Selection Of Thermocouple

The R or S type seem to be more accurate, why are considering the K type?

The concept in commercial dry block calibrators is that the displayed temperature value of the heated insertion dry block is calibrated at the factory (or subsequently by a calibration lab) to a stated accuracy by comparison to a reference thermometer with traceable certification.

The uncertainty expressed in the dry block's specification and cal cert is based up this calibration and traceability of the reference thermometer. A dry block calibrator is generally purchased with a calibration cert certifying the accuracy/unceratinty spec.

My question is why are you trying to compare to some other reference? Do you not trust the vendor's certification?

on internet the sensitivity of k type is 41uV/degree C and of S type 10uV/degree C and mentioned to use s type for higher range. i can not understand this

the temperature calibrator manufacturer has provided thermocouple but did not provide its calibration certificate. he has provided calibration certificate of calibrator ( which he calibrated with his own thermocouple). in my lab the accuracy with that thermocouple is about 4 degree C.

i think if i get class 1 it might give good result.

moreover pl guide a thermocouple accuracy has any relation with dia and lenght of thermocouple. i should buy small or big dia thermocouple.

R and S are essentially the same, maybe this will help...

http://en.wikipedia.org/wiki/Thermocouple

1. No-one issues a calibration certificate for an individual type 'k' t/couple.

2. Length and diameter depends on the application only.

in my lab the accuracy with that thermocouple is about 4 degree C.

How do you know this ? Compared to what ? Type 'k' means a clearly defined and constant characteristic within a specified temperature range. The fault may possibly lie within the temperature indicator, but unlikely (not impossible) the t/couple itself.

The t/couple junction may be slightly damaged due to mechanical shock, or you may be immersing an unsealed t/couple in conductive liquids.

3. Refer to your operating manual. The temperature bath has a calibration certificate, and it seems from your query that you wish to use a reference t/couple each time you calibrate other t/couples ... WHY ?

If you wish to use a reference thermocouple for a dry wall calibration unit, then an s type would be a logical choice. It has a temperature range that covers your calibrator and a fairly high accuracy accross that range. However, it's a secondary calibration and there really isn't any advantage in the 0.25% you'll maybe get by using the S type. If your calibrating K type thermocouples, use a K type reference. No use comparing apples with oranges?? As for the size thermocouple, it will depend on the size of the thermowell on your calibration unit. You want the most evenly distributed heat transfer between element and sensor. So a close tolerance is the key. There is also a thermal transfer compund that can be applied to the sensor/thermowell to ensure optimal heat transfer. Don't get carried away with infantecimal accuracies on field equipment and secondary calibrations. There has to be some amount of practicality applied in the field. Generally these instruments form part of a control loop which will have inherit errors anyway, which will nullify all your hard work in gaining perfect thermocouples. Hope this has been of some help.

I'm amazed by your very "interesting" questions regarding thermocouples and calibration.

I'd suggest that you do two things.

1. Ask Iris to help you identify the proper specifications and procedures for the instrumentation in your laboratory.

2. Purchase the specifications and procedures that Iris recommends, and then read them, over and over until you understand what they say and mean.

Good luck.

Idealy whole calibrating system (sensor +reader) should be10 times accurate than the accuracy to be gauranteed for the calibrated device.

Before counting on one thermocouple to check the accuracy of another thermocouple, I would consider using a Platinum RTD as my reference standard. I would also check with NIST to determine what they recommend.

My supplemet to what was said above..

1. At lower temperatures, up to 460'C [optionally..650'C] using PRT = Platinum Resistance Thermometer (or Standard Platinum Resistance Thermometer = SPRT) would give better accuracy than Thermocouple. There are SPRT up to 1000'C, but they are expensive.

2. Using the same type Reference Thermocouple as calibrated Thermocouple allows for comparison method. Using comparison (or replacement) You can reduce many sources of errors (milliVoltmeter scale error).

3. Gold - Platinum Au-Pt Thermocouples are the most accurate, but also the most expensive.

Re size: probably 0.4mm diameter would be near optimum, but it depends on diameter and length of the hole ( insertion depth) in Your block.