Resistance thermometers

Can electrical thermometers be calibrated?

It is not possible to calibrate resistance thermometers (e.g. measuring inserts).
Since electrical thermometers are normally connected to a measuring instrument or evaluation unit, it is only possible to calibrate the entire measuring chain.
Measuring inserts can, however, be subjected to a design test with a design test certificate.
Application area: e.g. resistance thermometers for mineral oil meters.

How does a resistance thermometer work?

The electrical resistance of a resistance thermometer's sensor changes with the temperature. As the resistance of measuring resistors to EN 60751 (2009-05) increases with rising temperature, we refer to it as PTC (Positive Temperature Coefficient). Pt100 or Pt1000 measuring resistors are normally used for industrial applications. The thermometers based around EN 60751 are defined in DIN 43735.

How high is the permissible vibration loading fo WIKA-Pt100 probes?

The standard WIKA measuring insert allows use up to 3g (amplitude). This corresponds to a loading of 6g, peak to peak, per DIN EN 60751 (58.86 m/s^2). In EN 60751 only 20-30 m/s^2 peak-to-peak is specified (1 g = 9.81 m/s^2). The vibration-resistant design is suitable for up to 20g peak-to-peak. Special designs, on request, up to 50g are possible.
(The values given above always apply to the vibration load directly at the measuring resistor.)

How is the accuracy class calculated?

Per DIN EN 60751 Point 5.1.3 Table 3 in °C

Class AA ± (0,1+0,0017 * t)
Class A ± (0,15+0,002 * t)
Class B ± (0,3+0,005 * t)
Class C ± (0,6+0,01 * t)

How large is the measuring error caused by the internal lead resistance with a Pt100 built into an MI cable with Cu internal wires in a 2-wire connection?

D=3 mm : 0.28 Ohm/m = 0.7 K/m (measuring error)
D=6 mm : 0.1 Ohm/m = 0.25 K/m (measuring error)
(D=external diameter of the MI cable)

How thick is the wall thickness of an MI cable?

Most manufacturers give a minimum wall thickness which corresponds to 10% of the external diameter of the MI cable.

What are 2-, 3- and 4-wire circuits?

They describe the number of wires with which the measuring resistor (e.g. a Pt100) is connected. While with the simplest 2-wire connection, the lead resistance can falsify the measuring result, this negative influence can be compensated within the 3- or 4-wire connection, and thus the accuracy of the measurement improved.

What are mineral-insulated (MI) cables?

Mineral-insulated cables for resistance thermometers consist of one or more copper wires that are embedded in highly-compacted magnesium oxide and sheathed in casing tube made from, for example, 1.4571 stainless steel.  For thermocouples, instead of copper wires, thermocouple cables suitable for the thermocouple type are used.  The most common standard sheath material for thermocouples is Inconel 2.4816.

What are the Callendar-van-Dusen-coefficients and how do I calculate these?

The Callendar-van-Dusen-coefficients are used to describe a polynomial function of the actual characteristic of a platinum measuring resistor. This can be stored in a transmitter and thus increases the accuracy of the entire measuring chain. To calculate the Callendar-van-Dusen-equation in the temperature range over 0 °C, the resistance at 0 °C and two other test temperatures are collected by comparative measurements.  Hence, the a and b constants are calculated.  For the negative temperature range, the inclusion of a measured value for another test temperature is needed in order to determine the d constant. One can, however, represent the characteristic curve of the platinum measuring resistor just as well mathematically using the polynomial equation per DIN EN 60751 with the constants A, B and C (see also WIKA data sheet IN 00.17, page 4) and also determine these by calculation from the measurement of 3 (or 4 at t < 0 ° C) test temperatures.  Similarly, one can convert the constants A, B, C into the Callendar - van Dusen constants.

What are the response times of the various measuring inserts?

The measurement of the response time is carried out in flowing water in accordance with DIN EN 60751 Section 4.3.3 and VDI / VDE 3522.

v=0,4m/s T start = approx. 20°C  T end = approx. 30°C
6mm 1xTyp K, isolated: T50= 4,0sec. T90=11,0sec.
6mm 1xTyp K, non-isolated: T50= 1,0sec.  T90=2,7sec.
6mm 1xPt100, thin-film sensor: T50= 8,5sec. T90=20,5sec.
6mm 1xPt100, ceramic sensor: T50= 7,0sec. T90=19,0sec.

What do the designations of Temperature Class mean?

The ignition temperature is the lowest temperature at which an inflammable mixture of gases can ignite at a flame, a hot surface or otherwise generated spark. Gases and vapours are divided into Classes in which the temperature of the surface must always be lower than that of the mixture. (T1 > 450 °C, T2 > 300 °C, T3 > 200 °C, T4 > 135 °C, T5 > 100 °C, T6 > 85 °C).

What do the Zones in explosion protection mean?

