Calibrating a temperature data logger

Thread Starter

drbenne

Joined Jul 30, 2013
48
I have some temperature data loggers that may need some calibration and received a response from the manufacturer on calibration procedures. Basically, I need a little help in understanding the procedures and results of calibration.

The procedure is as follows:
If you feed in zero mV and read off the temperature it should read exactly ambient. Enter this figure in the cell marked Lab Temperature (B11). This figures on the right hand side of each range are the mV values to feed in to the instrument to check the calibration.

If the resulting readings are not correct then there is a problem with the instrument. The only adjustment that is possible is to the reference voltage which is 4096mV. This is adjusted using pot RV2 also marked 4096. To check this, select the Meter function and press the Set button for 5 seconds. The analog circuits will then be switched on permanently. Measure the voltage across the ends of the pin header SK1 which is in the middle of the circuit board.

Note that in the attached spreadsheet, the column marked "Instrument reading" are the values obtained when calibrating an actual instrument.
Here is part of the spreadsheet of a sample logger that was calibrated:
(not shown: ambient lab temperature was logged at 23.1 °C in cell b11)




I guess my first questions are:

1) If the ref voltage is 4096 mV, why are the calibration values represented as negative uV?

2) When checking the ref voltage what does it mean by 'the analog circuits will be switched on permanently'?

These are Eltek 1000 series squirrel loggers if anyone is familiar with them.

regards,

Dave
 

wayneh

Joined Sep 9, 2010
17,152
1) If the ref voltage is 4096 mV, why are the calibration values represented as negative uV?
To make it extra confusing? My opinion is that changing the sign or the units - both of which are arbitrary choices - has no effect. Except to make one worry if they understand. Oh wait, maybe the unit change is as simple as the spreadsheet author or text author not knowing how to get the µ symbol into the sheet. Like how microfarads are usually given as mF on capacitors.

2) When checking the ref voltage what does it mean by 'the analog circuits will be switched on permanently'?
Can't help you there. What's a squirrel logger? Sounds more fun than a temp logger! :p
 

Thread Starter

drbenne

Joined Jul 30, 2013
48
Can't help you there. What's a squirrel logger? Sounds more fun than a temp logger! :p
haha, it's just the name of this particular series of loggers. They're manufactured in England... maybe they have fondness for squirrels or something?? :confused:

To make it extra confusing? My opinion is that changing the sign or the units - both of which are arbitrary choices - has no effect. Except to make one worry if they understand. Oh wait, maybe the unit change is as simple as the spreadsheet author or text author not knowing how to get the µ symbol into the sheet. Like how microfarads are usually given as mF on capacitors.
I guess I should've worded that differently. I can understand the units.

This is what I'm thinking:
The ambient was recorded at 23.1 C which would correspond to 0 mV being fed into channels 1-16. The negative μV values are referenced to the initial ambient temp reading.

I don't have the board in front of me right now, but I believe the 16 channels are divided into 4 separate multiplexers all being fed into a 12-bit ADC.

2^12 = 4096 (hence the ref of 4096 mV)

Does this sound about right and would 23 °C be a normal temperature to be calibrating most ADCs?
 

eeabe

Joined Nov 30, 2013
59
1) It sounds like the reference voltage is used for an analog to digital converter, and the calibration values are what you would measure on the SK1 header. If you adjust the pot to fine tune the 4.096V reference, you should see the output on SK1 change.

2) I'm guessing it just means it will turn on the analog circuit power for you to do the calibration and troubleshooting, and then it will go back to normal operation when you cycle power. It's pretty normal for sensors to turn on and off during normal operation to take samples and save power, but in that normal mode it would be impossible to troubleshoot with a DC meter -- you'd need an oscilloscope.
 
Top