Jump to content

3D printed RS mainspring winder project for hobbyists


Recommended Posts

Hi guys,

I have started a project in Github to generate all the possible types of mainspring winder sizes from a single base design using a Python script.
https://github.com/vishnu350/rs-mainspring-winder

The STL files can be downloaded from the release section:
https://github.com/vishnu350/rs-mainspring-winder/releases

This project is a 3D printable "Rising Sun" watch mainspring winder for hobbyists. Mainspring winding for hobbyists has always been a problem, as most of us cannot afford the $1000 Bergeon set. Or sometimes, the appropriate size just isn't available. This forces most casual hobbyists to resort to handwinding or other similar stunts, which is not ideal.

This mainspring winder is designed using FreeCAD to be easily 3D printed with no post-processing/finishing required. It will approximately cost $5/set by most 3D print service providers (use eBay to find someone near you). The design is parameterized, allowing you to dynamically change the size/dimensions of the winder and the entire design will be recalculated accordingly. The RS mainspring winder is a pleasure to use when constructed correctly, and can wind in both clockwise and counter-clockwise directions.

I have uploaded a video describing the 3D printed RS winder parts and how to use it.

Will appreciate any help or feedback as to further improve the design, as I actually do not own a 3D printer (I pay for each print). I also do not know the complete range of sizes that is required to be uploaded into the release section. Also, the optimal design parameters may still not be perfect, but it seems to work pretty well for me at the moment.

Thanks.

Edited by Zero
  • Like 6
  • Thanks 3
Link to comment
Share on other sites

The size I printed and tested with was for a 10.5mm mainspring diameter. I have no idea how the winder looks like for other sizes, but they looked OK on screen. Let me know if you test it out.

If you want to customize the winder, just measure the your mainspring/barrel dimensions with a caliper. Change these values in the parameter spreadsheet inside the rs-winder.FCStd file accordingly. Then you can export the STL files for 3D printing.

For printing, I use the service from these guys as they are nearby me: http://www.siegeadvancedmanufacturing.com

Link to comment
Share on other sites

5 hours ago, Sjk4x4 said:

I like your Staple - tab solution!

Now a person can buy a 1000+ winder set or a $300 printer and have their own.(Or order from a printing shop for less)

Yeah, the staple pin is an everyday item accessible to everyone. It works in both winding directions. It seems to be able to withstand the winding force, even after multiple uses. If it warps or breaks, it is easy to replace.

Link to comment
Share on other sites

17 hours ago, bklake said:

I'm going to print one and try it. 

Nice ?. Let us know if you tweak the parameters or did anything else to improve it. I can include it in the next release if it works well and is applicable to the majority of future users.

Some parameters I just arbitrarily chose, such as are the spring diameter shrink %, the barrel opening width, and the width/heights of the winder body, etc. I'm sure it can be optimized further, but I don't have a printer to keep testing them out. So the values i chose may seem pessimistic, due to my limited capacity for R&D.

Link to comment
Share on other sites

I'll also try your design. I don't have a 3D printer either, but I've contacted with someone that is printing the 4 pieces for me, and he will send them to me by mail.

Not having the tool yet, I don't fully understand how I should shape the anchor with the stapler pin, it's not clear on the video. I hope I'll figure it out when I have the tool with me.

Also I hope it's not a huge problem to have a barrel bowl a bit bigger than the watch barrel outer diameter and a housing barrel a bit smaller than the inner diameter. I'm talking of one or two tenths of a mm.

  • Like 1
Link to comment
Share on other sites

One of the main issues I faced, is if the plunger doesn't snugly fit inside the housing barrel, it wont be able to plunge correctly. The mainspring tail would get caught between the gap while plunging.

What I did was, I would always print multiple plungers and hope one of them would fit snugly. But this is not a good solution for everyone. Lubricating the mainspring and barrel helps a lot, but is still not a solid solution.

So I had a good think about it, and added three 'teeth' to the plunger. Low and behold all of my prints work regardless of how snug the plunger fits! Please print the latest version with this change (release v0.5 and above).

IMG_20210521_114042.jpg

  • Like 2
Link to comment
Share on other sites

15 hours ago, aac58 said:

Not having the tool yet, I don't fully understand how I should shape the anchor with the stapler pin, it's not clear on the video. I hope I'll figure it out when I have the tool with me.