Zone 0 (Category 1): permanent or long-term danger of explosions
Zone 1 (Category 2): hazardous atmosphere occasionally exists
Zone 2 (Category 3): explosive atmosphere exists only rarely, and then only for a short time

Zones 20, 21, 22 with the same meanings

What does „negative temperature coefficient thermistor“ mean?

Negative temperature coefficient thermistors conduct electricity better at higher temperatures than at lower temperatures. They are also known as NTC resistances (Negative Temperature Coefficient). Typically, NTC is used in the plastics and food and beverage industries.

What does "positive temperature coefficient thermistor" mean?

Positive temperature coefficient thermistors conduct electricity worse at higher temperatures than at lower temperatures. They are also known as PTC resistances (Positive Temperature Coefficient). Typically PTC are used in high-value temperature measuring points, e.g. in the chemical industry.

What does „Pt100“ mean?

Pt stands for Platinum with a nominal resistance of 100 Ohm at 0 °C (EN 60751).

What does the designation "1/3 DIN" mean with resistance thermometers?

IMPORTANT: The terms 1/3 DIN, and also 1/5 DIN and 1/10 DIN, have NOT been STANDARDISED!
By May 2009, with the introduction of the new DIN EN 60751, there was no standardised accuracy class better than Class A. Some manufacturers of resistance thermometers (including WIKA) have used these terms in order to supply customers with thermometers with a higher accuracy than Class A.What initially presented itself as a useful addition to traditional standard designation has proved to be, on closer inspection, woefully inadequate.
The typical question "1/3 DIN from what?" can be answered by the phrase "from Class B". Unfortunately defining "1/3 DIN B" makes the situation even less clear.
There are actually two ways of looking at this additional definition "from Class B".
1.) One fixes the increased accuracy to a specific temperature point: 1/3 DIN B at 0 °C.
2.) One defines a range in which this accuracy is valid: 1/3 DIN B 0 ... 50 °C.
The representation described in 2) carries an additional uncertainty. If one uses a Class B measuring resistance, so its characteristic curve has a defined pitch. In the example of 0 .. 50 ° C, a Class A measuring resistor would already deliver, at about 20 °C, a better result than 1/3 DIN B. Result: one must use a Class A measuring resistor here.. All of this "nebulosity" has ultimately led to the introduction of a new accuracy class. Since May 2009 the Class AA has been included in DIN EN 60751, which - now that it is standardised - makes the 1/3 DIN description superfluous.

What effect does poor insulation resistance have?

In accordance with DIN EN 60751 section 6.3.1 the insulation resistance between each measuring circuit and the sheath, at a minimum test voltage of 100 V DC, must not be less than 100 MOhm. Should the insulation resistance be too low, a measuring error occurs that causes the display of too low a temperature. In relation to a resistance thermometer (with sheathed cable) this results, with an insulation resistance of 100 kOhm, in a display error up to 0.25 Ohm and at 25 kOhm up to 1 Ohm. On all WIKA resistance thermometers, an insulation test with 500 V DC and an insulation resistance of > 1,000 MOhm is carried out, i.e. we test to a factor of 50 better than specified by the standard.

What is the permissible minimum bending radius for an MI cable?

VDI/VDE 3511 Sheet 2 recommends a radius of curvature R of ≥ 5 x D (D=external diameter of the MI cable), some manufacturers of MI cable even give ≥ 3 x D as the minimum bending radius.

What minimum insertion lengths are recommended, as a rough guide, for multipart thermowells (from tubing) in order to minimise the heat dissipation error?

for gaseous media: 15 ... 20 x thermowell tip diameter
for liquid media: 5 ... 10 x thermowell tip diameter
for solid media: 3 ... 5 x thermowell tip diameter
(these standard values are only valid for static mediaThe gap between the thermowell and measuring insert should be < 0.5mm)

Why has there been, for some time, a separation between the accuracy classes for "wire-wound resistance" and "film resistance" Pt100 measuring resistors?

In the past, no distinction had been made between the two basic types of measuring resistor and their temperature limits.  Practice, however, showed that film resistors (thin-film/chipset resistors) have a (not insignificant) deviation form the characteristic. This behaviour has been accommodated in DIN EN 60751:2009-5 through the splitting of the temperature ranges within the individual accuracy classes.

Why should Pt100 measuring circuits with reduced tolerance class A or AA per DIN EN 60751 be used in at least a 3- or 4-wire connection?

The 2-wire connection is not permissible for classes A and AA per DIN EN 60751 since here the internal lead resistance of the wires is added to the measured value. This will usually exceed the specified tolerance for the temperature sensor. A measurement of the cable resistance at room temperature and adjusting this in the transmitter (for example) is possible, but the temperature-dependent resistance of the inner conductor of the cable would still be added to the reading as an error.  Conclusion: A 2-wire circuit is not suitable for accurate temperature measurement.