Searching \ for '[OT]Thermocouple tips?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page:
Search entire site for: 'Thermocouple tips?'.

Exact match. Not showing close matches.
PICList Thread
'[OT]Thermocouple tips?'
1999\12\09@123831 by Dave Johnson

I need to learn to use thermocouples, and am looking for thermocouple
interface designs and/or chips to minimize the effort :-) I'm a
thermocouple newbie, so I'm looking for advice or suggestions from
veterans. I've been doing lots of reading about thermocouples, so I'm
getting a good picture for what's involved, and it's semi-hairy,
electronically speaking.

I've found two very helpful (though expensive) ICs: Analog Devices
AD594/5 (a monolithic cold junction compensator/amplifier package, looks
VERY handy) and Linear Technologies LT1025 (cold junction compensator,
also available as a "kit" with a matched amplifier).

Are there other ICs I should look at? Or should I simply "roll my own"
circuit with real good op amps, etc.? Cost is an issue, but time is even
more an issue, so right now I'm leaning toward the AD594 or something

Any advice/tips/war stories appreciated. Private email encouraged, since
this is really OT.

Dave Johnson

1999\12\09@182633 by Dwayne Reid

>I need to learn to use thermocouples, and am looking for thermocouple
>interface designs and/or chips to minimize the effort

I simply roll my own.  The main question is : what kind of accuracy do you
need?  How do you want to calibrate?  Can you tolerate the inherent
non-linearity of a K type TC?  Its a 7th order polynomial if you want
fractional degree accuracy.

I use a silicon diode to measure cold junction temperature - you wind up
with a large offset that must be subtracted.  You then calibrate gain if needed.

One series of TC transmitters we build (several thousand each year) gets
trimmed for offset during test - we just stuff a 1% resistor as determined
by the test jig.  Gain is not trimmed since I only need 1-2% accuracy.  This
transmitter is based upon the LM10 from Nat Semi - its got fairly low open
loop gain which limits the final accuracy anyways.

Another TC preamp uses software and eeprom to hold both offset and gain
calibration.  Its based upon a 12ce673's 8 bit a/d and uses dither to get 14
bit resolution (accuracy is only good to about 10-11 bits) and 24 bit
integer math.  A large portion of the dynamic range (25%) is eaten up by the
offset - I end up with 1 degree accuracy over the range of 0 - 650 F.  This
one is still under development - I expect the final version to be even more
accurate with some added software filtering.

In both cases (and in all my other pre-amps / transmitters), I use the -2.1
mV / degree C temperature response of a 1n4148 to measure cold junction
temperature.  This is divided down to 40.5 uV / degree C to match the K type
TC that I am using and summed with the TC signal.  My op-amp of choice is
now the AD8551 (brand new and just excellent).  Current designs that use the
LM10 will continue to do so but new designs are all gonna switch to the AD part.

This should get you started - ask more questions as they come up.


Dwayne Reid   <>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax

Celebrating 15 years of Engineering Innovation (1984 - 1999)

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
Do NOT send unsolicited commercial email to this email address.
My posting messages to Usenet neither grants consent to receive
unsolicited commercial email nor is intended to solicit commercial

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...