Jump to content

Nost

Recommended Posts

Hi again all,

since there are some really knowledgeable people here I hope you can impart some wisdom regarding repairing a bent hair spring. after getting this old movement working again with the help of this group (new mainspring), I need a little more advice.

after hooking this beaut up to  watch-a-scope it shows low amplitude and very erratic trace. I can see the hairspring is bent and need to start here in addressing the issue.

What tools are needed (not willing to spend £50 on hairspring levers at the moment)? I have basic tools  (enough to service a movement), but what are the essentials for working on a hairspring? i.e for removing from a balance safely what tweezers types/sizes are recommended for correction etc.

Also, is there an easy way to identify and obtain replacement hairsprings? I assume they are very interchangeable as it would not have been easy to make new hairsprings for every watch model.

 

 

Link to comment
Share on other sites

14 minutes ago, Nost said:

Also, is there an easy way to identify and obtain replacement hairsprings? I assume they are very interchangeable as it would not have been easy to make new hairsprings for every watch model.

With the exception of American pocket watch parts hairsprings typically come only with the balance wheel they were vibrated to. So that means if you need a new hairspring you need a balance complete unfortunately.

 

 

  • Like 1
Link to comment
Share on other sites

Only minor manipulation is practical at the begining.

Check if the collet is removable? Or show the collet.

If removable,  insert a razor balde under the collet, work the razor around to gradually raise it, go to thicker blade to and finally lift the HS. Eta showed a solution of some sort that loosens the collet so make removing easy.

I have seen highly skilled individuals manipulate( minor) the HS on the balance.

Once removed, rinse. Mount the collet on a needle, stick the needle on a white sheet of paper, lay on flat cardboard. 

High magnification and well dressed tweezers are must.

You should master any techniques you see.

Start with Makrs, two tweezers, show pix, ask for advice.

  • Like 1
Link to comment
Share on other sites

2 hours ago, JohnR725 said:

With the exception of American pocket watch parts hairsprings typically come only with the balance wheel they were vibrated to. So that means if you need a new hairspring you need a balance complete unfortunately.

 

 

Thank you for the heads up on that one... at least i can direct my search if it all goes pete tong.. I've found some (not cheap) but the the spring is non functional any way so if i balls it up, no loss..

Link to comment
Share on other sites

2 hours ago, Nucejoe said:

Only minor manipulation is practical at the begining.

Check if the collet is removable? Or show the collet.

If removable,  insert a razor balde under the collet, work the razor around to gradually raise it, go to thicker blade to and finally lift the HS. Eta showed a solution of some sort that loosens the collet so make removing easy.

I have seen highly skilled individuals manipulate( minor) the HS on the balance.

Once removed, rinse. Mount the collet on a needle, stick the needle on a white sheet of paper, lay on flat cardboard. 

High magnification and well dressed tweezers are must.

You should master any techniques you see.

Start with Makrs, two tweezers, show pix, ask for advice.

 

1 hour ago, Nucejoe said:

Replacement H/S may be sourced from material houses. Interchangable normally within the same family. 

You may have to vibrate and cut.

Should you come to vibrating, I work with you.

Thanks for your information, the collet is movable, as  i noticed the balance jewel wasn't within the banking pins when at rest... i made a slight alteration using my smallest driver and that helped with the alignment, but not the erratic trace.

After John's reply above, I think I'll attempt a dismantle and just look at the condition of the hairspring... I'll post a pic and get some thoughts.

Link to comment
Share on other sites

Fault at impulse jewel. Loose, dirty, damaged impulse jewel.

Did you see deftormity in HS coil or judging is based on erratic behaviour?

You can observe behavior of the HS from the movement side.fouling itself, sticking, rubbing on the cock or balance spokes are observable,  if you can video record it, slow replay reveals more.

  • Like 1
Link to comment
Share on other sites

8 hours ago, Nost said:

after hooking this beaut up to  watch-a-scope it shows low amplitude and very erratic trace. I can see the hairspring is bent and need to start here in addressing the issue.

I was thinking perhaps a different approach is needed for this watch? One of the problems with the message board is if you ask a question we will give you an answer but is the question the correct question for the problem of the watch?

