2265 116th Avenue N.E.

, Bellevue, WA 98004
Sales and Customer Support: (425) 453-2345
Finance and Administration: (425) 453-9489
Fax: (425) 453-3199
WWW: http://www.mstarlabs.com/

From the Microstar Laboratories web site

Temperature Measurement
Temperature Measurements using Data
Acquisition Processors
Off-the-shelf temperature measurement and display products are
excellent for what they do. They offer accuracy within a couple of
percent, sometimes better. And they are ready to go with all
necessary functions included. If they do what you want, you
should use them. But if they don't have exactly what you need,
they offer you few options. See how Data Acquisition Processors
can help.

Take a look at some sample multi-channel system drawings.

DAP 4000a System | DAP 5200a System | DAP 5216a System

Choosing Sensors

Sensors detect physical phenomena and convert these to

corresponding electrical signals that are relatively easy to
measure. Sensors that do not produce an electrical signal
naturally are supplemented with electronics to produce the
measurable signal.

Several temperature sensor types are available, each with its own strengths and
weaknesses. To help you make an appropriate selection, here are brief
descriptions of some of the most important temperature sensor types. DAPs are
intended to work with any sort of sensor that delivers a measurable electrical
signal.

www.mstarlabs.com/sensors/temperature.html Page 1
 Calibration

The term calibration is often used incorrectly, to mean applying a

standardized device curve. That assumes that a real device will
conform perfectly to the selected theoretical curve. In practice,
any real device will deviate from published theoretical curves,
which are only statistical composites. A true calibration process
determines how an actual device responds, given inputs
established by a reference standard.

Published standard curves are a good start, and sensor manufacturers try to
match them as close as they can, but actual devices deviate from the ideal.
Check your tolerance requirements. If you don't need best accuracy, keep things
simple, and stay with "typical" curves. But for better accuracy, calibrate. Check
the calibration and linearization page for more information.

Some DAPL processing commands can help you apply sensor calibration for
improved measurement accuracy.

• The INTERP command can represent nonlinear response curves of

arbitrary sensor devices over an extended range.
• The RTD command can apply nominal or calibrated conversion curves for
RTD devices.
• The THERMISTOR command can apply nominal or calibrated conversion
curves for thermistor devices.

Measurement

Temperature sensors tend to fall into two categories: those that

produce a voltage naturally as an indicator of temperature, and
those that produce some other physical response that must be
converted to a voltage for measurement.
Thermocouple

Example

• Natural output potentials.

The signals may be weak or strong. When weak, they are

vulnerable to external interference. When strong, they
typically require a power source to drive them, with RTD Example
potential dangers of power supply noise leaking into the
measurements. For both cases, differential signals help to
keep the measurements clean.
Thermistor

When working with natural voltages, you will typically Example

need:

www.mstarlabs.com/sensors/temperature.html Page 2
 • A SCALE or DAPL expression command to correct for gains and
offsets.
• A linearization command like THERMO or INTERP to compensate
for nonlinearity and calculate the corresponding temperature.

The article Thermocouple Measurement Applications on this site

describes a low-cost, low-speed 8 channel application using
thermocouples. It applies averaging for noise reduction, a conventional
cold junction compensation device, and standard conversion curves for
measurements of moderate accuracy.

Another article on this site covers accurate measurement with

thermocouples, using carefully-calibrated measurements and cold

junction compensation.

• Indirect voltage output.

Excitation power is typically required to convert the response into an

electrical signal.

For example, resistive devices are driven by regulated voltage or current

sources. You will typically need:

• A SCALE or DAPL expression command to correct for gains and

offsets.
• A command like DIVIDER or BRIDGE to determine the device
resistance from observed output voltage.
• A command like RTD or THERMISTOR to convert from a resistance
measurement to the corresponding operating temperature.

An article on this site describes an application using a thermistor in

conjuction with a thermocouple, for very accurate measurements of cold
junction temperature.

Another article on this site discusses an application using an RTD for

measuring moderately high temperatures with moderate accuracy, using
simple calibration procedures.

www.mstarlabs.com/sensors/temperature.html Page 3
 Better Temperature Measurements
for Systems
Temperature Measurements: More, Better, Faster
Here are some of the difficult temperature measurement problems
Data Acquisition Processor (DAP) boards can help with.

Cost | Packaging | Display | Integration | Diversity | Accuracy

| Speed | Noise

Problem: Fully integrated measurement devices are high

cost per channel.

Data Acquisition Processors have a higher initial cost, but

COSTS then a much lower cost per channel, which turns into an
economic advantage when you have lots of channels.
The cost of adding a channel is roughly the cost of the
extra wiring and sensor.
Problem: Independently packaged measurement devices
can fill a rack in a hurry.
PACKAGING Adding channels to a Data Acquisition Processor system
requires only the additional panel space for your
connectors.
Problem: The built-in display that is a strong advantage for
a single signal becomes a disadvantage when you have
lots of channels.
DISPLAY
Data Acquisition Processors can deliver the data for
display in any manner that you wish, including
software-controlled video graphics of your choosing.
Problem: Self-contained units offer limited options for
interconnection.

INTEGRATION Data Acquisition Processors bring measurements directly

into your host system on a high-speed data bus. From
there, you can apply the data management and networking
features of your host system to send data anywhere.

www.mstarlabs.com/sensors/temperature.html Page 4
 Problem: Fully integrated devices work with limited device
types, sometimes only with packages from one
manufacturer.

DIVERSITY Data Acquisition Processors work with any kind of

device that produces an accessible voltage signal. You
can account for the sensor type in your processing, mix
and match, then present results in a common
representation of your choosing.
Problem: Most integrated instruments use generic
"calibrations" that don't match sensors exactly.
ACCURACY DAPs allow true calibration. You can use generic curves
but are not limited to them. DAPs can apply individualized
corrections to each channel for best accuracy.
Problem: Self-contained equipment is usually not intended
to measure, deliver, or display results as fast as sensor
devices can operate.
SPEED Data Acquisition Processors are intended for operation at
rates orders of magnitude faster than the typical sensors.
They can process every channel at maximum rate and
deliver results immediately.
Problem: Filtering to reduce noise, from interference or the
sensors themselves, makes response slow.
NOISE
Data Acquisition Processors can use similar filtering, or a
IMMUNITY statistical approach. For example, measuring 1000 times,
very fast, and computing an average, reduces random
noise to about 3% of its original level.

www.mstarlabs.com/sensors/temperature.html Page 5
 Temperature Sensors Survey
Common Temperature Measurement Sensors
This page surveys the most common temperature measurement
sensor types.

Thermocouples | RTDs | Thermistors | Infrared | Solid State |

Bimetallic

Thermocouple

These sensors depend on differences in charge mobility in two

dissimilar metals to produce a potential difference. The developed
potentials are small, but measurable.
Thermocouple

Thermocouples are rugged, and tolerate a wide range of Example

temperatures, but they have many drawbacks, including low

signal levels, long-term stability, and noise. Their response time
tends to be very slow, but this often results from the packaging
more than the device physics.
Thermocouple

The measurement is actually temperature difference, not an Calibration

absolute temperature level, so a supplementary measurement is

required to establish the reference temperature. The resulting
temperature correction is called cold junction compensation.
Cold Junction

Despite their limitations, thermocouples are used with success Compensation

in a variety of applications. You can obtain moderate accuracy

using standard device curves with no calibration. Careful calibration can
improve accuracy to within a degree C or so.

www.mstarlabs.com/sensors/temperature.html Page 6
 RTD

RTDs, short for "resistive thermal devices," are built basically the
same way as wire-wound or thin-film resistors, but using materials
with relatively high levels of resistivity variation as a function of
temperature. They are generally between thermocouples and
RTD Example
thermistors in terms of speed, ruggedness, signal level and
temperature range.

If not severely stressed, they have good long term stability.

RTD Calibration
They cost more, but offer superior linearity and good accuracy
using standard curves without calibration. With individual device calibration, the
accuracy is even better.

Though most common metals exhibit RTD effects, platinum alloys have the best
range and performance, and are by far the most popular.

Thermistor

Negative temperature coefficient thermistors, the ones commonly

used for temperature measurements, change resistance
dramatically in response to temperature changes, but they have
some limitations.
Thermistor

Example

They are vulnerable to physical damage and chemical

contamination — for example, water can cause problems. They
require careful protective packaging, and this often limits how they
can be used. Response is moderately fast, but typically limited
because of the packaging. The thermal characteristics are highly Thermistor

nonlinear, so you must apply corrections that differ for every Calibration

device type. For full accuracy you must calibrate. They have a
limited range compared to other thermal sensors.

For the temperatures where liquid water can exist, thermistors usually
perform very well.

Infrared

These sensors include special optical and electrical components to detect

infrared blackbody radiation from a very specific location, without direct physical
contact.

www.mstarlabs.com/sensors/temperature.html Page 7
 These sensors are very useful for measuring extreme temperatures through
a viewing port, under conditions that would rapidly destroy other sensor types.

The accuracy, stability, and repeatability are not very good, so they need a lot of
attention.

The receiver/converter electronics scale the signal to convenient levels for

acquisition and recording.

Solid state

These devices use the thermal properties of semiconductor junction voltages to

detect temperature.

These devices are appealing because sensor, power-gain amplifiers, and

"linearization" can be fabricated on the same chip.

The drawbacks are that the operating range of the electronics limits the sensing
range, and any calibration other than offset adjustment is impractical.

These are extremely handy for measuring ambient temperatures around

circuit boards, and popular for "cold junction compensation" in combination with
thermocouples. Accuracy is moderate.

Bimetallic

We mention this one only in passing. A sensor bonded to a bimetallic strip

senses strain in proportion to temperature changes. This is too bulky, slow, and
vulnerable to mechanical and electrical interference for most applications, but
compatible.

www.mstarlabs.com/sensors/temperature.html Page 8
 Thermocouple Application
Fast, Accurate Thermocouple Measurement
This application example is based on a commercial HTST (High
Temperature, Short Time) pasteurization processing for milk. We
will cover only the temperature measurement aspects. The basic
HTST process must do the following:

1. Heat the milk quickly from a holding temperature to the

required temperature for destroying microbes and harmful
enzymes (for example 72 degrees C).
2. Holding the milk at this temperature, pass it through a Temperature

length of tubing at a closely controlled flow rate. Measurement

3. The milk exits the process at the far end of the tubing,
after the required time interval (for example 22 seconds).
4. If the milk is still at the required temperature at the exit
point, it is allowed to continue on for further processing.
Temperature

Sensors

Thermocouple

Calibration

Cold Junction

Compensation

Figure 1 - HTST pasteurization process

The processing is temperature critical. If the temperature is too

low, there are public safety and product quality concerns. If the temperature is
too high, energy is wasted and the product can be damaged. If instruments
respond too slowly, some inadequately treated milk could pass the exit valve
before a problem is detected.

