There's a reason companies build fancy laboratories for calibrating gages and instruments. They have little or no choice in the matter. Good metrology, ANSI and ISO standards all require controls on the calibration environment, otherwise the measurements often are useless.
This is especially so when it comes to high-accuracy instruments or low-accuracy but big or large range instruments. And when it comes to fixed-limit plug and ring gages, the measurement uncertainties will climb outside of any acceptable range if the environment is not controlled.
Calibration service providers are frequently asked to calibrate all types of masters, gages and instruments on-site and have a difficult time convincing customers that this is not always practical. Customers keep asking for this because someone out there will take the contract and let the measurement errors and uncertainties fall where they may. Of course, I'm assuming they know what measurement uncertainty is, which may be a bit optimistic on my part.
In some cases, the customer's only interest is in getting pieces of paper to show an auditor that everything has been calibrated. Under these conditions, whatever is written on the paper is of little interest unless the word "Reject" is prominently displayed. In the absence of this red flag, the instrument or gage will be put to use, irrespective of whatever the data indicates about its shortcomings.
Surface plates, optical comparators, coordinate measuring machines and other large pieces of equipment have to be calibrated on-site because of the problems involved in moving them to a more pristine environment and getting them back intact.
Invariably, on-site calibration results in higher measurement uncertainties because of the environment, masters and other customer equipment used. This can mean that a device may no longer be suitable for a specific application because of the increased uncertainties that result from on-site calibration.
In an effort to overcome on-site environmental problems, it is not uncommon for a mobile home to be converted into a mobile lab. Overlooked is the soak time for the items being calibrated. Thus, while a desirable environment is brought to your parking lot, the items to be calibrated may not be in that environment long enough to reach the correct temperature.
Many companies ask for on-site calibration because they have a one-of-a-kind when it comes to their instruments and can't spare them for a few days for outside calibration. This is a problem. The easiest way around it is to start acquiring some duplicate equipment. An alternative approach is to talk to your calibration source. Many will make arrangements so that if you bring the equipment in by noon on Friday, they can prepare it for calibration and let it stabilize over the weekend for calibration first thing Monday morning. A similar process can be applied for normal workdays so you are without the corporate jewels for a minimal amount of time.
Another reason companies want everything done on site is because they've let their calibration program slide and an audit is due within a few days. Fast, on-site calibration will not save your bacon if an instrument that is overdue for calibration was used to verify some parts. On the other hand, if the instrument has not been used beyond its calibration date, put it in quarantine and get it calibrated later.
I would be remiss if I didn't mention that some folks believe on-site calibration is cheaper than using an expensive laboratory; this may be true in some cases. But always remember that the final product, the data, is not the same or no one would go to the expense of building and maintaining such facilities. Many labs maintain service facilities or replacement parts for popular equipment they calibrate, and the benefits of having immediate access to these facilities and parts is not available on-site.
If you're going to spend the money to get your equipment calibrated and want the best value for your money-results that are useful with low measurement uncertainties attached to them-get the work done in a proper laboratory. Limit on-site work to those items you can't easily ship to such a facility.
Report Abusive Comment