Simon Nield wrote:
..........
>the circuit I was suggesting was:
>
>          v+
>          _|_
>          /|\    < diode!
>           |
>--R1--+----+------- output
>      |   _|_
>      R2  /|\
>      |    |
>------+   v-
>      |
>    0vref
>

Reminds me of an opamp ckt I use for measuring large AC voltages.
Calibrated to +/-50V, but OV tested by pluggin straight into 120VAC
mains:

                    +----82K-----+
                    |   | \      |
                    |   |  \     |
                    |   |   \    |
Vin-----1M---+--+---+---|-   +---+--- Vout
             |  |       |   /
             |  |     +-|+ / LF353
     1N4148 \/  --    | | /  TL082
     (2 ea) --  /\    |
             |  |     |
             +--+-----+
                |
               Gnd

When I first built this ckt, I thought the 1M input R alone
would protect the opmap from destruction by hiV, but was I wrong.
The 2 diodes to gnd shunt the input currents when Vout gets to
the Vcc/Vss rails and the feedback loop blocks up. Works with
Vin with hard gnd reference - not differential.

best regards,
- danM

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics