Justin, I agree with most of what you are getting at. However, calibration is legitimate for high accuracy instruments which are used to the limits of their specifications. This is because of ageing of critical components inside. Practical voltage references in portable instruments, for example, exhibit age-related drift which is not entirely predictable. To get ppm accuracy from them it is necessary to adjust them periodically to match sources which drift much less (i.e., primary standards). This is not necessary for 1% or even 0.1% accuracy for modern voltage references. Calibration can certainly be done poorly (like performing the cal on an instrument with a dead battery). Calibration is also often applied willy-nilly, even to instruments which do not need it (like your laptop example). In my view, it would be entirely legitimate to establish an internal company calibration test system, like you suggest. You get one high accuracy meter professionally calibrated once per year and then you use it to just check the calibration of all your other meters. However, someone would have to be in charge of doing that and keeping a record that it was done. Many companies decide it is better to just have a cal lab handle all of that. The risk that you are mitigating by this checking process is mainly that a meter could malfunction in a way which grossly affected the measurement accuracy but which was not otherwise evident. You are correct that there is really no need to perform a periodic adjustment on a general purpose multimeter or thermocouple interface and most likely the cal labs (at best) simply check these and they are always in spec so they just replace the cal sticker and do not make any changes. A well-managed cal lab should provide you with the opportunity to review the record of what they actually did to each instrument. Many will probably say something like "condition received: in spec. Adjustments made: none. Condition returned: in spec" They should NOT put a cal sticker over the battery compartment. Replacing the battery should NOT void the calibration. That is just a sloppy mistake on their part. Sean On Sat, Nov 16, 2013 at 10:54 PM, Justin Richards wrote: > OT, because I want to rant. > > I attempted to use on the very day a Fluke K-Type thermocouple multimeter > adapter is returned from calibration. I find the battery is flat (very fl= at > and very old) and do minor damage to the case to pop open to replace > battery and eventually discover that they have covered the screw over wit= h > a cal sticker. WTF. Does this mean technically it is out of cal because= I > have removed the sticker. How did they cal it, did they replace battery, > cal, then replace with old battery. > > They cover multimeter battery compartments with cal stickers, again, is i= t > out of cal if I need to change a battery. > > HP signal generators no output at some levels straight after cal. It cos= ts > big bucks to cal these items. > > We have so many items that get calibrated each year we never use. I > suggested we just cal one multi-meter as we mainly use them for go no go > and any reasonable tech will verify it operability before relying on it. > But oh no, "cant have an un-calibrated tool in the workshop, never know > what issues that could cause" was the response. If it needs precision > adjustment or commissioning we know we need model number/ serial number > /barcode/ cal date etc. > > I accept commissioning tests require calibrated documented procedures but > for day to day repair (which is often getting rough battery levels or > seeing if there is power or no power) it is over the top. > > Even had a fluke meter that was destined for disposal because the cal > report read "Unable to calibrate, 30mA fuse blown" and because the people > running with the cal from our end are not tech savvy just read "Unable = to > Calibrate". The effort required to save that meter from destruction was > far more than the $500 to replace it but for me it was the principal. I > was not popular. > > In Telstra they would calibrate the laptops. This practice has now stopp= ed > but what were they thinking. The argument was that it was a tool on the > tool asset list so needs to be calibrated. > > Part of commissioning tests involved measuring the height of a keyboard > that had to be between 900mm and 1200mm and i had to provide details of t= he > calibrated tape measure that I used to measure this. Yeah, right. > > So my blood boils when the topic of Cal comes up. > > Deep breaths... > > Justin > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .