When the inspection proves that these conditions have been properly provided for, it is necessary to slightly open the bankings so that there will be just a trifle of slide of each stone, on each tooth, after the locking takes place.
Extremely wide side shakes of the escape, pallet or balance pivots will sometimes cause striking of the roller jewel when conditions are otherwise correct, and these side shakes should not be very much beyond the extreme limits mentioned in this number. The fact of this feature, however, should not be construed as a recommendation that these pivots be closely fitted, for reasonable freedom is to be desired because it is positively necessary.
26. _Theory of Frictional Errors and the Isochronal Hairspring._
Theory teaches us in brief, that the position adjustment is made necessary principally because of frictional errors. It would therefore seem that if the watch was mechanically correct there would be little or no requirement for position alterations.
We are also advised that an isochronal hairspring is one which will cause the long and short arcs of the balance to be made in equal time and that to attain this, the center of gravity of the spring must coincide with the center of gravity of the balance and that a certain pinning point is necessary in producing this result.
Now if we have a watch of correct mechanical construction and fitted with an isochronal spring it would seem that a close rating timepiece would be assured.
27. _How Theory Works Out in Practice and What Isochronism Consists of._
Practical adjusting, however, proves that such is not the case, for even when the construction and alterations produce watches as nearly correct as scientific methods can determine, there is often considerable variation in the position rates. A twenty-four hour test in any position may prove that the long and short arcs are made in equal time showing the spring to be isochronous and yet the position variations have not been accounted for. In this connection experience proves that a spring showing a perfect isochronal rate may have its collet pinning point changed, in relation to the pinning point at the stud and that through such an alteration, a correction in positions can be obtained, without in the least disturbing the perfect isochronal rate.