Hi,
I'm reading and displaying a k-type thermocouple inside of a loop using c sharp. The displayed value jumps around in 2 deg C increments i.e 23.06 1 sec and 25.03 1 sec later. The T7 has 16 bits of resolution across -10 to +10 volts so if my math is correct the resolution for positive temp temp readings is 2^8 or 256 steps? How do I shift the full 16 bits of resolution to +10 V readings so I get 2^16 or 65,536 bits of resolution?
private string aiRegister = "AIN0";
private string tempSensor = "TEMPERATURE_DEVICE_K";
private double coldJunction = 0;
private double volts = 0;
private double tempK = 0;
private double tempC = 0;
LJM.OpenS("ANY", "ANY", "ANY", ref handle);
LJM.eReadName(handle, tempSensor, ref coldJunction);
LJM.eReadName(handle, aiRegister, ref volts);
LJM.TCVoltsToTemp(6004, volts, coldJunction, ref tempK);
tempC = tempK - 273.15;
Thanks,
Kevin
.png%3Fitok=PHGBxf6M "EEV Blog Reviews LabJack T7-Pro & Software Tools")
.jpg%3Fitok=B1YcuDFK)
.png%3Fitok=Ybt-OlgU "LabJack + JEDI One Can Create a Powerful Distributed DAQ Ecosystem")
The T7 has 16 bits of resolution across -10 to +10 volts so if my math is correct the resolution for positive temp temp readings is 2^8 or 256 steps?
2^15 is half the counts compared to 2^16, so you have 15-bits (32768 counts) of resolution from 0 to 10 volts.
More useful, though, is to look at the resolution tables:
https://labjack.com/support/datasheets/t-series/appendix-a-3-1
Since the range of voltage you get from a thermocouple is very small, you want to use the internal amp in the T7 to reduce your measurement span. I suggest a call to eWriteName to set using AIN0_RANGE=0.1:
https://labjack.com/support/datasheets/t-series/ain
Note that you are using the classic technique, whereas most people on the T7 would use the AIN-EF technique:
https://labjack.com/support/app-notes/thermocouples/t7
Also check out the main thermocouple app note if you have not already:
https://labjack.com/support/app-notes/thermocouples
Thank you for the help and pointing me to the documentation, bitten by RTFM. Problem solved, works great!
The temperature reading is stable when the labjack is connected via USB because it's grounded through connector to my computer.
The temperature reading is erratic and bounces around +/- 3 degrees when I switch to ethernet and power the labjack using the supplied walwart because it's isolated from earth ground.
Think I need a ground reference for this single ended temperature measurement, do I tie the labjack GND line to earth gnd with a 100 ohm resistor?
Thanks!
That is strange. It sounds like a sign that there might be a problem with your signal or wiring. First lets troubleshoot with nothing connected to the T7 except this single K type thermocouple (and power/comm as applicable). I'm guess you have it connected to AINx and GND? Do you still see the extra noise?
Confirm that channel is set to the proper range, which is typically the +/-0.1V range. Remove the thermocouple and instead jumper VS to AINx. VS is 5 volts so the reading should rail at just over 0.1 volts.