40 schunk: hysteresis of indicating instruments 



within a wide range by simply varying the point of appHcation 

 of the load upon the platform in a direction parallel to the plan 

 of the platform-supporting linkwork.^ 



4. As a consequence of the large frictional resistances, the 

 variance of the indications of such a scale is so great as to be 

 very easily observed and has, in common with that of many 

 other crude and ill-constructed instruments, hitherto been 

 deemed quite incapable of systematization or regularization. 

 (Such scales are in fact so "unreliable" in their indications when 

 read in the ordinary way, that their general use in trade is com- 

 monly prohibited by regulation.) 



In order to afford accurate control over the position of applica- 

 tion of the load, which, as has been stated, is a factor largely 

 determining the aggregate amount of hysteresis of scales of this 

 type, the standard weights used were suspended from below by 

 cords secured at opposite ends of a light horizontal bar super- 

 imposed on the platform. Projecting downward from the cen- 

 ter of this bar was secured a rubber stud or tip serving to localize 

 the load within a small and determinate area on the platform 

 while holding the bar itself above the platform and free from 

 contact with it. 



Since such scales are not artificially damped, gradual move- 

 ment of the platform and pointer to their positions of rest at 

 the several loads was provided for in the one direction by gen- 

 tle and careful appUcation of the weights manually, and in the 

 other by the use of the relatively slow motion of the rack-and- 

 pinion gear of a microscope stand by which a vertical rod was 

 gradually withdrawn from contact with the upper surface of 

 the platform after the removal of a given weight increment. By 

 these means it was possible to assure aperiodic, reasonably 

 steady, and shockless motion of the pointer, without overshoot- 

 ing of the rest point. The reasons and basis for this method are 

 discussed in detail in the first reference cited. Regularization 

 of operation of the scale, or the setting up of the cycUc state, 

 treated in the same reference, was carried out before beginning 

 the experiment by simply operating the scale a number of times 



2 See Bureau of Standards Technologic Paper No. 106. 



