Mosfet as shunt/mute device

[Ethan]

OK I give up! I have been trying to bias a mosfet as a mute switch.  I know RG has an article for a shunting circuit. but my application is a little different.  I just want to ground the output signal to mute it. I want to trigger a mosfet with a line off a PIC. I don't want to use a JFET cause a MOSFET has a lower Rds(on).  I don;t want to use an opto-fet cause it's too expensive.
I am all out of ideas, any help is appreciated.
-ethan

[R.G.]

I have been trying to bias a mosfet as a mute switch.  I know RG has an article for a shunting circuit. but my application is a little different.

It would help if you could describe how it's different. What results have you achieved?

A single MOSFET is not all that good for shunt muting, because in the reverse direction it just looks like a silicon diode, so it will clip the unmuted signal.

To get around that you have to use two MOSFETs in series, with opposing body diodes, and then activate the gates with a signal that enhances both of them. It gets complicated.

I just want to ground the output signal to mute it. I want to trigger a mosfet with a line off a PIC.

No problem there. I'm guessing that you want to have the PIC working with the same ground as the signal circuit, too, yes? It might make sense to use an NPN bipolar transistor as a shunt muter. Use a big capacitor in series with the collector, and pull the base more negative than the signal goes. This kind of thing is in all the Japnese hifi stuff. Or was.

I don't want to use a JFET cause a MOSFET has a lower Rds(on).

Rdson doesn't really matter all that much unless you're also trying to keep a very low output impedance on the output that's being shunted. If, for instance, the output being shunted has an internal resistance of 10K, then shunting by 10 ohms is 60db of muting. JFETs can get there easily enough. If the output doesn't have that much output impedance, then put a 10K in series with it.

As I said, all of these can be handled, but we need to know more of what you're trying to do to go a better job of advising you.

I don;t want to use an opto-fet cause it's too expensive.

They're under $3.00. A MOSFET is $0.50. The $2.50 pays for a lot of easy and quick.

If it's shunt muting the output of a PIC controlled footswitcher that you're doing, the JFET will work fine. All you really have to do is run the impedance down pretty low, under 100 ohms to contain the clicks. How do I know this?

[Ethan]

goal:
True bypass using relay with no pops.

quietest method so far:
omron dual coil relay with 2 3904 drivers (taken from RG) driven low or high from PIC

problem:
Still get faint ckicks;

proposed solution:
Shunt(mute) output to ground before switching relay.

method:
I figured using a MOSFET with 2ohm Rds on would give me more attenuation than using a JFET with 50-100 ohms Rds on.  I also figured that it would be easier to bias since I don't have to deal with a neg pinch off voltage of JFET.

My application is different from millenium bypass cause you are driving the gate of the mosfet very creatively w/leakage current from diode and then pulling it low through dc resistance of effect (usually pop-tie down resistor).     I am trying to drive the gate through +5v or 0v from PIC.

BTW, mouser has 2n7000 mosfets for $.10 but if it's too much of a PITA to make work, you're right  it's not worth it.  JAMECO has H11f3 for $1.59.

from "Millenium 2 shunt" article:
Question: In using a transistor as shunt device, what is the cut-off resistance of CE junction?
Question: I can't seem to figure out how transistor and the 100uf cap works.  I know it's blocking DC but there is no voltage on the collector.

I am pasting a link to couple of your shunt circuits that I modified to use PIC.

http://putfile.com/pic.php?pic=2/5611154637.jpg&s=x402

They simulate OK with multisim, but I haven't tried them for real yet.  Hoping you could take a  look and advise.  The mosfet circuit in the schematic does not work at all. It is there for illustrative purposes to to try explain what I was talking about earlier(sorta).    AC signal drifts to bias voltage or doesn't shunt signal.

Well that's it for now...