Thermocoupler offset?

Hey all I just got a new thermocoupler. I noticed that the original I was using had an offset of 23 degrees! I had to replace it and in the process I updated to the latest software. Anyway I tried to calibrate using the typical ice in water procedure. I had to reduce the reading by negative 5 degrees. The problem is when I take it out of the water and let it warm back up it only gets to about 60 degrees. I am unsure but this room isn't 60 degree air temp. Any thoughts or ideas on the best way to calibrate?

Thanks!

Neil

I always calibrate my things in boiling water. Who cares if the temperature reading is slightly wrong at room temperature, if the temperatures you are monitoring are needed to be accurate at 200ish degrees? Your pit is running 225 or better and you want to pull meat at or before 205 generally.

Keep in mind that the higher your elevation above sea level, the lower the boiling point will be. Google it to see what your specific elevation is and the boiling point at that elevation.

Also, I might be wrong but I think the thermocouples are less accurate at freezing temperatures? I know thermistors are less and less accurate at lower than room temperatures because of the resistance curve.

Sounds like either a problem with the components / connections around the TC amp, or a problem with the TC itself. There should really be no need to adjust any of the probe offsets under normal operating conditions.

Steve_M said:

Sounds like either a problem with the components / connections around the TC amp, or a problem with the TC itself. There should really be no need to adjust any of the probe offsets under normal operating conditions.

Click to expand...

I don't know the TC components were ordered preinstalled so I figured thats the best part of my heatermeter or the most reliable. I was happy to just have a 5 degree difference vs a 23 degree offset. I don't know about hot testing as it seems like there is a lot that can skew the results.

Thanks,

Neil

I'm not sure why you would prefer to test in a temperature range that is inconsequential to cooking, but that's up to you.

Put probe in pot of boiling water. Check that it reads 212 degrees. Offset to compensate. Altitude will change it but not a whole ton. At 5250ft (basically Denver) you're looking at 202 degrees.

There is a small amount of mostly-linear error in the measurement system that can have an effect. #1 is that the bandgap reference used in the ATmega isn't 1.100V (and Atmel assumes that you will manually calibrate for this). #2 is that the AD8495 has an allowable 3C error and 0.3% gain error. Generally it is better to do a single point calibration for the gain error which will also include a portion of the offset error.


The process of doing this is in the wiki under "Thermocouple Calibration" start by setting the offset to 0 and the mv/C to 5.0000 and then some quick math at boiling temp will adjust the "ramp" rather than just offsetting. You didn't say what the temperature you expected vs 60, so I can't speak about what's going on there.
https://github.com/CapnBry/HeaterMeter/wiki/Thermocouple-Amplifier-Testing#thermocouple-calibration

The standard thermistor probes generally do not need calibration because they don't use the bandgap reference and their responses are non-linear so you'd need to do multipoint correlations to calibrate to them. The probe offset option will be hidden away under an advanced menu in future webuis because it gives the user the false impression that a simple single point offset can calibrate their probes across the entire range.

Thanks Bryan - I calibrated the TC this morning and it's spot on now with the new mV/C value. Interestingly enough, all 3 of the thermistors are <1 deg of 211F (boiling point @ my altitude)