National Physical Laboratory

Can you provide any guidance or information as to the apparent difference in the results between the direct method and check plug method of screw gauge calibration? (FAQ - Length)

"I recently sent some screw gauges for calibration to an accredited laboratory, they found the gauges to be outside the specified limits when being measured by the direct method (i.e. measurement of individual parametric features such as pitch and taper etc.). I then sent the gauges to another accredited laboratory that certified that the gauges were within the specified limits using the check plug method (i.e. gauging with GO and NO GO plug and ring gauges). Can you provide any guidance or information as to the apparent difference in the results?"

This is a situation NPL has seen many times before. Check plugs and parametric measurements are not mutually exclusive. It is NPL's opinion that for checking master gauges (i.e. not products) both check plugs and parametric measurements have their place. Parametric measurements of pitch, taper, pitch diameter, minor and major diameter etc should take place first. Check plugs should finally be used to ensure that the measurements are consistent. The parametric measurements are more accurate but due to the fact that measurements are not made on 100 % of the gauge surface (the item may be gauged in two planes only) slight effects such as out of roundness and local bumps and burrs etc can be missed. The check plugs/rings pick this type of error up, as the gauges will not mate properly even though the parametric measurements suggest that the gauges are OK. Check plugs on their own cannot determine individual small errors in pitch, taper and diameter they only gauge the overall fit, for example if a standard requires cumulative pitch errors to be determined then this cannot be done using check plugs, also errors in pitch or flank angles can be compensated for by an increase in pitch diameter. We accept that for very small gauges parametric measurements may be impractical and gauging may be the only way, but for the majority of cases both have their part to play. Our recommendation is that you should first talk to the relevant accreditation body that carried out the measurements, as it is their job to sort out these issues and it is also useful for them to find out where things might be failing. Possible sources of the problem include the form of the gauge. It might be out of round or have problems as indicated above. Look at the results from the individual measurements to see if one parameter is close to its limit. If possible have the gauge measured at another position.

Last Updated: 1 Dec 2010
Created: 1 Dec 2010


Please note that the information will not be divulged to third parties, or used without your permission