John N. Power wrote:
> First of all, the maximum input current is 6 mA, which is 30 times
> 0.2mA. This is 1.5 decades. If one decade changes the gain by 5%,
> then the next 1.5 decades changes it 7.5%, not 20 times.

Where did I get 20 times, I don't know, sorry. BUT usual photosensing
App needs four decades. This will give us 20% guess (not specified)
error value. (One PICList member is great when discussing guessed
values - dead fish, right moon phase ...)=20


> Second, this amplifier is not meant for precision measurement of
> low currents. I suggested it only for use in a collision detector.
>The fact that there is a 5% maximum change in mirror gain at the low
> end of the scale does not mean that that rate is maintained over the
> entire range.=20

This could mean much worse situation. By the way, 20uA or 200uA=20
aren't lower end of the range, 0.00002 ua is. 6 mA input current will=20
cause tremendous temperature drift (unspecified).


>           I agree that a 20 times change in gain would be bad. I
> don't believe that it is that bad. The change in mirror gain is only
> given for the range 20uA to 200uA; that indicates that the effect is
> typical of the low current end of the scale.

That indicates nothing for < 20uA or > 200uA. One must suspect
worst case.

John, please don't take my post as an attempt to dig to you,=20
I mean this as usual EE discussion.

Best Regards,=20
Mike.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads