Why is RTD immune to electro-magnetic fields? [closed]

Question

Thermocouples may have added voltage & induction heating induced in a magnetic field, throwing off the measurements.

A Resistance Temperature Detector (RTD) is said to be a good alternative. Why is it more robust towards the said two issues?

Answer 1

Typically the voltages involved are an order of magnitude higher. A 100 ohm RTD is still relatively low impedance (though nowhere near as low as a typical thermocouple which is a few ohms typically).

A 100 ohm DIN RTD changes 0.385 ohms per °C average over 0~100°C so a 0.5mA or 1mA current will result in almost 200uV/°C or 400uV/°C output.

A type K thermocouple will have output of about 41uV/°C over that range, 1/5 or 1/10 as much.

Base metal thermocouples such as K, T, E, N etc. that are appropriate for this kind of measurement generally have one lead that is ferromagnetic so they can suffer eddy current losses (and thus heating) in a strong AC magnetic field.

Neither one is "immune", it's just easier to design an RTD circuit with a given level of susceptibility. You have to try hard to make it bad. Thermocouples also have more sources of error because of the low output (unwanted thermal EMFs, noise, drift and offset in electronics) and the need for an accurate temperature sensor (often something like an RTD) and isothermal provisions at transitions to provide cold-junction compensation.

Where thermocouples shine is (possible, depending on construction) fast response, mechanical ruggedness, and ability to handle temperature extremes. You can literally measure the temperature of molten iron with a thermocouple.

Answer 2

Besides what is mentioned in the other answers, another advantage of RTDs is that they can be driven and demodulated by AC current, which makes the readout immune to all disturbances which are either DC or at frequencies different from the readout frequency:

• mains interference
• switching PSU interference
• thermoelectric voltages
• induction voltages from moving machines

Answer 3

Probably the higher voltage drop is a contributing factor. The higher resistivity in the element might also reduce such heating in some cases.

I don't know of any test data offhand, and keep in mind it will depend critically on element size, wire material and thickness, excitation frequency, etc. Comparing two random devices of very different design won't be very meaningful. For example, a sensor of either type, in a stainless steel casing, will have the same heating response of the casing itself.