upr000329 116

14 INTERSTATE COMMERCE COMMISSION REPORTS. f enter aside from the inaccuracy o f the scale itself, but in our opinion • that particular source o f error should not exceed the lim it above named. OPERATION. N o matter how excellent the scale, how perfect the installation, or how painstaking the maintenance, i f the operation be fau lty the re­sult is error. This record leaves no doubt that the methods em­ployed in the use o f track scales fo r the w eighing o f carload freigh t are such as would often produce gross inaccuracies upon the most perfect scales. These faults consist p artly’ in the w ay the cars are placed upon the scales fo r weighing, and partly in the recording o f the w eight itself. Cars may be weighed, and are habitually weighed, both in motion and at rest. T h e y are sometimes coupled to other cars at both, ends when weighed, sometimes coupled at one end, and sometimes entirely uncoupled. D ifferent carriers have different practices in this respect, and frequently the custom is different at different points upon the line o f the same carrier. Upon no subject involved in this investigation has the testimony been so sharply conflicting as upon this. M a y cars be weighed in motion, or must they be spotted? M a y they be coupled when weighed at one or both ends, or must they be entirely uncoupled? These are the storm centers around which the fiercest discussion has raged. I t would seem to be a self-evident proposition, and is the opinion o f those witnesses who are best qualified to express one w ith no bias either way, that the ideal method o f ascertaining the w eight o f a car is to place that car upon the scale, allow it to come to rest, remove everything from contact w ith it so that it stands alone upon the scale, and, having done this, take the weight. I f the scales are accurate this w ill indicate upon the scalebeam the true w eight o f the car. I t is claimed, however, that as a practical matter no element o f error is introduced i f this car be coupled to other cars at one or both ends. Undoubtedly in many cases, perhaps in the great m ajority o f cases, i f a car were placed upon scales so arranged that this could be done, coupled at both ends to other cars, and allowed to come to rest, the w eight would be substantially the same as though uncoupled. Th is clearly would be so unless there were some pull up or down at one or the other end o f the car. B ut it is equally manifest that the possibility o f error always exists i f this car be attached to any other car by a coupling which is at all rig id or may be made rigid. Th e element o f possible error is greater i f the car be coupled at both ends than i f it be coupled at one end. Th e length o f most modern scales is such that cars can not be weighed coupled at both ends, especially i f the cars are o f the smaller 2 8 1. CL C. 1 I N RE W E IG H IN G OF F \ REIGHT BY CARRIER. 15 sizes, since the scale is so long that the attached cars often stand upon one end or the other o f the platform . Th e practice, therefore, o f w eighing cars coupled at both ends is not as common to-day as fo rm e rly ; indeed, it seems 'to be at the present time comparatively rare when any serious attempt is made to accurately weigh the car. I t is, however, still a very common practice to w eigh cars coupled at one end, and the results o f many experiments were introduced tend­in g to show that no substantial error resulted from this method. These experiments have fo r the most part been conducted by the carrier, but there is no reason to doubt that they were honestly car­ried out and that the results are correctly stated. Nevertheless, we are not prepared to accept the conclusion reached, and the reason w hy w ill be best shown by an examination o f one o f these tests. Th e most extensive, and perhaps the most satisfactory o f all tests o f this character, was conducted by the W estern W eigh in g and Inspection Bureau, and involved the weighing, lig h t and loaded, o f 10,000 cars in different parts o f the territory covered by the opera­tions o f that bureau during various periods o f the year 1912. The head o f that bureau testified that he aimed to have this w eighing done at points where competent weighmasters were in charge. The car was first placed upon the scales coupled at one end. A ft e r being weighed in this position it was uncoupled, so as to be entirely free, and again weighed. Th e weighmaster had particular instructions to see when the car was weighed that it did not bind at the point where it was attached to the other car. Since the only error which can arise by virtue o f the fact that the car is coupled is due to its binding with.the car to which it is attached, and since in most cases, i f proper attention is given, no such binding w ill occur, it follow s that the source o f error from w eighing in this manner was largely elim i­nated by the w ay in which the test was made. Th e real objection to. this method o f w eighing cars is that unless the very greatest care is taken there w ill be an up or down pressure at the drawbar, and that, as cars are ordinarily weighed in actual practice, the necessary attention is not given. A n element o f inaccuracy is introduced which is not liable, to be guarded against when cars are weighed coupled, but the possibility o f which is entirely removed when they are uncoupled. Th e question is not, Can cars be accurately weighed coupled at one end ? but, rather, W ill they be so weighed as the opera­tion is ordin arily performed? A n exhibit was filed g ivin g the result as to each car, and an examination o f this exhibit shows that in instances, although they are not numerous, marked differences occur between the coupled and the uncoupled weight. In one instance this difference was as great as 4,000 pounds. I t appears, therefore, that even w ith the greatest 28 I. C. C. !>r>(i21—13-*A -2