Also I hope it's not a huge problem to have a barrel bowl a bit bigger than the watch barrel outer diameter and a housing barrel a bit smaller than the inner diameter. I'm talking of one or two tenths of a mm.

I just used a wire cutter or pliers, and kept trimming the stapler pin until i got the size I wanted. It should fit roughly like how you see it in the video. If you get it right it will keep working without breaking.

The bowl size is OK if you choose to print a bigger one. You can use also use Rodico to help center it if it is too big. If you send me your dimensions, Ill try to include it in the next release. I'm still figuring out how to bundle all the different sizes in a neat way.

Edited by Zero
typo
  • Like 1
Link to comment
Share on other sites

I wonder if the bowl could be done away with?  Put a raised ring on the winder, slightly (approx barrel wall width) outside the recess where the spring sits, which is designed to have the top of the mainspring barrel fit inside it.  That will center the barrel when the spring is pushed it.  It'll be a bit easier to see what's happening too I think.

It only needs to be about half the barrel height, so that the teeth on the barrel no longer need to be accounted for.  Then put a matching groove on the winding portion.

  • Like 1
Link to comment
Share on other sites

That's a good suggestion. May need to think on it a little more. I didn't try your way because the width of the barrel wall was inconsistent across different manufacturers (although with the same mainspring diameter) . I did try variations of it though, by making a little ring on the inside, but it failed. These were my findings:

  1. The ring width/height was too small for the average 3D printer to properly print. It didn't turn out well.
  2. I tried reducing the inner mainspring barrel diameter so that this ring could be bigger. However this stresses the mainspring when winding, causing the tail to eventually break from fatigue. So that was just bad.
  3. I tried another idea variation where I made the bowl recess on the bottom of the winder base, but it felt uncomfortable on my fingers when winding.
  4. I tried putting the mainspring barrel into the winder base arbor for centering, but the arbor hole on the mainspring barrel usually smaller that the arbor itself, so wont fit.

So I gave up and just designed the bowl as it was the most accessible solution and would work for everyone, even with lower end 3D printers. Also, it cost me less than a dollar to print per piece. It does work very well, I haven't failed the plunging step even once with the latest prototype. It gives a satisfying 'click' when the mainspring pops in.

Also, I find the bowl very useful for disassembly/reassembly service, it kinda works as a staking tool thanks to the arbor recess hole. You can use it to pop the barrel cover safely back on as well. And it stays put when you do the wall lubrication, or if you choose to lube the spring after winding.

Link to comment
Share on other sites

6 hours ago, Fred said:

I printed it with PLA+ and it worked, I tried 4 times and it works as in the video, thank you for designing and sharing this.

Forget about hand winding mainsprings 😂😂

Nice! You got it to work on PLA+. Bye bye handwinding. No more losing bits of skin on your fingers! 😄😄

I'm curious to know more.
- Were you able to use the existing STL files, or did you customize the params and regenerate?
- What do you think of the clearance/fit for the mating parts?
- Did the diameter markings on top turn out well for your print?

Link to comment
Share on other sites

Hi, yes I used the existing STL files.

I think it’s good, it works well but was wondering if you could improve the winding base and make it modular so you can print multiple rod with different little holes and place a pin as the hook so can place it just where you want it, I’ll try to design something like that but here’s a really bad sketch of the idea. 
the diameter markings didn’t turn out well because I printed it upside down because of stringing issues, also I’m using masking tape as the printing surface but you can still tell the numbers. 
I’m attaching some pics.

thanks for sharing again and if you want me to print anything let me know.

 

 

6EBA78E3-1F77-4C0B-AF88-38E06893E7E6.jpeg

754CC3F6-63E7-4399-B82E-9624A163989B.jpeg

F5D37126-C88C-45FB-B077-94EA2BCFAF15.jpeg

Edited by Fred
  • Thanks 1
Link to comment
Share on other sites

5 hours ago, Fred said:

I think it’s good, it works well but was wondering if you could improve the winding base and make it modular so you can print multiple rod with different little holes and place a pin as the hook so can place it just where you want it, I’ll try to design something like that but here’s a really bad sketch of the idea.

A modular arbor pillar hole? I think that sounds like good idea to further improve the design. That way if users break the arbor pillar thingy, they can just replace it. However, I had trouble printing arbor pillars that were smaller than 2mm diameter, because it always came broken in the mail 😑. Maybe a better solution would be to find a way to securely slot in a steel nail from the bottom, if we can find something of the right size?

