Temperature Basics

Bill Bergquist, Sr. Applications Engineer Jeff Wigen, National Sales Manager

Bill Jeff

Agenda 2

Temperature Defined Temperature Scales Types of Temperature Sensors Thermocouple Types Temperature ranges Lead wire colors Standards RTD Temperature coefficient Construction Calibration Interchangeability Standards Identification of Sensor Types Sensor Selection - The Four Ps Calibration Preventive Maintenance

Temperature 3

What is temperature?

Measure of the average kinetic energy of particles in a substance.

Temperature is the result of the motion of particles. Temperature increases as the energy of this motion increases.

Physical quantity that is a measure of hotness and coldness on a numerical scale.

Temperature 4 Temperature of some things: ¾ Ice 32°F

¾ Dry ice −78.5°C (−109.3°F) at atmospheric pressure

¾ Liquid nitrogen -196°C

¾ “Fahrenheit 451” (Ray Bradbury) approximate auto-ignition point of some paper

¾ Sun 5700K (5430°C)

¾ Earth’s core 5700K (5430°C) (cooling at 100°C per billion years). The core may contain enough gold, platinum, and other siderophile elements that if extracted and poured onto the Earth's surface, would cover the entire Earth with a coating 0.45 m (1.5 feet) deep. ¾ Lightning bolt 50,000K

¾ Earth’s average surface temperature – good luck trying to figure out that one! ¾ In photography K (Kelvin) is used to denote color temperature. Match flame is about 1700K, sunshine is about 5400K.

Temperature Scales 5

Celsius The Celsius scale is widely used in industry Swedish astronomer Anders Celsius (1701–1744) with the Fahrenheit scale a close second. By international agreement the unit "degree Celsius" and the Celsius scale are currently defined by two different temperatures: absolute zero, and the Kelvin is almost exclusively used for scientific triple point of VSMOW (Vienna Standard Mean Ocean Water). and laboratory measurements. Triple point of water is defined as 273.16K or 0.01°C, the point at which water exists as a vapor, solid, and liquid.

A degree Celsius (or a Kelvin) is what you get when divide the thermodynamic range between absolute zero and the triple point of water into 273.16 equal parts.

In 1948, it was renamed Celsius because centigrade had other meanings in Spanish and French.

Temperature Scales 6

Kelvin A Kelvin or K does not have a degree symbol. One of the seven base units in the SI system of units named after the Belfast-born, Glasgow University engineer and physicist William Thomson, It is used by itself such as 273K. On 1st Baron Kelvin (1824–1907) calibration reports you will see the calibration Based on absolute zero which is 0K, or −273.15°C or −459.67°F. uncertainty expressed in millikelvin or mK so Theoretical point at which all thermal motion stops. 1mK is 0.001K. Fahrenheit Temperature scale based on one proposed in 1724 by the physicist Daniel Gabriel Fahrenheit (1686–1736) freezing of water into ice is defined at 32 degrees, while the boiling point of water is defined to be 212 degrees — on Fahrenheit's original scale the freezing point of brine was zero degrees.

Various Temperature Sensors 7

• Liquid displacement thermometers There are other types of sensors that have • Dial thermometers little practical use or are still in development • Thermistors • Optical (DTS, IR) such as the Johnson Noise Thermometer, • Thermocouples and a device that uses a laser as a measure. •RTD’s RTDs and Thermocouples Infrared

Dial Thermistors Bimetal coil

Thermometers 8

Liquid Displacement Dial Mercury-in-glass is probably the most well Both have . . . known liquid displacement type of Limited temperature ranges thermometer. Other types of fluid are used Limited accuracy with some used to deflect a needle that Industrial integrity indicates temperature. Introduce human error

Thermistors 9

• Very precise • Ceramic, silicon, or polymer resistor As temperature increases the resistance • High resistance values increases and can be correlated to a • Very limited temperature range temperature. • Often have drift error • Self-heating errors •Low price

