Measuring Temperature With Resistance: RTDs

August 10, 2017

In the last two publications, we talked about thermocouples.  As you recall, a thermocouple produces an electrical potential that is a function of the temperature gradient across a loop of two dissimilar metals.  Today we are going to talk a little about a different class of temperature transducer whose electrical RESISTANCE changes as a function of temperature.  There are two basic types of this class of temperature transducer; the Thermistor and ‘resistance temperature detector’ (RTD).  Like thermocouples, each of these devices has its
own pros and cons.

 

The basic principle behind the RTD comes from the discovery that the
electrical resistance of a metal changes with temperature.  This
discovery was made by Sir Humphrey Davy in Britain in the 1800’s
while he was experimenting with electricity and, by the way, more or
less inventing electrolysis.  According to Omega Engineering’s ‘The
Temperature Handbook’, this discovery by Davy was announced the same
year (1821) that Thomas Seebeck announced his findings that lead to the
thermocouple.  The year 1821, it seems, was a very important year in
the field of thermoelectricity!

 

What Davy found was that for all metals, electrical resistance
increases with temperature.  Another way of putting this, and indeed
the standard nomenclature for this behavior, is that metals have a
positive temperature coefficient, or PTC. Based on this finding, a new
class of temperature measurement instruments was developed.  The theory
is that if you can measure the resistance of a metallic conductor
placed at a location where temperature is of interest, you can back out
the temperature of that conductor.  Simple, right?  Wrong.

 

RTDs and thermistors have their own challenges.  Here are only a few:

 

1.  Just like with thermocouples where the electrical connection to the instrumentation device caused us difficulties, the electrical
connection of a resistance temperature detector device also introduces measurement problems.  Every wire that is connected to an RTD adds to
the resistance that must be accounted for.  Most measurement systems employ an RTD or thermistor element as one leg of a modified Wheatstone bridge circuit.  This gives additional sensitivity to the device and
allows for compensation of the connection wiring.  In addition to the
problem of wiring, because RTD’s and thermistors are powered devices,
self-heating becomes an issue.  The Omega publication ‘The
Temperature handbook’ has an EXCELLENT discussion on the various
connection options and difficulties with RTDs on pages Z-33 through
Z-36.  I will not attempt to reproduce their excellent work here for
two reasons:  1.  I won’t do NEARLY as good a job.  2.  As I
mentioned last week, you really ought to own this book!  (BTW, It is
free! – http://www.omega.com)

 

2.  In addition to the electrical considerations when deploying RTDs
and thermistors, sensor size will come into play just as it did with
thermocouples.  The RTD or thermistor must be placed in intimate
thermal contact with the UUT.  Convective and radiation losses can
affect the measurement accuracy.  Furthermore, the size of the RTD or
thermistor sensing element will affect the response time of the
measurements.

 

3.  The dependence of resistance on temperature for a metallic
conductor is not linear.  Resistance values must be converted to
temperature by the application of a non-linear polynomial curve-fit.
This requires computational power.

 

Ok, we have a device which, like a thermocouple, requires specialized
equipment.  Why wouldn’t you just use a thermocouple?  You will
remember from our last discussion that a thermocouple is not a terribly
accurate device.  The figure that most people bandy about is that your can get no closer than about 1 degree Celsius with a thermocouple.
RTDs on the other hand, are much more repeatable.  In fact, the
Platinum Resistance Temperature Device (PRTD) is used by NIST as a
temperature measurement standard.  Better repeatability means better
accuracy.

 

We have spent a lot of time talking about RTDs, but have said nothing about thermistors.  Thermistors work in exactly the same way as RTDs, but are typically made of some kind of semiconductor material, and
unlike metals, generally have negative temperature coefficients (NTC).
Furthermore, thermistors are EXTREMELY sensitive to changes in
temperature.  A thermistor might be orders of magnitude more sensitive
to temperature changes than an RTD.  VERY large changes in resistance
are produced with VERY small changes in temperature.  Thermistors are
also highly non-linear, and are useful only when the temperature range
of interest is very small.   Unfortunately, the electrical connection
difficulties are the same for thermistors as they are for RTDs.

 

Bottom line:

If you are interested in measuring temperatures more precisely than
you can with a thermocouple, an RTD might be the answer.  If you must
accurately measure MINUTE changes in temperature over a small
temperature range, a thermistor is probably the answer.  In either
case, however, just like thermocouples, you will require specialized
instrumentation and a good understanding of the device to make good
measurements.  RTD probes are generally more expensive than
thermocouple wire, but usually not cost-prohibitive.  They come in a
range of sizes and types for almost every application.  Thermistors are
usually more expensive still, and also come in a range of sizes and
types.

 

A word of caution!  Whether you are using thermocouples, thermistors or
RTDs, temperature measurements are not ‘easy’ to make.  With the
exception of a glass-bulb thermometer (which still has its place, by
the way), temperature transducers require specialized equipment and
specialized knowledge!  Read the manufacturer’s instructions for any
al all equipment that you buy.   In all cases, a good handbook or
catalog will prove enormously useful.  Once again, Omega Engineering is
pretty much the industry standard when it comes to temperature
measurements.

Share on Facebook
Share on Twitter
Please reload

Featured Posts

High-speed Automatic Frequency Controller for Tomotherapy Machine

June 6, 2013

1/1
Please reload

Recent Projects
Please reload

Follow Us
  • Google Places - Black Circle
  • Facebook - Black Circle
  • Instagram - Black Circle
  • LinkedIn - Black Circle
  • YouTube - Black Circle

info@cythsystems.uk

+44 (0) 845-519-7813

Oxford Centre for Innovation, New Rd,

Oxford, OX1 1BY

United Kingdom

  • Black Google Places Icon
  • Black Facebook Icon
  • Black Twitter Icon
  • Cyth Systems - Instagram
  • Cyth Systems - LinkedIn
  • Cyth Systems - YouTube

Privacy I Cyth Systems, Inc. & Cyth Systems Ltd. All Rights Reserved.

sales@cyth.com

+1-858-537-1960

9939 Via Pasar

San Diego, CA 92126

United States