Another thought is, since the arbor pillar hole would be quite close to the anchor hole, would both holes print correctly?

6 hours ago, Fred said:

the diameter markings didn’t turn out well because I printed it upside down because of stringing issues, also I’m using masking tape as the printing surface but you can still tell the numbers.

Hmm. I was never able to get it to print right either. Oh well, it is just text. We'll figure it out eventually 😆

6 hours ago, Fred said:

thanks for sharing again and if you want me to print anything let me know.

You're welcome! I had something that worked for me for a while now, and I thought to myself that it would be very selfish if I didn't share, as I am aware of others struggling with this problem as well.

Link to comment
Share on other sites

  • 2 weeks later...
On 5/21/2021 at 6:04 AM, Zero said:

I just used a wire cutter or pliers, and kept trimming the stapler pin until i got the size I wanted. It should fit roughly like how you see it in the video. If you get it right it will keep working without breaking.

The bowl size is OK if you choose to print a bigger one. You can use also use Rodico to help center it if it is too big. If you send me your dimensions, Ill try to include it in the next release. I'm still figuring out how to bundle all the different sizes in a neat way.

Hi there Zero! 

After a lot of problems with the courrier I got the first set of pieces to play with, I was very excited, but the joy lasted only two tries 😪

I've tried to cut the stapler to the best of my understanding. Then I made a first try and although the spring seemed to slip sometimes inside the housing, I managed to get all of it into the tool, even the bridle that was a bit tougher. However, when trying to take the winderbase off, the spring wanted to come with it instead of staying in the housing, so it didn't work this forst time.

I thought that maybe the problem was that the stapler was a bit too long, so I cut it a bit smaller. I tried to load the spring again into the housing, this time I could not get the bridle into it. When I took the base off, I saw that the rod had broken. I probably did something wrong. Any advise you can share is very welcome.

EDIT: Forgot to mention that when I received the pieces, the rod on the base couldn't get into the hole of the plunger. I sanded it in the middle (not in the zone that is now broken) until it could get inside the hole with no apparent friction, but maybe I should have sanded it a bit more (or a bit less, who knows) Next time I think I will try to look for a way to broaden the hole instead.

Unfortunately I only ordered one winder base, so I will need to order it again, and that will take some time.

Anyway, thank you very much for sharing this design, I'm sure I'll get it working soon!

20210609_192553.thumb.jpg.f4839a5aac1a3fbedc24c784eb14fd0a.jpg

 

Edited by aac58
  • Like 1
Link to comment
Share on other sites

18 hours ago, aac58 said:

I saw that the rod had broken

Alright, today I've done a bit of brico. After sanding the remains of the winder base arbour, I've replaced the broken arbour using a pin vise and a 2 mm drill bit and a 2 mm metal rod.

estrapada1.jpg.77b3b32e952b8642acde5f74d2fbf69d.jpg

estrapada2.jpg.81021a91a052c7b9d5a7a77a3d9e11e6.jpg

I tried again with a bad shaped Seiko 70xx spring I'm using for testing. For me it's really difficult to insert the staple into the spring hole. Or more specifically, I cannot, as there's no room for the staple between the arbour and the spring, and I think I deformed the spring even more. However, after trying afew times, I managed to close the base with the barrel housing/plunger set. The staple wasn't into the hole, but it was making enough friction to wind the spring into the housing.

This time, to take the base off of the housing I used my tweezers to prevent the spring to come off too, mostly the same way I've seen others doing with the real winders.

estrapada3.thumb.jpg.ec9cb96dfb8e5beb72391c48806e6d2d.jpg

Tranferring the spring to the watch barrel was a bit of try and pray. I recon that the barrel bowl is a bit too big for the barrel. I used the 11.8 barrel bowl, but in the printed thing I measure about 12.2 mm. I guess the 3D printers (or the printer used by the printing service I used) are not accurate enough, so I should have ordered a wider set of housings.

Nevertheless, the second attempt was successful (in the first the spring didn't get into the barrel).

estrapada4.jpg.216c8e3b0a8d3f5c392c78d7c253f448.jpg

I have to practice some more, and order more barrel housings, plungers and barrel bowls sizes, but if I get the hang of it it's going to be a great tool.

