I am currently calibrating PXI 4071 cards using two different versions of calibration executive. On version 3.3 I am getting failures for the 3A range 300mA reading at 1kHz for all cards I have tested thus far. On version 3.1.2 I test the same cards and receive failures at 70 V AC at 300kHz but none for the AC current values. I have tried adjustment on both systems and receive the same results. I have checked the output of both calibrators 5720A with 3.3 and 5700A with 3.1.2 I am using and the outputs are correct. The calibration interval on both are 90days as well as suggested by Calibration Executive. If I use the test panels on either system I receive valid readings that agree to within .2 mA with the autozero, null, minimum frequency of 500Hz or 1kHz, and filter at 8 readings on the current readings. I am trying to find a reason why these two systems disagree. Any input would be greatly appreciated.
Different versions of Calibration Executive could be using different versions of the calibration procedure. This means that certain values could have changed. To confirm that this is the cause, I would need Calibration Performance Test Data from both versions of Calibration Executive for the same card. I would recommend running the calibration again and making sure that the performance test data does not contain any personal information. In general, you should trust the latest version of Calibration Executive, since it has the latest in updates from National Instruments.
The 3A AC current limits have never changed on our manual procedure, so the behavior you're seeing is not expected. The manual limits have always been:
Can you manually test these limits using the 4071 calibration procedure to see if both machines yield the same limits under the same test conditions?
Can you provide us with a report of a failure on both versions of Cal Exec? Does the calibration fail during adjustment or the verification after adjustment step? Are the readings failing marginally outside the spec or grossly?
Here is a snapshot of the test point results that are failing. The out of tolerance indications are around the same magnitude for every card so far which is why the readings are suspicious.
700 Volts AC 300000 Hz 70.000 68.250 72.091 71.750 Failed 68.250 72.309 71.750 Failed
700 Volts DC 300000 Hz 70.000 68.250 72.087 71.750 Failed 68.250 72.308 71.750 Failed
3 A 1 kHz 300.00mA 299.10mA 297.43mA 300.90mA Failed 299.10mA 297.30mA 300.90mA Failed
The bold values are the before and after measurements.
If you measure the boards outside of Calibration Executive, say in Soft Front Panels, do the 'after' readings match up well with what Calibration Executive is reporting? Also, how many unique boards are you seeing this behavior occurring? Are there any boards that you have tried where you are able to successfully adjust 700 V AC/DC 300 Khz testpoint? If you are able to send a whole report we maybe able to see if there are trends in the measurements which may give us hints on what could be going on.
Any updates on the soft front panel measurements Brandon mentioned?
Any tests I perform on those units and the ones I have had in my lab since present different readings than are obtained in Calibration Executive. We have successfully adjusted a marginal amount of units (3 of 70) and even those barely passed reverification. This is not just localized to my lab either. Similar problems are present in other labs. I do not doubt my standards are correct. I am trying to narrow down the problem to Calibration Executive or the cards. The most pressing question for now is why does Calibration Executive disagree with the manual measurements.
Any chance you can provide a report of the failure from both versions of Cal Exec?
Also, I know this may sound unrelated, but what version of the NI-DMM driver are you using? I ask because there was a known issue with NI-DMM 3.0 (CAR# 1955588) that caused 407x DMMs to fail during calibration after performing a self-calibration on the device. It might be worth upgrading to a newer version of NI-DMM and see if this helps: