Logo of the Physikalisch-Technische Bundesanstalt

Realization of Torque

Working group 1.22

Details of the calibration facilities, their measuring ranges and uncertainties as well as adaptation conditions, especially in the 1.1 MN·m Torque Standard Machine can be found on the page Measuring Devices. Here you can submit a

Request for calibration

Remarks to calibrations and calibration certificates

The uncertainty stated in calibration certificates is the expanded measurement uncertainty obtained by multiplying the standard measurement uncertainty by the coverage factor k = 2. It has been determined in accordance with the “Guide to the Expression of Uncertainty in Measurement (GUM)”. The value of the measurand then normally lies, with a probability of approximately 95 %, within the attributed coverage interval.

Calibrations on the 1.1 MN·m/220 kN·m torque standard machine
are performed in only one mounting position of the calibration object. In principle, several mounting positions would be necessary to investigate - on the one hand - the influence of form and position deviations of the sensor and of the machine and - on the other hand - the influence of mechanical disturbing components (transverse forces, bending moments, axial forces which are due to, among other things, the deadweight of the system) on the measurement signal. However, the measuring device has no space-bound axis for the generation of the torque and thus no defined position deviations. During the measurement, the position of the lever on the driving side is actively influenced in such a way that the disturbing components are minimal. The signals required for this purpose are furnished by specific elastic coupling elements on the measurement side. Except for the form deviations of the sensor and its deadweight, all mechanical influences are minimized.
For experimental back-up of these assumptions, comparison measurements have been carried out with different sensors and in different measuring ranges with one or three mounting positions. It turned out that the repeatability error of the results in the case of one mounting position reproduced that of three mounting positions. This would justify a calibration with only two measurement series in one mounting position. For a better statistical back-up, calibration is, however, performed - in accordance with the procedure of DIN 51309 - with three measurement series (the first one of these with an additional repetition of the increasing series) in one mounting position, and the parameter "reproducibility" b(MK) is determined in this mounting position.

Calibrations according to DIN 51309:2005
The factor 4 in the indication for the creep is an estimate which has been determined by experiment. For its determination, results of creeping measurements obtained over 20 minutes with more than 50 different sensors were put into relation to values of the short-term creep which were available at the same time. Short-term creep is zero creeping during a waiting time of three minutes between the end of a preloading and the start of the subsequent measurement series. The factor was determined as a rounded integer amount, without taking a negative sign into account.
It turned out that the distribution of the ratios had a pronounced maximum at 4, although values up to 7 were also possible. No value was smaller than 2, approx. 96 % of the values were in the range between 2 and 6.
(Source: Peschel, D.; Röske, D., Determination of Creep Value Using Short-Term Creep, Proceedings of the XVth IMEKO World Congress, June 13-18, 1999, Osaka, Japan, vol. III, pp. 245-249)

To top