I like the metal arbour, I think I'll print a base with a 2mm hole instead of the original arbour. Maybe I should look for a thinner metal rod to allow for more room between it and the spring, so the staple can get into the spring hole.

Edited by aac58
  • Like 1
Link to comment
Share on other sites

Hi @aac58, nice work with the metal rod fix!

On 6/10/2021 at 1:56 AM, aac58 said:

However, when trying to take the winderbase off, the spring wanted to come with it instead of staying in the housing, so it didn't work this forst time.

When you finish unwinding, you should unwind just a little bit more and the staple will just unlatch itself and just come off. I think I forgot to mention this part in the video.

 

On 6/10/2021 at 1:56 AM, aac58 said:

When I took the base off, I saw that the rod had broken. I probably did something wrong. Any advise you can share is very welcome.

Most likely the winding stress has destroyed it, looks like your print service provider's plastic material may not be too good, unfortunately. Personally I did not have this issue even on bigger springs, I only broke it when I dropped them 😅 Maybe other members here can comment on using stronger materials.

 

On 6/10/2021 at 1:56 AM, aac58 said:

Forgot to mention that when I received the pieces, the rod on the base couldn't get into the hole of the plunger.

Thanks for mentioning this. I also seem to have this problem with the plunger (sometimes). I will add a little bit more fit clearance to solve this problem, in the next release.

 

6 hours ago, aac58 said:

The staple wasn't into the hole, but it was making enough friction to wind the spring into the housing.

Actually for smaller springs, this seems to work for me as well. However, I think it is best to get the staple into the hole. What I do is just use an Arkansas stone to file the staple until it gets small enough to enter the hole (if your staple is too big).

 

6 hours ago, aac58 said:

I like the metal arbour, I think I'll print a base with a 2mm hole instead of the original arbour. Maybe I should look for a thinner metal rod to allow for more room between it and the spring, so the staple can get into the spring hole.

I like this idea very much. Please share with us the metal arbour material you used. It looks like a nail? I was thinking maybe use springbars 😂

I can design a separate winder base with an arbor hole with hook holes that are side by side (close to each other) based on your suggestions.

Link to comment
Share on other sites

3 hours ago, Zero said:

Please share with us the metal arbour material you used

I bought it in a hobby shop. The tag says "2 mm x 1 m steel rod". BTW it was very hard to cut.

Estrapada0.thumb.jpg.1191fd2d197862fe1a82f0d241290e77.jpg

I have tested your tool some more times and I'm under the impression that the last turns of the spring (those with the smaller diameter) are affected by the compression. I say that because when I get the spring out of the barrel the last few turns are now closer to each other than before. The spring I'm using is in bad condition, I'll use a better (but still bad) one in the next tests. I'll try to post some photos tomorrow to show you what I mean.

Edited by aac58
Link to comment
Share on other sites

17 hours ago, aac58 said:

I have tested your tool some more times and I'm under the impression that the last turns of the spring (those with the smaller diameter) are affected by the compression. I say that because when I get the spring out of the barrel the last few turns are now closer to each other than before. The spring I'm using is in bad condition, I'll use a better (but still bad) one in the next tests. I'll try to post some photos tomorrow to show you what I mean.

This is expected as the arbor (2mm) is smaller than the actual arbor (~3mm), so the innermost portion of the spring compresses a little bit more. It becomes more obvious because we keep winding it again and again for practice. It seems to return to its original shape when you put it back into the movement barrel with its original arbor and wind it up normally. I also practice on bad springs so its hard to tell.

But I would like to solve this issue properly by providing winder bases ranging from 1mm to 3mm holes for a metal arbor, so we can all make it look like yours.

The question is, where do we find this metal arbor? Does the hobby shop sell with different diameters, eg. 2.5mm? Hmm... maybe try to use springbars.

Link to comment
Share on other sites

15 hours ago, Zero said:

The question is, where do we find this metal arbor? Does the hobby shop sell with different diameters, eg. 2.5mm? Hmm... maybe try to use springbars.

They have 3mm rods too but no 2.5, but I think you can use anything with the appropriate diameter as nails.

But, with a bigger diameter, perhaps the spring will be tighter to the arbour and it will be harder to extract from the base and keep it into the housing?

