Can I use BS170 as a "series" switch?

Started by earthtonesaudio, December 31, 2008, 08:43:58 AM

Previous topic - Next topic

earthtonesaudio

...Like the "series" photo-FET in this picture:
http://www.geofex.com/Article_Folders/automike/micmute3.gif

???

Or will the inherent substrate/drain "body diode" still be passing guitar level AC (of one polarity) even when it's supposed to be "off?"

R.G.

It's a silicon diode. It starts conducting when the voltage across it in the + direction is more than maybe 450mV. It has some capacitance, which varies with reverse bias.

I've had bad luck trying discrete MOSFETs as series switches because of this. The signal needs to be quite small to avoid the diode effects.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

earthtonesaudio

Thanks for the reply, RG.
I guess that leaves me with 3 options...

-Use two BS170's, joined at their sources, with drains pointing out, and deal with driving two separate gates (ugh! gross)

-Find a suitable 4-terminal replacement and tie the substrate to V+

-Use a CD4007 (might have a winner here)

R.G.

R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

earthtonesaudio

...If I could I would, but I can't.   :icon_neutral:

I've already committed to using n-channel MOSFETs elsewhere in the circuit, and I want to switch all devices simultaneously (or at least at approximately the same control voltage).

gez

#5
I once used MOSFETs for bypass in one of my stompboxes.  In that instance, the effect out and 'bypass' had two MOSFETs switched anti-phase.  One MOSFET's drain connected to bypass, the other to the effect out.  Both sources merged to an output buffer (biased, and everything DC connected).  Because the effect signal wasn't out-of-phase with the bypass signal, and because the effect out was the same amplitude as the dry signal (sort of), the source of whichever switch was open was always pulling the other MOSFET's source up/down, thus avoiding substrate conduction. 

In retrospect it was dumb of me as the effect was a trem, so with a fair amount of modulation there would still be the danger of a potential difference (the wrong way) across the MOSFET's channel...enough to cause conduction when bypassed (FX bleedthrough).  I got away with it, though. :icon_redface:
"They always say there's nothing new under the sun.  I think that that's a big copout..."  Wayne Shorter