"Research" Equipment

[62EE]

I work in the Air Force Research Laboratory as an electrical engineer (I did PMEL for about 10 years enlisted).  We are wrestling with this new category of TMDE, "Research" equipment.  Does anyone have any experience, tips, or advice on how to handle this category of equipment.  To me, my fellow engineers are using this as an excuse to not have to take equipment to PMEL--they think they know how to calibrate equipment.  What is the point of RES equipment other than to give AFRL an excuse to not pay AFMC PMELs for calibrations?

[OlDave]

I think I pretty much understand both sides of this question 62EE. As PMEL for 25 years and now metrologist for a research institute I kind of have to say "I feel strongly both ways".

Many times an item is purchased to be used in the development phase of a project or process and it has many capabilities or functions that we don't have any need for. I need the sensitivity, resolution, stability, impedance, or (insert your favorite word), but I really don't care a great deal about absolute accuracy or all the other whiz-bang features on it. So why should I have the unit calibrated if what I'm using it for may in actuality be a non-calibrated function of the instrument. And many times continuity of a test is much more important than breaking a test routine because of some arbitrary recalibration date that is set for no reason other than it's the default 1-year.

But as the test project evolves and this dedicated instrument is now used to refine a process, then the actual accuracy may become more critical. Or it may be reassigned completely to a different project with different needs entirely. Then it probably makes sense to calibrate it on a regular basis.

As an example, just yesterday I needed to test the uniformity and ramp rate of a heater block. I chose to use miniature type T thermocouples. Not a single one of them were calibrated. I did my testing and then tested the thermocouples with their readout devices in a calibrated bath at the 2 temperatures I was testing to and applied the necessary corrections to my results. Were the thermocouple calibrated? Kind of. Did it matter if the ice point was calibrated since my temperatures of interest were 37 and 56°C. No. Could I realistically have had this done at the local PMEL? Taking all my thermocouples and still attached readouts and having 2 temperature charted by the local lab that same afternoon? Not a chance!

But if I were to use those same thermocouples for routine work over an unspecified temperature range I would most likely have them calibrated regularly.

So you see it really depends more on the process and application, than the item itself in a research environment. How do you deal with that on a large scale though? That I don't have any answers for. The narrow "though SHALL calibrate everything" isn't the answer, nor is the "ahhh, who cares" attitude.

I would be interested in hearing how other folks have dealt with similar issues. Any feedback?

[scottbp]

When I was going through college for my EE degree, we used all kinds of test equipment for our classes, and the research labs had tons of equipment, and as far as I could tell, not a single one of them was calibrated, save for Geiger counters, dosimeters and other radiation monitoring equipment used in the life sciences and nuclear medicine labs; but that was because it was mandated by the radiation safety office, which had a higher mandate above them-the Nuclear Regulatory Commission. When I inquired about the other equipment, they said they didn't worry about it because it was for training purposes only, and the original calibration from the manufacturer was good enough. If an instrument was broke or the readings were off enough to where it was obvious, it was sent downstairs for repair, but there was no calibration stickers with due dates, because there was no mandate for it. They didn't manufacture anything so there was no need for an ISO style quality program. (It made me wonder how much research was done whose results were inconclusive due to bad data from uncalibrated instruments!  )