Thermocouples 10

Made by connecting two different metals to form a closed circuit. As temperature increases, the measured High durability voltage increases and does so in a Low initial cost Voltage change with temperature predictable manner. The thermocouple standards define the voltage vs. temperature

+ relationship for several types of thermocouple junctions. -

Thermocouples 11

2500

2156 2000

1500 1600 1400 1200 Temp 1000 700 F 500

32 32 32 -328 -328 -450 -328 Type T Type E Industrial RTD Thermocouples

Thermocouples 12

Types Most common • Type T: Copper-Constantan Red/Blue • Type J: Iron-Constantan Red/White • Type E: Chromel-Constantan Red/Purple • Type K: Chromel-Alumel Red/Yellow

High temperature types • R, S, B Platinum/Platinum-Rhodium 2640°F • W3, W5 Tungsten/Tungsten-Rhenium 4200°F Each type has a different temperature range and Voltage vs. Temperature relationship ASTM Standard E230 , ANSI MC 96.1, and IEC 584-3

Thermocouples 13

Junctions can be grounded, ungrounded or exposed.

Thermocouples

Advantages Disadvantages Very wide temperature Decreased accuracy vs. Thermocouples are typically used in an Range [1.2K to 2300°C] RTDs application where high temperature or high Fast Response Time More susceptible to vibration is present. Their durability is Available in small sheath RFI/EMI sizes Recalibration is difficult probably their best asset. Low initial cost Requires expensive TC wire Durable from sensor to recording device Difficult to verify Not as stable as RTDs

Resistance Temperature Detector 15

RTDs 3 common element styles The platinum resistance thermometer (PRT) Coil in the hole is the most widely used sensor type in Wire wound applications where highly accurate, Thin film repeatable and stable measurements are Resistor made from platinum, nickel, copper, or other metals required. Other metals are used but none have the wide temperature capability of platinum.

RTD 16

Basic Operation Resistance changes with temperature. As temperature increases, resistance increases in response. Small current is sent through the resistor element and electrical resistance is measured Performance defined by IEC 60751 and ASTM E1137

RTD 17

Basic Operation When specifying an RTD it is necessary to Temperature coefficient select the correct temperature coefficient to • Also called the Temperature Coefficient of Resistance or alpha match the readout instrument or controller. • Units are ohms/ohm/°C Failure to do so will result in an inaccurate • The average change in resistance per unit change in measurement. temperature between 0 and 100°C

• α = R100 -R0 / 100°C*R0

»R0 = resistance at 0°C

»R100 = resistance at 100°C For the current industry standard 100 ohm RTD the alpha is .00385 which means at 100°C the nominal sensor resistance is 138.5 ohms

RTD 18

Most common coefficients • 0.00385 – ASTM E1137 or IEC 60751 • 0.003902 – American • 0.003916 – JIS • 0.003925 - SPRT, Secondary SPRT Must match your instrument to the proper temperature coefficient of your sensor

RTD 19

Coefficient example A temperature is being measured with a sensor having a temperature coefficient of .003916 (JIS) but due to a sensor failure it was replaced with a sensor having a temperature coefficient of .00385. If the transmitter/controller is not recalibrated, at 100°C it will read 1.7°C low.

Interchangeability 20

RTDs are manufactured to have 100 ohms at 0°C. The two main standards in use today are the IEC 60751 and the ASTM E1137. Both have Interchangeability refers to the “closeness of agreement” between an actual PRT R vs. T relationship and a predefined R vs. T the same nominal R vs. T values but differ in relationship. defining some performance characteristics and the tolerances associated with the Defined by ASTM E1137 and IEC 60751 grades or classes of sensors. The Nominal R vs. T of ASTM and IEC standards are equivalent but interchangeability tolerances are the target tolerances are different. that manufacturers shoot for when building the sensing elements.

