Picture tour of replacing a staff on a Waltham 6/0s wristwatch, enjoy!

SpringDriven

Registered User
Dec 22, 2010
130
192
43
All,

I am servicing a Waltham 6/0s wristwatch that suffered a broken balance staff and I thought it would be interesting to post some simple photos of the steps involved. Here goes!

1. Remove the hairspring at the collet with hairspring levers, very carefully. Note the position of the stud in relation to the wheel.
20210402_140644.jpg

20210402_141143.jpg

2. Remove the roller and safety table, I have a nice tool in my K&D staking set for that. Note the position of the impulse jewel.

20210402_141851.jpg

3. Next is drive out the staff. This is a riveted staff, so the rivet needs to be broken to separate the staff from the wheel. K&D tool again.

20210402_145923.jpg

What is not shown in the above picture is the stake that will drive out the staff. What is pictured is supporting the hub of the wheel to prevent distortion as much as possible.

4. Inspect the wheel.

20210402_150210.jpg

20210402_150643.jpg

You can see the remains of the broken rivet in the above photos.

5. Compare the new staff to the old staff. Take measurements.

20210402_150454.jpg

20210402_150737.jpg

I included the above photo for working scale.
 
Last edited:

SpringDriven

Registered User
Dec 22, 2010
130
192
43
6. Now it is time to rivet the new staff to the wheel. I use a round headed punch followed by a flat punch.

20210402_151516.jpg

7. Inspect the rivet. (It seems I neglected to take a photo of this step.) Inspect the trueness of the wheel.

20210402_153244.jpg

8. Friction fit on the roller table. I took note of the position, it is important to put it back where it was as close as possible.

20210402_154257.jpg

9 Friction on the safety roller.

20210402_154627.jpg

I do a quick cleaning of the wheel at this point before poising, to ensure nothing is influencing the operation.

10. Poise the balance wheel. As it is practically impossible to place the roller table back in the exact same spot, the weight distribution of the balance will be disturbed and it will have a heavy spot which will affect the positional rate of the watch.

20210402_163936.jpg

This poising tool allows me to find the heavy spot on the wheel, which I can then remove material from the rim in that spot to correct.

20210402_163905.jpg

You can see a deep hole and a shallow hole next to it. I did not want to go any deeper in that hole. Now the weight of the wheel is evenly distributed.
 
Last edited:

SpringDriven

Registered User
Dec 22, 2010
130
192
43
11. Check the flatness of the hairspring in relation to the collet. Sometimes those hairspring levers nudge the hairspring where it is pinned to the collet when you remove it.

20210402_165314.jpg

12. Install the hairspring on the wheel.

20210402_170007.jpg

13. Install the wheel in the watch to check endshake and the flatness of the hairspring and the concentricity of the coils, and that the impulse jewel is in line with the escapement so that it is in beat. (This was the best perspective photo I could manage.)

20210403_154007.jpg

All set, repair is complete and ready for the cleaning machine!
 

svenedin

NAWCC Member
Jan 28, 2010
1,150
192
63
Surrey
Country
Region
Thank you for sharing that. Very educational for me and I’m sure many others. It’s not something I’ll ever be attempting but I’m interested to see how it is done. My grandfather was a watchmaker and replacing staffs was bread and butter work for him in the days before shock protection.
 
  • Like
Reactions: SpringDriven

gmorse

NAWCC Member
Jan 7, 2011
12,339
2,038
113
Breamore, Hampshire, UK
Country
Region
Hi SpringDriven,

Thanks for posting your series of pictures on a procedure for replacing a balance staff.

I'll leave aside the matter of removing the old staff in item 3, because that's been the subject of much debate here in the past, opinions are still divided on the matter and I expect others will comment on that.

I do have a question regarding item 10 however. I'm curious as to how removing and replacing the roller(s) from very near the centre of gravity of the balance affects the mass of the rim to such an extent that it's necessary to remove material, especially when the balance in question has adjustable quarter screws. I appreciate that it would be difficult or impossible to test the balance with a broken pivot for static poise before removing the broken staff, but the question occurs to me as the whether it was ever in static poise when it left the factory, and indeed whether it was in fact dynamically poised then. If it was, then it would most probably not have been in static poise to start with.

Regards,

Graham
 

Skutt50

Registered User
Mar 14, 2008
3,792
213
63
Gothenburg
Country
I have the same question as Graham!

I normally try the balance in the movement after changing a balance staff, to check the positional errors. Often the movement keeps good time as is.
 

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
Static posing can affect the factory dynamic poising, so like others I always check the performance before doing any poising, after changing a staff.

As noted there are very different views on punching out staffs, and it's not something I would ever personally do. This balance was an ideal candidate for removing the staff with alum, and other than that I would use a lathe to cut it out from the hub side.

Cheers, Al
 

SpringDriven

Registered User
Dec 22, 2010
130
192
43
All,

The position of the roller table does affect the poise of the wheel. Since you can not practically speaking put it back exactly where it was, the poise will change when a staff is changed. Putting it back in as similar a position as it originally was helps, assuming it was originally poised to begin with. Consider that the impulse stone is on a table on the outer rim of that table. Therefore it is mass that is not in the center of mass of the wheel where the staff is. Since it is not in the center of mass, it does influence the the weight distribution of the wheel. When you change a staff you change the position of this mass, and therefore change the distribution of mass in the wheel.

As for poising an unadjusted watch? You probably don't have to. Why do I do it? It is part of the process I do when I change the staff of a wristwatch.

Why is poising a thing? The mass of the balance wheel needs to be evenly distributed throughout the balance wheel. If this is done, then gravity pulls evenly on the wheel as it is in rotation. If the wheel has a heavy spot, then this heavy spot can influence the inertia of the wheel in the positions that gravity can pull on it, and change the rate in that position. Eliminate that heavy spot and the rate of the watch in various positions will be closer to each other. There are other things you have to adjust for positions, but this is just focusing on the poise of the balance.

Dynamic and static poising have similar end results, you can even do both. They are done at different points in the service. Dynamic poise can save time if you skip the static poising. However, both are a means to eliminate a heavy spot on the wheel that will affect the rate of the watch in that position.

Dynamic poising is done with the wheel in the watch, but this is not clean work when the watch has already been cleaned, assembled and lubricated. It can be done quickly though and is a good way to reduce the delta and achieve good timing results. The intent is to remove material from the rim with the wheel in the watch. From my perspective this is risky, so I prefer to poise statically during the repair phase.

Timing screws use inertia to change the rate of the watch, much like a figure skater spinning uses their arms to change their rotational speed. Arms out, they spin slowly, arms in they spin quickly. The opposing screws should be equal in length in relation to the rim of the wheel so that their effect on the wheel is equal at varying rates of amplitude, assuming that the wheel is in poise. Timing screws are not poising screws. They also are not always near a heavy spot and may have little effect in trying to reduce the delta to get the timing of a watch in a position better, and the results will vary as the amplitude varies.

As for driving out staffs? It is one way that I learned, and I have not sustained any issues doing it this way. I can support the hub of the wheel as such that no damage is sustained in driving out the staff. If I did have an issue where I was unable to support the hub of the wheel, I would seek an alternate method.
 
Last edited:

SpringDriven

Registered User
Dec 22, 2010
130
192
43
I will state for the open forum that the process and photos I shared here is the process that I choose to do. I took these photos to share with others some of the work involved, I thought it would be enjoyable to see.

There are other ways to achieve similar results and I am not saying my way is the only way. We have different means to achieve a repair that we become comfortable with.
 
  • Like
Reactions: darrahg

gmorse

NAWCC Member
Jan 7, 2011
12,339
2,038
113
Breamore, Hampshire, UK
Country
Region
Hi SpringDriven,

Thanks for the comprehensive statement of your views, which I respect, but I fear we must agree to differ on this subject, because my understanding of dynamic poising is at variance with yours.

Regards,

Graham
 

SpringDriven

Registered User
Dec 22, 2010
130
192
43
Hi SpringDriven,

Thanks for the comprehensive statement of your views, which I respect, but I fear we must agree to differ on this subject, because my understanding of dynamic poising is at variance with yours.

Regards,

Graham
I see. Yes, after reading that link, I agree. Our views are at a variance.

I would like to ask one question. How do you regulate a watch, if you use timing screws to perform a dynamic poise, whether that is moving them or adding weight to the wheel? I am not 100% certain that if you moved a timing screw to achieve a better delta in the watch during dynamic poising, which is an adjustment, that you would want to touch it again to regulate the rate. Would you then not undo the work you did adjusting for poise?

Do you only move the regulating pins? What if you need to make a big adjustment? Do you finish the service with the regulating arm (assuming the watch has one) somewhere in the fast or slow zones and not in the middle, or way out of the zones? If it is a freesprung balance and you can ONLY regulate it by adjusting the timing screws, what do you do?

This is the only place that I can find variance in views that I can logically ask a question. I am not here to argue or create conflicts. My way is not the only way obviously and I am not saying anyone is wrong. But I am curious about my above question.

Perhaps the one thing I can do for myself is to do an experiment with roller tables and changing their position on the wheel to determine if it changes the poise of the wheel. I am under the impression that any time you restaff a wheel, any factory adjustment of that wheel is gone. The staff due to manufacturing tolerances has slightly different mass, the impulse jewel will be in a ever so slightly different position. These all have an influence on the wheel, there is no returning to the factory poise. All you can do is minimize the change by putting the roller table back in as same a position as the universe of variables will allow. This is what I believe, have witnessed etc. But I could do an experiment to verify the roller table having an influence.
 
Last edited:

gmorse

NAWCC Member
Jan 7, 2011
12,339
2,038
113
Breamore, Hampshire, UK
Country
Region
Hi SpringDriven,

One of the prerequisites for this approach to regulation is that the movement should be in as perfect a state as possible, which I believe includes the correct balance spring for the balance, pinned so that the rate is right in a horizontal position to start with. I think your proposed experiment would be a worthwhile exercise, as would some conversation with the author of the document, Dewey Clark, (DeweyC).

Regards,

Graham
 

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
I see. Yes, after reading that link, I agree. Our views are at a variance.

I would like to ask one question. How do you regulate a watch, if you use timing screws to perform a dynamic poise, whether that is moving them or adding weight to the wheel? I am not 100% certain that if you moved a timing screw to achieve a better delta in the watch during dynamic poising, which is an adjustment, that you would want to touch it again to regulate the rate. Would you then not undo the work you did adjusting for poise?

Do you only move the regulating pins? What if you need to make a big adjustment? Do you finish the service with the regulating arm (assuming the watch has one) somewhere in the fast or slow zones and not in the middle, or way out of the zones? If it is a freesprung balance and you can ONLY regulate it by adjusting the timing screws, what do you do?

This is the only place that I can find variance in views that I can logically ask a question. I am not here to argue or create conflicts. My way is not the only way obviously and I am not saying anyone is wrong. But I am curious about my above question.
Not sure where you get the idea that screw on the balance can't be used for more than one purpose. Clearly if you are adding or removing weight to make a timing adjustment or to "center" the regulator, you need to do so equally on both sides. If you are making an adjustment for poise, obviously you only add or remove from one side.

There are times when I've done both at the same time when the regulator is way off to one side, so I'll add a 20 seconds washer on one screw and 40 on the opposite.

I am under the impression that any time you restaff a wheel, any factory adjustment of that wheel is gone.
Not really. If you are careful, you can maintain factory poise when you change a staff. Let me give you an example...this is an Omega Cal. 1002 I serviced last year for a customer:

Omega vintage DeVille 3_0116.jpg

Unfortunately, the staff wasn't properly riveted, so the wheel was loose on the staff, and would rotate freely on it. The roller table and balance spring were solid on the staff, but the wheel wasn't as you can see in this video:


I didn't have a reference from the roller jewel to the balance obviously, because the balance was spinning. The only thing I had was a reference between the angle of the roller jewel and stud:

Omega vintage DeVille 3_0078.jpg

So when I changed the staff, I made sure to make those two features to the same angle relative to each other, and I looked on another balance of a slightly different caliber to determine where to put the roller jewel in relation to the poising slot and balance arms, and started there.

The result was a Delta of 5.6 seconds over 6 positions on the first try. If you want to move roller tables around to get an idea of the effect on poise, then let us know what happens. All I am saying is that I don't automatically static pose the balance after staffing. Yes, this is the way many people say to do things, but for me personally, it's worth putting the balance in the watch and seeing how it runs before I decide if it needs that or not.

Cheers, Al
 

DoughBoyWatches

Registered User
Jan 5, 2021
73
19
8
33
Country
this was awesome, I have done larger friction fit staffs for waltham/elgin 3/0 and 0, but I always had a hard time with the smaller 6/0 8/0 riveted staffs. Question how did you drill the hole? could a balance screw undercutter come in handy here as well or only if heavy spots happens to be by a screw?
 

karlmansson

Registered User
Apr 20, 2013
2,738
166
63
Linköping, Sweden
Country
this was awesome, I have done larger friction fit staffs for waltham/elgin 3/0 and 0, but I always had a hard time with the smaller 6/0 8/0 riveted staffs. Question how did you drill the hole? could a balance screw undercutter come in handy here as well or only if heavy spots happens to be by a screw?
It was interesting to see the process and the photo documentation is really high quality. So thank you for that SpringDriven!

That being said, I have a couple of issues with it as well. Although you pop the extreme edges of the rivet off, the remainder of the mushroomed balance seat will stretch and/or score the inside of the hole in the balance making the replacement staff a looser fit, leading to more potiential poising and truing issues. Should this need to be done more than once the cumulative error would spell trouble.

My preferred method is to turn the old staff off in the lathe from the hub side. If you make a female cone with the graver in the hub the last material of the hub will pop off like a washer and the balance can then be removed from the side of the staff that isn't riveted. The reason for turning from the hub is in part that the material there will not be as tight of a contact as on the riveted side and also not work hardened to the same degree as the rivet, making it easier and more controllable to turn.

Also DoughBoy, please read Dewey Clarkes document that is linked to in Grahams post above before drilling into a balance. I think you will get a clearer understanding of what that material removal can lead to. A balance that is in poise statically will almost certainly not be as good of a time keeper as a balance that has been dynamically poised for a certain escapement, and therefore appears to be out of poise when poised statically. You would be undoing someone elses painstaking regulation work only to have to do it yourself again.
That being said, I use a very small 120 deg point spade drill when the need does arise. I do, however, prefer reversible, non-destructive ways of poising when poising. Washers being my preferred go-to. Only on smooth balances would I consider drilling, and on those it usually comes down to them being damaged by improper poising attempts in the first place. Smooth balances are usually present in watches from the 1960s and onwards and the industrial poising standards from that time on has been rather spot on across the board. So if it ain't broke: don't fix it. (There are fantastic time keepers from earlier on than that but they were either specifically made to that standard and most have passed under the nose of many a watchmaker and enthusiast over the years. So chances are that material has been removed and then washers are the perfect solution.)

Regards
Karl
 
Last edited:
  • Like
Reactions: gmorse

DoughBoyWatches

Registered User
Jan 5, 2021
73
19
8
33
Country
It was interesting to see the process and the photo documentation is really high quality. So thank you for that SpringDriven!

That being said, I have a couple of issues with it as well. Although you pop the extreme edges of the rivet off, the remainder of the mushroomed balance seat will stretch and/or score the inside of the hole in the balance making the replacement staff a looser fit, leading to more potiential poising and truing issues. Should this need to be done more than once the cumulative error would spell trouble.

My preferred method is to turn the old staff off in the lathe from the hub side. If you make a female cone with the graver in the hub the last material of the hub will pop off like a washer and the balance can then be removed from the side of the staff that isn't riveted. The reason for turning from the hub is in part that the material there will not be as tight of a contact as on the riveted side and also not work hardened to the same degree as the rivet, making it easier and more controllable to turn.

Also DoughBoy, please read Dewey Clarkes document that is linked to in Grahams post above before drilling into a balance. I think you will get a clearer understanding of what that material removal can lead to. A balance that is in poise statically will almost certainly not be as good of a time keeper as a balance that has been dynamically poised for a certain escapement, and therefore appears to be out of poise when poised statically. You would be undoing someone elses painstaking regulation work only to have to do it yourself again.
That being said, I use a very small 120 deg point spade drill when the need does arise. I do, however, prefer reversible, non-destructive ways of poising when poising. Washers being my preferred go-to. Only on smooth balances would I consider drilling, and on those it usually comes down to them being damaged by improper poising attempts in the first place. Smooth balances are usually present in watches from the 1960s and onwards and the industrial poising standards from that time on has been rather spot on across the board. So if it ain't broke: don't fix it. (There are fantastic time keepers from earlier on than that but they were either specifically made to that standard and most have passed under the nose of many a watchmaker and enthusiast over the years. So chances are that material has been removed and then washers are the perfect solution.)

Regards
Karl
ahh timing washers forgot about those. I really hate swapping staffs but sometimes when Iam bored i do it on scrap balances just to get my practice up. Still very painstaking. I would also love to see a video on truing a balance I think the calipers I have arent very good I always have trouble with it.
 

Brian C.

Registered User
Jun 9, 2001
1,249
4
38
NH
Country
Static posing can affect the factory dynamic poising, so like others I always check the performance before doing any poising, after changing a staff.

As noted there are very different views on punching out staffs, and it's not something I would ever personally do. This balance was an ideal candidate for removing the staff with alum, and other than that I would use a lathe to cut it out from the hub side.

Cheers, Al
I always cut the staffs out. Punching them out can widen the hole in the balance.
 

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
I always cut the staffs out. Punching them out can widen the hole in the balance.

This is not true and in fact the converse was proven to me in Switzerland. We were taught to remove staffs either by use of the Molfres (which leaves a "rivet" on the balance underside that still needs to be punched/pressed out), or by pressing/punching out from the top. Whether it was a Rolex glucydor balance or a nickel balance, the replacement staff fit perfectly. This was a class of 12 people with a minimum of 3 balances each. Not one problem.

I have never seen any analysis to support the claim of probable (inevitable??) cross arm deformation. A couple diagrams purporting to represent the supposed effect with dotted lines, that is all. No statistical studies, no analyses from an ME.

The strain on the rivet exceeds the force needed to deform the crossarm by quite a bit. Further, by virtue of its being rivetted, it is already under strain making the force needed to break the rivet even smaller. And like riveting a new staff, whacks with a big hammer are not even used (we learned we could rivet with the back of a tweezer while holding the staking tool on our leg).

Now of course, just like any other procedure, someone who is care-less can choose a hole too large and have the balance off center in the hole and damage it. But then, that same person would be care-less with a graver.

Not at all saying you (or anyone else) personally have to do it differently. Just pointing out factual information. A lot of stuff has been written and repeated by authors; like insisting balances should always be statically poised after a new staff has been installed.
 
Last edited:

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
I know Dewey insists that this doesn't cause damage, and that his experiences at Wostep prove it. I remain unconvinced, as last time we talked about this it was established that the testing done there wasn't meeting any standard for a "statistical studies, or analyses from an ME", just that they were able to rivet the staff on the balance fine. There were no measurements of the hole in the balance taken before and after riveting a staff and punching it out. There was no proof that the hole wasn't affected, only that it wasn't affected enough to be a problem by doing this once.

Another poster here had posted these photos the last time we went around in circles on this subject - the rivet is not just a thin section at the very top of that portion of the staff - the staff is enlarged in a flaring cone shape:

StaffRivet.jpg

When you punch the staff out, the top portion breaks off, but there is still an enlarged portion that has to be forced through the hole - this is a staff that was punched out:

Sheared staff rivet.jpg

You can see that the shearing isn't always even.

