Timegrapher reads slow, watch runs very fast?

[cortman]

Hi all,

I have a puzzling problem with an Elgin 16s Grade 211 model 7 pocket watch I'm servicing.

The service went well- the only part that required replacing was the hairspring, as it broke at the stud during the cleaning. I ordered a new one from Cas-Ker, matching the exact Elgin part number (769).

This is the part I ordered: https://www.jewelerssupplies.com/elgin-16s-hairspring-03-4254-959.001.0008.html

This is the parts list I cross-referenced: https://pocketwatchdatabase.com/search/result/elgin/9704768/parts/Springs

The re-installation went perfectly and the spring is operating correctly, with no sticking or abnormal behavior.

I put the watch on my timegrapher, and it automatically selected a beat rate of 19800, which surprised me as I thought these models were 18000 bph, per this link: http://home.elgintime.com:8080/elgintime/GnumLookup?GR=211

After adjusting for beat error, the timegraph showed much what I expected- low amplitude (some bushings were rather worn) and therefore losing some time, with a consistent trace.

Upon observing the handset in action over a 24 hour period, though, I found that it was *gaining* almost 20 minutes a day. This in spite of the timegraph continuing to consistently read about 1 minute slow.

I spent some time checking and observing the hairspring again, but I found no sign of sticking or rubbing on the balance bridge or the balance wheel arms.

I checked the handset with a timelapse video to make sure it wasn't loose or skipping or something, and the action was smooth and as it should be.

I checked the balance wheel action with a slow motion video, and by counting the vibrations as accurately as I could I think it actually is running at 19800. Video here: https://photos.app.goo.gl/Vz7V91bUwFt27LFSA

Do any of you know what would cause the watch to run so fast, while giving a low reading on the graph? And what could be causing it to run so fast, when it should be running slow?

Thanks!

Edited December 13, 2022 by cortman

[Marc]

My understanding is that the oscillation frequency of the balance is dependent on the combination of balance wheel and hair spring parameters, although I'm happy to be corrected if I have this wrong. If all else is in order then my guess would be that the H/S is too strong (or possibly too short) resulting in lower amplitude and a high enough rate for the TG to put it in the 19,800 range.

The TG thinks it is a 19,800 because it is hearing 5.5 ticks per second, however the train is set up for only 5 ticks per second (18,000bph) so when it gets 19,800 beats in an hour the hands will have moved 66 minutes. This should result in a gain of 2 hours and 24 minutes in a 24 hour period so the numbers don't add to what you are seeing but I suspect that the TG is not getting particularly clean or consistent data to base its values on.

Try manually setting the TG to 18,000bph and see if the trace corresponds more with what the hands are telling you.

Did you vibrate the new H/S to the balance before you fitted it into the watch?

[cortman]

11 minutes ago, Marc said:

Try manually setting the TG to 18,000bph and see if the trace corresponds more with what the hands are telling you.

I have done this, actually- it produces an odd "tick-tick-TICK, pause, tick-tick-TICK, pause" and gives up after a bit without recording any data. It sounds completely out of pace with the actual ticking I hear from the watch.

Quote

Did you vibrate the new H/S to the balance before you fitted it into the watch?

No I did not, I purchased a complete hairspring assembly with collet and stud already attached.

Quote

My understanding is that the oscillation frequency of the balance is dependent on the combination of balance wheel and hair spring parameters, although I'm happy to be corrected if I have this wrong. If all else is in order then my guess would be that the H/S is too strong (or possibly too short) resulting in lower amplitude and a high enough rate for the TG to put it in the 19,800 range.

Since the hairspring I purchased is the correct, original model for this watch, would I be safe in assuming the watch has had the balance wheel replaced at some point with one non-original? Would I be able to slow down the rate  using balance screws?

Thanks for the reply!

[Marc]

3 minutes ago, cortman said:

Since the hairspring I purchased is the correct, original model for this watch

Individual hair springs are normally matched to individual balances since there will always be slight variations from one H/S to the next, and one balance wheel to the next. These variances can combine to either cancel each other out or to produce a significant error in the rate. This is probably why they are sold as "balance complete" which is a balance and H/S which are matched to each other.

8 minutes ago, cortman said:

Would I be able to slow down the rate  using balance screws?

You may be able to slow things down by adding weight but I'm not sure how that would impact amplitude. My gut feeling says it won't be good but it would be good to hear from someone with a more solid understanding than I have.

[RickTock]

Details are a little fuzzy in my brain, but I had a similar situation with a pocket watch that was badly magnetized. Demagnetizing put the beat rate back where it should be on the Timegrapher and addresses the massive daily time gain. Worth checking if you haven't already.  Best of luck.

[Klassiker]

Check the number of teeth in the train to be sure of the bph. I suspect it should be 18000, and your new hairspring / balance combination is beating much too fast.

[nickelsilver]

American watch manufacturers offered replacement hairsprings in several "strengths". I'm pretty sure assortments were available back when the companies were still in business, which would be handy as there's really no way to know which strength you need without trying it out. Even with a good match it might be necessary to add timing washers or something to get things right on.

 

Sets of timing washers usually give a rough idea of the rate change for a pair of washers for a certain size watch/balance, and for a given size you will usually see that there are 3 choices, 1min, 2min, 3min. In your case you might be able to fit 6 pairs of 3min washers (to opposing screws by pair, and spread over 6 pairs of screws if possible). That's a lot of timing washers, but it might be your only option.