Interchangeability 21

4 IEC Class B Note that the ASTM standard has slightly ASTM Grade B 3 tighter tolerances for the two grades of 2 sensors. All RTDs are built with the tightest IEC Class A ASTM Grade A 1 tolerance at 0°C and as the temperature diverges from 0°C the tolerance increases. 0 -300 -200 -100 0 100 200 300 400 500 600 700 800 The vertical line on the graph represents 0°C Tolerance (±°C) Tolerance -1 ASTM Grade A IEC Class A and the tolerance on the y axis is expressed -2 in ± °C from nominal. ASTM Grade B -3 IEC Class B

-4 Temperature (°C)

Interchangeability 22

Standard Tolerance Defining Equation¹ These equations can be used to calculate the ASTM E1137 Grade A ± [ .13 + 0.0017 | t | ] ASTM E1137 Grade B ± [ .25 + 0.0042 | t | ] interchangeability at any temperature. Note IEC 607512 Class AA2 ± [ .1 + 0.0017 | t | ] that the temperature t is an absolute value in IEC 60751 Class A ± [ .15 + 0.002 | t | ] °C. The resultant is the interchangeability in IEC 60751 Class B ± [ .3 + 0.005 | t | ] ± °C. 2 2 IEC 60751 Class C ± [ .6 + 0.01 | t | ] Note 1: | t | = absolute value of temperature of interest in °C Note 2: These tolerance classes are included in a pending change to the IEC 60751 standard.

RTDs

Advantages Disadvantages Very stable output More limited temperature Linear and predictable range [-200°C to 500°C] Easy to verify and High initial price recalibrate Slower response time than High accuracy a thermocouple No special wires required Less durable than a for installation thermocouple

Identification 24

RTD or Thermocouple? Lead wires • RTD has two, three, or four leads per sensing element • TC has two leads per junction • RTDs typically have red, white, green or black leads (not defined by the standards) • TC colors match thermocouple type – red (common), yellow, purple, blue, white Resistance check • RTD will have about 109 ohms between leads • TC typically less than 1 ohm Continuity check • TC grounded junction has path from leads to case Magnet test • RTD leads are not magnetic – usually copper • TC type J has one iron lead which is highly magnetic

Selection and Application 25

How do we decide which technology to use? Accuracy and stability are almost always Thermocouple Exhaust gas preferable characteristics to have in a Injection molding temperature sensor and for that reason I Bearings recommend that an RTD be selected unless Refinery the environment or process characteristics RTD Pharmaceuticals dictate another technology. Those factors are Fuel custody transfer discussed on the following slides. Chemical Tire /rubber

Selection 26

Factors to consider There are other factors to consider other than Placement Protection these but for most applications these will get Performance you the measurement and service life from Price the sensor that you desire. Service life

Placement 27

Two options Surface mount For a pipe there are two options, surface Immersion mounted or immersion. The two parts on the left are just two of a wide variety of surface mount sensors available and the two on the right are examples of the two most common styles of immersion sensors.

Placement 28

Surface mount Installation of a surface mount sensor can be accomplished with a hose clamp, tape, or an adhesive.

Placement 29

A few additional styles of surface mount sensors. Whether RTD or thermocouple they almost all look alike but are very different inside.

Placement 30

Surface mount - positives Installation is very easy to do and is low cost. Easy installation No flow obstruction Accuracy suffers though and measurement Low cost accuracy requirements may not be achieved. Surface mount - negatives Requires insulation for best accuracy Minimal protection from ambient conditions Difficult to calibrate Measures pipe surface

Placement 31

Immersion An immersion sensor overcomes the negatives of the surface mount and in most cases dramatically improves the measurement accuracy.

Placement 32

Immersion (no thermowell) Fast response Direct immersion of a small diameter sensor Low cost gives an accurate measurement at low cost. Short immersion It is not always possible or desirable though because of maintenance or durability considerations. Limitations Durability Maintenance Strength

Protection 33

Process connection These are just a few of the several styles available. Any piping connection either has been or could be adapted to a thermowell process connection.

