Re: MP-1 Hum, Hiss, Troubleshooting

On Mon, 26 Nov 2012 23:31:11 -0600, Benjamin Catchings
<jonahcrock@???> wrote:
> I have gone through the whole thing on a couple of occasions and
> replaced my mods (the three caps in the power section from the noise
> mod, the 3 caps in the tube section, 3 or 4 transistors from the tube
> section and a couple on the main board in the same area, about 6 of
> the opamps including the headphone, effects loop, etc - U4, U8, U9,
> U10) and did a little better quality work on it than I did the first
> time.

Hi Ben,

I feel your pain!

So you've done some homework already. New IC's, filter caps and such.

For this kind of troubleshooting, it benefits you to have an
oscilloscope, or at least an audio probe (a small audio amp
with a speaker or headphones, and multimeter probes for poking
into circuits). But you can do a lot without the tools.

The only things I have yet to replace are the tubes (both are
> lighting up fine and don't appear to be damaged at all from visual
> inspection), the 6 diodes in the power section or the 4 in the tube
> section and the transformer.

You can largely rule out the possibility that the problems are
confined to the tube board, by reproducing the problems using
the S.S. channel. (Unless the channel switching itself is affected.)

> Is it likely that my problem is the transformer or would it even be
> coming on if the problem was there? Are the diodes more likely to be
> the culprit? Or is there something else that more commonly causes this
> symptom?

I would investigate this as an overvoltage event affecting the
front jack's FET buffer circuit.

The front input of the MP-1 goes into a JFET transistor stage, which
is followed by another bipolar transistor stage. After that stage,
the OD1 level is applied, and the signal goes into the buffer provided
by an op-amp on the U4.

People don't seem pay attention to that JFET circuit; it's not
the target of any popular mods or upgrades that I know about,
and the "stock gain mod" even talks about bypassing it, which is
a dumb idea that will ruin the unit for any guitar with passive
pickups.

An over-voltage into the front input jack would likely fry the JFET
transistor.

Here is a test: does the volume level of the noise respond to
changes in OD1? What if you turn OD1 all the way to 0.0?

Another test: does it make a difference if something is
plugged into the front jack or not?

Another test: what happens if you plug into the rear jack?

Another test: what if you put a plug into both at the
same time?

When nothing is plugged into the front jack, then the
FET buffer is disconnected; it does not feed into the OD1
stage. However, the rear jack circuit *is* connected at
that time. The rear jack is based on the U11 op-amp. That
might be fried, because as you insert a plug into a jack,
the momentary contact could send the voltage there.

The rear jack itself is also switched; it is shorted out of
the way when nothing is plugged in. But that is before the U11
op-amp. If the U11 op-amp is fried, it could spew noise.

When you plug into the front, the output of U11 is then
disconnected; it does not feed into OD1.

If the noise goes away when you plug something into the
front, that would indicate that U11 is fried and spilling
noise.

If the noise is only present when you plug something into
the front, that points to the FET stage.

If the noise is different, but OD1 turned to 0.0 can
quiet it, that suggests both the rear jack U11 stage
and the front jack FET circuit are toast.

Here are some other ways to subdivide the MP-1 externally
(without probing for signal into circuits).

* Enable the effect loop and then plug a dummy plug into
send. Does it quiet down? Then the noise is not in
the output. You can further verify this by injecting
some signal into effects return and see that it cleanly
comes out on the output.

* Always check whether a noise comes out the same
way from all possible outputs: output A, output B,
headphones, effect send.

* Always check which control the noise responds to:
OD1, OD2, master? Rear pots? Solid-state versus
tube?