To repliers -

> ditch the LM385 and put a pair of potential dividers  to reduce the 
> voltage and increase the opamp gain to compensate.

Michael, I hadn't appreciated a problem with CM (must get more
familiar with op amps), and the breadboarded prototype appeared to
work OK. Not until I made a PCB did I notice the meter deflection.
At the time the innards were out of the case so there was no dial for
visual reference

As there is gain available in following stages I could reduce the front
end gain to < 1 and compensate later

> At 10V supply, the common-mode range is 0.2 to 9V minimum.
> This is measured from the negative supply, so max CM voltage in
> circuit is -5+9, or 4 V

Ah, got it

> I saw the part number TLC something and assumed that the original 
> poster had checked that the common-mode voltage was within the 
> operating range of the chip

Dwayne, ditto

> If you have 12 volts available, why not power the opamp from +12 and
> -5 volts? You couldn't use the TLC271 since it's limited to 16 volts max,
> but you could use an LM358 or do the whole circuit in a quad LM324

Bob and Mark, not keen on hacking the PCB to fit dual amps. Reducing
the input voltages to the differential amp is what I'll try first

Thanks
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist