Hairspring overcoil bend

[SndChsr]

Good day folks,

So I'm working on a watch I recently bought off ebay and I noticed that when inserting the balance back into watch, it would immediately over bank and hit one of the banking pins when the balance swung  clockwise. 

It didn't take long to realize (though I've missed this previously) that this hairspring is missing an overcoil! Someone who had "worked" on this watch before me simply fitted a brand new hairspring (as well as made a self fashioned copper pin to hold the hairspring stud in place as seen in the photo), out of the box but didn't bother forming an overcoil.

Having never done this myself, my question is, how does one determine where the overcoil bend begins? I'd like to give this a try and I know that this is a very difficult task, but nonetheless, one has to learn sometime right? So yeah, what is the formula in determining where to begin the overcoil bend on a hairspring?

Thank you for any help on this.

SndChsr.

Edited June 6, 2021 by SndChsr
Bad spelling

[LittleWatchShop]

This is way above my pay grade currently.  However, I did remember seeing this covered in some book I had.  Indeed...Practical Watch Repairing by Donald de Carle.  Chapter 13.

[SndChsr]

3 hours ago, LittleWatchShop said:

This is way above my pay grade currently.  However, I did remember seeing this covered in some book I had.  Indeed...Practical Watch Repairing by Donald de Carle.  Chapter 13.

Thanks for your input. Incidentally, my "The Watch Repairer's Manual" just showed up. It doesn't quite explain the issue I'm having here, but does have handful info that may lead me to a solution.

Fingers crossed. Something tells me I'll be browsing Ebay soon for a new hairspring 

[JohnR725]

12 hours ago, SndChsr said:

that this hairspring is missing an overcoil!

always helpful if you would tell us what you're working on? Then a picture would be nice like the balance bridge. This appears to have a single roller and the reason I bring up the pictures are some of the early watches single roller in particular might have had a over coil hairspring.

[clockboy]

Not sure it needs an overcoil. As John says a pic of the movement or ID of the caliber would be helpful. It might be a spring that is too strong. 

[Nucejoe]

A side close up of oscilator helps or you can judge yourself:  longer staff is to provide more space for the overcoil. Exceptions are high grade movement in some thin luxury watches like Piaget, Jaeger Lecoultre.

 

[SndChsr]

Thanks for your replies gentlemen. It is an Elgin, grade 50, single roller I believe.

More info here

I'm attaching some photos with the spring removed because I'm currently making a new brass pin to hold the stud in. 

I was under the impression that all hairsprings require some sort of overcoil. My apologies for my ignorance. Lots yet to learn!

Thank you for all the valuable info in your replies.

 

[SndChsr]

I just realized I'm ummmm...."not so bright".

From the pocket watch DB

 

Well, this pretty much answers my question

Thank you again for your input everyone! Lots and lots to learn!

[JohnR725]

59 minutes ago, SndChsr said:

I was under the impression that all hairsprings require some sort of overcoil.

this answer is incorrect for variety of reasons? Early American watches typically used flat hairsprings because are easy to make. But they went to over coil hairsprings for timekeeping reasons. But there's a problem with them to get the over coil in exacting location it's easier to pre-make them. This means if you're missing one it's easy to replace it. I'm attaching notes from the 1915 parts catalog that explains how you would do that in other words you have the timing screws.

But as watchmaking got better we don't need over coils typically. So the modern flat hairspring each one is vibrated to the balance wheel. So if you have a modern flat hairspring and detritus swap you'll have timing issues unless you have a way of changing the timing screws.

Now I would've agreed with you that this is probably an over coil and I couldn't tell from the picture but I did look up the hairspring and as you already discovered its flat.

A lot of times with the early flat hairspring is because they didn't have modern timing machines yet they weren't invented they left the hairspring  a little longer so it's not pin exactly where it's supposed to be. So initially pen it at the very end and see if you're running slow. In which case you can move a little more in and speed the watch up until you get close to where you're supposed to be. This is where if you have access to modern timing machine that can measure beats per hour doing this is really quite simple.

 

[SndChsr]

3 minutes ago, JohnR725 said:

this answer is incorrect for variety of reasons? Early American watches typically used flat hairsprings because are easy to make. But they went to over coil hairsprings for timekeeping reasons. But there's a problem with them to get the over coil in exacting location it's easier to pre-make them. This means if you're missing one it's easy to replace it. I'm attaching notes from the 1915 parts catalog that explains how you would do that in other words you have the timing screws.

But as watchmaking got better we don't need over coils typically. So the modern flat hairspring each one is vibrated to the balance wheel. So if you have a modern flat hairspring and detritus swap you'll have timing issues unless you have a way of changing the timing screws.

Now I would've agreed with you that this is probably an over coil and I couldn't tell from the picture but I did look up the hairspring and as you already discovered its flat.

A lot of times with the early flat hairspring is because they didn't have modern timing machines yet they weren't invented they left the hairspring  a little longer so it's not pin exactly where it's supposed to be. So initially pen it at the very end and see if you're running slow. In which case you can move a little more in and speed the watch up until you get close to where you're supposed to be. This is where if you have access to modern timing machine that can measure beats per hour doing this is really quite simple.

 

Thank you John. Very valuable info and very much appreciate the effort you put in to post it. As my post just prior to yours shows, the Pocket Watch DB lists this particular grade as having a flat hairspring. So I assume that no overcoil is required.

This particular hairspring looks to be a new replacement. In this case, as you state, I assume it would need to be vibrated? Because if this is the case then I'm stuck as I don't have a Luthy and do to their prices, probably won't have one for a looooooong time!

[JohnR725]

11 minutes ago, SndChsr said:

This particular hairspring looks to be a new replacement. In this case, as you state, I assume it would need to be vibrated? Because if this is the case then I'm stuck as I don't have a Luthy and do to their prices, probably won't have one for a looooooong time!

technically yes realistically no?

You have a hairspring in the watch right now temporarily pin it to the stud and put the hands-on and see what it's doing. If you're lucky you'll be running slow that means they hairspring is too long. If it's running fast that means you're going to need some screws.

Except I'm forgetting something?

20 hours ago, SndChsr said:

So I'm working on a watch I recently bought off ebay and I noticed that when inserting the balance back into watch, it would immediately over bank and hit one of the banking pins when the balance swung  clockwise. 

minor problem here? Over banking I have a video below let's make sure that's really what's happening and I'm going to assume it's not what's happening? Or I'm going to assume we have terminology wrong? so I don't think the above quote is exactly right and for now I'm going to ignore it.

okay since I remembered something and quoted above a minor confusion here's a problem I'm not sure if I actually understand what you're saying versus you looking at the video and finding your terminology is wrong. When you put the balance in the watch with no power you need to verify that the watches visually in beat? We've covered this on the message board somewhere else I can cover it again if you need that.

Let's make an assumption I hate assumptions by the way but let's do it anyway. Let's assume that somebody unpin the hairspring and shoved something. Ross is going to assume that they hairspring wasn't pinned at the end which I said is typical with vintage watches. This allowed them some room to move things around and then they would correct it with their timing screws. But if you change the pending point of the hairspring it changes the beats of the watch. So if you put the balance wheel in the watch in the watch is grossly out of beat because somebody pin they hairspring at the very end it would mean that I have a wild interpretation of what you're saying the balance wheel is going to be resting with the fork on one of the banking pins your watch as such is grossly out of beat.

So providing I'm interpreting correctly it's back to temporarily pin they hairspring at the very end see if it's resting on one of the banking pins will explain how to put the watch in beat them not good do that until we know where we stand or you can do a search in the group. Then with the hands on the watch running we can see how fast or slow it is and then we'll figure out what we have to do. Because as a guess when you look at the parts catalog you can get a flat hairspring weather was a flat hairspring or over quell hairspring they didn't vibrated to the exact watch. Which meant she would still use the timing screws to bring it into time.

So hopefully this all makes sense

https://youtu.be/DxeM85XRTbU

 

 

 

 

[Nucejoe]

Overcoil, Notice the distance between the balance and cock. 

[JohnR725]

7 minutes ago, Nucejoe said:

Overcoil, Notice the distance between the balance and cock. 

yes, that was something that was bothering me from the pictures up above. But the parts reference book still says is a flat hairspring. if you go to the pocket watch database and look at the other watches in that category trying to find a better picture with the hairspring I snipped out something I'm attaching it. it does look like a flat hairspring. The ancient Elgin flat hairsprings and others have almost zero terminal curve. In other words they hairspring just barely comes out goes to the regulator pins and goes to the stud.

the really one of verification all it has to be done is the balance wheels put in the watch they hairspring just stuck through the regulator pin it doesn't have to be pinned and just get us a decent picture I think you'll find that the regulator pins do go through the outer coil. It was an over coil the hairspring pins would have to be really short .

 

 

 

[Nucejoe]

2 hours ago, SndChsr said:

This particular hairspring looks to be a new replacement. In this case, as you state, I assume it would need to be vibrated? Because if this is the case then I'm stuck as I don't have a Luthy and do to their prices, probably won't have one for a looooooong time!

Join me as I will be vibrating a flat hairspring on a balance wheel without a luthy in a few days. 

I am sure plenty of discussions will pour in to optimize it to a decent balance complete.

Regs 

Joe

 

 

[SndChsr]

4 hours ago, JohnR725 said:

technically yes realistically no?

You have a hairspring in the watch right now temporarily pin it to the stud and put the hands-on and see what it's doing. If you're lucky you'll be running slow that means they hairspring is too long. If it's running fast that means you're going to need some screws.

Except I'm forgetting something?

minor problem here? Over banking I have a video below let's make sure that's really what's happening and I'm going to assume it's not what's happening? Or I'm going to assume we have terminology wrong? so I don't think the above quote is exactly right and for now I'm going to ignore it.

okay since I remembered something and quoted above a minor confusion here's a problem I'm not sure if I actually understand what you're saying versus you looking at the video and finding your terminology is wrong. When you put the balance in the watch with no power you need to verify that the watches visually in beat? We've covered this on the message board somewhere else I can cover it again if you need that.

Let's make an assumption I hate assumptions by the way but let's do it anyway. Let's assume that somebody unpin the hairspring and shoved something. Ross is going to assume that they hairspring wasn't pinned at the end which I said is typical with vintage watches. This allowed them some room to move things around and then they would correct it with their timing screws. But if you change the pending point of the hairspring it changes the beats of the watch. So if you put the balance wheel in the watch in the watch is grossly out of beat because somebody pin they hairspring at the very end it would mean that I have a wild interpretation of what you're saying the balance wheel is going to be resting with the fork on one of the banking pins your watch as such is grossly out of beat.

So providing I'm interpreting correctly it's back to temporarily pin they hairspring at the very end see if it's resting on one of the banking pins will explain how to put the watch in beat them not good do that until we know where we stand or you can do a search in the group. Then with the hands on the watch running we can see how fast or slow it is and then we'll figure out what we have to do. Because as a guess when you look at the parts catalog you can get a flat hairspring weather was a flat hairspring or over quell hairspring they didn't vibrated to the exact watch. Which meant she would still use the timing screws to bring it into time.

So hopefully this all makes sense

https://youtu.be/DxeM85XRTbU

 

 

 

 

Hello John,

The problem is me all along. I completely misdiagnosed the problem. Having taken the hairspring out of the equation, I put the balance wheel back in its place, along with the balance cock so it runs freely without the hairspring. Immediately I noticed that the problem is not a lack of coil (though that had me wondering as I was under the impression that an overcoil is needed on ALL hairsprings), but rather the problem resides with over banking. When the balance wheel swings clockwise it is immediately stopped and has not free motion. 

I see that the pallet jewels are engaging the escape wheel too high due to a restriction of the banking pins being too close together. So, total failure in diagnosis on my part. Mea culpa!

Thank you for your excellent explanation. I'll deal with the banking pns first (unless I'm wrong to do that) and then I'llI will follow your instruction, pin the hairspring into the stud and see what my timegrapher reveals as far as accuracy is concerned.

BTW, I did not know that was your youtube channel! I've been a regular there and your videos are very informative...as are others who are active members of this forum.

I'll be at this all day today. Hopefully I can figure this mess out. If not, please be patient with me as I may ask more questions.

Again, thank you for all the time and effort put into your answers.

[SndChsr]

4 hours ago, Nucejoe said:

Overcoil, Notice the distance between the balance and cock. 

Thank you! So much to learn...sigh

[SndChsr]

2 hours ago, Nucejoe said:

Join me as I will be vibrating a flat hairspring on a balance wheel without a luthy in a few days. 

I am sure plenty of discussions will pour in to optimize it to a decent balance complete.

Regs 

Joe

 

 

Hello Joe, will this be live on YT? I would love to see that. Especially if done without the use of Luthy!

[Nucejoe]

3 hours ago, SndChsr said:

Hello Joe, will this be live on YT? I would love to see that. Especially if done without the use of Luthy!

Hi, A walk through on WRT.  Discusions/ Question, advice/help are always useful to improve ideas.

Joe

 

 

 

 

[JohnR725]

3 hours ago, SndChsr said:

TW, I did not know that was your youtube channel! I've been a regular there and your videos are very informative...as are others who are active members of this forum.

I suspect something didn't quite come out in the text that I had it's not my YouTube channel belongs to a school I believe.

3 hours ago, SndChsr said:

I see that the pallet jewels are engaging the escape wheel too high due to a restriction of the banking pins being too close together. So, total failure in diagnosis on my part. Mea culpa!

 

3 hours ago, SndChsr said:

I'll deal with the banking pns first (unless

out of curiosity do you know what the banking pins are used for?

[david]

SndChsr,

Imagine a perfectly balanced   beam with equal   lengths past a fulcrum.   As long as the beam is perfectly balanced the beam will just sit there.  When you move the beam so the length of one side is longer than the other side, the beam will tip. Now replace the concept of a beam with a watch hairspring perfectly balanced around a staff. As long as the coils of the hairspring are the same distance from the center of the axis, the pivots will (theoretically) ride in the center of the jewell holes resulting in zero (again theoretically) friction.  Now you wind up the hairspring, by rotating the balance wheel, and the  coils are no longer equidistant from the axis of the staff.  This pushes the top pivot against one side of the jewell hole and the bottom pivot against the other side of the jewell hole. This creates friction which interferes with the oscillation of the balance wheel. A "perfect" balance wheel will oscillate in the same period of time regardless of the amplitude.  When friction is introduced, the balance wheel no longer oscillates in the same time period as the amplitude changes. As the mainspring runs down, the force applied to the impulse jewell is reduced, making the friction on the pivots more difficult to overcome.  The result is the rate of oscillation changes with the change in force.  This is called isochronal error.  To deal with this issue the overcoil was invented to keep the hairspring coils more even as the balance wheel  pulsed back and forth.  The overcoil configurations were figured out at the watch factories through laborious trial and error.  The best a watch repairer can hope for is to find a diagram for their particular balance wheel  and hairspring configuration and form the overcoil to match the factory pattern.   

david

Edited June 8, 2021 by david

[nickelsilver]

There are texts on overcoil forms. Some come up gradually and gentlty bend in toward the regulator pins, some come up and abruptly streak over. In some cases abrubtness is necessary to avoid hitting the center wheel for example. The curve upwards can be a couple of rather severe bends or a series of gentle bends.

 

End of the day you can make an "overcoil", then check your isochronal error, which is effectively checking the rate at about full wind and about 24h after in one position, imagining that everything is adjusted "nsk nsk" all around. If, at low wind, the rate is gaining, adjust the body of the overcoil away from the balance center. If losing, adjust it toward the balance center. It works and feels like magic when you do it. Of course you throw it out of beat when doing this and you correct that.

 

This is advanced timing stuff but if you're comfy with hairsprings you suddenly become a wizard, and that's motivation to become comfy with hairsprings. A watch with overcoil, while seemingly more daunting than a flat spring watch, offers much more adjustment for fine timing. You can send your 10 dollar flea market find to Besançon haha.

[SndChsr]

20 hours ago, JohnR725 said:

I suspect something didn't quite come out in the text that I had it's not my YouTube channel belongs to a school I believe.

 

out of curiosity do you know what the banking pins are used for?

Hello John, as far as banking pins go, yes I do understand their purpose. Limit side to side motion of the pallet as well as the amount of contact of the escape wheel tooth with the pallet jewel. The reason why I suspected the banking pins as being the culprit is because I've had a previous case where the movement of the pallet was grossly limited and the pins were the culprit.

