I have voltage offsets on three channels and I do not know how to get rid of them. These channels are on an AM25T connected to a CR1000 running OS 27. Nothing is connected to the EX port on the AM25T as I'm making no thermocouple measurements. On the AM25T, the 'L' and ground terminals for these thermopile measurements (e.g., pyrheliometer, pyranometer) are jumpered. I am doing full 60 Hz integrations and input reversal. I have no Calibrate() instruction in the code and there is no compile-time background calibration warning error. Both instruments are less than two years old. I have another redundant CR1000, AM25T, and instrument set that do not have these offsets.
At night, when they should read 0, these channels read:
Pyrh1_mV = 0.06 mV or 0.064 mV
Pyrad1_mV = 0.04 mV
Pyra1_mV = 0.033 or 0.037
CalGain(10) and CalDiffOffset(10) in the Status table correspond to the differential measurement, 25 mV range, and 60 Hz integration combination, but CalDiffOffset(10) = 0, so that does not appear to be the cause. The Lithium battery voltage is 3.500221 V and the CR1000 temperature is 32.7 C, which both appear normal. My program instructions and calibration coefficients from the status table are as follows:
Const AM25TDiffCh = 1
Const MUXClkPort = 4
Const AM25TResetPort = 6
Const AM25TIsInRev = True
Const FullPyrSetl_us = 10000
Const FullAM25TInteg = _60Hz
AM25T(Pyrh1_mV,1,mV25,2,AM25TDiffCh,-1,AM25TRefTemp_C,MUXClkPort, _
AM25TResetPort,0,AM25TIsInRev,FullPyrSetl_us,FullAM25TInteg,1,0)
AM25T(Pyrad1_mV,1,mV25,3,AM25TDiffCh,-1,AM25TRefTemp_C,MUXClkPort, _
AM25TResetPort,0,AM25TIsInRev,FullPyrSetl_us,FullAM25TInteg,1,0)
AM25T(Pyra1_mV,1,mV25,4,AM25TDiffCh,-1,AM25TRefTemp_C,MUXClkPort, _
AM25TResetPort,0,AM25TIsInRev,FullPyrSetl_us,FullAM25TInteg,1,0)
Status table, non-zero calibration constants:
CalGain(1) = -1.345442
CalGain(2) = -0.6781959
CalGain(8) = -0.3390979
CalGain(9) = -0.06753144
CalGain(10) = -0.006698291
CalGain(11) = -0.002019229
CalGain(12) = -0.000675727
CalSeOffset(2) = 6
Thanks for any help.
Hi,
it often happens to us with solar radiation sensors and believe it is a technical issue with these sensors, especially if thermopile (I think as a matter of heat exchange).
Try putting a short circuit in place of the sensor to see if this happens anyway.
We solved it, with a filter in the logger program, so below a certain level of mV and if is night, then the value is set to zero.
If it is a question of the sensor there are more complicated solution, the type using the internal temperature sensor or ventilation systems of the body sensor.
Smile
Thank you, but I should have specified, these are likely not net-IR offsets as these instruments are pyrheliometers and black-and-white pyranometers which theoretically do not have them. I also doubt it is a temperature effect as we have an identical set of instruments on the same tracker, connected to a different CR1000, and we see no offsets.
I did just try installing a jumper between the H and L and also between the H and L and the L and ground, but I still see the offsets. However, I did notice that the channel originally read 0 and then after about 5-10 s (we are doing 1 s samples), it grew to the aforementioned offset values.