Calibration and Verification of Multicomponent Dynamometers in the Meganewton Range

Nowadays multicomponent dynamometers are required with capacities in the meganewton range, for a number of purposes ranging from improvement of existing force standards to routine checks on material testing machines. Owing to size constraints, elastic element configuration is almost invariably limited to single-billet style; advantage is taken of current strain gage technique for first order compensation right at bridge level of the influence of the main error originating sources. Transducer construction is therefore well within the reach of most laboratories having the required proficiency in strain gage work; commercially available instrumentation is more than adequate for excitation, signal conditioning and recording as required. Data processing is no problem either thanks to the ubiquitous microprocessor. The catch however is that for measurement exploitation purposes the relationship between applied load components and output signals must be known within a level of uncertainty consistent with the requirements of the problem at hand. As these may not infrequently entail evaluation of minute components in view of their compensation or elimination as for as possible, accurate assessment of some second order terms may then be necessary. At medium to low force/moment capacities, that is within the range of large wind tunnel testing work, say up to the order of 105 N and 104 Nm, the problem is only one of time, and cost. Facilities and procedures are available capable of yielding routinely all reasonable requirements at state of the art level. The picture however is a completely different one when the limits of these facilities are exceeded, and the outlay of the huge sums of money required to make a large capacity installation is out of question. In these conditions one finds that but a subset of the component combinations required for regular second order model fitting and validation may be applied, and educated guesses only instead of hard facts are available as basis for estimation of some effects. Theoretical analysis and accumulated experience are drawn upon to offset at least to some extent the lack of specific experimental evidence