At 09:56 PM 08/08/2012, you wrote:
>Hi all,
>
>I've put this page together for PIClist
>
>http://www.piclist.com/techref/io/sensor/0-10mA_remote_lm335.htm
>
>This is a circuit I use for a couple of applications and it works fine

Since your immediate problem has been identified, I'll make a
few more comments (welcome or not!):-

I REALLY don't like that 5.1V zener- aside from the temp drift
which you noted, you'll get a huge sensitivity to power supply
voltage- maybe 10 degrees per volt (WAG). A 1% change in the
5.1V is 2.73 degrees change at the output. How about replace it
(or both zeners) with a TL431 and a couple of resistors? It
will be maybe two orders of magnitude better with the same part
count and similar cost.

Another way to improve it without substantial changes would be
to use the 9V to power the 5.1V zener (reduce both resistors).

Another suggested improvement would be to lose the 10K/10K
divider on the LM335 output and double the 2.73V offset, then
adjust the 27K/3K divider to 12K/3K (x5). That will reduce
the influence of offset and offset drift in the summing amplifier
and the offset buffer to half. Not a huge issue since you're
trimming to zero with a pot anyway.

For some op-amps, a bypass cap on the power supply might be prudent.

BTW, you should probably state the range as offset a bit from zero
(like 1=B0C to 50=B0C) since it won't actually get down to 0mV even with
a load resistance to ground.

Care with that quick 'fix' of a 4xx ohm resistor to ground-- it can draw
enough current to pull the 9V supply well out of regulation, which
could lead to  'interesting' behavior modes (at best, not making
it quite to 5V out, at worst oscillating or something like that).


>TIA
>
>Joe


--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.