Trouble with timing a Waltham model 1900 after hairspring replacement

[AndrewSi]

Hello everyone-

I have a 0S “Lady Waltham” model 1900 movement which I’ve just completed a rebuild of the balance- new staff to replace broken pivots and new hairspring to replace mangled one. The hairspring is a NOS 20277 “light” rather than the preferred “medium” but that’s all I was able to source.

I was able to ensure that the new hairspring is reshaped slightly to correct the radius for the overcoil segment affected by the regulator pins, and have also made certain that the hairspring is not interfering with the balance cock or balance arms, and is breathing relatively evenly all around so that coils aren’t touching each other in operation. In other words, everything looks pretty good and the balance appears to oscillate reasonably well.

When I put it on the grapher, it fails to lock on if I let it auto-determine the rate. If I manually set it to 18k, it runs a little longer before it gives up and tries to lock on again. I’m able to get it to briefly show an amplitude in the 180+ range and a sub-1ms beat error.  When I manually time the movement against a quartz watch I can see that it’s running about eight seconds slow per minute. Regulator position doesn’t make a significant difference. I am surprised that it’s that slow and with a pre-studded NOS hairspring it doesn’t seem like it should need to be shortened that significantly in the first place to get into the right range, should it?  At this point I’m not sure whether the failure to lock on and get a stable timing reading is due to the excessively slow operation or whether there’s another cause within the escapement, but I don’t see anything else obviously wrong in the pallet action the impulse jewel interaction with the fork, interference from the single roller guard pin, or anything like that.

Any thoughts or guidance as to the possible effects of the light hairspring on the timing or other sources of the failure to get stable timing on the machine would be greatly appreciated!  Thanks-

Edited July 2, 2024 by AndrewSi

[praezis]

Hello,

no need to search in distant places when you - I suppose - know the reason: you simply installed a wrong hairspring. Too weak hairspring makes a watch running slow. In your case 6s/min = 3 hours 12 minutes / day.

A timing machine can catch the beat rate if it is off by max. 15 min/day.

Solution is a correct hairspring (if that exists at all), making one (more than difficult by many reasons), or start butchering: you can adapt the balance wheel to your hairspring by removing weight. Remove a pair of weight screws and test if and how much it helped. Maybe looks a bit naked in the end 

Oh - I forgot the famous donor movement to rob a balance complete.

Edited July 2, 2024 by praezis

[AndrewSi]

I suspected that might be the cause but I’m definitely still a newbie and it’s good to get some confirmation. I’ll have to look for a donor and see whether I can address it that way.  Appreciate the reply-

[JohnR725]

16 hours ago, AndrewSi said:

mangled one

yes the problem of mangled hairsprings fortunately with American pocket watches you have replacement hairsprings. But they don't put a little star after the replacement hairspring and explain something in the fine print?

Classically with modern balance wheels the hairspring is individually vibrated to every single balance wheel. Which is why for the most part she would never can find a replacement flat hairspring. There are some minor examples out there where you can but for the most part you cannot.

But if you have an over coil hairspring because it was made to very exacting specifications then it comes first in the balance wheel is matched. this is why if you look carefully in the Waltham parts catalog or any other watch catalog for that matter their replacement timing screws. Even Rolex had replaceable hairsprings at one time with timing screws to match the balance wheel to your shiny new hairspring.

Then changing the strength introduces interesting new issues? Like the strengths of the hairspring was to match the various mass or simplistically weight of the balance wheel. So in addition to now having a match your balance wheel to its new hairspring the also have to change the mass of the balance wheel to correspond to the hairspring itself. Strangely enough I did this with an Elgin watch a week or two ago it's good that I have lots of Elgin timing screws.

just re-reading what you wrote above no you do not change or modify the hairspring its pre-made to exacting characteristics yes you have to change the balance wheel to match the hairspring.

9 hours ago, praezis said:

Maybe looks a bit naked in the end 

one time we had a incoming pocket watch that looked basically like that there were almost no timing screws at all. The owner of the store told the customer we would fix this problem but I did not. Because whoever had a bad hair spring day did an outstanding job of matching the balance wheel to the hairspring and the timekeeping was absolutely perfect and the watch ran perfect. If everything seems to be perfect it's just fine even if it was practically naked.