[sdiy] expo accuracy? or integrator accuracy?, or both?

[Czech Martin]

Not really. CA3140 has a zener diode ESD protection network.
Without that you would most probably destroy the input
very soon. This means leakage, I have measured 30-40pA
at room temperature.

Therefore most CMOS types are not better than JFET types.
The only CMOS type with close to theoretical leakage
is the LMC6001 (?), fA! But I think the gain
is not so good.
Only A grades are tested for that, and suppliers do
not carry them. This test seems to take 3 min,
and you have to pay for that (testing cost much more than die
cost). ESD protection goes to 2000V HBM
(I would really like to know how they do this,
perhaps by creating virtual 0V across leaky protection
network junctions).

OTOH: the OPA111 (or so) is JFET, but has very low leakage
and good gain /bandwidth as Ian Fritz pointed out...


m.c.

-----Original Message-----
From: Scott Bernardi [mailto:sbernardi at attbi.com]
Sent: Freitag, 7. Februar 2003 15:05
To: Synth-DIY list
Subject: Re: [sdiy] expo accuracy? or integrator accuracy?, or both?


I'll comment on one of these error components - input bias current. The input bias
current of a JFET type opamp (TL7x) comes from the leakage of a reverse biased pn
junction, and that tends to double every 10 degrees C or so.  The CMOS input types
(CA3140 et al) do not suffer from this, and so are generally chosen for VCO
integrators.


>
> >-amplifier input bias current (easy to understand,
> >  steals current, will get worse with temperature,
> >  leads to shift of some Hz down, parts with less then
> >  30pA are expensive and sometimes slow)
>
> Yes, increases with temperature are quite dramatic. Watch for specs given
> at just room temperature vs. those given over a range when comparing
> integrators.
>