From my own personal standpoint, I don't punch out staffs. To me, I see a risk here, and it's one not worth taking when there are much easier and completely safe methods of removing staffs. I removed one last night actually - the balance with the staff was sitting in the alum solution, and the staff was removed while I enjoyed my dinner. Today I'll install the new staff.

I always have another customer's watch that I can be working on while the staff is soaking, so I have no need to remove a staff immediately. I can be doing other things in the shop while the staff is dissolving in alum, and on the rare occasions where it's a balance that can't go in alum, it only takes a few minutes to cut it out on the lathe.

I don't agree that punching them out is risk free, and no risk is better than some risk. But to each his own...

Cheers, Al - engineer turned watchmaker.
 

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
I know Dewey insists that this doesn't cause damage, and that his experiences at Wostep prove it. I remain unconvinced, as last time we talked about this it was established that the testing done there wasn't meeting any standard for a "statistical studies, or analyses from an ME", just that they were able to rivet the staff on the balance fine. There were no measurements of the hole in the balance taken before and after riveting a staff and punching it out. There was no proof that the hole wasn't affected, only that it wasn't affected enough to be a problem by doing this once.

Another poster here had posted these photos the last time we went around in circles on this subject - the rivet is not just a thin section at the very top of that portion of the staff - the staff is enlarged in a flaring cone shape:

View attachment 648382

When you punch the staff out, the top portion breaks off, but there is still an enlarged portion that has to be forced through the hole - this is a staff that was punched out:

View attachment 648383

You can see that the shearing isn't always even.

From my own personal standpoint, I don't punch out staffs. To me, I see a risk here, and it's one not worth taking when there are much easier and completely safe methods of removing staffs. I removed one last night actually - the balance with the staff was sitting in the alum solution, and the staff was removed while I enjoyed my dinner. Today I'll install the new staff.

I always have another customer's watch that I can be working on while the staff is soaking, so I have no need to remove a staff immediately. I can be doing other things in the shop while the staff is dissolving in alum, and on the rare occasions where it's a balance that can't go in alum, it only takes a few minutes to cut it out on the lathe.

I don't agree that punching them out is risk free, and no risk is better than some risk. But to each his own...

Cheers, Al - engineer turned watchmaker.
Al,

I never said a care-less person could not screw up a balance. However, there are no data to support the position that a deformed balance is inevitable.

Anyone here can find examples of balances ruined by a care-less person using a graver. I have seen arms all but torn off because the graver caught a screw. There are risks no matter what procedure is used.

It is like when I was working as a researcher in health. Yes there are mortality risks involved with many medications. But what is the relative risk of the drug compared to non treatment?
'
All I know is that a minimum of 36 balances were not deformed by a class of 12. I have never deformed a balance. And this is the method taught in the Swiss schools.

In light of the lack of data that refutes my observations, I will continue to point out that the notion that cutting out a staff is less problematic than punching/pressing it out has not been established by any systematic study.

As I pointed out, above, just because a notion has been published (and recycled) does not make it true. Do you remember the article in HT (when Fried was the technical editor, no less), that promoted that if a staff shoulder is too short, "Just grind away the bottom of the balance"? I never did see a retraction.

If something is discrepant with my prior knowledge, I look for systematic data. In the absence of such data, I stick with my prior knowledge.

It is fine to suggest cutting out a staff as a way to remove a staff.

But I would never assert that one method is better than the other unless I had empirical data to support that position.
 

karlmansson

Registered User
Apr 20, 2013
2,738
166
63
Linköping, Sweden
Country
I’m neither swiss trained, nor an engineer so take what I write for what it’s worth. But when the same mechanical principles are present as in metal working procedures (stamping, broaching and press forming) I would think it reasonable to deduct that the same result will apply. Even without an empirical study on the matter. It would be interesting to see how a gauge pin fits a balance hole after punching the rivet through. Maybe it’s not noticeable when restaffing the first time but when forcing something of a larger size through a smaller hole, something has to give.

Regards
Karl
 
  • Like
Reactions: Al J

Brian C.

Registered User
Jun 9, 2001
1,249
4
38
NH
Country
I’m neither swiss trained, nor an engineer so take what I write for what it’s worth. But when the same mechanical principles are present as in metal working procedures (stamping, broaching and press forming) I would think it reasonable to deduct that the same result will apply. Even without an empirical study on the matter. It would be interesting to see how a gauge pin fits a balance hole after punching the rivet through. Maybe it’s not noticeable when restaffing the first time but when forcing something of a larger size through a smaller hole, something has to give.

Regards
Karl
This is not true and in fact the converse was proven to me in Switzerland. We were taught to remove staffs either by use of the Molfres (which leaves a "rivet" on the balance underside that still needs to be punched/pressed out), or by pressing/punching out from the top. Whether it was a Rolex glucydor balance or a nickel balance, the replacement staff fit perfectly. This was a class of 12 people with a minimum of 3 balances each. Not one problem.

I have never seen any analysis to support the claim of probable (inevitable??) cross arm deformation. A couple diagrams purporting to represent the supposed effect with dotted lines, that is all. No statistical studies, no analyses from an ME.

The strain on the rivet exceeds the force needed to deform the crossarm by quite a bit. Further, by virtue of its being rivetted, it is already under strain making the force needed to break the rivet even smaller. And like riveting a new staff, whacks with a big hammer are not even used (we learned we could rivet with the back of a tweezer while holding the staking tool on our leg).

Now of course, just like any other procedure, someone who is care-less can choose a hole too large and have the balance off center in the hole and damage it. But then, that same person would be care-less with a graver.

Not at all saying you (or anyone else) personally have to do it differently. Just pointing out factual information. A lot of stuff has been written and repeated by authors; like insisting balances should always be statically poised after a new staff has been installed.
Dewey, you're probably right. I've only been repairing watches for over 40 years. I'm still learning all the time.
 

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
I’m neither swiss trained, nor an engineer so take what I write for what it’s worth. But when the same mechanical principles are present as in metal working procedures (stamping, broaching and press forming) I would think it reasonable to deduct that the same result will apply. Even without an empirical study on the matter. It would be interesting to see how a gauge pin fits a balance hole after punching the rivet through. Maybe it’s not noticeable when restaffing the first time but when forcing something of a larger size through a smaller hole, something has to give.

Regards
Karl
This is exactly my point Karl.
 

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
I’m neither swiss trained, nor an engineer so take what I write for what it’s worth. But when the same mechanical principles are present as in metal working procedures (stamping, broaching and press forming) I would think it reasonable to deduct that the same result will apply. Even without an empirical study on the matter. It would be interesting to see how a gauge pin fits a balance hole after punching the rivet through. Maybe it’s not noticeable when restaffing the first time but when forcing something of a larger size through a smaller hole, something has to give.

Regards
Karl
Karl,

Many "thought" experiments do not hold when confronted with reality. That is the beauty of the scientific method for discovering truth.

It makes no difference to me how people remove staffs. I simply point out that there is risk to the assembly no matter how one does it, And those risks increase when a care-less person is involved.

AS I am sure you were taught in evaluating surgical literature, you must evaluate your prior knowledge against new knowledge, even using what you know about the author. Have his prior publications been credible?

The new knowledge here is the first hand report of data that at least 36 balances were restaffed in a controlled environment using the press/punch technique with no ill effects. There is also the report by the same author that he has used this technique exclusively for over 10 years with no ill effects.

If you have reason to doubt the integrity of the author, then you should in fact reject that report as suspect in the face of what you think you know.

But in no event, is the author stating his method is the only/best/official way to do things. I simply am not dogmatic and have never told anyone "my way is the only way". I have enough life experience to know there are many ways to "skin a cat".

You might find this post of interest: https://mb.nawcc.org/threads/critical-thinking-and-watchmaking.178701/
 
Last edited:

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
Al,

I never said a care-less person could not screw up a balance. However, there are no data to support the position that a deformed balance is inevitable.

Anyone here can find examples of balances ruined by a care-less person using a graver. I have seen arms all but torn off because the graver caught a screw. There are risks no matter what procedure is used.

It is like when I was working as a researcher in health. Yes there are mortality risks involved with many medications. But what is the relative risk of the drug compared to non treatment?
'
All I know is that a minimum of 36 balances were not deformed by a class of 12. I have never deformed a balance. And this is the method taught in the Swiss schools.

In light of the lack of data that refutes my observations, I will continue to point out that the notion that cutting out a staff is less problematic than punching/pressing it out has not been established by any systematic study.

As I pointed out, above, just because a notion has been published (and recycled) does not make it true. Do you remember the article in HT (when Fried was the technical editor, no less), that promoted that if a staff shoulder is too short, "Just grind away the bottom of the balance"? I never did see a retraction.

If something is discrepant with my prior knowledge, I look for systematic data. In the absence of such data, I stick with my prior knowledge.

It is fine to suggest cutting out a staff as a way to remove a staff.

But I would never assert that one method is better than the other unless I had empirical data to support that position.
Hi Dewey,

Let me say that I don't care at all what method you use to remove staffs. So if you want to continue punching them out, by all means do so.

My approach to watchmaking is to first do no harm. To me, forcing a staff through a hole, hoping that the rivet breaks off cleanly, and knowing that there will be a larger part of the staff that gets pushed through the hole, does not match this approach.

I've demonstrated that for Glucydur and nickel gilt balance wheels, there is a 100% safe method of removing the staff using alum. There is no risk at all - zero, zilch, nada. I'll use this method every time I can, because in my view there is absolutely no compelling reason to use a method that involves risk, then there is a risk free method.

You have advocated for punching out every time this comes up, and you use your own experience, and an appeal to authority (WOSTEP) every time, and that's fine but it's not the type of evidence that you demand of others.

But in reaction to this:

"I always cut the staffs out. Punching them out can widen the hole in the balance. "

You said this:

"This is not true and in fact the converse was proven to me in Switzerland."

As much as you talk about the need for engineering reports, statistical analyses, and empirical data, you didn't actually prove that punching out the staff doesn't enlarge the hole. If you did, please provide the relevant data to support this - measurements of the holes before and after would be a very good start.

Based on what you have said previously, what you did "prove" is that it's possible to do it once without enlarging the hole enough to prevent a staff from being installed - these are of course two very different things.

I'm not saying that you can't remove a staff by punching it out and still be able to rivet a new one in place. But I think recognizing the risk in doing so, and the potential to enlarge the hole, is prudent.

Cheers, Al
 
  • Like
Reactions: Jerry Kieffer

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
Al,

I am not criticizing you. There are many ways to skin a cat. AS has oft been said; the watch does not care how the part was made as long as the fit and finish are correct.

I understand you object to the method for which I reported actual data. Not anecdotal data; not "in my opinion".

I understand you reject a method taught in modern Swiss watchmaking schools.

I am more circumspect than to claim my way is the only way. There are risks involved in either method, especially when a care-less person is involved.

Just like "poising a balance immediately after staffing" was once accepted as the only way to do things, it is my opinion that caution is needed when claiming one method of removing staffs is superior to another. There are no scientifically derived data to support that one method presents a higher relative risk than the other.

I stand by my data, and I have not seen any evidence to cause me to reconsider my conclusion and opinion.

This discussion may be best continued off line.
 
Last edited:

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
Al,

I am not criticizing you. There are many ways to skin a cat. AS has oft been said; the watch does not care how the part was made as long as the fit and finish are correct.

I understand you object to the method for which I reported actual data. Not anecdotal data; not "in my opinion".

I understand you reject a method taught in modern Swiss watchmaking schools.

I am more circumspect than to claim my way is the only way. There are risks involved in either method, especially when a care-less person is involved.

Just like "poising a balance immediately after staffing" was once accepted as the only way to do things, it is my opinion that caution is needed when claiming one method of removing staffs is superior to another. There are no scientifically derived data to support that one method presents a higher relative risk than the other.

I stand by my data, and I have not seen any evidence to cause me to reconsider my conclusion and opinion.

This discussion may be best continued off line.
Dewey,

This is very simple. The "data" you reported doesn't actually support the claim you made that the hole is not enlarged. It's not any more complicated than that.

You continue to claim risks in "either" method, and I agree that there are risks in removing a staff by punching it out or turning it out. However you completely ignore the third method using alum that is 100% risk free. If you believe there are risks to sing alum, please elaborate on what those are.

It isn't about objecting to or rejecting a particular method, it's about the claim being proven or not - it hasn't been.

I am fine to agree to disagree, and let others judge for themselves based on what has been presented.

Cheers, Al
 

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
AL,

You apparently do not understand the definition of data. Data are quantified observations that are subject to replication. There is no equivalency between sporadic anecdotal reports from uncontrolled conditions and data derived from a controlled environment.

There is no need for you to take this tack. You seem intent on getting into a catfight in which I will not engage. NOr will I engage in a discussion here on what is termed falsifiability of science (for the unfamiliar, it refers to the ability to test reality via data).

Whereas I am not attacking your method, you seem intent on attacking the method taught in modern Swiss watchmaking schools simply because it is not aligned with your belief structure. No one cares what you choose to believe.

And to be clear Al, you are not criticizing MY method. You are actually claiming the current method taught by the Swiss is destructive. Interesting.

BTW, do you know Rolex is a member of the foundation that supports these schools and was the source for the over 120 balances destroyed by my class. THAT is the way to gain experience. No way anyone can afford to "experiment" on such a scale with Rolex material they have to buy.

Readers trying to make sense of this particular thread may benefit from reading this thread:

 
Last edited:

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
AL,

You apparently do not the definition of data. Data are quantified observations that are subject replication,

There is no need for you to take this tack. You seem intent on getting into a catfight in which I will not engage.

Whereas I am not attacking your method, you seem intent on attacking the method taught in modern Swiss watchmaking schools simply because it is not aligned with your belief structure. No one cares what you choose to believe.

And I really do think you should bring it off line.

Readers trying to make sense of this particular thread may benefit from reading this thread:

Dewey,

There is no need for you to make personal attacks. I do in fact know the definition of data. You have presented data, but as I said clearly:

"The "data" you reported doesn't actually support the claim you made that the hole is not enlarged."

I have no intent on getting into a "catfight". You have interpreted your experience in punching out staffs to mean that it doesn't enlarge the hole, simply because you were able to rivet a staff in. I think we all know that you can still rivet a staff in a hole that is oversized.

My belief system is that in order to prove an assertion, you need relevant data to back it up. I haven't seen that yet from you.

As I said, I admit that you can punch out a staff and still fit a new staff. That wasn't the question - the question is does it enlarge the hole.

Cheers, Al
 

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
Al,

I made no personal attacks. I merely pointed out definitions and reality. Your claim to know more than Swiss educators. Maybe you do and maybe you don't. But I have no way of judging that and make no claim one way or the other.

You made your arguments, and I made mine. You have presented no evidence to cause me to rethink my conclusions and I never did try to convince you the Swiss way was better than yours. Time to let it go.

If you truly feel I engaged in a personal attack, we should discuss it off line.

And to be clear, I only stated that using the Swiss method an original material staff fits correctly. Meaning it requires pressure to fit the balance onto the staff shoulder, rivets correctly with the back of a tweezers with the staking tool held on your leg, is centered and poise not disturbed.

Once again, readers trying to make sense of this thread may want to read:

 
Last edited:

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
Al,

I made no personal attacks. I merely pointed out definitions and reality.
Well to be clear, suggesting that I "do not understand the definition of data." is something I consider to be a personal attack.

Your claim to know more than Swiss educators. Maybe you do and maybe you don't. But I have no way of judging that and make no claim one way or the other.
I have never claimed such a thing. I am not privy to what "Swiss educators" have told you with regards to punching out staffs and the enlarging of holes in balances. I only know what you have stated here, that enlarging of the hole has been proven to not happen.

As you stated the last time we talked about this, the holes were never measured to see if they were enlarged or not, you just restaffed them.

The thread is here:

Balance staff replacement tooling | NAWCC Forums

Quoting you from that thread:

"As was said and described, we punched out a staff from a nickel balance and restaffed it. No one had an issue."

The claim remains unproven.

Cheers, Al
 

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
Al,

To be clear, your misuse of the term does indeed suggest you did not know the definition of "data". I cannot fathom any other explanation for the way you used the term.

Precisely what is "unproven"? That the Swiss teach this method? That I did not observe the effective use of this technique? That you know more than the Swiss educators? That your technique is the only one that is "correct" That one method is superior to the other?

You have no claim to a personal attack. Had you, I am certain you would have referred this discussion to the moderators.

Once again, I refer readers trying to make sense of this barn circling to:

 

SpringDriven

Registered User
Dec 22, 2010
130
192
43
I would like to take a moment to point out one small but significant piece of information about balance staffs and the holes in the balance wheels they are riveted to. I think this may be overlooked, due to whatever reasons that may be.

The hole in the balance wheel is a larger diameter than the hub diameter of the balance staff that will be riveted to the wheel. If this were not the case and the diameters were the same size, it would become a friction fit staff and have no need to be riveted.

Yes there are risks. But considering the fact that the hole in the wheel is already larger than the riveted hub on the staff, one of the perceived risks is mitigated.
 
  • Like
Reactions: DeweyC

karlmansson

Registered User
Apr 20, 2013
2,738
166
63
Linköping, Sweden
Country
I have to agree with Al here, we are discussing apples and pears. The data you have shared Dewey is that there seemed to be a good fit and that all 36 balances that were subjected to this procedure took a new riveted staff without issue.

Is that really the question at hand though? If a repair can be done once without issue? Or if there is indeed damage being done to the balance hole? I would still like to see a comparison with a microscope and gauge pins. Comparing first time restaffs and measuring outcome with “no issues” doesn’t really tell me a lot about data analysis, if we are indeed intent of holding this comparison to academic standards.

In healthcare we often talk about “number needed to treat” and number needed to harm” when evaluating a treatment or procedure. In those cases I would consider an inclusion number of 36 a very small population.

Regards
Karl
 

SpringDriven

Registered User
Dec 22, 2010
130
192
43
I will follow up. I have used a balance I have from an ETA 2896. While I did not restaff it, I did check the factory static poise with the hairspring removed. It was statically poised, the mass of the wheel was evenly distributed from ETA.

I made a mark on the rim of the wheel in the approximate location that the impulse jewel was in. Placing it on the jaws of my poising tool in random positions, it did not move, meaning there was no heavy spot on the wheel for gravity to pull on and cause the wheel to rotate on the jaws. If I am wording this correctly, the mass of the wheel was evenly distributed around the center of mass. Gravity has an equal pull on the circumference of the wheel when in a vertical position. I do hope my words are correct in this statement.

Then I removed the roller table and placed it on in a random position, approximately 180 from where the impulse jewel originally was, and rechecked the poise of the wheel.

It was no longer in poise, it had a heavy spot. It did not gain mass, but the mass of the assembly was no longer evenly distributed throughout the wheel.

I removed the roller table and placed it back on with the jewel very close to where it was in poise before. It still had a heavy spot but it was much less severe.

Moving the roller table will change the distribution of mass on the wheel. That was my observation.
 

Jerry Kieffer

NAWCC Fellow
NAWCC Member
May 31, 2005
2,926
550
113
wisconsin
Country
I know Dewey insists that this doesn't cause damage, and that his experiences at Wostep prove it. I remain unconvinced, as last time we talked about this it was established that the testing done there wasn't meeting any standard for a "statistical studies, or analyses from an ME", just that they were able to rivet the staff on the balance fine. There were no measurements of the hole in the balance taken before and after riveting a staff and punching it out. There was no proof that the hole wasn't affected, only that it wasn't affected enough to be a problem by doing this once.

Another poster here had posted these photos the last time we went around in circles on this subject - the rivet is not just a thin section at the very top of that portion of the staff - the staff is enlarged in a flaring cone shape:

View attachment 648382

When you punch the staff out, the top portion breaks off, but there is still an enlarged portion that has to be forced through the hole - this is a staff that was punched out:

View attachment 648383

You can see that the shearing isn't always even.

