I have a DOD FX50 overdrive and when the effect is in bypass mode
I can hear a distorted signal in with the clean signal.
Is there any way to help this problem?
I have tried different JFETs but it didn't help.

Any ideas?

Adam

Lots of them.

In a JFET bypassed pedal, the bypassing relies on the ability to turn the JFETs on and off; for N-channel JFETs, that means that the sources and drains are biased up to some DC voltage less than the battery but more than the Vgs(off) of the JFET. That lets you turn the JFET off by pulling it down to ground and on by letting it float up to the source voltage, which it does all by itself if you just open the drive on the gate by inserting a diode in series with the gate in the correct direction.

When these kinds of things don't fully turn off, it means that one or more of the JFETs is not getting turned fully off. That means one or more of the following is true:
(a) the JFET gate can't be pulled down enough by the driver
(b) the JFET has drifted and needs more pulldown voltage than it used to
(c) the diode in series has gone bad and is no longer letting the driver pull the gate low enough
(d) the source and drain bias voltage is too low to allow a grounded drive voltage to turn off the JFET
(e) the driver circuit is dingled and can't pull the gate low enough through the diode.

The debug process is as follows:
Measure the source voltage on the JFET. Is this high enough? (look at the schemo)
If it's high enough, measure the voltage on the gates of the JFETs in both effect and bypass settings. Does the voltage equal the source bias voltage in one position, and more importantly, does the gate voltage go to one diode drop above ground when the JFET is supposed to be off.
If it doesn't go low enough, measure voltage on both sides of the diode when the JFET is supposed to be off. Is the diode open? If so, replace.
If not, does the driver pull the cathode to ground? If not, the driver is bad.