[JohnR725]

On 12/13/2022 at 5:57 AM, cortman said:

exact Elgin part number (769)

American watch companies are interesting with part numbers. So for instance Elgin has an 18 size balance staff numbered 857 that's a part number that's a exact part number. But the staff comes in four different versions and every single staff has a variation in size and that's not a pivots size variation every single staff is different in each of the four categories. Then every single one of those has pivots size variations. All with one exact part number which translates into how many different parts?

On 12/13/2022 at 6:50 AM, cortman said:

would I be safe in assuming the watch has had the balance wheel replaced at some point with one non-original?

No that's not a good assumption at all for one thing look at the balance wheel it usually has the serial number scribed on it which should agree with the serial number of the watch but that's not the problem.

2 hours ago, nickelsilver said:

American watch manufacturers offered replacement hairsprings in several "strengths". I'm pretty sure assortments were available back when the companies were still in business, which would be handy as there's really no way to know which strength you need without trying it out. Even with a good match it might be necessary to add timing washers or something to get things right on.

 

Sets of timing washers usually give a rough idea of the rate change for a pair of washers for a certain size watch/balance, and for a given size you will usually see that there are 3 choices, 1min, 2min, 3min. In your case you might be able to fit 6 pairs of 3min washers (to opposing screws by pair, and spread over 6 pairs of screws if possible). That's a lot of timing washers, but it might be your only option

Yes this subject is come up before and I've explained this before but I'll explain it again.

On modern watches or watches with flat hairsprings each hairspring is vibrated to the exact balance wheel it's on which is why hairsprings are never available separately it's a balance complete or nothing.

On watches with an over coil hairspring to get the hairspring in the very specific shape it needs to be the easiest way to do that is to pre-make a hairsprings. That means the balance wheel has to be matched to the hairspring.

Then the various strengths of the hairspring were determined by how heavy basically the balance wheel is. So if you have the right hairspring for that balance wheel than it should be the right strength it just won't match for timekeeping. Then timing washers probably aren't going to cut it you going to need timing screws which were available at one time.

Then yes the watch is 18,000 beats per hour so you have to set the timing machine to that the problem is ill be way out of range and then the timing machine will be unhappy and not show the proper results. The amplitude on your display is unacceptable but it may be it's being influenced by the rate in the wrong setting such doing the wrong calculation so until you can get the rate close I'm going to assume the amplitude is probably not right on the machine very likely none of the rest of the numbers are going to be right

You'll notice in the attached image what does it say about timing screws? It even tells you how to figure out which timing screw you need although just as a reminder make sure your balance wheel is round. Often times with handling a bimetallic balance wheels they get squeezed that will screw up timekeeping to a certain degree. But not as much as changing the hairspring you just need some timing screws.

Then for Elgin part numbers just if you want to verify there is the link below yes it does look like that's the right hairspring now you just need to get some timing screws.

http://www.elginwatchparts.com/

 

 

[Nucejoe]

 It be nice if someone here had a damaged balance wheel with good big screws to donate to your watch, as you are very likely to need heavier screws to slow this balance by 20 min per day.

 

[JohnR725]

3 hours ago, nickelsilver said:

American watch manufacturers offered replacement hairsprings in several "strengths".

There is something that was bothering me at all of this? The early hairsprings had strengths of week medium and strong which is based on the balance weight. But if you look at the images off the parts listing it tells us the strength. So ask he have a strength number but is still going to need to use timing screws to bring it into time.

Then remember I said the American watch companies well one number fits all basically sort of so I snipped out a section of the 1915 Elgin parts catalog and you'll notice that your hairspring number came with various strengths. Then where you purchase they hairspring from they didn't indicate which strength that came in. So in addition to the hairspring has to match the balance wheel it apparently came in a whole bunch of different strengths. The worst-case if your strength was too far off the watch just won't run well you can still get at the time. Because as a reminder the hairspring strength aspect has to do with the hairspring has to unlock the escapement. The mainspring is the balance wheel to kick that they hairspring pulling the balance wheel back is what unlocks escapement to allow the watch to run so if your strength gets too far off you could conceivably have running issues.

But you seem to your running and you do of the hairspring so we can work with that.

Oh and how hard would it be to get an Elgin timing screw? Well just to search online Or look at eBay like for instance

https://www.ebay.com/itm/284556703567

Oh and if you do a search on eBay for timing screws just be careful that there for the right size.

Then we have this because of the way the Elgin watch factory was liquidated basically the entire factory was spread out across the countryside which is why new old stock is still available because it was such a huge quantity of it you can get assortments like this and if you look carefully those are timing screws. So I have assortments like this and I just use the screws out of them and they work fine the only catch is you don't get defined timing so you'll have to work a little harder At it as opposed to predefined screws

https://www.ebay.com/itm/304579739370

So basically you just have to look at all the Elgin screws on eBay there may be timing screws on the same site that you got your balance hairspring from mighty and look their.

 

[cortman]

Thank you very much, all, for enlightening me on the hairspring/balance relationship- I did not know vibrating was such an essential part of it. I've learned a lot in this thread. I have opted for simply buying a balance complete and dropping it in instead of vibrating and re-pinning the hairspring.