All I know is that thermocouples are a "pain" because they put-out very low voltages and when you amplify anything noise can be a problem. ...Theoretically, thermocouples are great because the temperature/voltage relationship depends on physics and unchanging laws of nature.
"Aaccuracy" can be calibrated. Noise & drift can be more difficult.
Personally, I'd use a solid state (semiconductor) temperature sensor if it works in the application.
Hi,
Why are you using K type and for what use, temperature range?
An accuracy of +/- 2degC is what you expect with K type, they are designed for hundreds of DegC measurement.
If you are only measuring under 200DegC then use another tempsensor.
Hi,
Can I suggest on the protobaord, you connect each of the modules to the power rails directly.
In the pic I can only see one or two connections so you are daisy chaining the power.
Can you please post a copy of your circuit, in CAD or a picture of a hand drawn circuit in jpg, png?
I would absolutely love it if someone knew of a better sensor/shield.
The goal is that it will go in water pipes, there are 4 and if the temp gets to high +100 we know the heater is stuck on and if it gets below 32 we know the heater went out.
I have about 200 thermal couplers with the markings:
TC-21128s l=192* Type K
I do not need to use them but figured since I had them laying around I would.
Thermocouples are not the sensor of choice for low temperatures , the voltage output is low and you rely on the performance of the cold junction compensation .
I’d second the DS18B20 , it’s digital
And gets around most of the accuracy issues associated with multiple shields and your poor layout using a prototype board