The basic application requirements for the temperature measurement are:

1. Operating range of 55 degrees to 85 degrees C.

www.mstarlabs.com/sensors/temperature.html Page 9
 2. No physical damage from out-of-range temperatures.
3. Total measurement accuracy +- 0.5 degrees C.
4. Long term drift +- 0.2 degrees C.
5. Maximum settling time 4 seconds.

Selecting a sensor
A high quality type T thermocouple probe is selected. Thermistors have good
response characteristics but are vulnerable to moisture and damage. RTDs have
better stability and linearity, but slow response times — perhaps this is because
the entire RTD has to settle to thermal equilibrium before a reading is accurate,
whereas only a thermocouple probe tip needs to settle to temperature.

High quality type T thermocouple probes can be calibrated to an accuracy of

about 0.1 degrees C. These use very fine gage wires that can reach a thermal
equilibrium quickly. To prevent chemical interaction with the milk products there
must be a protective cover. This will degrade the thermal response rate
significantly, but if the resulting time constant remains under 0.5 seconds, the
settling time requirement is met easily.

The main challenge with a thermocouple sensor is the very low signal level,
difficult to isolate from background noise.

Thermocouple cold junctions

Thermocouples have a fundamental limitation that they measure a temperature
difference rather than an absolute temperature. To know the temperature of the
milk, it is necessary to know the reference temperature at the thermocouple cold
junction. The usual approach of measuring the cold junction with a solid-state
device located on a termination board near the cold junction terminals is not
satisfactory for this application, because the solid-state sensors have a baseline
error of +- 0.5 degrees C, allowing no room for device calibration errors,
measurement noise, or any other error source.

Figure 2 - Thermocouple measurement without reference junction

www.mstarlabs.com/sensors/temperature.html Page 10
 The issue of how to obtain an accurate cold junction measurement is covered in
a separate note. We will merely presume here that there is a separate
temperature measurement involved, and that the cold junction absolute
temperature is known within +- 0.1 degrees C.

Signal conditioning and sampling

An MSXB065 differential signal
conditioning and filtering board
amplifies the low-level thermocouple
signal, applying a gain of 25 and
raising the signal range to
approximately 0.1 volts. Despite
best efforts to shield the signals
from noise, some interfering noise
will leak in and add to the inherently
noisy readings from the Figure 3 - MSXB 065 signal conditioning
thermocouple. A fourth order board
Butterworth filter tuned to a 500 Hz
cutoff will remove all high frequency random noise from the measurements.

Applications with less stringent accuracy and speed requirements can route
thermocouples directly to a termination board, with an onboard CJC circuit, and
breadboarding space for termination networks.

The excitation and signal lines required for the thermocouple and supplementary
cold junction sensors are provided through the 37-in panel connector of the
MSXB 065. A balanced termination network on the board provides a ground
reference voltage so that the thermocouple junction does not "float" to a high
common mode voltage.

We will choose a DAP 5000a for the

sampling and data processing. A
DAP 5000a provides good support
for numerical filtering operations,
plus processing capacity for
additional monitoring functions not
covered here.

With less stringent measurement

requirements, you would probably Figure 4 - DAP 5000a board
choose a DAP 840.

www.mstarlabs.com/sensors/temperature.html Page 11
 For supporting more channels and more complex calibration curves, you would
probably choose a DAP 5200a.

The 100 millivolt signal level is still too small to digitize with good precision.
Configure the input sampling of the Data Acquisition Processor to apply an
additional gain of 40. This will raise the signal range to about 4 volts, within the
usual +/- 5 volt input range used by most Data Acquisition Processor
configurations. In the input configuration, your input channel definition will look
something like the following:

SET IP1 D1 40

Over the temperature range of 55 degrees to 85 degrees, the response levels of

the thermocouple range from 2.251 to 3.385 millivolts, so the range of variations
seen by the converters are 2.251 to 3.385 volts. This is 1.134 volts out of the full
10 volt span — 1858 converter counts out of the 16384 counts available with a
14-bit converter. That gives a temperature resolution of 30/1858 = 0.02 degrees.
If we presume that the last two bits of the converter are compromised by
converter chatter and nonlinearity, that still yields measurement accuracy better
than 0.08 degrees C.

Processing and digital filtering

Sampling at 1500 samples per second guarantees that any signal passing
through the 500 Hz lowpass filter is accurately represented, including
lower-frequency noise. We can use the following command to smooth the signal
and reduce the sampling rate to a more friendly 150 samples per second for
processing.

FIRLOWPASS(IP1, 10, PREDUCED)

After filtering, all noise beyond 75% of the Nyquist frequency (75 samples per
second x 0.75) are gone. In particular, all 60 Hz interference from power sources
is gone.

The 150 samples per second can still exhibit residual random noise in the 10 to
50 Hz range. We can apply digital filtering to remove most of that, with care. The
lower the cutoff frequency, the more time delay in producing results. The 10 Hz
cutoff is selected to keep this filtering delay relatively small. We can apply a 4th
order inverse Chebyshev filter with 46 dB stopband rejection to remove the
higher frequencies sharply while preserving the low-frequency temperature
variations accurately. 10 Hz is 0.1333 of the Nyquist frequency of 75 Hz, so we
specify the following digital filter.

www.mstarlabs.com/sensors/temperature.html Page 12
 CHEBYINV(PREDUCED, 4, 0.1333, 0.005, PFILT)

Calibration and conversions

The data stream PFILT above will have clean measurements of thermocouple
junction potential, but these results must be converted into the desired
temperature measurements. We know that for a precision type-T thermocouple,
the response is almost linear in the limited operating range. If we measure the
thermocouple's output potential at the limits of the operating range (55 degrees
and 85 degrees C, call these measured values POT55 and POT85), the following
DAPL expression can convert the measured junction potential to the
corresponding temperature anywhere in the range.

LINTEMP = \

((POT85-PFILT)*55.0 + (PFILT-POT55)*85.0)) \

/ (POT85-POT55)

The measurements used for the calibration

include any effects from offsets and gain DAPL expressions look like
deviations in the amplifiers. So in additional to simple arithmetic — but
accounting for the thermocouple response, they actually define
the calibration above also covers the complete processing tasks
measurement hardware. The usual approach that apply conversion
of applying standard curves does not take operations to your entire
measurement errors and sensor variations data stream. Learn more.
into account.

Unfortunately, the simple linear mapping as described above isn't completely

satisfactory. From the standard type-T characteristic curve, we can learn that
there is a maximum deviation of about -0.20 degrees from a linear curve in this
range. We can easily correct for this effect, using an additional DAPL expression.

TADJUSTED = LINTEMP + ( -0.20 * 4.0 * \

(PFILT-POT55)/(POT85-POT55) * \

(POT85-PFILT)/(POT85-POT55) )

You can verify that the second-order correction is zero at the ends of the ranges,
and equal to the maximum -0.20 at the center of the range. The residual error in
the corrected temperature reading is just a few hundredths of a degree. Thus,
conversion errors are almost insignificant. Total system error depends almost
exclusively on the quality of the signal and calibration.

The final temperature measurements are available 150 times per second, with a
delay of a few milliseconds after receiving each temperature update. Total

www.mstarlabs.com/sensors/temperature.html Page 13
 response delay depends almost exclusively on the temperature probe time
constant.

Parts list (parts for thermocouple termination and cold junction

measurement not included)

DAP
Analog Cable
Half IE with backplane
MSXB065 signal conditioning board

www.mstarlabs.com/sensors/temperature.html Page 14
 RTD Application
RTD Measurements for Roaster Control
Producing a rich, aromatic cup of java requires a blend of art,
science, and good coffee beans. A critical step is the roasting of
the beans. There are hundreds of complex organic chemicals that
produce the flavors and aromas. Roasting brings out some of
these, and produces others, in a manner highly dependent on
timing and temperatures. Good roasting techniques can produce
quite acceptable results from relatively low-grade beans; bad
roasting can produce quite dreadful results from the best
high-grade beans. Temperature

Measurement

Profile roasting
Green coffee beans are placed into a drum or chamber, where
they are agitated or tumbled as they are roasted by hot air. There Temperature

Sensors

are four controllable variables:

1. temperature of the hot air

2. air flow rate
3. temperature of the beans RTD Calibration
4. stir rate

The most successful and repeatable control strategy is called

profile control. Rather than trying to hold specific temperatures,
the temperature targets are treated as a continuously changing
trajectory. The amount of heating at "each level" depends on the
rates of temperature change.

Temperature sensing
The critical variable is the temperature of the beans. Other variables can help to
improve the control of the critical variable. The heated air is easier to measure,
but it can differ from the bean temperature by as much as 200 degrees F, so it is
not a sufficient indicator by itself.

A thermal probe measuring the temperature within the coffee beans provides the
most important feedback. The probe must withstand the pounding as beans are
tumbled. Because of its rugged sheath, this kind of probe will not respond

www.mstarlabs.com/sensors/temperature.html Page 15
 quickly, but the temperature profile will not change very fast either, and a
reasonable balance is maintained.

Either a thermocouple or RTD probe could work for this application. We will
select an RTD probe because it is reliable, accurate, stable, easy to use, and
well suited for an operating temperature range of 100 to 500 degrees F.

RTD electrical connections

RTDs are often driven by specialized, regulated current source devices, but most
of the time all you really need is an accurately-regulated voltage source and
accurately-measured load resistor. Connect them in a voltage divider
configuration.

Figure 1 - RTD in simple divider network

The voltage across the RTD changes roughly in proportion to the resistance
change of the RTD, which changes roughly in proportion to changes in
temperature. The nonlinearities are simple and can be computed exactly, to yield
accurate measurements.

