Analog Multiplexer for Thermocouples

[Reloadron]

Here's the point I get stuck on for calibration - if the thermocouple is ±2.2C, then you can calibrate for better results. But what if they change thermocouples? You would have to ship it and say it's ±1C as long as you never mix up your thermocouples, correct? Is that realistic?

Thanks for the reading material! I'm just getting started on this so I'm doing a bit of reading on it all!

Yes, when we look at measuring temperature with a thermocouple we look at the measurement plane. We have the thermocouple which has an uncertainty and we have the system reading the thermocouple which also has an uncertainty. The allowable error is the algebraic sum of the uncertainty. I can also order a thermocouple (or make my own) and have it calibrated and charted. Since a thermocouple can't be adjusted to some tolerance we chart them for calibration. I may use AWG 20 wire and request calibration at 5 cardinal points between 0 and 100 degrees C for example. I can then offset my display or indicating device for the known TC error at any point. I could for example also note my error at 150 C (using your temperature) and figure my error into my reading.

Thermocouples, while very good are not quite high accuracy devices. They aren't bad but aren't great either. If you multiples several each can have its own error. If I want to improve things I call a TC manufacturer and specify what I want and have my thermocouples made from the same spool of wire. Here in Cleveland, Ohio I dealt with Marlin Manufacturing prior to retiring. I called and they made me whatever I wanted. I also used Omega but Marlin was local and did good custom work. Anyway, yes, you change the thermocouple in a system and everything can change.

Depending on application other sensors can be used like the PRT as an example. A good PRT can be very, very accurate.

Ron

 

[BuffaloEngineer]

Yes, when we look at measuring temperature with a thermocouple we look at the measurement plane. We have the thermocouple which has an uncertainty and we have the system reading the thermocouple which also has an uncertainty. The allowable error is the algebraic sum of the uncertainty. I can also order a thermocouple (or make my own) and have it calibrated and charted. Since a thermocouple can't be adjusted to some tolerance we chart them for calibration. I may use AWG 20 wire and request calibration at 5 cardinal points between 0 and 100 degrees C for example. I can then offset my display or indicating device for the known TC error at any point. I could for example also note my error at 150 C (using your temperature) and figure my error into my reading.

Thermocouples, while very good are not quite high accuracy devices. They aren't bad but aren't great either. If you multiples several each can have its own error. If I want to improve things I call a TC manufacturer and specify what I want and have my thermocouples made from the same spool of wire. Here in Cleveland, Ohio I dealt with Marlin Manufacturing prior to retiring. I called and they made me whatever I wanted. I also used Omega but Marlin was local and did good custom work. Anyway, yes, you change the thermocouple in a system and everything can change.

Depending on application other sensors can be used like the PRT as an example. A good PRT can be very, very accurate.

Ron

Thanks! What device did you use when you wanted to calibrate them? Or would you mostly have that done by someone else?

 

[Reloadron]

Thanks! What device did you use when you wanted to calibrate them? Or would you mostly have that done by someone else?

We did them in house. Below 300 F we used an oil bath and above 300 F (300F to 2,000F) we used a furnace designed for TC Calibration. The laboratory standards were PRTs (Platinum Resistance Thermometers). The oil bath used an oil made by Dow Corning, I forget what it was. During my later years I managed to avoid much of the calibration. The furnace was maybe 5 feet long with a tubular working zone and 3 controllers. We likely had a thousand thermocouples in the entire operation between furnaces and freezers, everything done to Mil-Standards and assorted ASTM specifications. Considerable time and money went into things. Thermocouple calibration can range from simple to a complex affair of comparison. The only thermocouples sent out of house were a few lab standard types which were Type S. Thermocouples become a science unto themselves. It really comes down to your specific applications and temperature ranges. Some have a sheath some don't. In some furnace applications the sheath material is critical. Oxydizing environments are a good example and vacuum furnaces.

Ron