Protection 34

Connection heads Attach extension wires A connection head is the best method and Protect sensor from ambient provides a convenient place to attach lead conditions wires or to house a local transmitter.

Protection 35

Numerous styles and materials from plastic to aluminum are available. Some carry ratings for use in hazardous atmospheres.

Protection 36

Transmitters Heads provide protection for transmitters and local indicators.

Protection 37

Hazardous atmosphere Hazardous atmospheres require an RTD and connection head assembly that carry an appropriate rating. A word of caution -- just the addition of a rated head to any sensor does not make the whole assembly rated. The entire assembly must be tagged and identified as having the desired rating.

Protection 38

Options Nipple Head is attached to the thermowell and Nipple - coupling sensor typically with a pipe nipple. The most Union versatile is the union connection. It allows • Calibration easy removal of the sensor from a • Replacement thermowell.

Performance 39

Consideration Thermocouple RTD Here are some general performance Accuracy at 32°F Standard limits: ± 4°F* Grade B: ± 0.54°F (± .3°C) Special limits: ± 2°F* Grade A: ± 0.27°F (± .15°C) specifications for RTDs and thermocouples. As you can see the RTD has quite a large Calibration Limited to in-situ calibration - Easily recalibrated for longer service life and traceability accuracy advantage over the thermocouple. - Matching transmitter improves performance Stability Dependent on wire Average drift is ± 0.06°C after homogeneity and process 1000 hours at 400°C. conditions Repeatability Highly dependent on process Less than ± 0.04% change in characteristics ice point resistance after 10 cycles -200 to 500°C.

*Types J and K. Types T and E special limits are ± 0.9°F

Performance 40

Cost of inaccuracy High accuracy insures product quality and • Process Fluid: Water efficient use of your energy dollar. • Flow Rate: 25 GPM • Control Temperature: 100 °F • Energy Cost: 2.9¢ / KW-hour

Annual Cost of Energy Per °F Error $923 / year

Performance 41

Time response A tapered thermowell will have a 3 to 4 times TIME RESPONSE Direct Immersion RTDs DIAMETERS slower response than the ¼” diameter direct 2.5 seconds 1/4” - 1/8” immersion sensor. This can be a big factor in

4 to 6 seconds 1/4” accuracy for processes that are changing temperature rapidly. The sensor needs to be 6 to 8 seconds 1/2” - 1/4” fast enough to keep up with the process. Thermowells 22 seconds Stepped

26 seconds Tapered

Performance 42

Time response – surface mount more dependent on housing style and mounting surface than immersion types Surface mount sensors will have a slower Surface Mount RTDs time response that an immersion style. They Ranges from are affected more by the surface they are 8 to 67 seconds mounted to rather than by the sensor design.

Thermowells Surface Mount TCs

8 to 67 seconds depending on junction type

Performance 43

Transmitters Converts resistance or voltage to a Adding a transmitter can improve accuracy variable current of 4 to 20 ma when a long run of lead wire is required. Lead wire > 250 feet (+0.16°F/100 ft) They also provide a more robust signal that is Accuracy less susceptible to interference from electro- Matching Lead wire magnetic or radio frequency interference. Robust signal RFI/EMI Local display

Selection Guide 44

Situation Thermocouple Wire Wound RTD Thin Film RTD

Accuracy/Stability X Mod. Temp (-50 to 200°C) X X High Temp (-200 to 500°C) X Higher Temp (over 500°C) X Time Response (< 6 sec.) X X Long-term Stability X High Vibration (g level) X X Extra High Vibration, Shock X X Critical Temp. Application X

What is Calibration? 45

• Calibration is performed to verify sensor/instrument performance.

• Calibration is the process used to insure that a sensor/instrument maintains specification over time and changing ambient conditions.

• Calibration is the process used to maintain traceability of parameters with reference to national/international standards.