From my own personal standpoint, I don't punch out staffs. To me, I see a risk here, and it's one not worth taking when there are much easier and completely safe methods of removing staffs. I removed one last night actually - the balance with the staff was sitting in the alum solution, and the staff was removed while I enjoyed my dinner. Today I'll install the new staff.

I always have another customer's watch that I can be working on while the staff is soaking, so I have no need to remove a staff immediately. I can be doing other things in the shop while the staff is dissolving in alum, and on the rare occasions where it's a balance that can't go in alum, it only takes a few minutes to cut it out on the lathe.

I don't agree that punching them out is risk free, and no risk is better than some risk. But to each his own...

Cheers, Al - engineer turned watchmaker.
Al
This discussion sometimes comes up in class when demonstrating a procedure to grind off a staff hub on the lathe per attached link.


To demonstrate any effects of driving out a staff, I first grind off the hub and remove the staff from the balance. From this point the balance center hole is measured with gage pins. The staff is then reinserted in the balance and driven through where the hole is then remeasured with gage pins. In more cases than not, the hole is enlarged to so some degree depending on ones standards. Having done this a fair number of times, you learn that some staffs are more heavily staked than others. While a lightly staked staff may have no effect on a balance, a heavily staked one can enlarge a hole beyond the comfort zone I suspect of some. While nothing is guaranteed, I make every effort possible to assure that the balances utilized are original and have not been restaffed.

Personally, when remeasuring balances,I can except a slightly looser fit of the same pin but not next size up pins.

Jerry Kieffer
 
  • Like
Reactions: Al J

karlmansson

Registered User
Apr 20, 2013
2,738
166
63
Linköping, Sweden
Country
Karl,

Many "thought" experiments do not hold when confronted with reality. That is the beauty of the scientific method for discovering truth.

It makes no difference to me how people remove staffs. I simply point out that there is risk to the assembly no matter how one does it, And those risks increase when a care-less person is involved.

AS I am sure you were taught in evaluating surgical literature, you must evaluate your prior knowledge against new knowledge, even using what you know about the author. Have his prior publications been credible?

The new knowledge here is the first hand report of data that at least 36 balances were restaffed in a controlled environment using the press/punch technique with no ill effects. There is also the report by the same author that he has used this technique exclusively for over 10 years with no ill effects.

If you have reason to doubt the integrity of the author, then you should in fact reject that report as suspect in the face of what you think you know.

But in no event, is the author stating his method is the only/best/official way to do things. I simply am not dogmatic and have never told anyone "my way is the only way". I have enough life experience to know there are many ways to "skin a cat".

You might find this post of interest: https://mb.nawcc.org/threads/critical-thinking-and-watchmaking.178701/
I mean, it’s not a thought experiment as such. Not any more than saying that the balance didn’t sustain damage without looking at the finish in the hole, which I don’t think was mentioned.

I have my own experience of what pushing a burr through a bore does to surface finish and dimensions. That’s very real and noticeable and not at all a thought experiment. Why anything else would apply to a burr on a hardened steel staff and a softer balance is beyond me.

Regards
Karl
 
Last edited:
  • Like
Reactions: Al J

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
I have to agree with Al here, we are discussing apples and pears. The data you have shared Dewey is that there seemed to be a good fit and that all 36 balances that were subjected to this procedure took a new riveted staff without issue.

Is that really the question at hand though? If a repair can be done once without issue? Or if there is indeed damage being done to the balance hole? I would still like to see a comparison with a microscope and gauge pins. Comparing first time restaffs and measuring outcome with “no issues” doesn’t really tell me a lot about data analysis, if we are indeed intent of holding this comparison to academic standards.

In healthcare we often talk about “number needed to treat” and number needed to harm” when evaluating a treatment or procedure. In those cases I would consider an inclusion number of 36 a very small population.

Regards
Karl
Karl,

I have in fact proven this to myself. It is easily done. Take a balance for which you have a number of staffs. For me it was a single roller Hanmilton 16s. I also have a couple customers I restaffed at least twice and those balance assemblies behaved as they did in the past.

I do agree that there has been some attempt to change the discussion from what I observed and actually said, to measuring IDs. By all means, if this keeps someone up at night, they should measure.

I would not be surprised if the Swiss institutes have studied this ad nauseum. But I have no idea how to access such research. What I do know is that the Swiss are data and results driven. They do not teach and prescribe based on untested opinion.

As a practical matter, the staff should be a close fit; but a not driving fit. The rivet collar should just peak over the arm and you should see no space between the arm and the staff's balance post. I am not sure what else matters.

Now, let us assume you see a female cup on the top of the balance arm. Was that an artifact of pressing the staff out or a misplaced touch of the graver? What about that female cup on the bottom? How did that happen?

This is why the only way to claim one method is superior to another is through a scientific approach. Given that no such research is currently accessible, it is not logical to claim there is only one right way. And all I did originally was to point that out (see my first response), to point out that this is the way they teach it Switzerland, and that I observed enough instances to be satisfied this is a prudent method.

What one does with my report is up to them. As Clark Gable said: "Frankly Margeret, I don't give a ......"

Karl, for the record, my last real job was as the lead researcher for the American Association of Medical Colleges on a $5mil study to assess how to improve the process of undergraduate medical education (in the US this is the 4 years before residency). I do in fact know how to calculate sample sizes needed at various confidence levels. And it is NOT as simple as a number. It depends upon the robustness of the design and the size of the actual population being considered.

And while the cost of a wrong conclusion is much higher in medicine, I have seen surgical procedures that became established with an even smaller sample size.

This is why reproducible data is important. Yes, I am well aware that one data point doth not a trend make. But my observations of actual results in a controlled setting are more scientifically powerful than conclusions drawn from anecdotal data derived in unknown circumstances.

Springdriven, I know we have drifted far afield from your post, and I apologize for my part in that.
 

SpringDriven

Registered User
Dec 22, 2010
130
192
43
I mean, it’s not a thought experiment as such. Not any more than saying that the balance didn’t sustain damage without looking at the finish in the hole, which I don’t think was mentioned.

I have my own experience of what pushing a burr through a bore does to surface finish and dimensions. That’s very real and noticeable and not at all a thought experiment. Why anything else would apply to a burr on a hardened steel staff and a softer balance is beyond me.

Regards
Karl
Karl,

You see evidence of burrs and mushrooming of hubs as perception that this is a destructive process to the balance wheel. I have no means to refute the possibility of these things damaging a balance wheel other than to say this.

The hole in the balance wheel is already larger than the diameter of the riveted hub, which is why riveting is required. The size difference of the two is to the point that mushrooming alone is not enough to friction a staff to a wheel. Metal must be pushed over and onto the wheel to perform the rivet process and join the two together.

Please consider that even a mushroomed hub, MAYBE is a similar diameter as the hole in the wheel. If the mushroomed hub was the same diameter as the hole in the wheel, it would cause no harm pushing it out, this is normal friction. How do you perceive a friction fit staff to work? If the mushroomed hub was LARGER than the hole in the wheel, the damage you are worried about was already done in the previous rivet process, no amount of care or process to remove the staff will change this. The hole was already enlarged. This mushroomed head will not enlarge in the process of pushing it through, so it can not be larger than the hole currently is.

So I do not believe there is any concern with damaging the diameter of the balance hole when pushing a staff through it in the process that is used to break the rivet. It simply can not happen due to diameters and friction. I of course can not account for a poor quality not properly hardened staff, this probably the largest risk involved with driving out a staff. When the rivet does not break. However in this case the watchmaker should feel resistance entirely different than normal and not continue to PUSH through. A quick twist, or light tap is all that is needed to break a rivet depending on the tool used. If doing such does not result in an immediate felling of breaking through a wall and then no resistance, and you feel resistance instead it would be the watchmaker now causing harm to the wheel by FORCING the staff through, the rivet did not break and probably will not break. Time to seek an alternate method.

As for burrs? I can not account for them. I have not experienced this as a problem. I will not say it is not possible. I will say it is unlikely to greatly effect the outcome of a restaff. Yes it is sad if it did happen.

I will not say someone should cut out a staff, grind out a staff, push out a staff or use a chemical. They all come with risks. Please choose the method that works for you, that you are able to best mitigate the risks with, and that you can best overcome any issues with.
 
Last edited:

SpringDriven

Registered User
Dec 22, 2010
130
192
43
Springdriven, I know we have drifted far afield from your post, and I apologize for my part in that.
No need to apologize, I was not aware of the watchmaking atmosphere here when I did this post. It is therefore entirely my fault, not anyone else.

To be honest, more people took (my perception) offense to me static poising the wheel than how I removed the balance staff.

But the removal of a staff is much easier to discuss, so it became the hot topic.

I would love to be able to educate my position on static poising and dynamic poising, but it requires an understanding of moment of inertia, angular velocity, math, or observable data, none of which I currently have, so I have not pushed back, nor do I really want to at this time without the former. I can not discuss this with math, as I am not amazing with math. I can not show observable data, because I do not have the opportunity to do the work, take photos and discuss in a timeframe that is reasonable. But my understanding is different. And like my experiment with moving the roller table, I probably will do an experiment like this with timing screws and poising.

But remember everyone, for every experiment performed in watchmaking, it is so very difficult to prove that the results are not influenced by something else. A piece of lint for example during the process... At what point of making an adjustment does not influence lubrication etc. Which is why math usually wins, but requires everyone to be on the same math field, from my perspective anyways.
 
Last edited:

karlmansson

Registered User
Apr 20, 2013
2,738
166
63
Linköping, Sweden
Country
Karl,

I have in fact proven this to myself. It is easily done. Take a balance for which you have a number of staffs. For me it was a single roller Hanmilton 16s. I also have a couple customers I restaffed at least twice and those balance assemblies behaved as they did in the past.

I do agree that there has been some attempt to change the discussion from what I observed and actually said, to measuring IDs. By all means, if this keeps someone up at night, they should measure.

