Bypass 3pdt Footswitch Causing Oscillation

[idy]

I built a clone of a Wampler black 65, a four stage FET "amp-like" overdrive on veroboard.  I had a problem with oscillation and spent a long time tweaking the trimmers on the FETs and etc. The problem only happened when I had all four stages going (the "boost" setting) and volume and gain maxed. I moved wires, took the board out of the box, etc. etc. This didn't happen on the bread board. To get rid of the oscillation I had to turn the trimmers to less than optimal settings, getting a weak signal. I tried a new layout on vero, same problem.

Finally I thought, since this only happens with volume maxed, it must not be a fault on the board since the volume is just a voltage divider connecting board to output jack (or the footswitch..).
I took the wire from the Board, connected to it to the out jack, no problem, no oscillation. I could set all the trimmers to the "sweet spots" and everything was OK.
I put the volume back in, (no footswitch) no problem. Good sound.
I took the wires from the volume and the out jack and as I even brought them near the footswitch terminals...whistling oscillation!
Everything is good until the wires touch the 3pdt! I tried replacing the switch, no help. I tried an alternate wiring scheme: I usually wire the switch with the LED on the left hand pole, then the board in, then the output, so I tried putting the LED on the middle pole, still no help.

The switches I have used for many projects without problems before. I won't name supplier, they are the blue ones and don't even say country of origin.
Anyone else experience this? Better switch will fix it? Any tricks?

[Pyr0]

Have you tried using shielded cables, especially for the input side ?

[armdnrdy]

I had a similar problem with my very first build. Since then I've used RG-174 coax cable (very small diameter and very flexible for coax) for my I/O connections on high gain circuits. I haven't had a similar issue again. (but I've had many, many others!!)

Use the center conductor for your signal wire and attach one side of the braided shield to ground. RG-174 conductor size is 26 awg which is fine for what we do. I make up the cable by stripping back the outer jacket, sorting out the braid with a piece of resistor wire that I cut off from a past build, (see guys, you throw those things away! they're tools!) solder a short piece of sranded wire to the shield, and then finish it with a bit of small shrink tubing. It's extra work but it looks neat, and works!

I took a break in writing this and Pyro shot the same thing in.

[idy]

Thank you.  I ordered some RG174. In the mean time I took some shielded two conductor radio shack cable (stranded two conductor with braided shield), pulled the conductors out and slid the braid and outer insulator over my wires: in and out jacks and in and out board, grounding one end. Oscillation gone. I see there is some craft to making a neat improvised shield connector.

[armdnrdy]

Glad to see it was an easy fix.

When I route my own boards, (which is always) I often add extra ground pads by the inputs and outputs so I can attach the shield wire (that I soldered to the braid) to ground. If adding a ground in those locations isn't viable, then I take a ground from the board and star ground everything in the enclosure. (jack grounds, any pot or switch grounds, and the shields)

[idy]

I did get the shielded cable. It improves matters but not enough. I have built three and all feedback when the FETs are in the sweet spot.And I can hear that thngs really sound better in that spot, the overdrive sounds more substantial.
I've tried a few work-arounds like changing the voltage divider after the first stage for a trimmer. This helps but there is not enough difference with the boost engaged. Really taking the shielded idea to the maximum is hard because there is also a "boost" switch off board that adds that first stage, and tiny runs of shielded seem awkward. I am using jacks on the north face of the box, so there is a lot of cable in the box. Jacks near the 3pdt might help.

[armdnrdy]

tiny runs of shielded seem awkward.

Awkward as it may be, if it might solve the problem.....try it!

I think that adding shielded cable period! is a pain in the butt and awkward, but it sure beats noise generated along the wires in a high gain device!