In profile parameters, I have set to F on P36, and ran a reset. I am using a IR sensor with a LCD display. IR value set to F, so I can check it against DT80 LCD display. S

ync of temp starts out ok, and then drifts where, DT80 is almost 15 higher than the IR sensor. Using K thermocouple. and set that on the IR sensor. Do you have any suggestions on what might be causing this?

Best regards,
Dean Phillips

Good morning Dean,

Is the wire from the IR sensor K type thermocouple wire? Does the sensor put out the temperature referenced to zero? Is the dataTaker isothermal?

A thermocouple putout a Voltage that is proportional to the temperature difference at each end of the thermocouple. This means we need to accurately know the temperature at the point the thermocouple connects to the dataTaker. How does the IR handle the reference temperature compensation?

Cheers,
Roger

Roger,

Yes, I am using twisted shielded K type Thermocouple wire: http://www.omega.com/pptst/EXPP-K-TWSH-UL.html. I'll have to call the mfg regarding the temp ref to 0. I don't see anything in the manual.

Wikipedia definition:

An isothermal process is a change in which the temperature of the system stays constant: T = 0.

So, for testing it is in a room. For data acquisition, it is in a car. I'll be reviewing the FAQ as well. I'll check with mfg and send you a post tomorrow.

Best regards,
Dean

Roger,

Isolated the TC wire away from power supply wire. Ran mV output on sensor and it gave me some range problems. Then I reset the Sensor to TK. Found out I had the TC wires reversed. Temp dropped from 15 deg difference, to same temp within .2 deg of sensor.

I can live with that TC wires are very finicky, you have to make sure they are really tightened down. Sensor must have ambient greater than 0 degF. So no problem for outside use.

One thing I did find out. I can set the sensor Mv range from 0 degF to 100 degF. So that each volt equals 20 deg. I know there are some pretty elaborate compensation equations. http://www.dataforth.com/catalog/pdf/an501.pdf

But what about just a simple CV for mV, then variables =1CV. 2CV=(I don't know how this would be done), it is sending out MV.

Any ideas?

Thanks again.
Dean

Good afternoon Dean,

Ok, I read that document and now my head hurts. So I read it again and realized that all that math is to correct for the thermocouple with voltage referenced to 0 deg C.

But if your device has been set up to operate between the temperatures of 0 Deg F (0 VDC) to 100 Deg F (5 VDC) then the thermocouple effect is going to be so insignificant we can forget it.

Proof:
K Type TC @ 0 deg F = -0.692 mV
K Type TC @ 100 deg F = 1.520 mV

Sensor scaling 1 deg F = 50*1000/100 = 50 mV. Therefore 1.52 mV maximum error due to thermocouple effects = 1.52 / 50 = 0.0304 Deg F

Now to the dataTaker stuff:

1. Set up a span S1=0,100,0,5000"degF"
2. Read the sensor as a high Voltage channel. e.g 1HV
3. Apply the span to the channel as a channel option. e.g 1HV(S1)

Cheers,
Roger