Perhaps I did not explain my issue eloquently...which is not unusual for me

I noticed that the swing of the balance wheel on this particular movement is severely limited. At first, I thought that absence of the overcoil is to blame as it may somehow hinder the overall movement of the balance wheel. In retrospect, that was a completely wrong if not ridiculous analysis on my part. 

Then I thought it may be the banking pins...again, limiting the movement of the pallet. Well, having taken a very close look at this, I now see that the balance wheel freely swings counter clockwise (even with the hairspring removed). When trying to swing the balance wheel clockwise, it stops almost at the point where the pallet fork is in the center and correctly aligned with the impulse jewel, as if the balance were correctly aligned to be set in beat. In other words, the balance stops almost in the dead center, between the 2 pins. The impulse jewel is correctly aligned inside the pallet fork. My only explanation would then be pallet jewels which are misaligned (too long?). In fact, looking at the escape wheel and its engagement with the pallet jewels, and referring to the Watch Repair Manual, the jewels indeed do seem to be engaging the escape teeth too high. 

I'll try to make a video of it and post it on YT. If this were the case, I assume re-positioning and re-schellacing the jewels?

Thanks again.

BTW, the way you had worded your post, I was under the assumption that the dude in the watch school video was you lol. My bad comprehension rearing its ugly head!

[SndChsr]

1 hour ago, nickelsilver said:

There are texts on overcoil forms. Some come up gradually and gentlty bend in toward the regulator pins, some come up and abruptly streak over. In some cases abrubtness is necessary to avoid hitting the center wheel for example. The curve upwards can be a couple of rather severe bends or a series of gentle bends.

 

End of the day you can make an "overcoil", then check your isochronal error, which is effectively checking the rate at about full wind and about 24h after in one position, imagining that everything is adjusted "nsk nsk" all around. If, at low wind, the rate is gaining, adjust the body of the overcoil away from the balance center. If losing, adjust it toward the balance center. It works and feels like magic when you do it. Of course you throw it out of beat when doing this and you correct that.

 

This is advanced timing stuff but if you're comfy with hairsprings you suddenly become a wizard, and that's motivation to become comfy with hairsprings. A watch with overcoil, while seemingly more daunting than a flat spring watch, offers much more adjustment for fine timing. You can send your 10 dollar flea market find to Besançon haha.

I've actually been buying tons of damaged movements just so I can become comfortable working with hairsprings. Spent a good amount of time straightening them out so I can become proficient at it! In this case however, I totally misdiagnosed. The hairspring is not the cause of the issue I face here. 

Thank you for the suggestion though. It's all knowledge that I'm happy to absorb!

[david]

Nickelsilver,

Thanks for sharing the information. Here is a video put out by the Lititz Watch School showing one way of forming the bend to raise the overcoil.

https://youtu.be/4ddWRl_NVTE 

Physically forming the bend  is a small part of the problem  Here is a diagram of one overcoil design.

With this particular design Point A is where the upward bend starts and point B is where the upward angle terminates putting the overcoil above and parallel to the coil. From point B to point C is a continuation of the spiral curve of the hairspring.  From point C through point D terminating at point is a curve that would be categorized as a mathematical spline.  If you are a math buff these curves are approximately  defined as F(x)=AX^3+BX^2+CX+D. If not, no problem as these curves are formed by trial and error.  Point E is the location of the regulator pins.  From point E to the stud the curve must be a circular radius. As  The curve can be  adjusted at point D toward or away from the collet to  help reduce the isochronal error.

This is a Stephan Pahlow video showing the use of overcoil charts to obtain the desired overciol pattern.

https://youtu.be/Qlcb3czu5_s

david

 

Edited June 9, 2021 by david
added additional information

[JohnR725]

3 hours ago, SndChsr said:

as far as banking pins go, yes I do understand their purpose. Limit side to side motion of the pallet as well as the amount of contact of the escape wheel tooth with the pallet jewel.

strangely enough it does all of those things but I appear to be confused? I was under the impression that the banking pins were to control the fork horn clearance? Then once you get that adjusted you can adjust the jewels to get the proper locking depth.