I have recently been completing an investigation into a temperature transmitter failure, of which the manufacturers I do not wish to mention. The failure was found during a calibration of the transmitter where the output had "drifted". I understand that these things happen, and I have had it in the past. Is there any explanation as to why transmitters in general drift? I know it could be a mixture of mechanical issues i.e. diaphragm stress' and electronic issues i.e. resistance changes. However, I was wondering if there was any definite answer for certain transmitters?

Thanks,

If you are saying that the device has drifted permanently then that could be due to faulty components inside the transmitter. All temperature transmitters use temperature compensation components and failure of one of these would cause permanent drift requiring re-calibration. All components inside the transmitter are affected by variations in temperature and it is the compensating components and circuit design that stops these variations from causing larger swings in accuracy.

Have you checked that the transmitter is operating within its maximum operating temperature?

alamatec,

Thanks for the response. Yes the transmitter is operating within its maximum operating range. You have given me a clearer understanding of the possibilities of what could go wrong electronically within the transmitter.

Thanks.

I agree with almantec on this, I had a similar experience with a batch of rail mount temperature transmitters on a Generator years back, and we had literally 40 out of 40 drifting up to 6 Degrees in 3 months.

Before we realised the problem we had been reducing load on the machine to bring the stator bar temperatures down as in the Production Managers view "You cant just have 40 failing" so there must be a fault in the stator bars.

We did an intense investgation into it and discovered that the transmitters (they were not programmable) were originally designed and calibrated for 0 - 100 DegC Type K thermocouples, the manufacturer, realising he didnt have the right stuff when he one the job for putting them in, just changed some components and then then recalibrated for 20 - 120 Dec C Type T thermocouples, the bottom line was that the inferior components used in the modification and the calibration adjustments literally on their max seemed to be the problem, and we ended up with a mass replacement.

What made it worse it that we had to drag that information out of him.

Not sure you would have the same problem, but just sharing my experience.