Selecting a platinum RTD probe with a 500-ohm nominal resistance (PT500

characteristic), its alpha approximation (usually given as 0.00385 per-unit per
degree C) tells us that the resistance changes from about 570 ohms at 100
degrees F, to 1000 ohms at 500 degrees F. If we select a series load resistance
of 12K and excite the network with a +5.0 volt source, the voltage across the
RTD will vary from about 0.2 to 0.4 volts. Amplify by a factor of 10 to get the
range 2.0 to 4.0 volts, for excellent measurement resolution.

www.mstarlabs.com/sensors/temperature.html Page 16
 Accuracy of the measurement will be affected by tolerances in the resistance
values:

1. The load resistor value can vary from nominal.

Measure the actual resistance value with an accurate ohmmeter.

2. The RTD base resistance can vary from nominal.

Measure the actual RTD resistance and its temperature as accurately as
you can. Then calculate the RTD base resistance at the reference
temperature using the standard device curve. For a standard Pt500
device:
Tf = measured temperature in degrees F

Rmeas = measured RTD resistance in ohms

Tc = (Tf - 32) * 5/9

Rbase = Rmeas / (1.0 + 0.003908 * Tc - 5.775e-7 * Tc
2)

A DAP board's voltage reference can supply the 1/2 milliampere of current
required to power the resistive network, while holding the reference level of +5.0
volts closely. If properly calibrated, the DAP will have negligible offset and gain
errors.

Processing the measurements

The following shows how the calibration and conversion data are applied in the
DAPL system.

CONSTANT RTDbase float = 500.2

CONSTANT LOADohms float = 12088

CONSTANT Vexcite float = 5.0

VECTOR TypePtRTD float = ( 0.0,0.003908,-5.775E-7,0,0,0,0 )

The input declaration sets the amplifier gain to 10 for measuring the RTD voltage
(and some other temperature measurements). The measurement rate is
established by setting the time interval in microseconds between sampling
events. The rate of data arrival can be used as a timing base for driving all
automatic processing.

IDEFINE measurement

CHANNELS 4

SET IP0 S0 10 // RTD voltage

SET IP1 S1 10 // heated air

SET IP2 S2 10 // environment

SET IP3 S3 // humidity

...

TIME ...

www.mstarlabs.com/sensors/temperature.html Page 17
 The processing includes commands that convert the captured voltage
measurements into the corresponding temperature measurements. Two of these
are DAPL expression tasks, and two are specialized downloadable commands
available on this site.

PDEFINE processing

// Correct for amplifier gain

RTDvolts = IP0 / 10.0

// Compute RTD resistance

DIVIDER( RTDvolts, LOADohms, Vexcite, RTDohms )

// Convert RTD resistance to equivalent temperature

RTD( RTDohms, RTDbase, TypePtRTD, TEMPRdegc )

// Convert temperature to Fahrenheit

TEMPRdegf = (9*TEMPRdegc/5)+32

// Perform control algorithms here...

// Select data for host displays here...

END

More than just measurements

Clearly, if monitoring one temperature were the complete story, a DAP system
would provide more capability than you need. But an automated roasting process
needs many other things to be successful. In addition to the basic
measurements, DAPs can support the following:

• Supplementary measurement channels

• Regulated sources
• Trajectory generation
• Device outputs
• Interactive control algorithms
• Local process monitoring
• Remote data display and data logging

For supporting integrated measurement and control activity on multiple channels,

the ability to incorporate RTD measurements along with other data capture and
processing can be a real advantage. With most of the low-level details already
covered, higher-level systems can concentrate on use interactions, profile
configuration, and data management.

www.mstarlabs.com/sensors/temperature.html Page 18
 Thermistor Application
Precise Measurement Using a Thermistor
This page describes an application using an individually-calibrated
thermistor to obtain highly accurate measurements of ambient
temperature. This is an important side-problem for the HTST
pasteurization application on this site: specifically, the temperature
measurement of the thermocouple sensor used by that system
does not completely determine the process temperature. The cold
junction temperature must be measured accurately as well. Many
of the application configuration issues are predetermined to satisfy
the requirements of that system. Better than 0.1 degree C Temperature

accuracy is required for the cold junction measurement in order to Measurement

meet the stringent total measurement error requirements.

A sensor is needed that can measure absolute temperature with

good stability and accuracy. Within the thermally protected
Temperature

environment where the thermocouple is terminated: Sensors

• Temperatures changes are small.

• Temperatures changes occur very slowly.
• The measuring device is subjected to minimal operating
stress. Thermistor

Calibration

This is a perfect place to take advantage of a thermistor's

excellent responsiveness and accuracy.

Hardware details
The pasteurization application uses an MSXB 065 signal conditioning and
filtering board. The application reserves one of the channels on this board for
cold junction measurements.

The MSXB065 and the thermocouple

termination boards reside in an MSIE001
half-size standard rack enclosure. One slot is The boards in the Channel
used by the interface board connecting the Architecture conform to
Data Acquisition Processor (DAP) board to Eurocard formats, so you
the analog backplane. One slot is used by the can easily obtain boards
MSXB 065. There is at least one slot for a that fit within the enclosure
thermal barrier, and then a slot used by a and can meet EMC
custom thermocouple termination board. The emissions standards.

www.mstarlabs.com/sensors/temperature.html Page 19
 custom termination board provides a thermally stable environment, where the
temperature can be measured accurately at the thermocouple cold junction
terminals.

A cable must be built to connect the MSXB 065 to the thermocouple termination
board.

Figure 1 - cable between boards

The thermocouple termination board is basically just a set of connectors. On the

outside of the end panel, there are two connectors: a 37-pin D connector
attaches to the signal conditioning board, and a thermocouple connector
attaches to the thermocouple wires and their shield. On the inside, the board is
blank and supports a terminal strip.

It is important for the thermistor to be mounted physically as close to the

thermocouple wire termination as possible to avoid a temperature offset. One
way to do this is to physically bond the thermistor body to the connector on the
inside of the panel. Route the thermistor leads to connect at the terminal strip.

Beyond the point where the thermocouple wires terminate, use well-balanced
conductors to connect to the terminal strip. On the other side of the terminal strip,
connect to the excitation source and the differential measurement pins from the
signal conditioning connector. The following diagram summarizes the electrical
connections.

www.mstarlabs.com/sensors/temperature.html Page 20
 Figure 3 - thermistor connections on termination board

Thermistor characteristic calibration

The MSXB065 board is configured for gains of 25 to detect very small
thermocouple potentials. After applying the gain, the results must remain under
+5 volts to be within the range of the Data Acquisition Processor. That means,
5/25 = 0.2 volts is a reasonable upper bound on the voltage across the
thermistor when its resistance is maximum.

We pick a thermistor with a nominal value of about 1K at 25 degrees C. It

happens to have a maximum value of about 2K in the operating range. To stay
below 0.2 volts when the excitation voltage is +5 volts, we need a load resistor of
about 50K. Choose standard value 56K for the loading resistor. Using a precision
ohmmeter, determine the actual value. We will assume a measured value of
56.18K.

Assume an ambient temperature range 10 to 40 degrees C. The operating

environment is unlikely to vary by this much, because milk must be refrigerated
most of the time, and regulating the ambient temperature helps regulate the
storage temperature. Even though this is a restricted range, the thermistor
characteristic is nonlinear over this range and we need a correspondingly
nonlinear characteristic to evaluate its response. See the thermistor calibration
page for more information about how to do a simple three-point calibration. Here
is the calibration data set. We will use the first, center, and last terms for an exact
fit and evaluate the approximations at the other two points to check the
interpolation accuracy.

www.mstarlabs.com/sensors/temperature.html Page 21
 TemperatureObserved voltageResistance
(degrees C) (gain 25X) (Ohms)
10 4.274 1989
18 2.974 1369
25 2.190 1002
32 1.625 740
40 1.175 533
Steinhart-Hart Equation Three-Point Fit

a = 1.6901e-3 b = 2.3284e-4 c = 1.6663e-7

Verifying at midpoints:

for R = 1369 --> T = 18.01 (compare: 18.00)

for R = 740 --> T = 32.05 (compare: 32.00)

The model is able to fit the data to better than 0.1 degree, which makes it
possible to keep the total measurement error within 0.5 degrees.

Excitation for the thermistor

Thermistors are not self-powering. An excitation is required to force a current
through it, and the resulting voltage tells us the resistance value.

The MSXB 065 does not have a reference voltage source in the ordinary sense.
It does, however, have an alternative that is almost as good. When jumpers are
installed at position J8 (this is the default configuration) the outputs of the DAC
(digital-to-analog converter) devices on the DAP are routed through the
MSXB 065 to pins on its 37-pin "input connector." (This makes the term "input
connector" a little awkward when these signals are outputs!) We only need one of
these signals to use for the excitation. There are a few minor hazards.

1. When you use a DAC device as a voltage source in this way, without any
means to latch and hold the signal level at the receiving end, the DAC
output must remain undisturbed. For example, if you use a DACOUT
processing command that reads its value from a variable, the value of
that variable must never be changed.
2. There is limited drive current. With loading of over 50K ohms driven by a
voltage of +5 volts or less, there is about 0.1 mA, well below this limit.
3. Special configuration is required to set up the DAC to operate this way.

Excitation source configuration

A digitizer count of 32768 corresponds to the full input range of +5.0 volts (so
numerically reachable output levels are one tick short of this). Suppose we pick a

www.mstarlabs.com/sensors/temperature.html Page 22
 4.0 volt excitation, below the maximum range, to drive the loaded thermistor.
Then the following calculates the desired digitized excitation level.

(4.0 / 5.0) x 32768 = 26214 counts

In the DAPL configuration, declare a source pipe to deliver this value to the DAC.
Insert the data into that pipe.

PIPE EXCITATION WORD

FILL EXCITATION 26214

In the processing section, define a task that moves this value to the selected
output DAC.

DACOUT( EXCITATION, 0 )

When this task begins to execute, it will move the specified output level to DAC0.
After this, the DACOUT task will be starved for new data and unable to change the
output DAC further. The DACOUT task goes to sleep permanently, and does not
get in the way of other processing.

While the converter electronics are very precise, you cannot be sure that small
offsets or losses aren't affecting the voltage actually appearing across the
thermistor and loading resistance network. You can improve accuracy by
measuring the actual voltage with an accurate high-impedance voltmeter. If
necessary, adjust the value loaded into the EXCITATION pipe, upward or
downward (each level change of 16 adjusts the 12-bit converter output by one
"quantum" on the DAP5200a). Adjust until the measured value is as close to a
perfect 4.000 volts as possible.

