Our full technical support staff does not monitor this forum. If you need assistance from a member of our staff, please submit your question from the Ask a Question page.


Log in or register to post/reply in the forum.

CR1000 compilation error "self calibration failed"


niranjan Oct 20, 2011 05:10 PM

Hi all,

While sending a program (which was written in CR BASIC) from logger net to Cr1000 the device showing an error such as "device self calibration failed and compilation failed". If we turn OFF the cr1000 and turn it ON back few times it works.
I am getting this error very frequently.

Can somebody provide me solution for this?

Thank you.

Regards
Niranjan


Dana Oct 20, 2011 06:28 PM

Most likely an attached sensor is outputting a voltage on an analog channel beyond what the datalogger can handle. Try removing sensors one-by-one to find the offending sensor.

Dana W.


niranjan Nov 2, 2011 07:11 PM

Hello Dana W,

Sorry for late reply.

1) Actually I am using 6 Omega T type thermocouple in differential mode and 4 ITI heat-flux sensors in single ended mode with approx a wire length of 100feet each. These sensors will produce voltage at rate few uV to mV. I think voltage applied to the analog channels is not going beyond the rated voltage. Except this is there any other cause that data logger raise such error.

2) The same program compiled few times with minor changes and the same connections were there on logger and it ran for 1 complete day. Later I changed the averaging time and I tried compiled the same error got raised and after many trials finally program got compiled. From last few weeks the logger kept it ON and today I try to see the data logged so far there are "NAN"'s in the data on random days for all channels and on few random days the data is valid (i.e the data is correct on all channels).

Please let me know any solution where I can exactly diagnosis the problem.

Thank you.

Niranjan


Dana Nov 2, 2011 10:55 PM

I would start by looking at grounding. It sounds to me like transient voltages might be coming in on the datalogger. I'm not the best one to advise in that, but you may want to refer to Section 7 of the current CR1000 manual on proper grounding.

Did you try removing sensors one-by-one to see if there is a problem sensor?

If the problem continues after you have verified sensors and proper grounding, you will need to contact the Campbell Scientific office that serves your area. The error message can also indicate a problem with the datalogger hardware itself, but given that the error is sporadic, I would still suspect some type of voltage spike rather than a hardware problem.

Dana W.


TweedleDee Nov 2, 2011 11:02 PM

Are you using the C (open input check) option for the range code?

Thermocouples tend to pick up a static charge. The C option pulls the inputs to ground before making the measurement, dissipating the static charge.


niranjan Nov 3, 2011 06:28 AM

Hello Dana,

Problem got resolved when I removed thermocouples which are approx 100mts in length and the sensors are used to measure outside roof temperature. What can be the problem? Is the length of wire matters here or any specific reason.

Hello TweedleDee,

Thank you for response. I have not tried this C option. Where can I find the information about C option.
I will try and get back to you.

Regards
Niranjan


niranjan Nov 3, 2011 06:51 AM

Hello TweedleDee,

As you said I am already using the option C in my program, so it might not be the problem. Following is line for measuring the thermocouple in my program.

TCDiff(Temp_C(),1,mV25C,1,TypeT,PTemp_C,True,10000,_50Hz,1,0)

Regards
Niranjan


TweedleDee Nov 4, 2011 04:30 PM

It appears that Dana's assesment is correct. You probably have different ground planes at the logger and the end of the sensor (it sounds like the sensor is grounded).

Try isolating the thermocouples from earth ground at the sensor end.


niranjan Nov 4, 2011 04:47 PM

Hello TweedleDee,

Thermocouples are configured in differential mode.
I am assuming that with differential configuration there is no matter of grounding issues.

Regards
Niranjan


aps Nov 7, 2011 09:55 AM

You will only get good measurements differentialy if both high and low voltages are within the logger common mode range. By this I mean if either the high or low input to the logger are outside the common mode voltage of the logger (a little over +/-5 V) for the CR1000, you will start to get errors in measurement initially and for wider range of input voltages all analogue measurements can be upset.

Inicidentally once you have resolved this issue you can reduce the analogue input range to 7.5 or even 2.5 mV and get higher resolution.

Log in or register to post/reply in the forum.