IIW History 1948-1958

- 210

3. WORK OF THE COMMISSION. A. Dynamic calibration of fatigue testing machines .

When Commission XIII started " ·ork , its members realised that the fe\\· experimental results from different sources ,,·hich \Yere available \\'Ould, for many reasons, be hard to compare. They were therefore soon a\1·are of the necessity of trying to unify such results in order to allo1Y of comparatiye studies being later undertaken. This was in particular the case for the calibration of fatigue testing machine: . To compare the results of fatigue test s undertaken on different machines and in different laboratories, it 1vas necessary to know on what bases and by what me– thods the accuracy of these machines was evaluated . For this reason , at the Copenl1agen assembly in 1953, the Commission laid clown the conditions which t he methods used for the calibration of the machines should satisfy. In particu– lar, no inertia effects should intervene (11·hence the necessity to use resist ance strain gauges) nor should it be necessary to know exactly the gauge factor ; fin– ally, the method should be a null method. These recommendations led to the adoption , for the study of the calibration of machines, of the method described by lVIr. H. ROBER'l'S (1) . As a result of the application of this method , the members of the Commiss ion were better able to realise the mistakes made with their machines and the machine makers themselves were encouraged to reconsider this matter very ca refully . At the Zurich assembly in 1955, the Commission had the great satisfaction of learning that a general agreement had been reached on new guarantees of accu– racy, admissible tolerance of the load range being narrmYed t o a maximum of ± 2% fo r all hydraulic pulsating machines to be built from then on. This is an interesting example of the happy influence which the II\V can exercise in the technical field; all users of machines uI ll1is kind have been able to benefit from the progress thus made. B. Fatigue tests on welded assemblies . If it was necessary and urgent to study the problem of the calibration of testing machines, it 11·as also necessary, to be able to compa re the results of fatigue tests, to lay down general rules covering, on the one h and, the specimens themselves and, on the other, the conditions of fatigue t esting. With regard to the form of sp ecimens, the Commission considered at the Goteborg assembly in 1952 a design with enlarged ends necessary fo r testpieces which have no zone weakened as the result of a weld. _.\. de ign \\·ith a curve of increasing curvature was the subj ect of a recommendation of the Commission ; this design makes it possible to avoid inopportune fractures in the enlarged ends, and this by means of a relatively simple machining apparatus. \\ ith regard t o specimens containing a weld, it is not necessary to use enlarged ends since a prismatic specimen lends itself very well to t ests. Thi matter has remainecl from year to year on the agenda and, in 1957, the Commission wondered if it was r eally useful to take, as a basis of comparison fo r tests on welded assemblies, the fatigue strength of unwelded plate or other products. If the reply to this ques– ti on 1vere negative, the design \1·ith a curve of increasing curvature would ob– ,·ionsly Jose its intere. t. \;\hen the form of specimens 1rns bei11g tudied, it was recognized that, if a specimen \\·ere eccentrically mounted in the fatigue t esting machin e, stress con– centration s at least· equal to those in the connecting curve between the parallel section and the enlarged end s would result. As insuffici ent was known about

( t ) • illetallurgia »( Th e British j o11mal of .l1etals), .-\ugust 19y2 , pp. 10 7 t o l T-f.