To get a better feel I read the prior messages related to this watch and? So is this your first watch learning watch repair? Then what was the condition of the watch before you started repairing it? I noticed you misplaced one of the end stones did you find that?

then it be really helpful if we could see the watch-a-scope Print out. That is preferably into positions like dial down and crown down

 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Felsa_690

  • Like 1
Link to comment
Share on other sites

2 hours ago, JohnR725 said:

I was thinking perhaps a different approach is needed for this watch? One of the problems with the message board is if you ask a question we will give you an answer but is the question the correct question for the problem of the watch?

To get a better feel I read the prior messages related to this watch and? So is this your first watch learning watch repair? Then what was the condition of the watch before you started repairing it? I noticed you misplaced one of the end stones did you find that?

then it be really helpful if we could see the watch-a-scope Print out. That is preferably into positions like dial down and crown down

 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Felsa_690

Thanks for taking an interest John,

This is my second watch, I am new to watch repair, my first was a slava 2427.

I have no baseline data for this watch as the mainspring, second sweep pinion and crystal were broken. It was sold as a non runner.

I have replaced all broke parts and now have life in the watch but as you can see in the vid, the hairspring is compressed on one side (the spirals are not concentric. I can see the spring is deformed at the regulator and this is where I'm guessing there is a problem (though admittedly there may be other issues but I'm unsure of any others until I've sorted the hairspring out.

I'm unable to slow down the vid bit I  don't think the  watch o scope trace will be of any help until the spring is corrected/replaced (though I have attached it in case it can provide any clues.

just to note, I was curious, as the only way to set the beat is through adjusting at the collet I thought I'd have a look, I made a slight adjustment as the impulsed jewel was not within the banking pins. My adjustment was as improvement but still I don't fell that is there is a beat error, That it is significant at this time due to the larger issue of the hair spring (and what ever else may be going on.).

Thanks for your continued input

felsa 690.PNG

Link to comment
Share on other sites

Starting off with the timing results first as they are helpful if they're correct? I like the program and it works nicely but it be nice if you verify with another watch that it actually is working and can pick up watches. We've had another discussion were somebody had questionable results because they weren't getting a good clean signal from the watch. Then if the signal is correct when you get random dots the numbers mean nothing. Timing software/machines can only run correctly if  they have a nice clean signal and this signal is not nice.

It would be nice before undertaking hairspring work to  find a disposable watch to practice with. Even a disposable balance wheel just to get a feel for how easy it is to bend a hairspring.. It takes a lot of practice to be good at bending hairsprings..

Your video is interesting and misleading perhaps. Misleading because were looking at  your hairspring doesn't look centered but is that really an issue? So the hairspring needs to only be attached to the stud,,  collet and go between the regulator pins. If it's off-center it's not an issue  as long as is not rubbing bumping touching anything else. I'm attaching an image  of your video notice the curvature from the stud through and we can't see the regulator pins?  It would be nice to have a slightly angled view  to see the curvature better.

To understand what I'm getting at I'm attaching a link to a video.. As a reminder in the video he comments that  they hairspring cc gets in this condition because the students didn't properly form their terminal curve. You had a properly formed terminal curve and casually it looks like it's not quite right? So we need to figure out why it's not right and if it is not right before attempting to fix the problem.. Usually the place that  hairsprings get bent at is the stud to because it's an easy place for watchmakers to bend their hairsprings accidentally.

Then recovering too many subjects  and if you're having a terminal curve problem that needs to be addressed before putting the watch in beat. I checked the service bulletin for your watch  it appears to have a fixed stud mounting. A lot a modern watches have floating stud's they can be moved to put the watch in beat. Visually you should build a verify whether your watches in beat without the timing machine and then use the timing machine the verify you got it correctly. o it in the watch will have the pallet fork between the banking pins but that doesn't guarantee the watches in beat. Technically the roller jewel has to be centered  in the slot of the fork which is really hard to see so we usually go by the pallet fork should be centered between the banking pins. I can explain more about that but let's see what we can do about they hairspring and  getting a better signal on the timing machine.

https://www.youtube.com/watch?v=4EYk787sjAM

 

 

hs-tc.JPG

Link to comment
Share on other sites

16 hours ago, JohnR725 said:

Starting off with the timing results first as they are helpful if they're correct? I like the program and it works nicely but it be nice if you verify with another watch that it actually is working and can pick up watches. We've had another discussion were somebody had questionable results because they weren't getting a good clean signal from the watch. Then if the signal is correct when you get random dots the numbers mean nothing. Timing software/machines can only run correctly if  they have a nice clean signal and this signal is not nice.

