Ken Roach
Lifetime Supporting Member + Moderator
Story Time
I worked in a factory where we provided a Fluke 754 documenting process calibrator as part of our machine, complete with a long document describing every step we took to calibrate and document our critical temperature sensors.
The customer promptly hid the Fluke in a desk and used their own calibrator, an ancient Yokogawa unit that required you to carry a flask of icewater for a zero point. In the summer, there was a helper whose whole job was to run back to the lunchroom and requisition ice.
They also used another Yokogawa voltage source unit to check our limit sensors, but half the time when they connected it they would leave the output selector at 24V and damage the Thermocouple input module on the PLC. This led to them complaining bitterly about what junk the PLCs were. This wouldn't have happened if they'd been using the Fluke and following our instructions. It certainly never happened our other customers, using identical I/O systems.
Eventually they figured out how to use RSLogix 5000 to attempt to calibrate the thermocouple modules they were damaging, insisting that they weren't damaged but rather mis-calibrated. Now the PLC calibration was off, too.
I visited the site with another process calibrator and set up a bench lab to compare my tools to theirs. Sure enough, that icewater-dependent Yokogawa unit was off by about 3 degrees F. I proved it with another independently calibrated millivolt source.
"Ken, that's simply not possible. This is our master reference instrument, and it is operated only by our master calibration technician. He has been doing this since before you were born. It is not possible for him to be wrong."
I know that this process won't be damaged by having all the temperatures be a few degrees too hot, so I simply gave up. But it was an important lesson on the weakness of objective fact when it runs up against cultural deference to authority.
I worked in a factory where we provided a Fluke 754 documenting process calibrator as part of our machine, complete with a long document describing every step we took to calibrate and document our critical temperature sensors.
The customer promptly hid the Fluke in a desk and used their own calibrator, an ancient Yokogawa unit that required you to carry a flask of icewater for a zero point. In the summer, there was a helper whose whole job was to run back to the lunchroom and requisition ice.
They also used another Yokogawa voltage source unit to check our limit sensors, but half the time when they connected it they would leave the output selector at 24V and damage the Thermocouple input module on the PLC. This led to them complaining bitterly about what junk the PLCs were. This wouldn't have happened if they'd been using the Fluke and following our instructions. It certainly never happened our other customers, using identical I/O systems.
Eventually they figured out how to use RSLogix 5000 to attempt to calibrate the thermocouple modules they were damaging, insisting that they weren't damaged but rather mis-calibrated. Now the PLC calibration was off, too.
I visited the site with another process calibrator and set up a bench lab to compare my tools to theirs. Sure enough, that icewater-dependent Yokogawa unit was off by about 3 degrees F. I proved it with another independently calibrated millivolt source.
"Ken, that's simply not possible. This is our master reference instrument, and it is operated only by our master calibration technician. He has been doing this since before you were born. It is not possible for him to be wrong."
I know that this process won't be damaged by having all the temperatures be a few degrees too hot, so I simply gave up. But it was an important lesson on the weakness of objective fact when it runs up against cultural deference to authority.