Filtering on the CJC channel

Filtering is not strictly necessary for measuring the thermistor voltage on the CJC
channel. With the high sensitivity of the thermistor and the highly protected
environment, it experiences little noise disturbance. However, it remains true that
any signal component above 500 Hz cannot possibly be useful cold junction
temperature information. Hardware filtering is used for the noisy thermocouple
measurements, and there is no harm in filtering the less-noisy thermistor
measurements the same way.

www.mstarlabs.com/sensors/temperature.html Page 23
 Figure 3 - Filtering channels on a MSXB065 board

This application only uses one thermocouple channel, but it could be expanded
to measure additional channels. As long as all cold junctions are maintained at
the same temperature, you only need one cold junction measurement.

Configuring the measurements

In the HTST pasteurization application, the input channel IP1 is used for
measuring the thermocouple, while channel IP0 is reserved for measuring the
CJC thermistor. No additional gain is required for the CJC measurement, so its
gain level is set to 1 for acquisition. Its samples alternate with the samples of the
thermocouple. For 1500 samples per second on each channel, the time interval
between sampling events is 333.33 microseconds.

SET IP0 D0 1 // CJC thermistor

SET IP1 D1 40 // thermocouple

TIME 333.3

www.mstarlabs.com/sensors/temperature.html Page 24
 The DAP 5000a processing capacity is useful
for independently processing the CJC
temperature readings in parallel with the You can export the
thermocouple readings. Changes in the cold configurations you develop
junction temperature will be very slow, so with DAPstudio software,
simple averaging that reduces each set of 10 and run them on your DAP
thermistor measurements to one averaged independent of DAPstudio
measurement will do the trick. Multi-tasking is or other GUI packages.
automatic in the DAPL system.

// Noise and rate reduction by filtering

AVERAGE(IP0, 10, PAMBIENT) // cold junction

FIRLOWPASS(IP1, 10, PREDUCED) // thermocouple

Converting measured voltage to resistance

First, you need to know the thermistor resistance. To determine this, you need to
convert the measured voltage from scaled converter ticks into physical voltage
units. The range limit 32768 corresponds to 5.0 volts. The gain of 25 applied by
MSXB065 board amplifiers must also be compensated. The following conversion
can be used.

VTHERM = PAMBIENT * (5.0 / 32768) / 25

Now, use the following voltage divider formula to combine the measured voltage
VTHERM, the known 4.0 volt excitation level, and the known loading resistance
56.18K, yielding the value of the thermistor resistance in ohms, to high accuracy.

RTHERM = 56180.0 * VTHERM / (4.0 - VTHERM)

Converting resistance to temperature

Each time a thermistor measurement value is received, it is converted to the
equivalent temperature using the calibrated response curve. Probably the easiest
way to do this is the THERMISTOR command, available on this site. Make sure
that the SENSORM module is installed on your system. The conversion
coefficients are specified in a DAPL vector.

VECTOR CALIB0 FLOAT = \

(0, 1.6901e-3, 2.3284e-4, 1.6663e-7)

www.mstarlabs.com/sensors/temperature.html Page 25
 Later, in the processing section, specify this
device model as you convert the
measurements into temperatures. Another useful way to apply
device-specific calibrations
is the DAPL system's
SCALE command.

THERMISTOR(RTHERM, CALIB0, PCOLDJUNC)

To complete the processing, combine the cold junction temperature result with
the temperature difference result from the thermocouple.

TADJUSTED = ... // apply thermocouple calibration

TEMPERATURE = TADJUSTED + PCOLDJUNC

Now you can route the stream of TEMPERATURE measurements to any location
you wish, test the range and signal alarms in real time, collect statistics for data
logging, etc.

www.mstarlabs.com/sensors/temperature.html Page 26
 Calibrate Temperature Sensors

What is sensor calibration?

When using temperature sensors, you are actually measuring a
voltage, and relating that to what the operating temperature of the
sensor must be. If you can avoid errors in the voltage
measurements, and represent the relationship between voltage
and temperature more accurately, you can get better temperature
readings. How much effort is worthwhile depends on the
application's error tolerances.

There are four adjustments that a good calibration can provide. Temperature

Measurement

• OFFSET
All voltages are measured with respect to a reference. All
devices operate at some operating voltage. Any
displacements in these voltages, or any consistent errors
Temperature

during measurement, will produce consistent errors that Sensors

affect all measurements. Offset corrections make these

errors as small as possible.

• GAIN
The voltage that you measure is not really the voltage Thermocouple

present on the sensor device. Amplifiers and attenuation Calibration

between the sensor and the digitizing converter change

the signal level. To recover the sensor information, you
must restore the data to the original level accurately.
Uncorrected gain errors tend to produce measurement RTD Calibration
errors that change consistently across the operating
range.

• LINEARIZATION
The relationship between measured voltage and sensed Thermistor

temperature is in general nonlinear and dependent on the Calibration

physical properties of each sensor type. Over a limited

range, a simple linear function is often a sufficient approximation, but a
more complicated curve is necessary to describe the relationship
accurately. Generalized curves defined by standards help, but they won't
match any individual device perfectly. For best accuracy, you need to
calibrate, and adjust the coefficient values of the conversion function.

www.mstarlabs.com/sensors/temperature.html Page 27
 After applying the offset, gain, and linearization corrections, the results might not
be in the most useful form. A good follow-up step is the following:

• UNIT SCALING
Convert the results to a common and useful representation. For example,
present all temperature measurements in degrees C.

Calibrating linearization curves

Calibration is a process of aligning what your formulas say with what real devices
actually do. This involves taking some accurate measurements.

You can't control what a sensor does directly. It responds based on its physical
properties. But you can to some extent control what your sensor measures. You
can establish a set of temperature levels that span the operating range, and
measure those levels with a laboratory-grade temperature standard. For each of
those measured temperature points, observe the response level of the sensor. If
you construct a curve that passes through these points, you will have a very
good calibration specialized for the individual sensor.

Given the temperature vs. voltage data set, treat the sensor readings as noisy
input values, and the temperature measurements as the corresponding output
values to be produced. Taking multiple measurements and averaging them helps
to obtain the best possible data quality for calibration.

From here, there are two ways that you can go.

1. Explicit function. Use a processing command that applies a "calibrated

conversion function" for each measurement. This can be a very generic
calculation like a DAPL expression or a GENPOLY command, or it can be
a specialized conversion such as the RTD command.

2. Piece-wise linear approximation. Select a set of representative points

along the curve and approximate the curve with straight lines "point to
point." Code the "breakpoints" as vectors and supply them to the DAPL
command INTERP, which can then locate the appropriate line section
and evaluate the function at intermediate points.

For more information about the linearization curves most commonly used for
temperature sensor calibration, check the following.

• Calibrate thermocouples
• Calibrate RTDs
• Calibrate thermistors

www.mstarlabs.com/sensors/temperature.html Page 28
 Calibrating Thermocouple Sensors

Calibrate Thermocouples
The potential produced by a thermocouple as a function of
temperature difference between its two ends is very nearly linear,
over a very wide range. The following curve is the "standard
response curve" for a type-K thermocouple.

Temperature

Measurement

Calibration

Overview

Thermocouple

Example

The deviations from perfect linearity are significant, however. If

you compare the curve to a best fit line, you will find the following
differences over the full temperature range.

Cold Junction

Compensation

www.mstarlabs.com/sensors/temperature.html Page 29
 The differences are significant enough that most applications need a correction
to obtain acceptable measurement accuracy. For moderate accuracy, you can
assume that standard device curves match your actual thermocouple exactly,
and use the standard response curves. This can yield a measurement accuracy
within a few degrees. If you calibrate, you obtain maximum measurement
accuracy for your individual thermocouple device, and also correct for any offset
and gain errors present in your measurement system.

Linearization Functions for Thermocouples

The most common way to linearize the thermocouple response is to use a

relatively high-order polynomial function. The function is dominated by the linear
terms, but higher-order terms add the local curvature for a better fit. Standard
linearizing curves are constructed in this way.

Individual thermocouples are very stable and repeatable, even if they do not
match the standard curves exactly. Over a limited temperature range, it is
possible to reduce the measurement error to about 1 degrees C using a
well-calibrated lower-order polynomial curve.

The nonlinearity in thermocouple characteristic curves is almost unmeasurable

over a bounded temperature range. Piece-wise linear approximations are a good
alternative to the explicit function models. The DAPL system's THERMO command
uses the piecewise curve strategy, matching the standard curves within a fraction
of a degree.

Calibrating a polynomial model

You will need to determine the order of polynomial to use. For a reduced
temperature range (a few hundred degrees) a third or fourth order polynomial fit
should provide sufficient conversion accuracy. For example, the following
third-order polynomial matches the standard curves for a type-K thermocouple

www.mstarlabs.com/sensors/temperature.html Page 30
 with error less than 0.1 degrees C through the range -45 degrees C to +150
degrees C.

T = -0.01897 + 25.41881 V - 0.42456 V2 + 0.04365 V3

where V is voltage in units of millivolts

and T is temperature in degrees C

You can modify the coefficients slightly to make the conversion curve match an
individual device's response for best accuracy. The calibration data must be
captured very carefully, but the process is straightforward.

1. Establish stable temperatures throughout the full operating range that you
intend to cover.
2. Measure each temperature carefully using an accurate measurement
standard.
3. Take many measurements of the noisy thermocouple potential at each
temperature point, averaging these measurements to reduce noise.
4. Collect the temperature-potential data pairs, and apply a least-squares
curve fitting analysis to obtain a polynomial that best predicts the
temperature given the junction potential.

Calibrating a piece-wise linear model

It typically takes about 20 linear pieces to adequately approximate thermocouple
standard curves. For better accuracy, more pieces should be used. Assume that
the number of sections is increased to 40, then reduce the number of pieces in
proportion to the fraction of the total range your application needs. So for
example, a type-K thermocouple has a range of about 1600 degrees C. A
temperature range of 200 degrees C is 1/8 of that, so a calibrated model with
about 40/8 = 5 pieces should produce good conversion accuracy.

Knowing that the response will be nearly linear for each of the pieces, you don't
need to measure everywhere as you do with a polynomial fit. For an N-piece
model, measure at the two ends of the temperature range, and also at N-1
roughly evenly-spaced intermediate points. Use the same precautions as with a
polynomial fit, allowing temperatures to stabilize, and taking multiple
measurements at each selected point for noise reduction. You can then code the
monotonically-increasing voltage terms as a vector of input x-values, and the
corresponding temperature levels as a vector of y-values. Provide these two
vectors to the INTERP command to evaluate your piecewise curve at any point.

www.mstarlabs.com/sensors/temperature.html Page 31
 Calibrating cold junction effects
You can read more about the principles and application of cold junction
compensation in another page on this site.