It would be nice before undertaking hairspring work to  find a disposable watch to practice with. Even a disposable balance wheel just to get a feel for how easy it is to bend a hairspring.. It takes a lot of practice to be good at bending hairsprings..

Your video is interesting and misleading perhaps. Misleading because were looking at  your hairspring doesn't look centered but is that really an issue? So the hairspring needs to only be attached to the stud,,  collet and go between the regulator pins. If it's off-center it's not an issue  as long as is not rubbing bumping touching anything else. I'm attaching an image  of your video notice the curvature from the stud through and we can't see the regulator pins?  It would be nice to have a slightly angled view  to see the curvature better.

To understand what I'm getting at I'm attaching a link to a video.. As a reminder in the video he comments that  they hairspring cc gets in this condition because the students didn't properly form their terminal curve. You had a properly formed terminal curve and casually it looks like it's not quite right? So we need to figure out why it's not right and if it is not right before attempting to fix the problem.. Usually the place that  hairsprings get bent at is the stud to because it's an easy place for watchmakers to bend their hairsprings accidentally.

Then recovering too many subjects  and if you're having a terminal curve problem that needs to be addressed before putting the watch in beat. I checked the service bulletin for your watch  it appears to have a fixed stud mounting. A lot a modern watches have floating stud's they can be moved to put the watch in beat. Visually you should build a verify whether your watches in beat without the timing machine and then use the timing machine the verify you got it correctly. o it in the watch will have the pallet fork between the banking pins but that doesn't guarantee the watches in beat. Technically the roller jewel has to be centered  in the slot of the fork which is really hard to see so we usually go by the pallet fork should be centered between the banking pins. I can explain more about that but let's see what we can do about they hairspring and  getting a better signal on the timing machine.

https://www.youtube.com/watch?v=4EYk787sjAM

 

 

hs-tc.JPG

@JohnR725

I've compiled some more information, mainly pictures to attempt to clarify the situation....

Some better images of the hairspring including the regulator pins (the spring appears to be seated correctly), however there is definitely a deformation of the spring in the termianl curve as you correctly pointed out. I think the spring is fouling on itself where the coils are tighter.

I have attached an image of the graph from the felsa which indeed looks odd (though I did alter the regulator to see if positively or negatively affected the trace. Apparently, according to the graph, it had a negative affect as the software cannot find a consitant 'tick' now.

I have also attached a raw data image from a AS2066 movement that I have yet to work on (just to show the reliability of the setup).

I do have some scrap slava 2428 movements that i used to source parts fo the 2427 I renovated. I could use these to practice manipulating hair springs. I have found some balance completes for this movement (watch also shares parts with the felsa 1560). forseably, if i do make a mess, replacments are available (though I would like to avoid that if possible.

I have seen the video you posted, and watching the beat alter with the deformed terminal curve is the reason I was asking about mainspring manipulation in my OP.

What are your thoughts? Should I be looking elsewhere?

allaine as2066.PNG

felsa graph.PNG

IMG_20190811_125852.jpg

IMG_20190811_125932.jpg

IMG_20190811_130128.jpg

IMG_20190811_130153.jpg

Link to comment
Share on other sites

Looking at your pictures I found what I was looking for so I've snipped out the interesting ones. Looking at the images attached you'll notice that your hairspring is touching the backside of the regulator pin. It also looks like your terminal curve is distorted. In the video I gave you he clearly shows how to make the terminal curve but you have to be careful in that each watch is different. So for that I went looking on eBay to see if anyone had nice picture is of a balance complete. Currently on eBay there are quite a few balance completes a couple have pictures this one was the nicest picture.

Personally for fixing this I would do it in the watch. That's because the problem is in the watch. Looks like the balance complete has a nice gradual terminal curve which I think yours still has. The typical problem and it's the most common problem for watchmakers is distortion in the around the stud.

The preferred method of practicing with the hairspring would be to practice on a disposable watch. Practice distorting the hairspring bending it then fixing the problem until you get good at it. But as there are replacement balance completes on eBay you can practice with this watch what's the worst you're going to do? I notice they even have movements on eBay.

f 690.JPG

hs-3.JPG

hs-2.JPG

hs-1.JPG

  • Like 2
Link to comment
Share on other sites

@JohnR725

Thanks for your time and explanation. I've identified the 2 probable areas that will need altering though I will analyse the spring with the regulator in different positions before starting, and as you suggest this should be done in situ. My dexterity is pretty good and as you say, given the worst outcome, there are replacement balance completes.