Why 46

Initial Calibration New plant or equipment commissioning Calibration should be performed when Verify vendor data – shipping and installation damage • Insure accuracy of measurements starting up a new facility or if a new piece of Recording data equipment is added. This insures that the Ongoing Calibration Minimize and control random and systematic errors instruments have not been damaged during Compare and complement the quality and reliability of measurements by shipping or installation and provides a comparison to international standards • Provide traceability to national standards, (e.g. NIST) baseline for comparison to subsequent • Meet Regulatory Requirements (FDA, NRC) Quality System requirements calibrations. • Ensure consistent product quality • Safety Cost • Poor accuracy = wasted $$

When 47 When? Is one of the most frequently asked questions about calibration. There is no RTD has drifted How do you know? generic answer for when; it all depends on Frequency the process and comfort level of the invested Process dictates the calibration cycle parties. Process conditions that affect the Probe drift probe drift rate and the product value are the •Vibration two main considerations. Consult with the • Shock • Temperature manufacturer(s) of your system and they •Cycling should be able to help you estimate the long Product value term accuracy of their equipment based on Complete 5 cycles w/o shift then double the interval your process conditions. After that it’s up to you to pick a frequency that meets your comfort level and that of any 3rd party watch- © Burns Engineering Temperature Basics dogs that oversee your production facility. You may decide because of product value that a calibration is performed before and after each batch. Then if you desire, after 5 cycles are completed without a significant shift, the calibration cycle could be doubled.

Terminology 48

• ITS-90 = International Temperature Scale of 1990 • IPTS-68 = International Practical Temperature Scale • TPW – triple point of water 0.01°C or 273.16 K

•R0 = resistance at 0°C • SPRT = standard platinum resistance thermometer • Dewar = insulated container • IR = insulation resistance • K – Kelvin temperature scale (used for ITS-90) • mK or milliK = .001 K • NIST – National Institute of Standards and Technology • NVLAP – National Voluntary Laboratory Accreditation Program • A2LA – American Association for Laboratory Accreditation

How - Temperature Scales 49

Evolution of standard temperature scales IPTS-27 IPTS-48 IPTS-68 ITS-90 Beginning in 1927 the International Bureau of

ITS-90 (International Temperature Scale) Weights and Measures decided that a better Released in 1990 standard was required for temperature and The official international scale In better agreement with thermodynamic values than the IPTS-68 the International Practical Temperature Scale

ITS-90 vs. IPTS-68 was born. Since then about every 20 years ITS-90 • Uses TPW the scale has been refined to improve • Most accurate accuracy. In 1990 the name changed to • Complex equations IPTS-68 International Temperature Scale and the • Simpler equations • Less accurate equations defining the R vs. T relationship • Callendar-Van Dusen equation became more accurate.

How - Calibration Options 50

Factory Calibration Options Calibrating an RTD and adjusting the readout Matched Calibration • Matched with other probes or transmitter accordingly is a cost effective • Matched to a transmitter method to improve measurement system Multiple Point Calibration accuracy. This eliminates most of the RTD • -196, -38, 0, 100, 200, 300, and 420 °C interchangeability tolerance and can also Matched to a Temperature Readout (meters) minimize other instrument errors inherent in the system.

Calibration 51

Ice Bath • Easy to Produce The ice bath is the easiest and most accurate • Uncertainty ±.002°C method of checking an RTD. Addition of a stirring motor insures even temperature throughout the insulated Dewar. Ice is made from pure water, crushed, and packed into the Dewar. Purified water is added to fill in the gaps. Too much water and the ice will float which is not desirable.

Calibration 52

Triple-Point-of-Water Cell The triple point of water (TPW) cell may be the most commonly used type of fixed point and is used in ITS-90 calibrations. Water can exist as a solid, liquid, and vapor at 0.01°C and this device creates this temperature.