I would not be surprised if the Swiss institutes have studied this ad nauseum. But I have no idea how to access such research. What I do know is that the Swiss are data and results driven. They do not teach and prescribe based on untested opinion.

As a practical matter, the staff should be a close fit; but a not driving fit. The rivet collar should just peak over the arm and you should see no space between the arm and the staff's balance post. I am not sure what else matters.

Now, let us assume you see a female cup on the top of the balance arm. Was that an artifact of pressing the staff out or a misplaced touch of the graver? What about that female cup on the bottom? How did that happen?

This is why the only way to claim one method is superior to another is through a scientific approach. Given that no such research is currently accessible, it is not logical to claim there is only one right way. And all I did originally was to point that out (see my first response), to point out that this is the way they teach it Switzerland, and that I observed enough instances to be satisfied this is a prudent method.

What one does with my report is up to them. As Clark Gable said: "Frankly Margeret, I don't give a ......"

Karl, for the record, my last real job was as the lead researcher for the American Association of Medical Colleges on a $5mil study to assess how to improve the process of undergraduate medical education (in the US this is the 4 years before residency). I do in fact know how to calculate sample sizes needed at various confidence levels. And it is NOT as simple as a number. It depends upon the robustness of the design and the size of the actual population being considered.

And while the cost of a wrong conclusion is much higher in medicine, I have seen surgical procedures that became established with an even smaller sample size.

This is why reproducible data is important. Yes, I am well aware that one data point doth not a trend make. But my observations of actual results in a controlled setting are more scientifically powerful than conclusions drawn from anecdotal data derived in unknown circumstances.

Springdriven, I know we have drifted far afield from your post, and I apologize for my part in that.
Also for the record: I have no real stake in this. I will continue using my preferred method, in part because of my above arguments but also because, from the same arguments, I have not bothered with getting the tools for punching a staff out.
As you pointed out in your thread about critical thinking in horology it serves to be critical of many practices. Some might work well for some balances and some less well for others. I would refer back to Jerrys post about different degrees of riveting on different staffs. I´m also a bit surprised and confused about the amount of assumption on others behalf that you argue ("I would not be surprised if the Swiss institutes have studied this ad nauseum. But I have no idea how to access such research. What I do know is that the Swiss are data and results driven. They do not teach and prescribe based on untested opinion.", " Precisely what is "unproven"? That the Swiss teach this method? That I did not observe the effective use of this technique? That you know more than the Swiss educators? That your technique is the only one that is "correct" That one method is superior to the other? ") while asking for evidence of others. I don't think anyone is doing any guess work here, obviously this discussion has been ongoing for decades. And has sparked heated debates for at least as long as I've been a member of this forum. It would however be interesting to see some actual measurements and studies of surface finish when comparing one method to the other.

Again, you are comparing two different outcomes. One is the hard endpoint of restaffing once across several calibers of the same type. The other is a surrogate measure of how the hole in the balance is affected by one procedure or the other. Jerry provided some anectodal evidence (as is yours Dewey) about looser fits after punching out.

Karl,

You see evidence of burrs and mushrooming of hubs as perception that this is a destructive process to the balance wheel. I have no means to refute the possibility of these things damaging a balance wheel other than to say this.

The hole in the balance wheel is already larger than the diameter of the riveted hub, which is why riveting is required. The size difference of the two is to the point that mushrooming alone is not enough to friction a staff to a wheel. Metal must be pushed over and onto the wheel to perform the rivet process and join the two together.

Please consider that even a mushroomed hub, MAYBE is a similar diameter as the hole in the wheel. If the mushroomed hub was the same diameter as the hole in the wheel, it would cause no harm pushing it out, this is normal friction. How do you perceive a friction fit staff to work? If the mushroomed hub was LARGER than the hole in the wheel, the damage you are worried about was already done in the previous rivet process, no amount of care or process to remove the staff will change this. The hole was already enlarged. This mushroomed head will not enlarge in the process of pushing it through, so it can not be larger than the hole currently is.

So I do not believe there is any concern with damaging the diameter of the balance hole when pushing a staff through it in the process that is used to break the rivet. It simply can not happen due to diameters and friction. I of course can not account for a poor quality not properly hardened staff, this probably the largest risk involved with driving out a staff. When the rivet does not break. However in this case the watchmaker should feel resistance entirely different than normal and not continue to PUSH through. A quick twist, or light tap is all that is needed to break a rivet depending on the tool used. If doing such does not result in an immediate felling of breaking through a wall and then no resistance, and you feel resistance instead it would be the watchmaker now causing harm to the wheel by FORCING the staff through, the rivet did not break and probably will not break. Time to seek an alternate method.

As for burrs? I can not account for them. I have not experienced this as a problem. I will not say it is not possible. I will say it is unlikely to greatly effect the outcome of a restaff. Yes it is sad if it did happen.

I will not say someone should cut out a staff, grind out a staff, push out a staff or use a chemical. They all come with risks. Please choose the method that works for you, that you are able to best mitigate the risks with, and that you can best overcome any issues with.
I have very little experience with friction fit staffs but my understanding is that a smooth staff is friction fitted into a hardened bushing that is itself riveted to the balance. That would be very different thing from materials with sharp edges and dissimilar hardness. I would refer to the images above posted by Al that shows the magnified cross section of a riveted staff. The rivet would shear off where the stress is high enough to fracture the steel, presumably along the very edge of the balance hole as that would be the fulcrum of the bending and shearing forces. Regardless of the hole being slightly undersized, the rivet is flared out to take up that space. So I mean, it's not as if we are creating that space again when punching the rivet out. And as for the hole being enlarged when riveting, wouldn't that deformation follow the conical shape of the rivet as shown in Al's images? And only by driving that cone through the balance would it be imparted on the hole all the way through?

As for which process actually causes the deformation, I can't think of any other method more suited to finding that out than Al's alum method. Which, by all measures, seems to be by far the safest on non-steel balances. No risk of tearing up the hole, no risk of slippig with the graver. Just a little waiting time. Which I can imagine is a problem in the service houses and something that is attempted to be worked around by the Swiss schools. After all, time is money. Not implying that anyone is cutting corners. Just that if there is little or no evidence that one way is better than the other (which I still have not seen. Any actual evidence that is.), why not choose the one that is quicker? I'm not in a rush though as I'm not making a living from this.

Regards
Karl
 
  • Like
Reactions: Al J

SpringDriven

Registered User
Dec 22, 2010
130
192
43
I can however as I have done in earlier posts ask some observable questions in relation to dynamic poising and static poising.

I will state that my understanding does not relate to vintage thought, or bi-metallic cut rim wheels. I just want to get that out of the way.

No one has to answer any questions, they are just meant to be thoughtful exercise.

1. Why are the screws on a screwed balance called timing screws and poising screws?

a. What does moving a timing screw do to the rate of a watch?

b. What is the relationship between poise and timing?

c. What is the difference between poise and moment of inertia used in timing?

2. Why are the poising screws all of different sizes in length on short threads in seemingly random positions, with the timing screws opposite each other equal in size with long threaded sections?

3. If the hairspring collet and stud have a timing influence that acts similar to a poising error (I understand this as fact). Why would you want to fight TWO poising errors as a watchmaker during the dynamic poising if you skip first static poising the wheel? It is not possible that the wheel is statically poised after performing a restaff without first verifying. So you would be fighting a poising error from restaffing and a similar to a poising error from the hairspring, in the position that math says this hairspring influence will happen. Why would you want to fight two errors?

4. What are you supposed to do with a wheel that has no screws?

5. What are you supposed to do with a freesprung balance when you do not have the terminal curve of the hairspring to change rate?

6. As in question five but the freesprung wheel only has two timing screws 180 from each other, and no other screws on the rim?

I ask questions four and five and six because I hope it makes some think about how can there be two different ways to dynamically poise a wheel?

Mass dealing with poise and moment of inertia dealing with timing is the premise of the questions.

All I can think of for now.
 
Last edited:
  • Like
Reactions: DeweyC

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
Springdriven,

First we need to decide whether we are talking about one position timepieces (chronometers) or precision pocket watches which assume random positions.

Marine chronometers are adjusted to temperature with the balance masses; poise is essentially ignored. That is taken care of by the gimbals. The screws at the ends of the crossrams are for timing.

The masses on the balance rim of a pocket watch are for poising and ballpark timing. Since a pocket watch is in a controlled environment (pocket), position is the preeminent concern for adjustment. So on most pocket watch balances, the poising screws are there for positional adjustment and temperature adjustment is a secondary concern.

The screws on the rim may not be named well. They actually perform 3 duties: timing, temp adjustment and positional adjustment. This is why it took 6 or so months to get a production watch out the door. Every time the balance was touched it had to checked for at least 6 days against the vertical positions (not sure what was done about temp after the initial screw loading).

These screws can be brass or gold. The higher the mass (coupled to an appropriate balance spring) the more stable the oscillator.

Timing screws are the screws at the end of the crossarms and can also be found in the middle of the rim. When there are four they are called quartering screws.

The goal of proper technique in restaffing is to avoid disturbing the original adjustments. So we mark the roller position (w/ ink) and remove and restaff the balance. We check truing in the flat and in the round at 30X. If we are flat and round, the balance, when the roller is replaced, should be in the same adjustment as it came into our hands.

Proper technique avoids creating work for ourselves.

Before we make ANY changes to the balance, we check it in the serviced watch. There is no point in touching the assembly until you know what you got.

This starts by checking its isochronal rates (not to be confused with isochronism of the balance spring over which we have no control). We do this with the watch in a horizontal position and comparing the rates at 180 degrees amplitude and 260 or higher. The goal is to equalize those rates. This can be done in a watch with regulator pins by changing their spacing.

The reason we do this is to not confuse variations in isochronal rate with the change of rate when the watch loses amplitude in the vertical positions.
Then we check the watch in the vertical positions. Not just its rate; but its amplitude as well. The amplitude should be the same in all tested vertical positions. I personally allow 10 degrees. Dramatic changes in amplitude indicate the watch needs to be reexamined and the fault found. No point in going further.