I will need a good second pair of tweezers, I've heard brass is a good choice as it is a little softer than steel. I do not need to rush this as I wish to learn and rushing will not accomplish anything... other than another set of tweezers, is there anything else I should consider to make the job easier?

Inkedtrminal curve_LI.jpg

Link to comment
Share on other sites

Usually when I'm doing hairspring work I like to use number five tweezers. They have really nice long and narrow tips.

Then it looks like from your photograph you have identified the areas that don't look quite right. The other thing that was bothering me the one that's closest to the stud and it may be an optical illusion almost looks like it's twisted their.

Link to comment
Share on other sites

Googling I found you a PDF at the link below it's on page 16.

Then I really should find you a picture of this but the other method is to not to remove the hairspring. Some companies Rolex for instance supplies special tools for this which were not going to have and are exceedingly expensive even if we did. So what they do is to push the staff out and once the staff is free of the balance wheel they continue to push the staff right through the Hairspring collet.

 

 

http://www.awci.com/wp-content/uploads/ht/2006/2006-05-web.pdf

  • Like 1
Link to comment
Share on other sites

5 hours ago, CaptCalvin said:

Can you share this solution with me? ETA collets are a bear to handle. I even had one crumble on me when trying to remove 

Saw it used in a video by ETA, just a droplet applied to the collet and lifted the collet with tweezers.

Sorry don,t have the link nor the video title. Lets hope @HSL  can remember which video I am talking about,  he is a data base in person.

Link to comment
Share on other sites

52 minutes ago, Nucejoe said:

Sorry don,t have the link nor the video title. Lets hope @HSL  can remember which video I am talking about,  he is a data base in person.

Sorry I can´t seem to recall a video with this but I'm sure there are one out there.

To make adjustments on a hairspring with a fixed stud is a dying art and can be pretty tricky.
As i look at the Pictures here the spring and especially the endcurve is not very satisfying, but that is just one of the problems.
If you look at the regulator pins they also are in a poor shape both the index pin and the boot seems to be bent to much inwards . As a rule one should keep a space between them like twice the hairspring thickness so the spring can "Breath". Adjusting these after excessive bending been done might easily break them off so be very carefull.

To get the collar of there used to be a tool you fitted into the collar slot and then gently turned it around whilst liting upwards.
Before this operation were performed you put a dropp of very thin oil on the balance staff just above the collet.
I never bought this tool since I learned to do this operation with a small screwdriver. One probably can make one perfectly shaped if one is willing to sacrifice a screwdriveer blade.

There are however Tools for lifting it out there like the Bergeon 5430 levers. For tuning and adjusting the collet position Bergeon 30017-TP.
When you adjust the collet position you just adjust the beat error and that probably is to be done after fixing the rest.
Unfortunately I'm not at home so I could take nice Pictures of the task but maybe I post some later.

Before you take it of you might record it's position right now on the balance staff in relation to the impulse pin so one doesn't have to start from scratch to find a good starting position.
 

Link to comment
Share on other sites

10 hours ago, Nucejoe said:

 I will google for video the first chance. Check this how he loosens the collet through the collet slot.

https://www.Youtube.com/watch?v=4EYk787sjAM

Done few of these myself. Favorite kind of collet to deal with. Just need an oiler and they come off without fuss. No need for razor blades levers or sweat. Unfortunately these don't seem to be in fashion these days.

Really hope you find that video. It would really save me from any more sweat and tears haha.

Edited by CaptCalvin
Link to comment
Share on other sites

On 8/12/2019 at 12:59 AM, JohnR725 said:

Usually when I'm doing hairspring work I like to use number five tweezers. They have really nice long and narrow tips.

Then it looks like from your photograph you have identified the areas that don't look quite right. The other thing that was bothering me the one that's closest to the stud and it may be an optical illusion almost looks like it's twisted their.

I will soon be ordering some tweezers (on my next pay cheque) and have a 'bash'. 

Are there any tips on doing this while the spring is in situ? All of the videos I've seen remove the spring entirely... From what I can see, the terminal curve, while simple, seems very gradual in its arc with an ill defined starting point with no seemingly defined bend (other than the one leading to the stud). 

Should I attempt to correct the stud curve first? 

Link to comment
Share on other sites



×
×
  • Create New...