There are some practical limitations on the accuracy you can achieve when using
the LT1025A temperature measurement devices available on MSTB009
termination boards, and optionally available on MSXB037 analog expansion
boards.

1. The baseline temperature measurement error of the chip is 0.5 degrees

C.
2. Even when you bring the thermocouple wires all the way to termination
board terminals, physically close to the location of the temperature
measurement chip, it is difficult to maintain perfectly uniform
temperatures across the termination board. You are doing very well if you
can hold this error to an additional 0.5 degrees C.
3. Calibrating the sensor chips would require controlling the board
environment to a number of stable and accurately measured operating
temperatures. This is probably infeasible.

The bottom line: any errors in reading the cold junction temperature will
contribute directly to errors in the final temperature measurement. Expect at least
1 degree error from cold junction measurements when using the on-board
temperature memeasurement circuits.

For better accuracy, you will need better control and calibration of your cold
junction measurements. An application using a separate calibrated temperature
sensor for best cold junction measurement accuracy is described in an
application note on this site.

Thermocouple applications that do not calibrate cannot do much better than

the THERMO command provided by the DAPL system. For evaluating
customized piecewise-linear curves, you can use the INTERP command,
also provided by the DAPL system. For evaluating polynomial models, you
can use the generic polynomial evaluator command GENPOLY available on
this site. Or for a special case that also covers cold junction compensation,
you can use the THERMOPOLY command available on this site.

www.mstarlabs.com/sensors/temperature.html Page 32
 Cold junction compensation
Thermocouple Cold Junctions
The terms hot junction and cold junction, as applied to
thermocouple devices, are mostly historical. You don't need to
have any junctions to get thermocouple effects. If you heat one
end of a metal conductor and hold the other end at a constant
reference temperature, two important things occur.

1. Heat flow. There is a thermal gradient, so heat flows from

the hot end to the cold end. With small-gage thermocouple
wire, very little thermal energy actually reaches the cold Temperature

end, and the thermal gradient is typically not constant Measurement

along the wires because of heat loss.

2. Seebeck effect. Energetic electrons at the hot end diffuse

toward the cold end, pushing less energetic electrons
Calibration

along with them, resulting in a higher static potential at the Overview

hot end relative to the cold end. The larger the

temperature gradient, the larger the potential difference.
(There are additional contributing effects when dissimilar
materials are joined.)
Thermocouple

Example

In practice, it is difficult to measure the Seebeck effect directly.

When you attach measurement probes, there is a thermal
difference across the probe leads, producing additional
thermocouple effects that interfere with the measurements.
Thermocouple

Classical thermocouple loop configuration Calibration

To make the thermal effects measurable, two different metal

conductors are used. They must be chemically, electrically, and
physically compatible. They produce different electric potentials
when subjected to the same thermal gradient.

In the classical configuration, the dissimilar thermocouple wires are welded

together at the measurement end (hot junction), and again at the reference end
(cold junction), forming a loop. The hot junction assures that the potential at that
point matches in the two metals. Immersing the reference-end junction in an
ice-water slurry assures that the temperature gradients are the same across both
materials. The ice-water slurry establishes a reference temperature at 0 degrees
C.

www.mstarlabs.com/sensors/temperature.html Page 33
 Figure 1 - dissimilar metals form loop, two junctions

Welding the thermocouple wires at the cold junction also equalizes the potentials
there. To make the potential difference observable again, it is necessary to break
the loop. Pick a location in one of the thermocouple wires where the temperature
matches the temperature of the measurement leads. Break the loop there, and
attach matching leads to the two sides of the gap to measure the potential.

Figure 2 - potential difference where loop is broken

• By maintaining uniform temperatures where the leads connect, thermal

gradients are unaffected.
• By avoiding thermal gradients across lead wires, stray thermocouple
effects are kept small.
• By matching the leads well, any residual effects cancel out of differential
measurements.

Cold junction in practice

Maintaining an ice water slurry and actual cold junction is rarely feasible.
Typically, the cold junction is omitted, and the potential is measured directly
across the two terminal ends of the thermocouple wires at ambient temperature.
For historical reasons, we speak of the terminal ends of the thermocouple wires
as the cold junction, despite the fact that there is no longer an intentional
junction. (For the same historical reasons, we refer to the measurement junction
of the thermocouple as the hot junction even if it is used to measure below-zero
temperatures.) The measured potential indicates the temperature difference
between the hot junction point and the unknown cold junction terminals. To
complete the temperature measurement, you must determine the terminal
temperature in some manner.

www.mstarlabs.com/sensors/temperature.html Page 34
 Figure 3 - omitting the physical cold junction

Cold Junction Compensation

There are two commonly used approaches.

1. Simulate the potential effects that would result for a thermocouple wire
pair between the terminals, at their measured temperature, and another
junction at a reference temperature of 0 degrees. Measure the potential
across the thermocouple wire pair in series with the simulated potential.
Apply the linearizing curve to the sum, thus obtaining an estimated
absolute temperature directly. This is known as cold junction
compensation. Usually, the simulation is done electronically with
specialized integrated circuit devices.

Figure 4 - electronic cold junction compensation

This approach makes two approximation errors, one for estimating the
cold junction temperature, and one for approximating the effects on
junction potential. Beyond what is already built into the electronic
simulation, calibration is tricky and probably limited to offset adjustment.

2. Independently measure the temperature of the cold junction. Measure the

thermocouple potential and apply conversion curves to determine the
temperature difference across the thermocouple. Then add the known
cold junction temperature to the measured temperature difference to
determine the absolute temperature measurement.

www.mstarlabs.com/sensors/temperature.html Page 35
 Figure 5 - independent cold junction measurement

This approach uses one less estimate, but it still depends on accurate
measurements of the cold junction temperature.

LT1025A solid state temperature measurement devices are available on

MSTB009 termination boards and optionally available on MSXB037 analog
expansion boards for measuring the cold junction temperature when
thermocouple wires connect directly to the terminals on the boards. Any errors in
reading the cold junction temperature will appear directly as errors in the final
temperature measurement. The readings are 10 millivolts per degree centigrade
absolute temperature. Install a header plug on your board to activate temperature
measurements that you can route directly into the THERMO command provided by
the DAPL system.

www.mstarlabs.com/sensors/temperature.html Page 36
 Calibrating RTD Sensors
Calibrate RTDs
RTD devices depend on the fact that metals increase resistivity
approximately in proportion to temperature. Their thermal
characteristics are conventionally described by a base resistance
value and a conversion characteristic that equals 1.0 exactly at 0
degrees C.

R = R0 ( 1.0 + ka * T + kb * T2 + ... )

At T = 0 degrees C, the measured value of R equals the base Temperature

resistance value R0. Immersing the RTD sensor in an ice water Measurement

bath is a way of measuring this parameter directly. RTDs are

manufactured with relatively tight tolerances, and many
applications can assume that the nominal device parametes are
sufficiently accurate.
Calibration

Overview

Though inexpensive materials can be used for making RTDs,

platinum RTDs have a response that is close to perfectly linear. A
common approximation is to draw a "straight line curve," starting
from the nominal normalized resistance 1.0 at temperature 0
degrees C, extending to the normalized resistance for the device RTD Example
at 100 degrees C.

www.mstarlabs.com/sensors/temperature.html Page 37
 Figure 1 - RTD characteristic and alpha approximation

The illustration above exaggerates the curvature so that there is something to

see. For platinum RTD devices, the slope of the constructed line — the alpha
parameter value — is 0.00385 per-unit resistance change per degree C. The
difference between this straight-line approximation and the exact curve is about
0.4 degrees, with the maximum error occurring at roughly 50 degrees C. If this is
already within an acceptable tolerance, there is no point in doing any additional
calibration. This is what makes RTDs so popular.

For accurately modelling RTD response through its full temperature range of a
few hundred degrees, a higher order polynomial is used. For platinum materials,
a polynomial of order two is usually sufficient for full measurement accuracy. For
less expensive but less linear RTD devices, you might need a polynomial order
as high as six.

Unfortunately, a conventional RTD characteristic maps temperature to

normalized resistance. This is usually the reverse of how you want to use it — to
map the normalized resistance into a temperature. Inverting the curve equation is
no problem for a first or second order polynomial, but for higher-order
polynomials you must either use iterative methods or abandon the standard form.

The steps to determine the temperature of an RTD, using a calibrated

characteristic curve in conventional form, are as follows:

1. Measure operating resistance in ohms.

2. Divide by the normalizing resistance factor.
3. Solve the RTD equation in the inverse direction to determine the

temperature corresponding to the normalized resistance.

www.mstarlabs.com/sensors/temperature.html Page 38
 Usually it is most efficient to compute RTD characteristic curves directly. As with
any sensor, it is possible to select corner points at intervals along the
characteristic curve and represent the inverse curve with a piece-wise linear
approximation, and evaluate it using the INTERP command. Intervals of about 50
degrees C should work well.

Operating over a wide temperature range, you will probably do well to measure
the RTD resistance using a simple voltage divider network and unity amplifier
gain.

Figure 2 - RTD in voltage divider network

Operating over a limited temperature range, you might want to consider a bridge
configuration. If the bridge is reasonably well balanced, you can measure the
differential voltage across the bridge with a gain amplifier to improve
measurement resolution. A linear mapping from differential voltage to
temperature will then yield an accurate conversion.

www.mstarlabs.com/sensors/temperature.html Page 39
 Figure 3 - RTD in bridge network

You can use the DIVIDER command or the BRIDGE command for deriving
RTD resistance from voltage measurements. Both commands are available
on this site. You can use the RTD command, also available on this site, for
the inverse calculation of RTD temperature from normalized resistance. The
RTD command also provides extensions, not discussed here, for
multiple-part curves covering sub-zero temperatures.

www.mstarlabs.com/sensors/temperature.html Page 40
 Calibrating Thermistor Sensors

Calibrate Thermistors
Thermistors have the advantage of a very high sensitivity to
temperature changes, but the disadvantage of an aggressively
nonlinear characteristic. Here is a characteristic curve showing the
resistance of a typical negative temperature coefficient thermistor
device over a temperature range from 0 to 100 degrees C.

Figure 1 - Typical thermistor curve

Temperature
Measurement

Calibration
Overview

Thermistor
Example
As you can see, the value changes from over 15k ohms to under
100 ohms. The change is most rapid at low temperatures, giving
great resolution for determining the corresponding temperature
values there. At the other end of the range, resistance levels
change relatively less with temperature and measurement
resolution is relatively poor.

