When headlights are aimed, even with the help of modern aiming equipment, the aiming is not absolutely correct nor does the "correct" aim given by one instrument necessarily agree with that given by another which works on a different principle. Headlights which have just been "correctly" set by means of one of those instruments may therefore be rejected at a testing station unless the tolerances allowed by the testing station are sufficiently wide. A tolerance range for horizontal aim of about 1 degree centred on the straight ahead direction is shown to lead to the rejection of between 3 and 50 per cent of lamps when existing equipment and procedures are used. Vertical aim though essentially more accurate than horizontal, is complicated by the effect of changes of loading. It is expected that even here, with the same tolerance range of 1 degree the rejection rates would be rather high for vehicles which show large deflections with full loading. If the tolerance range is not centred on the "correct" aim large rejection rates -50 per cent and more may be inevitable unless those who set the lamps deliberately ignore the aiming instructions in order to diminish the change of subsequent rejection. Such a situation is to be avoided.
Abstract