Oh, did I mention we demagnetize the watch complete prior to checking any rates? I generally assume it is part of the watch service . Like proper lubes.

Now we check the rates in each vertical position. All eight (checking at 45 degree intervals). The exact procedure is written up elsewhere. Suffice it to say with the proper system it is easy to identify the heavy spot causing rate differences and to add mass (washer) to the opposite screw.

This is what is meant by dynamic poising.

There are some variations. Early Hamiltons did not use timing screws. They used sets of balance screws to bring the watch into the predetermined range of the regulator. I am trying to track down how they pulled this off. It looks like they did several hundred thousand pocket watches this way. I detect the change to timing screws around 1910. The reason the balance screws vary in size is because at the factory they were what was used to impact positional adjustment. If the balance was slow, smaller screws were used.

Hamilton developed charts for impacts of the screw heights and washers for the M21. I suspect they had something similar for the early Hamiltons.

Here it is best to make positional adjustments with timing washers. First, it is reversible. Secondly, as proved to the class in Neuchatel, there is a point at which you remove too much mass and the watch can be timed to only one position.

BUT, I have found the difference in rate on these early Hamiltons can vary over 60 seconds depending upon the pin spacing. Sometimes pin spacing is all I need to get rid of some thick timing washers.

On watches with quartering screws, all positional adjustments can be made using those screws. There is no need for timing washers.

I have restored a number of Hamiltons to factory performance simply by changing the staff and removing in excess of 5 pairs of timing washers.

The saddest case is when someone followed the instructions of he who shall not be named to file the underside of the screws (GOLD no less). In those cases I have to find a donor balance for new screws so as to preserve the S/N scribed under the balance arm.

Freesprung balances that are out of positional adjustment should be checked to ensure the assembly is correct. For example, on Hamiltons the roller should be placed dead center between the crossarms when you have oriented so that you can read the serial number. On Hampden 18s size the roller is oriented along one of the crossarms so that you can infer its position relative to the banking pins.

If the roller orientation on the freesprung balance is correct, then adjust the quartering screws.

If only timing screws are present then use tining washers. The only watch I encountered with no poising screws is Rolex. But that has quartering screws.

There are "cheats" that can be done (and were done for trial preparation) that involve altering the pivots. But those are not considered permanent.

There are adjustments that can be made to the spring; overcoil, flat and helical. For these it is best to refresh from Gibri and Jendritski. But these are generally made at the factory and/or for special trials. Can take many months.

This is why the balance spring (and the whole balance assembly) should not be treated like a sack of potatoes.

Sadly, had the balance assembly been properly treated all along, there would be little for us to do. I will never understand why people had to make changes prior to checking the post service performance. Talk about data free decisions!

If I get a watch with decent amplitude, I actually do a pretest. This provides me with information about what to look for during service and to document for myself the improvement in performance generated by my work. Takes about 10 minutes.

As far as poising the collet, this was first discussed by Gibri and I think I have seen later references to it. But as you may know, the inertial impact of the collet is trivial in comparison to that of the balance rim due to the difference in radii.

I would expect its impact would be swamped by the forces due to the natural escapement errors of the lever escapement. I would think the asymmetrical excursion of the balance spring would be much more influential for example. But I never saw any data comparing the relative impact of collet poise.

Or even be swamped by the inertial impact on the free ends of the balance arms. Guillaume addressed this with his chronometer balance cut square in the middle of the arms to shorten the length of the lever arm. This design is still used in the Poljot chronometer and is found in all Nardin Chronometers that I know of.

Dan Munroe is also interested in the impact of collet poising and may be doing some research on this.

Todays wrist watches, with auto wind to keep constant power and FLAT balance springs, modern materials, are remarkable timekeepers. For our exams we had to service "prepared" (read wrecked) humble ETA 2892s in 3 hours. IIRC, all of us finished with all ten positions having a variance no greater than 3 seconds per day.

So I suspect research into improving mechanical oscillators has been relegated to the back room.

Do you realize I built and installed a 9 step 3 stringer (cut by me) staircase today in addition to writing? Busy day.
 
Last edited:

SpringDriven

Registered User
Dec 22, 2010
130
192
43
Springdriven,

First we need to decide whether we are talking about one position timepieces (chronometers) or precision pocket watches which assume random positions.

Marine chronometers are adjusted to temperature with the balance masses; poise is essentially ignored. That is taken care of by the gimbals. The screws at the ends of the crossrams are for timing.

The masses on the balance rim of a pocket watch are for poising and ballpark timing. Since a pocket watch is in a controlled environment (pocket), position is the preeminent concern for adjustment. So on most pocket watch balances, the poising screws are there for positional adjustment and temperature adjustment is a secondary concern.

The screws on the rim may not be named well. They actually perform 3 duties: timing, temp adjustment and positional adjustment. This is why it took 6 or so months to get a production watch out the door. Every time the balance was touched it had to checked for at least 6 days against the vertical positions (not sure what was done about temp after the initial screw loading).

These screws can be brass or gold. The higher the mass (coupled to an appropriate balance spring) the more stable the oscillator.

Timing screws are the screws at the end of the crossarms and can also be found in the middle of the rim. When there are four they are called quartering screws.

The goal of proper technique in restaffing is to avoid disturbing the original adjustments. So we mark the roller position (w/ ink) and remove and restaff the balance. We check truing in the flat and in the round at 30X. If we are flat and round, the balance, when the roller is replaced, should be in the same adjustment as it came into our hands.

Proper technique avoids creating work for ourselves.

Before we make ANY changes to the balance, we check it in the serviced watch. There is no point in touching the assembly until you know what you got.

This starts by checking its isochronal rates (not to be confused with isochronism of the balance spring over which we have no control). We do this with the watch in a horizontal position and comparing the rates at 180 degrees amplitude and 260 or higher. The goal is to equalize those rates. This can be done in a watch with regulator pins by changing their spacing.

The reason we do this is to not confuse variations in isochronal rate with the change of rate when the watch loses amplitude in the vertical positions.
Then we check the watch in the vertical positions. Not just its rate; but its amplitude as well. The amplitude should be the same in all tested vertical positions. I personally allow 10 degrees. Dramatic changes in amplitude indicate the watch needs to be reexamined and the fault found. No point in going further.

Oh, did I mention we demagnetize the watch complete prior to checking any rates? I generally assume it is part of the watch service . Like proper lubes.

Now we check the rates in each vertical position. All eight (checking at 45 degree intervals). The exact procedure is written up elsewhere. Suffice it to say with the proper system it is easy to identify the heavy spot causing rate differences and to add mass (washer) to the opposite screw.

This is what is meant by dynamic poising.

There are some variations. Early Hamiltons did not use timing screws. They used sets of balance screws to bring the watch into the predetermined range of the regulator. I am trying to track down how they pulled this off. It looks like they did several hundred thousand pocket watches this way. I detect the change to timing screws around 1910. The reason the balance screws vary in size is because at the factory they were what was used to impact positional adjustment. If the balance was slow, smaller screws were used.

Hamilton developed charts for impacts of the screw heights and washers for the M21. I suspect they had something similar for the early Hamiltons.

Here it is best to make positional adjustments with timing washers. First, it is reversible. Secondly, as proved to the class in Neuchatel, there is a point at which you remove too much mass and the watch can be timed to only one position.

BUT, I have found the difference in rate on these early Hamiltons can vary over 60 seconds depending upon the pin spacing. Sometimes pin spacing is all I need to get rid of some thick timing washers.

On watches with quartering screws, all positional adjustments can be made using those screws. There is no need for timing washers.

I have restored a number of Hamiltons to factory performance simply by changing the staff and removing in excess of 5 pairs of timing washers.

The saddest case is when someone followed the instructions of he who shall not be named to file the underside of the screws (GOLD no less). In those cases I have to find a donor balance for new screws so as to preserve the S/N scribed under the balance arm.

Freesprung balances that are out of positional adjustment should be checked to ensure the assembly is correct. For example, on Hamiltons the roller should be placed dead center between the crossarms when you have oriented so that you can read the serial number. On Hampden 18s size the roller is oriented along one of the crossarms so that you can infer its position relative to the banking pins.

If the roller orientation on the freesprung balance is correct, then adjust the quartering screws.

If only timing screws are present then use tining washers. The only watch I encountered with no poising screws is Rolex. But that has quartering screws.

There are "cheats" that can be done (and were done for trial preparation) that involve altering the pivots. But those are not considered permanent.

There are adjustments that can be made to the spring; overcoil, flat and helical. For these it is best to refresh from Gibri and Jendritski. But these are generally made at the factory and/or for special trials. Can take many months.

This is why the balance spring (and the whole balance assembly) should not be treated like a sack of potatoes.

Sadly, had the balance assembly been properly treated all along, there would be little for us to do. I will never understand why people had to make changes prior to checking the post service performance. Talk about data free decisions!

If I get a watch with decent amplitude, I actually do a pretest. This provides me with information about what to look for during service and to document for myself the improvement in performance generated by my work. Takes about 10 minutes.

As far as poising the collet, this was first discussed by Gibri and I think I have seen later references to it. But as you may know, the inertial impact of the collet is trivial in comparison to that of the balance rim due to the difference in radii.

I would expect its impact would be swamped by the forces due to the natural escapement errors of the lever escapement. I would think the asymmetrical excursion of the balance spring would be much more influential for example. But I never saw any data comparing the relative impact of collet poise.

Or even be swamped by the inertial impact on the free ends of the balance arms. Guillaume addressed this with his chronometer balance cut square in the middle of the arms to shorten the length of the lever arm. This design is still used in the Poljot chronometer and is found in all Nardin Chronometers that I know of.

Dan Munroe is also interested in the impact of collet poising and may be doing some research on this.

Todays wrist watches, with auto wind to keep constant power and FLAT balance springs, modern materials, are remarkable timekeepers. For our exams we had to service "prepared" (read wrecked) humble ETA 2892s in 3 hours. IIRC, all of us finished with all ten positions having a variance no greater than 3 seconds per day.