Curve forms are available that describe the nonlinear shape of the thermistor
characteristic quite well. The most commonly used form is the Steinhart-Hart
Equation. The resistance measurement of the thermistor is not normalized, so
just use the measured value of Rt in ohms. Manufacturers can provide typical
values of the ka, kb, and kc coefficients, or you can calibrate these values for
better accuracy.

Steinhart-Hart Equation

www.mstarlabs.com/sensors/temperature.html Page 41
 Thermistor Linearization Curves
It is relatively easy to calibrate your own response curves, if you have an
accurate temperature measurement standard. Convert the temperature values to
Kelvins, and invert. Take the corresponding measured resistance values and
compute the natural logarithm. Now, fit the coefficients of a third order polynomial
in the log-resistance values to best match the inverse-temperature values.

For the following example, three points are selected, two close to the ends of the
operating range and one near the center. We know that measurements will not
be completely accurate, so artificial errors have been inserted into the data to
result in temperature errors of magnitude 0.1 degrees C with alternating sign at
the three measured points.

Data with artificial 0.1 degree errors added

ResistanceTemperature
10500 9.13
3200 35.56
700 77.02

Powers of log-resistance are collected in a matrix, and the inverses of

temperature in Kelvins are collected in a vector. The model coefficients ka, kb,
and kc are obtained by solving the following matrix equation.

3-point Actual
Term
Fit Curve
ka 1.236*10 1.283*10-3
-3

kb 2.453*10-42.362*10-4
kc 4.389*10-89.285*10-8

www.mstarlabs.com/sensors/temperature.html Page 42
 Both of these formulas produce curves that are virtually indistinguishable from
Figure 1. The following shows the differences — the calibration errors — that
resulted from the data errors deliberately included for the the 3-point fit.

Figure 2 - Fitting error in degrees

Deviations of 0.1 degrees appear, as we know they should, where they were
injected at the locations of the measured points used for the fit. At intermediate
locations, the fit error is well behaved. We can conclude that the fit is about as
good as the measurement errors that went into making it — but don't extrapolate
much beyond the range that you measure. To reduce sensitivity to noise during
calibration try the following steps.

1. Calibrate over a range just a little wider than the range you intend to use.
2. Select some points very close to the limits of the range you intend to use.
3. Take multiple measurements at each point and average to reduce
random noise.
4. Consider using more than three points, and determining best-fit

coefficients using least-squares methods.

Balancing measurement resolution

The linearization takes care of the problem of interpreting the highly nonlinear
response, but not the problem of uneven measurement resolution.

If the range is not too large, you can balance the resolution significantly by
measuring in a voltage divider configuration. Power the thermistor from a
regulated voltage supply, connect the other end to ground through an accurately
measured load resistance, and observe the output voltage where the thermistor
and load resistor join.

www.mstarlabs.com/sensors/temperature.html Page 43
 Figure 3 - Voltage divider network

The goal is to obtain a relatively uniform relationship between temperature and

measured voltage. The linearization curves will take care of the rest. The
following shows the relationship between temperature and measured voltage
with a load resistor that is about half of the nominal room-temperature resistance.

Figure 4 - Flattened thermistor response in divider network

The slope doesn't change much through the operating range. This is very
different from the drastic nonlinear behavior you see in Figure 1. How does this
work? The voltage divider has a saturating characteristic that responds less as
thermistor resistance grows. The growth and saturation effects approximately
balance.

Measure a temperature using a thermistor device in the voltage divider

configuration by doing the following.

www.mstarlabs.com/sensors/temperature.html Page 44
 1. Calculate the current flow from the measured voltage and accurately

known load resistance.

2. Determine the thermistor resistance from the voltage across it and the
known current.
3. Apply the Steinhart-Hart equation, either with nominal values provided by
the manufacturer, or with adjusted values determined from calibration, to
obtain the temperature reading.
4. Optionally: convert temperature units from Kelvins to degrees C or

degrees F.

You can use the DIVIDER command, available on this site, for computing
the resistance value given the measured voltage level in a voltage divider
configuration. You can use the THERMISTOR command, also available on
this site, for computing the Steinhart-Hart curves using typical or calibrated
coefficients.

www.mstarlabs.com/sensors/temperature.html Page 45
 Module SENSORM :: BRIDGE
DAPL Operating System | Processing Commands

Convert measurements from a balanced bridge network to determine resistance.

Syntax

BRIDGE( VIN, VS, BALANCE, RLOAD,

[RCOEFF, LTMP,] [GAIN,] ROUT )

Parameters

VIN
Input data pipe with bridge imbalance voltage measurements

FLOAT PIPE

VS
Nominal or measured excitation voltage driving bridge network

FLOAT CONSTANT | FLOAT PIPE

BALANCE
Nominal or measured balancing network gain

FLOAT CONSTANT

RLOAD
Nominal or calibrated load resistance, ohms at 0 degrees C

FLOAT CONSTANT

RCOEFF
Temperature coefficient of resistance of load resistor, ohms per degree C

FLOAT CONSTANT

LTMP
Input pipe with load temperature measurements, degrees C

www.mstarlabs.com/sensors/temperature.html Page 46
 FLOAT PIPE

GAIN
The gain used to measure the VIN signal

FLOAT CONSTANT

ROUT
Output resistance measurements, in ohms

FLOAT PIPE

Description

The BRIDGE command converts the differential voltage readings from VIN to
measure an unknown resistance in a bridge network configuration. Current from
a known voltage supply drives a load resistance, and then passes through the
unknown device to the drain voltage. On the other side of the bridge network, two
known resistors form a voltage divider to establish a reference voltage. The
differential measurements of voltage between the measurement and reference
sides of the bridge are used to calculate the values of the unknown resistance,
with results placed into the ROUT pipe, one output value for each input value.

The differential input measurement will reject "common mode" voltages, so it

does not matter whether the bridge circuit is excited by balanced plus/minus
supplies or a single-sided supply, as long as the common mode voltage is within
a measurable range. Bridge configurations are useful when the measured
changes in resistance are relatively small and ride on top of a relatively large
constant level. The differential voltage can be measured with gain to improve
resolution. If a gain other than 1.0 is used, either specify it as the value of the
GAIN parameter, or correct for the gain prior to sending the VIN data to the
BRIDGE command.

If the supply voltage is very well regulated, it can be specified as a constant VS

parameter. For best accuracy, measure the actual supply voltage accurately
rather than assuming a nominal value. If the supply is subject to small but
relevant variations, measure its voltage simultaneously and pass these
measurements in units of volts through a VS pipe, with one supply voltage
measurement per divider voltage reading. If you have balanced plus and minus
supplies, specify the supply-to-supply voltage.

Ideally, the voltage dividers on the reference and measurement sides of the
bridge produce exactly the same voltage at a suitable reference operating point
within the normal operating range. Obtaining a perfect balance is hard to achieve

www.mstarlabs.com/sensors/temperature.html Page 47
 and not really necessary. Measure the resistances in the balancing network
accurately, then set the BALANCE parameter equal to the ratio

BALANCE = Rg / (Rs + Rg)

where Rs is the balancing resistor on the positive supply side

and Rg is the balancing resistor on the negative supply or
ground side

If measurement accuracy is not critical, use nominal values of the resistors to

determine the value of the BALANCE parameter. Good balancing resistors will
not cause excessive power supply loading, and will establish a zero differential
reading near the center of the operating range.

Resistors in the balancing network are presumed to be located in a controlled

operating environment where small temperature variations affect both balancing
resistors the same, hence the ratio remains very stable. If the loading resistor is
also located in a controlled environment, it too can be presumed to maintain a
consistent operating temperature. Specify an accurately measured value of the
RLOAD parameter at the stable operating temperature, in ohms. If accuracy is
not critical, use the nominal loading resistor value.

When load temperature is not so well controlled and there is significant thermal
variation in the loading resistor, use the optional RCOEFF and LTMP parameters.
Set the RCOEFF parameter to the temperature coefficient of load resistance in
ohms per degree Centigrade. Adjust the RLOAD parameter if necessary so that it
equals the correct loading resistance at 0 degrees Centigrade. Independently
measure the load temperature in units of degrees Centigrade, and send these
readings to the BRIDGE command through the LTMP pipe. The BRIDGE
command will adjust the value of the loading resistor prior to each conversion.

For each input voltage measurement, the reference voltage on the balancing side
of the bridge is equal to the BALANCE ratio times the excitation voltage. The
voltage on the active side of the bridge equals this reference voltage plus the
measured differential voltage. The voltage between the positive source and the
measurement point appears across the known loading resistance, so the
measurement-side current can be computed. Using the value of this current, and
the known voltage on the measurement side of the bridge, the unknown value of
resistance can be calculated.

Examples

BRIDGE(IP2, 5.0, 0.5, 1000.0, R2)

www.mstarlabs.com/sensors/temperature.html Page 48
 Voltage measurements are taken for a bridge configuration in which nominal
values of components are used. The voltage at the active divider junction,
relative to the balancing network junction, is obtained from the differential input
sample channel pipe IPipe 2. The excitation voltage is a nominal 5.0 volts.
The nominal balancing ratio with equal-value balancing resistors is 0.5. The
loading resistor is 1K, equal to the nominal operating value of the measured
resistance. Default measurement gain of 1.0 is assumed. The computed
resistance values are reported in pipe R2.

BRIDGE(IP2, 4.959, 0.5025, 1001.8, 0.087, TLOAD, 10.0, R2)

The same as the previous configuration, except that all components are
calibrated and the loading resistor value is compensated for temperature
variations to obtain maximum measurement accuracy. The supply voltage is
measured at 4.959 volts. The balancing resistors are not perfectly matched and
their gain ratio is 0.5025. The nominal 1K loading resistance is measured at 0
degrees Centigrade where it has the value 1001.8 ohms. The loading resistor
value is observed to increase by 8.7 ohms over a 100 degree temperature swing,
so the temperature coefficient is 0.087 ohms per degrees C. Independent
measurements of the operating temperature of the load resistance are provided
by pipe TLOAD. The measurements use an amplifier gain of 10.0. The results of
the resistance calculations are returned in pipe R2.

See also:

DIVIDER

www.mstarlabs.com/sensors/temperature.html Page 49
 Module SENSORM :: DIVIDER
DAPL Operating System | Processing Commands

Convert measurements from a voltage divider network to determine resistance.