Comparison Calibration 53

Most common method Comparison of unknown to known sensors Multiple sensors can be calibrated at the same time Equipment Meter, Standard PRT, Recorder, etc. (system) • All add to uncertainty level The standard PRT should have an accuracy at least four times greater than the unit under test

Comparison Calibration 54

More practical and less expensive than fixed point temperature calibration Comparison calibrations can be performed in a laboratory or in the field. High accuracy Laboratory can be obtained with careful selection of Typical uncertainty: 0.001°C to 0.01°C equipment. Durability is as important as Very high accuracy reference resistance bridge, standard PRT, calibration baths, etc. accuracy when used for field calibrations. • Uses some fixed point temperatures Equipment that cannot stand up to field use Field will drift quickly and not give the expected Typical uncertainty: 0.05 to 0.5°C measurement uncertainties. Accurate reference meters, secondary PRTs, baths or dry-wells Instruments are field compatible

Standard PRTs 55

Specifications This is NOT the type of device to use for field Very fragile Use mainly in laboratory calibrations. It is extremely fragile and very environments expensive, about $10k with calibration. Highest accuracy, high repeatability, low drift -328 to 1983°F (-200 to 1084°C), accurate to ±.0018°F (±.001°C)

Anatomy of an SPRT 56

Photo of an SPRT element inside its quartz sheath.

Anatomy of an SPRT 57

A little closer look.

Anatomy of an SPRT 58

Oooops Yes, the quartz sheath does break very easily -$10K and is one of the few things duct tape won’t fix!

Preventive Maintenance 59

Corroded terminals can cause high resistance in the leads

3-wire circuits are susceptible – accuracy depends on each conductor having exactly the same resistance Terminals clean and tight Terminal block clean and dry, secured to head Wires are tinned, or terminated with spade lugs Connector pins connect firmly and are clean Use gold plated pins in a high quality connector

4-wire circuits also compensate for some poor maintenance Compensate fully for all lead wire resistance in the circuit

Preventive Maintenance 60

The good and bad The connection head shown in the lower right corner had a lot of dust in it that caused some electrical leakage between the terminals resulting in a bad temperature reading.

Preventive Maintenance 61

RTD Fan blowing on sensor location Something as innocuous as a fan blowing Radiated heating or cooling from nearby equipment Insulation covering the external portions of sensor can cause a measurement error if it is Sunlight - solar heating right where you don’t want it Thermowell directed on the external portions of the Bore cleaning Heat transfer compound temperature assembly. Product buildup on wetted portion Cracks in flange weld or leaky gasket RTD bottoms in well and spring loads Controller RTD temperature coefficient is set correctly in controller 3 or 4 wire circuit connected correctly with correct wire type Transmitter Wires connected securely Check output at zero and span

Preventive Maintenance 62

RTD Check insulation resistance 50 VDC > 100 megohms at 25°C Check ice point resistance 100 ±0.12 ohms or ±0.06 ohms RTD and transmitter matching Frequency of checks – process dictates the intervals

Connection head Wire insulation Shielding General condition, corrosion, discoloration, threads, cracks Corrosion on terminal connections Water inside the connection head Conduit seal for hazardous atmospheres

Transmitter Wires connected securely Check output at zero and span

Replacement Checklist 63

RTD Choose the correct temperature coefficient. Most common is a .00385 conforming to IEC 60751 or ASTM E1137 Interchangeability – choose class A for better accuracy 3 or 4 wire – 4 wire provides better accuracy Choose correct length to match thermowell or provide significant immersion to avoid stem conduction – for a direct immersion probe minimum immersion = 10x probe diameter + sensitive length

Thermowell selection Corrosion Erosion Wake frequency and strength Time response Immersion length

Summary

There are many types of temperature sensors. RTD and thermocouple are the most widely used in industry. RTD is the most accurate and TC is the most durable When selecting a sensor remember the 4 Ps Performance Placement Protection Price Periodic calibration is necessary to maintain measurement accuracy

BE educated

Watch for upcoming RTDology® events

View presentation notes from previous sessions on our website at: www.burnsengineering.com/RTDology