debugging shorted traces.

[TheWinterSnow]

So I just got around to assembling a new OD pedal from my own design.  Tested the voltages and everything was ok there, but when I went to plug it into the guitar rig it was nothing more than a buzz machine.  After plenty of poking around I found that I have a short to ground on on the low pass filter after the clipping section after the resistor, so the second half of the filter and the input to the tone stack op amp is shorted to ground.  I got the board made from OSH park and I cannot see any visible traces of the solder mask stop being too large and causing shorts to the ground plane, I even checked the other two unpopulated boards and they check out fine.  The only thing I could think of is that there may have been too much heat soldering on the opamp and therefore the input is shorted to ground.

The tone circuit being used is nearly identical to the Boss SD-1 variant, except the output is going to a 100K volume pot.  The only difference is that I am using a 25K pot that is reading 20K (25K in parallel with the 10K feedback resistor and the 100K volume pot) which indeed indicated that the input if that circuit is shorted.  The 10K feedback resistor is reading 9.2K and it is a 1% tolerance.  The 100K volume pot is even more strange, it is reading 25K across terminal 1 and 3 when up all the way, and 10K when down all the way.  The range of one terminal to the whiper is 0 to 25K and the other terminal is 0 to 10K.  Again really making me thing that there is a massive short on the input pin and a smaller but significant short on the output pin.  I just cannot see any locations where a short is likely.  I have rinsed the board off with rubbing alcohol and a brush that I used for removing solder flux.

You can get an idea of the circuit by looking at the boss SD-1 schematic http://www.hobby-hour.com/electronics/s/schematics/boss-sd1-super-overdrive.gif, the short is at the junction of R7 and C4.  In my design, remove, C7 and R10, VR2 is 100K and VR3 is 25K.

What are the methods for finding the exact location of a short when you can't visually see where it is at?

[GibsonGM]

You can (very carefully) remove the IC and see if you still have a short.  That will rule out the IC, if it still exists.

Take voltages (post them here) of each active component (Q's and IC) - sometimes a person accidentally connects a + connection to ground (thinking Q1, here).    You may see an issue in the voltages, like your input could be grounded....

I test for continuity between + and - , and if there IS continuity, there is a short.    To definitively locate it, on a relatively simple circuit, you could jumper your - probe to ground, and use the + probe (set to continuity beeper) to go thru each component.  Looking at your schematic, you'll note where you SHOULD have continuity to ground, and where not....eventually, you should find the one that is NOT supposed to be grounded, and there you have it.    Should not take you very long to go thru, especially if you take the IC out first.   

You can also use the IC socket and test "pin to ground" - I assume you should only have ground at the chip's ground connection.  Also test that you only have your supply voltage on the pin you should!   Only apply power when measuring the voltage, NOT when doing continuity/resistance checks....

This should get you started. Also, in the main "build your own stompbox" forum, see the thread "Debugging: what to do when it doesn't work" for the standard way to present your voltages, etc.....it will help us to help you.   

[TheWinterSnow]

I know what trace it is shorting at, I just don't know the exact location where the trace is shorting at.  On the schematic the short is taking place at the junction of R7 and C4, meaning the input of the opamp has continuity to ground.  That is the only trace that is doing that and that trace has an infinite resistance to all other nearby pins.  Actually it has a 150K resistance from ground to the power rails.

Anyway, I made the board too small and I can't do any desoldering work, I will make a new one, screw SMD and 2.54mm spacing, I don't have the equipment to do rework on that.  Back to big bulky through holes that are spaced out.

[GibsonGM]

Tone control wiper goes to ground, could be something over there....R19 does, too, and there could be a place to look.
Or just re-do it, like you said....SMD, UGH!!!!   But we'll all be going over to it in time....altho I have a lifetime of parts in stock, so maybe not.
Good luck on a new etch!

[TheWinterSnow]

Secondary question as I finish up on the new board design.  Could making both layers on a two sided board where components connecting to ground are connected to both layers cause ground loop issues?  Do you typically leave one layer with a ground our and the other with none?