So I suspect research into improving mechanical oscillators has been relegated to the back room.

Do you realize I built and installed a 9 step 3 stringer (cut by me) staircase today in addition to writing? Busy day.
I admit that was a lot of effort to write on top of housework. I will do my best to answer in kind as I am able, which means I am working on answering my own questions the best that I am able.
 

Al J

Registered User
Jul 21, 2009
673
98
28
Canada
Country
Region
Precisely what is "unproven"? That the Swiss teach this method? That I did not observe the effective use of this technique? That you know more than the Swiss educators? That your technique is the only one that is "correct" That one method is superior to the other?
Your claim that punching out the staff doesn't enlarge the hole. As I posted above:

In reaction to this:

"I always cut the staffs out. Punching them out can widen the hole in the balance. "

You said this:

"This is not true and in fact the converse was proven to me in Switzerland."

You can try to steer it away from your claim as much as you like, but the claim you made is not supported by the data you provided. I have acknowledged a few times already that staffs can be punched out, and a new staff can successfully be riveted back in place, so telling me that I don't agree with that is somewhat disingenuous, at least that is the only way I can interpret it.

I do agree that there has been some attempt to change the discussion from what I observed and actually said, to measuring IDs.
Dewey, with all respect, you were the one who brought up the idea that ID's are not changed by punching the staff out.

I'm all for having a well reasoned technical discussion, but it's quite difficult if you won't at least acknowledge a basic set of facts, or even what you have said.

Cheers, Al
 

DeweyC

NAWCC Member
Feb 5, 2007
2,404
949
113
Baltimore
www.historictimekeepers.com
Country
I admit that was a lot of effort to write on top of housework. I will do my best to answer in kind as I am able, which means I am working on answering my own questions the best that I am able.

Springdriven,

Cool beans. I will read it with coffee or beer; whichever is is correct for the time.
 

SpringDriven

Registered User
Dec 22, 2010
130
192
43
I have written quite a bit. But as I write I realize I am not easily able to say "just press the I believe button", and I do not have the full scientific knowledge to back up my understanding on hand.

I enjoy sharing, which is the reason I originally posted this thread, to share a process that I enjoy as a watchmaker, and that I thought others would enjoy seeing as well. Be prepared, it probably will not be the only time I do this! My intent was never to come here and say "do it my way", but only to share some of the work I do.

I have decided that it is best if I simply say, this is not what I learned in school, and until I am able to be an instructor, that I am able to best write and instruct what I have learned, I should not get on the podium.

So I will though share my background. It is my sincere hope that my credentials help everyone understand that although you may have a different way that you do procedures, I too have my way, and it is the way that I was taught, the way that I understand, and I am recognized for the way I do my work. In no way am I saying anyone is wrong, as I just said earlier, I lack the ability to demonstrate in a way as I learned in school, I can not properly teach static and dynamic poise as I learned. If I can not properly convey my understanding, I can not say your understanding is not right. But I can say I learned differently.

I was Active Duty in the US Navy for 20 years and then I was retired. I joined when I was 17 and a Junior in High School as I knew it was what I wanted to do, and went to boot camp after I graduated from High School and turned 18. While I was in the Navy I started my career as a Data Systems technician specializing in the Link 11 system and ended my career as an Electronics Technician (as the DS rate was phased out), my primary specialty as an IFF technician, but I also worked on GPS navigation systems, etc. I had quite a few NECs. I spent nine years of my career overseas.

Since my Grandfather showed me an American Pocket watch when I was five I was hooked on watches pretty much my whole life. While I was getting close to being retired from Active Duty, I was considering a career in watchmaking. And as you can tell I took that leap. It is not uncommon in America that most professional watchmakers are second careers. In America it is not common to grow up wanting to be a watchmaker, as it is in Europe, and the training infrastructure is simply not the same as Europe for watchmaking. Watchmaking was not a career option for me out of High School so it is also a reason why it is a second career for me.

I applied and was accepted to the Lititz Watch Technicum in Lititz, PA. It was a two year school, with more than 4000 hours of training. The first year concentrated on micromechanics, where I learned how to use a lathe, drilling, hand filing, sawing, hardening and tempering steel, working with brass, friction, rivets, screws, threading, bevels, line finish, polishing steel, sanding, burnishing, boring, broaching, finishes... I can go on and on. I had a clock project first, where I took a common Hermle clock movement and converted it to a straight line clock.

I had to make drawings for all of my early micromechanic projects. One set in a 2D CAD program, one set by hand drawing. They had to be to scale, which means I had to properly measure things. The instructor graded the drawings for accuracy. For the clock I made the bridges, and I learned various ways to set the spacing of the pivots to ensure proper depthing of the wheels and learned the importance of positioning, which allows you to take it apart and put it back together and the pivot holes will be lined up properly every time. One of the pivot holes for example I had to do through math and triangulation only. So swinging arcs with a compass and placing a divot on that point and drilling a hole. I made the pillars for the bridges, one of the train wheels needed a new arbor to meet the endshake requirements of the bridges. I made the click, the dial, the hour markers, the hands, the hubs for the hands. And I made from scratch a Graham deadbeat escapement for the clock. So I made the escape wheel and the pallet and the pallet stones.

Once that project was finished I made a 3/4 bridge for a ETA 6497, and in my case the balance bridge as well. I learned about various ways to transfer holes from mainplate to bridge, turning recesses, boring holes for jewels. I can go on here as well.

I am not a master of micromechanics, but I learned a considerable amount. In the first year I also learned to service quartz watches and quartz theory. And I also learned to service basic automatic watches. The last portion of the first year was learning refinishing of cases and bracelets, the various wheels and compounds used to remove damage, maintain case shape and polish, or appropriate line finish and how to apply it.

The first year of school is fairly intense. There is far more than I mentioned.

Second year of school was dedicated to watchmaking theory and real life repairs and the service of the ETA 7750 chronograph. I learned to vibrate hairsprings, make overcoils, adjust hairspings in the flat and concentric coils. Adjustments. Regulations. Poising static and dynamic. Various calibers, adjusting hammers on vintage chronographs. Real life repairs included case repairs, maintaining water resistance, crystals and gaskets, case tubes and crowns. I bored out ovalized pivots and friction fit jewels in watches to make repairs, I burnished pivots. I wrote up job estimates for the repairs I did for approval. My work was not just focused on modern watches but also vintage... Again I can go on. The second year was just as intense, manipulating hairsprings is not easy work.

I successfully graduated from my school.

I currently have a SAWTA certification and a AWCI CW21 certification for watchmaking. I am level 40 certified with Jaeger LeCoultre. I am level 30 certified with Panerai. I am level 30 certified with IWC. I am level 40 certified with Cartier. I am a level 40 Rolex plaque account in my store. The industry recognizes my certifications, provides me brand and caliber specific training and allows me to perform warranty work on their products, sells me parts and they provide a brand warranty with my work, meaning if I service your watch and it has a warranty error, you can take it to them to be fixed under warranty, you don't have to come back to me. That is a high level of trust right? I started my career focusing on vintage service, before gaining the modern brand certifications.

So after writing my background in watchmaking I can say this. As I learned about Isochronism, rate independent of amplitude, I learned it was a lofty ideal not able to be fully realized. As example friction is an enemy of Isochronism, and you simply can not eliminate friction, so you always do your best to minimize friction.

In school in poising a wheel, I did not once ever move a screw to adjust for poise. Ever. Not static or dynamic. Moving screws is for changing the moment of inertia in a wheel. It does not eliminate a poise error as I understand. The mass of the screw is only moving further away from the center of mass or closer to it, but the poise error, which is a heavy spot in the wheel that can be pulled on by gravity in a vertical position is still there.

To have poise means that the mass of the wheel is evenly distributed around the circumference of the wheel in relation to the center of mass. It has no "heavy spot" for gravity to pull on in the vertical positions. As the wheel moves in the direction of gravity, it accelerates due to the influence of the heavy spot. As it moves away from the influence of gravity it slows down. This affects rate. Eliminate that heavy spot and the wheel has one less influence in the search for Isochronism, gravity being one of the enemies of Isochronism. I eliminate a heavy spot through poise, redistributing mass throughout the wheel, either by adding mass or subtracting mass on the rim of the balance, depending on the type of balance I am am working with, screw vs screwless, or position of the heavy spot, which screws are not always at the appropriate location.

I learned dynamic poising is necessary if after static poising I still have a poise error in the rate of the watch for the positions it is adjusted for during the timing phase of the service. I find the heavy spot by timing methods and I eliminate that heavy spot on the rim of the wheel.

There is much left unsaid here, as I am being extremely basic in my writing.

As moving screws is not a method I have ever used for poising, I can not comment on the effectiveness, and as you can see the methods I learned are different from what is accepted here.

Unfortunately I am not able to fully convey my understanding as I was taught. So I have to conclude with, the way I learned in school is the way that I service a watch, and I don't feel that it is wrong by any significant means. I believe someone said in this thread there is more than one way to skin something...

I will end with my passion in watches since becoming a watchmaker has focused on early watchmaking, as I enjoy all the many ways they overcome the enemies of Isochronism, or how they made advancements in technology. I am always carrying one of my various pocket watches in appreciation for what my Grandfather showed me when I was very young that lead me to this point today.
 
Last edited:

SpringDriven

Registered User
Dec 22, 2010
130
192
43
Do you still have the modified straight line clock? I would love to see it.
Of course. It is unfortunately in storage, because I moved to a location where I lack a nice place to set it up. But I will share some photos I took of the project while I was in school.

Please consider it is not amazing. It was a project to learn skills and techniques and demonstrate what you learn. It also allows you to learn many micromechanic skills in one project before advancing to smaller tolerances in watches.

 

Forum statistics

Threads
165,007
Messages
1,435,873
Members
85,921
Latest member
Authenthic
Encyclopedia Pages
1,101
Total wiki contributions
2,873
Last edit
Weekly News 7/7/19 by Tom McIntyre