[sdiy] mosfet/cmos opamp

[Ian Fritz]

At 06:50 AM 10/28/2010, Harry Bissell wrote:
>Do you find the CA3140 to be much better  the TL07x (or 08x) series
>in a sample hold application. I have used both but have not noticed that
>much difference.

My notes indicate that I got .4 mV/sec drift corresponding to a leakage 
current of 2 pA.  This was considerably better than what I got with an 
OPA134 (spec'ed at 5 pA input bias current).  The OPA602 (1 pA spec) worked 
pretty well, but it costs over $5.  I didn't measure with a TL07x, but they 
are spec'ed at 65 pA typical, 200 pA max.

My circuit uses a 30 usec sampling pulse to allow sampling rates well up 
into the audio range, so getting the droop down is pretty critical.  A 
circuit with a guard ring is required to get the best performance. On the 
SBB the measured leakage was a couple of pA larger.  :-)

Ian