.

The resistance I measured between the pins was with the circuit unplugged.

I see how my measurements are wrong, when I test resistance between V+ and pin 5 I'm getting -7 BUT that's because I have my black probe on the V+ and my red probe on pin 5! I didn't think this would make a difference when just measuring resistance without power but apparently it does. With my red probe on V+ and black on pin 5 I'm reading 58.5k

Sorry! I'll have to re do my original measurements tonight when I have time. 

[jymaze]

It should not make a difference for the resistance measurements with the IC off the socket at least. You just don't want to inadvertently counter-polarize C5 and C7 by having the red lead on the ground.

I was thinking about something too: Make sure that the tracks on the left and right edges of the vero are not short-cutting to the case. Usually I add a good bevel so the tracks don't reach the border of the board.

You know, your vero looks good, but the measurements are all whacky, I really have no idea why. As Jok3rx said, check your off-board wiring again, and maybe pull the whole thing off and try to have the pedal work with minimal wiring in the open at first with foolproof wiring (one battery, the 2 jacks, no switching, with the pots just hanging there).

It is easier to debug that way, and you can take great pictures for us to see. Then when it is all fixed (hopefully we will get there), you can rebox it.

[Pyr0]

Some of those solder joints don't look solid especially on the RHS. Could possibly have a dry/cold joint or two. The solder should be coming up the component legs, try re-soldering them and make sure you heat the component legs so that the solder wicks up them.

[mikestahlme]

I pulled it all out of the box and it's still doing the same thing. I'll try going over all the solder joints again and post some pictures.

I'm still trying to figure out the neatest way to wire this into the box, using the least amount of wire possible. I wired this with all the parts already in the box. Is it better to just wire it up outside the box?

[jymaze]

Usually I wire with everything already in the box, giving just enough slack to be able to bundle the wires together with some zip-ties so it looks less messy in the end.