tempus fugit wrote:

The 2 switches in question are the Channel select (at the bottom of the
schem - the rest of it is to the left, which I'll post tomorrow - sorry
again about that), and the volume boost (closer to the top). The reverb is
not switched remotely, since I leave it on all the time. I misposted in
saying that it is the effects 1+2 footswitch, since it is in fact the
channel select and volume boost (effect 1) that I'm actually working with.
The rhy LED is the one that is still (very) dimly lit when I switch. I
should also note that when I use a MAX4662 anaog switch
(http://datasheets.maxim-ic.com/en/ds...61-MAX4663.pdf) (Ron = 2.5 ohms)
that the LED is completely out when switched off. I didn't relocate Q7 and
Q8 outside the amp; the LEDs are on the front panel, and if I am correct in
their purpose, they are the drive transistors for the LEDs.

You mentioned the MOSFETs as a possible alternative - this is what got me
posting in the first place, along with the other post about transistors. I
had thought of replacing my BJTs with MOSFETs, but then learned that they
shouldn't have any voltage on them before they were powered up. SInce the
possibility exists that the amp may get turned on before the pedalboard (the
switching system), that would put 12v on the drain with no voltage on
anything else until the pedalboard was switched on. Looking at the circuit,
though (which would be basically the same as my BJT circuit - using a
transistor as a switch, only with much higher resistor values) I couldn't
see why that would be an issue, since I wouldn't have any voltage at the
gate unless I was actually switching the MOSFET on anyway.

I've included a partial schematic of my transistor switching circuit. The
ones in question are in the lower left.

Okay, I think I see what is causing you trouble. When the
select switch is closed, there is a 1 diode drop of voltage
at the right end of D4. And that one diode drop is applied
to the series combination of D5 and the base emitter drop of
the two transistors Q7 and 8. If the switch is replaces
with a saturated transistor, you add its couple tenths of a
volt drop to that diode drop (while it carries the LED
current to light D3). So the bases really do not see zero
volts, or even necessarily half of the diode + drop. This
allows them to both leak just a little collector current.
Add either another diode in series with D5, or a base to
emitter leak resistor, say, 22k, across Q7 and Q8, to keep
them completely off till the full drive comes through. Then
any ordinary transistor like a 2N3904 should work in place
of the switch.