Syntax

DIVIDER( VIN, VS, RLOAD, [RCOEFF, LTMP,] [GAIN,] ROUT )

Parameters

VIN
Input measurements of divider voltage

FLOAT PIPE

VS
Nominal or measured excitation voltage driving divider network

FLOAT CONSTANT | FLOAT PIPE

RLOAD
Nominal or calibrated load resistance, ohms at 0 degrees C

FLOAT CONSTANT

RCOEFF
Temperature coefficient of resistance for load resistor

FLOAT CONSTANT

LTMP
Input pipe with load temperature measurements, degrees C

FLOAT PIPE

GAIN
Gain used to measure input signal VIN

FLOAT CONSTANT

www.mstarlabs.com/sensors/temperature.html Page 50
 ROUT
Output resistance measurements, in ohms

FLOAT PIPE

Description

The DIVIDER command converts voltage readings to measure and unknown

resistance. The unknown device to be measured is placed in a voltage divider
network with an accurately-known loading resistor. The divider network is driven
by an accurately-regulated voltage source, and the voltages VIN are observed
at the junction between the unknown element and load resistor. The results, in
ohms, are returned through the ROUT pipe, one result per divider measurement
input value.

The input divider voltage measurements are received through the VIN pipe.
These measurements have units of volts. By default, this command assumes that
the voltage measurement are obtained using an amplifier gain of 1.0, but if you
use a higher gain, specify it as the GAIN parameter.

If the supply voltage is very well regulated, you can measure it once and specify
the value as a constant VS parameter. If you use the regulated voltage from the
data acquisition processor for this excitation, you could specify 5.0 volts and omit
the calibration measurement. If the excitation is subject to small but potentially
relevant variations, make simultaneous voltage measurements of the supply
voltage, in units of volts, and provide this second input stream to the DIVIDER
command through the VS pipe, one supply voltage measurement per divider
voltage reading.

www.mstarlabs.com/sensors/temperature.html Page 51
 Specify the loading resistor value RLOAD parameter in units of ohms. Sometimes
a nominal resistor value is close enough, but for better results enter an
accurately- measured value.

Most of the time, variations in operating temperature are insignificant, and the
load resistance can be presumed constant. For these cases, omit the RCOEFF
and LTMP parameters. However, if significant temperature variations are
anticipated, the DIVIDER command can adjust the effective load resistance to
compensate for temperature-dependent changes. Set the RCOEFF parameter to
the temperature coefficient of load resistance in ohms per degree Centigrade.
Adjust the RLOAD parameter if necessary so that it reports the correct loading
resistance at 0 degrees Centigrade. Measure the load resistance temperature in
units of degrees Centigrade and send these measurements to the DIVIDER
command through the LTMP pipe, one temperature reading per divider
measurement.

For processing each input value, the difference between the known excitation
voltage source and the measured divider junction appears across the known
loading resistor, hence the divider current can be computed. This known divider
current passes through the unknown element to produce the measured voltage,
so its resistance can be computed, producing the results placed into the ROUT
pipe.

Examples

DIVIDER(IP2, 5.000, 10000.0, ROUT)

Read the voltage at the junction between a load resistance of exactly 10K ohms
and an unknown resistive sensor when a 5.000 volt excitation voltage is applied
across the divider network. The voltages across the resistive sensor element are
measured with gain 1 and received from input sample channel pipe IPipe 2.
The computed resistance values are reported in pipe ROUT.

DIVIDER(IP2, 4.968, 10008.0, 0.42, PAmb, 4.0, ROUT)

This example is the same as the previous configuration, except calibrated for
maximum measurement accuracy. The pre-measured supply voltage level is
4.958 ohms is specified as a constant. The measured load resistor value is
10008 ohms. The nominal load resistance is observed to increase by 42 ohms
over a 100 degree temperature swing, so a temperature coefficient 0.42 ohms
per degrees C is specified. The measurements of divider voltage are captured
using an amplifier gain of 4.0. The operating temperature is separately measured
and provided in the PAmb pipe. The computations use a temperature-adjusted
value of load resistance for each conversion.

www.mstarlabs.com/sensors/temperature.html Page 52
 See also:

BRIDGE

www.mstarlabs.com/sensors/temperature.html Page 53
 Module SENSORM :: GENPOLY
DAPL Operating System | Processing Commands

Generic polynomial conversion function.

Syntax

GENPOLY( VIN, [NORDER,] VCOEFFS, VOUT )

Parameters

VIN
Input data values

FLOAT PIPE

NORDER
Order of the polynomial function

WORD CONSTANT

VCOEFFS
Coefficients of the polynomial

FLOAT VECTOR

VOUT
Output data values

FLOAT PIPE

Description

The GENPOLY command is a general purpose mapping from a stream of input

data received from pipe VIN, to produce a stream of output data in pipe VOUT.
One common application of this command is mapping measurements of an
arbitrary nonlinear sensor into the equivalent measurement in the desired
physical units.

www.mstarlabs.com/sensors/temperature.html Page 54
 The polynomial mapping is determined by the VCOEFFS vector. The coefficients
specified in this vector start with the zero-order term, followed by the first-order
term, followed by the second-order term, etc. up to the order specified by the
NORDER parameter. If you omit the NORDER parameter, the command will count
terms and assume the corresponding maximum order.

Using only a first-order polynomial, this command becomes equivalent to a

SCALE command. From the opposite point of view, this command covers the
functionality of the SCALE command at the same time that it performs nonlinear
mappings. When using this command for "linearizing" the response of a
nonlinear sensor, it can at the same time provide gain and offset corrections.

Examples

VECTOR VCOEFFS FLOAT = ( 1.0, 3.085e-3, -5.707e-7 )

CONSTANT ORDER WORD = 2
...
GENPOLY( PIN, ORDER, VCOEFFS, POUT )

Take values X from the PIN pipe one at a time, and for each one apply the
second-order polynomial mapping

Y = 1.0 + 3.085e-3*X + -5.707e-7*X2

to obtain the values Y that are placed into the output data stream POUT.

See also:

SCALE, RTD, THERMOPOLY

www.mstarlabs.com/sensors/temperature.html Page 55
 Module SENSORM :: RTD
DAPL Operating System | Processing Commands

Convert resistance measurements to the corresponding temperature for an RTD

device.

Syntax

RTD( RIN, VMODEL, TEMPOUT )

Parameters

RIN
Input resistance values in ohms

FLOAT PIPE

VMODEL
Coefficients for polynomial conversion model

FLOAT VECTOR

TEMPOUT
Output data pipe, temperatures in degrees Centigrade

FLOAT PIPE

Description

The RTD command converts resistance values received from the RIN pipe,
using the device model specified by the VMODEL vector, and delivering the
corresponding output temperatures in degrees C to the TEMPOUT pipe.

Though their temperature response is almost linear, RTD devices need a

nonlinear conversion for full accuracy. The device model is a polynomial curve,
with the number of terms required dependent on the material type used to
fabricate the RTD. Devices have standardized conversion curves, but actual
devices don't always match the standardized curves perfectly. For best accuracy,

www.mstarlabs.com/sensors/temperature.html Page 56
 you can fit a polynomial curve to measured resistance and the corresponding
temperatures, thus calibrating your own response curves.

The following conversion characteristic for an RTD device is suitable for wide
temperature ranges and a variety of RTD device types.

R = R0 ( 1.0 + c1 T + c2 T2 + c3 T3 ...)

The RTD command supports terms up to order 6. The R0 term corresponds to

the nominal resistance of the device at 0 degrees C. For example, a type JPT200
RTD will have a nominal resistance of 200 ohms, and a perfectly manufactured
device will have a base resistance R0 exactly equal to 200. The zero-order term
of the conversion polynomial is always normalized to 1.0, so it is not stored, and
the base resistance value is recorded in that location.

Some curve forms provided by manufacturers and standards bodies will not
correspond to the general polynomial form. For these, you will need to adjust the
coefficients. An important example is the platinum PT device series. It uses a
conversion polynomial of the Callendar - Van Dusen form

R = R0 ( 1.0 + a1 T + a2 T2 + a3(T-100) T3 )

The appropriate conversions to the general polynomial form are:

c1 = a1
c2 = a2
c3 = -100.0 a3
c4 = a3

There is an additional complication with standard platinum RTD devices. The

IEC751 standard specifies a different polynomial for temperatures below 0
degrees C and above 0 degrees C. To allow for this, and also to allow you to
calibrate your own piecewise polynomial curves, the RTD command allows
multiple polynomial curve sections. The nominal RTD resistance is the same for
all sections. In the second, third, subsequent pieces, the zero order term location
is used to store the temperature level above which the evaluation scheme
switches to that piece of the composite curve. The additional 6 coefficient terms
for that piece follow the temperature break-point level term.

The following table provides coefficients that you can use for typical device types.

www.mstarlabs.com/sensors/temperature.html Page 57
 Device type c1 c2 c3 c4 c5 c6
Platinum
(PT family, IEC751)
above 0 degrees
3.9083e-3 -5.7750e-7 0.0 0.0 0.0 0.0
C
below 0 degrees
3.9083e-3 -5.7750e-7 4.183e-10 -4.183e-12 0.0 0.0
C
Platinum
(JPT family, RC-4, 3.9787e-3 -5.8686e-7 4.167e-10 -4.167e-12 0.0 0.0
SAMA)
Nickel
5.485e-3 6.6650e-6 0.0 2.805e-11 0.0 0.0
(limited range)
Copper
4.270e-3 0.0 0.0 0.0 0.0 0.0
(limited range)

To summarize the layout of the device characteristic vector, the terms are
specified in the following sequence.

1. Initial conversion curve

1. Base resistance at 0 degrees C
2. Coefficients c1, c2, c3, c4, c5, c6
for first piece of conversion curve
2. Additional sections of the curve, as needed
1. Temperature above which curve section is used
2. Coefficients c1, c2, c3, c4, c5, c6
for this piece of conversion curve

When the operating temperature range is within interval T=0 to T=100 degrees
Centigrade, and maximum accuracy is not required, the simplified "alpha
coefficient" model is often used. The alpha coefficient is the slope of a straight
line that matches the nonlinear conversion curve at the points T=0 and T=100. If
you substitute the alpha coefficient for the c1 coefficient in the conversion data,
and set all of the other coefficients to zero, the conversion is linear and fast, with
maximum conversion error a fraction of a degree near the center of the range.
This often is good enough.

