Some number of years ago I was called in to investigate a situation where a batch of springs (4.5-in diameter) had been rejected by our QA department because they had the wrong spring constant. The vendor of the springs disputed our findings and had their own test results to support their claim.

Normally, we have a pretty strict calibration program and our gage calibrations are traceable to NIST (National Bureau of Standards, back then). Indeed, I found that our spring gage had been sent out to have its calibration checked and certified. Our test results had come from a different organization that had used an Instron machine.

So, I talked with the test person who ran the Instron machine. He showed me how he ran the test and got the same results that he previously got.
"How do you calibrate the Instron?" I asked.
"I push this button," he said, pushing a button on the Instron machine.
"What about an external standard," I asked.
"You don't need one," he said, "it's internally calibrating."

Sure enough, an external standard showed that his Instron was out of calibration. I think Instron would probably have said that the test person needed to have his understanding recalibrated.