Electronics > Metrology

Fluke 5100B Calibration

<< < (5/5)

Verifying the performance of one 3458A with another 3458A has two problems:
1. The test uncertainty ratio of the calibrated 3458A to the uncalibrated 3458A (unit under test) is unlikely to be good enough to verify if the unit under test is in spec unless you apply lots of guardbanding, and hence are very likely to find it out of spec. See Fluke Appnote Calibrating the Hewlett-Packard 3458A DMM with the Fluke 5720A Multifunction Calibrator for more information on this. Try going through the math and seeing what you get. The 3458A calibration manual talks about having a 3458A option 2 (high stability reference) calibrated to the 90 day specifications, and that barely allows verification to 1 year limits.

2. You can't just connected two 3458As together and check if they match. You also need a voltage / current / resistance with the right value and frequency (in case of AC). And though the signal doesn't need to be accurate (the monitoring DMM can take care of this), the signal needs to be stable for the duration of the measurement. I've tried comparing two meters (not 3458As) against unstable sources and finding the relation over time between the two meters, but it's a hassle and no standards body would approve. The 5100B could provide most of the signals, except high frequency AC, but I highly doubt the stability and noise will be good enough. Maybe for 6.5 digit, but not for 8.5 digit.

best to check the 24h stability of 5100b with your 3458a at first, also the linearity can be checked


[0] Message Index

[*] Previous page

There was an error while thanking
Go to full version