For wider temperature ranges, the conversion curves are not in the most useful
form. An RTD curve yields resistance given the temperature, so calculating
temperature requires evaluation of the inverse characteristic. The evaluation is
almost perfect for low-order characteristics and moderate temperature ranges.
For devices with significant nonlinearities at temperature extremes, the
conversion error is typically better than 0.1 degrees C.

Examples
www.mstarlabs.com/sensors/temperature.html Page 58
 VECTOR vPT100POS FLOAT = ( 100.0,
3.9083e-3, -5.7750e-7, 0.0, 0.0,
0.0, 0.0 )
...
RTD(PRESIST,vPT100POS,TEMPR)

A type PT100 RTD is used, but temperatures never go below 0 degrees C. The
characteristic terms for the sub-zero piece of the standard conversion curve are
omitted. Otherwise, the conversion curve in vector vPT100POS matches the
IEC751 standard for this device type. The nominal 100 ohms base resistance at
0 degrees C is used. When the RTD command executes, measurements of
resistance in ohms are obtained from pipe PRESIST. Each resistance value is
converted to the corresponding operating temperature of the RTD device by
solving the inverse of the device characteristic. The resulting temperatures in
degrees C are placed into pipe TEMPR.

VECTOR vPT100 FLOAT = ( 99.86,

3.9083e-3, -5.7750e-7, 4.1830e-10, -4.1830e-12,
0.0, 0.0,
0.000,
3.9083e-3, -5.7750e-7, 0.0, 0.0,
0.0, 0.0 )
...
RTD(PRESIST,vPT100,TEMPR)

The same example, except that temperatures can range both below and above 0
degrees C. The resistance of the RTD device is calibrated by measuring
accurately at 0 degrees C, and the observed value 99.86 ohms is specified as
the base resistance. The first set of polynomial terms conform to the the IEC751
standard for temperatures below 0.000 degrees. For temperatures above 0.000
degrees, the conversion processing switches over to the second set of terms.

See also:

DIVIDER, BRIDGE

www.mstarlabs.com/sensors/temperature.html Page 59
 Module SENSORM :: THERMISTOR
DAPL Operating System | Processing Commands

Convert resistance measurements to the corresponding temperature for a

thermistor device.

Syntax

THERMISTOR( RIN, VMODEL, TEMPOUT )

Parameters

RIN
Input resistance values in ohms

FLOAT PIPE

VMODEL
Coefficients for Steinhart and Hart thermistor model

FLOAT VECTOR

TEMPOUT
Output data pipe, temperatures in degrees Centigrade

FLOAT PIPE

Description

The THERMISTOR command converts resistance values received from the RIN
pipe, using the device model specified by the VMODEL vector, and delivering the
corresponding output temperatures in degrees C to the TEMPOUT pipe.

The device model used is the Steinhart-Hart equation. This equation is an

empirical formula capable of conversion accuracy within a small fraction of a
degree. There are no standardized conversion curves, but most thermistor
manufacturers will provide Steinhart-Hart coefficient values that work over a
suitable range with their devices.

www.mstarlabs.com/sensors/temperature.html Page 60
 The form of the Steinhart and Hart equation is

T = 1.0 / ( a + b ln(R) + c ln3(R) )

where R is the measured device resistance in ohms, and the a, b, and c

terms are the model coefficients. The THERMISTOR command uses a slight
variation of this

T = 1.0 / ( a + b ln(R) + c ln3(R) ) - 273.16

where the final 273.16 term converts the temperature units from Kelvins to
degrees Centigrade. For maximum conversion accuracy, you can calibrate the
curve for the individual device you are using. Measure the actual resistance at
three well-selected temperature points representative of your operating range,
insert these values into the Steinhart-Hart equation form, and solve for the
coefficient values.

To support a very wide temperature range, the THERMISTOR command supports

a multiple-piece device model. Each piece of the model is encoded into the
VMODEL parameter as a break-point temperature in degrees C, followed by the
three Steinhart-Hart coefficients to use at that temperature and beyond. The
model pieces are encoded in order from lowest temperature range to highest
temperature range. The first piece is the default that provides conversions at low
temperatures, so its break-point term is ignored and can be set to zero. Most
applications will use a one-piece model. Multiple-piece models only work for
negative temperature coefficient thermistors, which are the ones that most
applications use.

Example

VECTOR THERM44007 FLOAT = ( 0.0, 1.285E-3, 2.362E-4, 9.285E-8 )

...
THERMISTOR(PRESIST, THERM44007, TEMPR4)

Manufacturer-provided coefficients for a model 44007 thermistor are used, and

specified in the vector THERM44007. Because this is a one-piece model, the
temperature breakpoint between pieces is unused, and the first term is set to 0.
Measurements of the temperature-dependent thermistor device resistance in
ohms are obtained from pipe PRESIST. Each resistance value is converted to
the corresponding operating temperature of the thermistor using the
Steinhart-Hart equation with the coefficients from the THERM44007 vector,
producing the corresponding temperature results in degrees C. The temperature
results are placed into pipe TEMPR4.

www.mstarlabs.com/sensors/temperature.html Page 61
 See also:

DIVIDER, BRIDGE

www.mstarlabs.com/sensors/temperature.html Page 62
 Module SENSORM :: THERMOPOLY
DAPL Operating System | Processing Commands

Calibrated temperature conversions for thermocouples.

Syntax

THERMOPOLY( VIN, CJT, [NORDER,] VCOEFFS, TEMPC )

Parameters

VIN
Input thermocouple potential measurements

WORD PIPE | FLOAT PIPE

CJT
Input cold junction temperature measurements

FLOAT PIPE

NORDER
Order of the conversion polynomial function

WORD CONSTANT

VCOEFFS
Coefficients of the conversion polynomial

FLOAT VECTOR

TEMPC
Output results, temperature Centigrade

FLOAT PIPE

Description

www.mstarlabs.com/sensors/temperature.html Page 63
 This command is an alternative to the THERMO command provided by the DAPL
system for systems that need calibrated temperature measurements for
maximum accuracy. Input values are received from pipe VIN. The polynomial
characteristic defined by vector VCOEFFS is applied to each input value,
producing the corresponding thermocouple temperature difference.
Thermocouples do not return an absolute temperature, so independent
measurements of the "cold junction temperature" must be provided in pipe CJT,
one cold junction temperature for each thermocouple measurement. The
measurements are combined and the resulting absolute temperatures in degrees
C are placed into the TEMPC pipe.

For determining the temperature, the usual way to interpret the potential across a
thermocouple to apply published conversion curves as determined by standards
organizations. Manufacturers do their best to produce devices match these
standard curves, but the match is never perfect. Individual devices are very
repeatable, however. If operated consistently over a bounded range, without
severe thermal stresses, it is possible to apply adjusted curves and get a total
measurement accuracy within about 1 degrees C. The THERMOPOLY command
supports adjustable conversion curves; in contrast, the THERMO supports only
predetermined standard curves.

The nonlinear conversion mapping is defined by a polynomial. The polynomial

coefficients are defined in the VCOEFFS vector. The coefficients start with the
zero-order term, followed by the first-order term, followed by the second-order
term, etc. up to the order specified by the NORDER parameter. If you omit the
NORDER parameter, the command will count terms and assume the
corresponding maximum order. A third to fourth order polynomial is typically
sufficient to span an interval of a few hundred degrees C to full accuracy. A
second order polynomial is sufficient over a lesser range, around 100 degrees C
or so. A first order polynomial is the equivalent of offset and gain corrections
using the SCALE command.

The measured thermocouple potential is very small, so it must be amplified to

obtain a signal level large enough to digitize. Small offset or gain errors are
amplified along with the signal and can have a significant effect. The polynomial
of the THERMOPOLY command can fully account for these effects at the same
time that it corrects for nonlinearity. It can also cover any required scaling
changes related to converting from A/D digital readings to physical units.
Consequently, the THERMOPOLY command does not need a SCALE command to
condition the signal prior to conversions.

Examples

www.mstarlabs.com/sensors/temperature.html Page 64
 VECTOR VCF FLOAT = ( -0.01897, 25.41881, -0.42456, 0.04365 )
...
THERMOPOLY( PKIN, PCJC, VCF, PTOUT )

Use a fast third-order approximation for a type K thermocouple to measure water

and steam temperatures (in the range 0 to 200 degrees C). Take junction
measurements captured (with a gain of 1000) from pipe PKIN. The third-order
polynomial function defined by coefficient vector VCF is applied to the input data
to determine temperature differences. Read the corresponding ambient
temperature values from the PCJC pipe and combine with the temperature
difference to produce the final output temperature measurements in pipe PTOUT.

See also:

SCALE, GENPOLY, THERMO

www.mstarlabs.com/sensors/temperature.html Page 65
 Copyright (c) 2008, Microstar Laboratories, Inc.
All rights reserved.

Microstar Laboratories, Data Acquisition Processor, DAP, DAP 840, DAP 4000a, DAP 5000a, DAP 5016a, DAP 5200a, DAP 5216a, DAP
5380a, DAP 5400a, iDSC 1816, DAPcell, DAPserver, Accel, Accel32, DAPL, DAPL 2000, DAP Measurement Studio, DAPstudio, DAPcal,
DAPlog, DAPview, and Channel List Clocking are trademarks of Microstar Laboratories, Inc.

This document presents proprietary information regarding Microstar Laboratories products. The information is provided "AS IS" and may be
subject to change without notice. You are granted no intellectual property rights in the information nor in the products. Microstar
Laboratories ASSUMES NO LIABILITY WHATSOEVER, AND DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO
INFORMATION PRESENTED, WITH OR WITHOUT USE OF MICROSTAR LABORATORIES PRODUCTS. Microstar Laboratories
MAKES NO CLAIMS OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR
INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.

Any performance specifications were determined in a controlled environment, dependent on component parts that are themselves subject
to unannounced specification changes by their respective manufacturers. Actual results may vary. Performance information is provided "AS
IS" with no warranties or guarantees expressed or implied by Microstar Laboratories regarding suitability of the information for determining
actual performance for any specific application.

Microstar Laboratories products are not intended for use in medical, life saving, life sustaining, critical control or safety systems, or in
nuclear facility applications.

Other names and brands may have claims as property of others. Microstar Laboratories is not responsible for the performance or support of
third-party products mentioned in this document, and does not make any representations or warranties whatsoever regarding these devices
or products.

www.mstarlabs.com/sensors/temperature.html Page 66