[sdiy] MOSFET cap discharge

[Ian Fritz]

Hi listers --

We often hear that MOSFETs shouldn't be used for the cap discharge switch 
in a saw VCO, because they have too much leakage current.  Indeed, the spec 
sheet for the BS170 (for example) lists a max drain cutoff current of 500 
nA.  Pretty bad.

But if you look in Horowitz and Hill, they say that the channel resistance 
in the off state is usually over 10000 M, i.e., 10^10 Ohm.  So with  1V 
across the channel, leakage should be under 100 pA.  How bad is it in 
practice?  I have heard that some people have had success with MOSFET 
discharge switches.

So I wanted to try this, and I also wanted to try to make a VCO with a pair 
of dual opamps (for simplicity) and with a 1V saw amplitude (to keep a 
reasonable switching speed).  The circuit I ended up with is posted here:
http://home.earthlink.net/~ijfritz/xfer.htm

The front end uses a precision device (OPA2227) to reduce offset errors and 
drift.  The ramp/reset circuit uses an OPA2134.  This has been used before 
in a VCO integrator, and with bias current in the pA range and a 20 V/us 
slew rate it works quite well.  As a comparitor it's not so hot, but using 
the MOSFET switch and keeping the ramp voltage low keeps the reset time 
down to about 600 ns (see photo of  the reset part of waveform). Note that 
the voltage divider feeding the FET gate keeps the gate near 0 V in the off 
state, so the comparitor only needs to slew a couple of volts to turn the 
switch on.

With the current source disconnected, the oscillator free-runs with a 
period of about 5 min (600 s or 3 mHz).  This corresponds to a total 
leakage current of only 40 pA.  Moreover, when the MOSFET is removed the 
ramp rises at about the same rate!  So the MOSFET leakage is under 
something like 20 pA.  40 pA of leakage is a bit large for the 2134, but 
not out of specs.  There is probably also some leakage from my grungy old 
whiteboard.

Capacitive coupling is also an issue with MOSFET switches and you can see 
some ringing on the waveform.  Most of that can be filtered out in 
combination with whatever amplifier is used to raise the signal up to the 5 
or 10 V level.

This VCO tracks quite well.  With correct tracking set at the octaves 
starting at 200 Hz and 5 kHz, tracking is within 0.2% up to ~16 kHz.  Above 
that it goes a bit sharp.  (If the comparitor is replaced with an LM311 the 
high end goes flat, so it must be opamp phase shifts that make it go sharp.)

Anyway, don't be afraid to try MOSFETs!

   Ian