Edited by aac58
Link to comment
Share on other sites

This is quite and interesting project, and welcome in that I don't have a winder and can't see spending tons at this (beginning) point in my hobby horology. I DO have a Prusa i3 MK2S and a Creality 3 Pro. I'll have a look at GitHub for the STL files, and print from PLA and PETG, and let you know how it works out. I would also be interested in the design with the metal arbor, if we can source some good rods (or springbars?). I'm working on practice movements at present, the Seagull 6498 and 2824....what size would I print for those?

Imp

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.




  • Recently Browsing

    • No registered users viewing this page.
  • Topics

  • Posts

    • somewhere in the universe although as I'm answering this I remembered which discussion group. Somebody had a similar question loss basically expanded it to the watch would totally disintegrate in my think like 30 years. I'd have to go back and find the original discussion if I can the person seem to think that they hairspring would disintegrate like you describe and just metal on metal wearing out the entire watch. Except of course we all work on watches over 30 years and they obviously do not disintegrate. Plus somebody came along and explained why hairsprings do not have the same issues as mainsprings. Although hairsprings do end up with watchmakers insisting on bending and playing with them and torturing them etc. and that obviously is not good for the metal at all. one of the problems that you're having here it is what is the purpose of the test that started this discussion? Let me go and snip out the original image as you can see from images above this watch is horrible. Or is it? What is the purpose of the images up above in other words what exactly would use this test for? The real purpose this test is show the effect of amplitude on timekeeping. Or specifically you're looking for mechanical issues that are causing fluctuations in amplitude which unfortunately shows up  with timekeeping. or basically everything affects timekeeping but amplitude is affected by the mechanical characteristics of the watch from the mainspring to the balance wheel and unfortunately as it's a mechanical watch your always going to have power fluctuations. so how do we rule out unacceptable fluctuations versus the natural characteristic of the watch? Usually if you can find a repeating pattern you can narrow it down to the offending components for instance I'm attaching a PDF. on the second page of the PDF it talks about 21st-century equipment versus paper tape timing machines. Then they give an example of timing problems solely caused by a faulty component. although off you have a user's manual for a paper tape machine it does explain that you can find faulty components by looking at the variations on your paper tape seeing how often they repeat and do the same thing without the fancy software. Even though it was claimed that you couldn't do that in other words you couldn't find a pattern? One of the problems that comes up with modern LCD-based timing machines versus software is limited screen size. In other words it makes it very hard to look for patterns you'll see variations in numbers but it's hard to tell what's going on which is why the display above is really nice to see if there is a problem. for instance here's a paper printout from a witschi timing machine it does look distorted because I changed the speed at which the image would move across the screen. In other words I was trying to figure out a way to extend the screen to being much longer as I was looking for a pattern as you can see there doesn't appear to be a pattern at all so basically we end up with a watch that I cannot time at all they cannot really figure out what the problem is and I actually cannot find a pattern even begin the figure out where the problem might be. Plus I agreed to service this watch for free as I was going to use it for the purposes of a lecture. In other words it's a nice railroad grade pocket watch and I wanted to show before how horrible amplitude is and how wonderful it looks after serviced and after servicing it looked exactly the same still horrible. Then I used software for a clock timing machine and came up with this interesting image one a minor problems we have with time plots and  how they look is that they all do things a little bit differently. So this was occurring approximately every five minutes. Then we need another chart then I replaced something in the watch and we now get this one of the things that I was always bothered with was if I had put the hands on after servicing would the watch have Time? Because the pattern was repeating the watch would average that out may  it would have Time. Oh and what did I change somebody had swapped the mainspring barrel for something different were getting a binding between the mainspring in the center wheel pinion. so the problem you're having is what exactly is the problem? the purpose of the test image is to look for mechanical problems causing amplitude problems. Because it's a mechanical watch your always going to have variations so are the variations in this watch abnormal or normal for this watch? Once we eliminate the mechanical issues beyond it's a mechanical watch then you can work on timing issues. for timing issues I recommend going back to the normal display that were used to and make sure you have your averaging times set correctly. In other words while the graphical display is basically real time years of the numbers are averaged over time. Anywhere from 20 to 40 seconds depending upon whose specs you're looking at. So basically they will average out the problems were seeing on this time plot.   one of my amusements with students that go to the same school is that you have different instructors. So this gives you different experiences like what exactly is tight anyway as I don't remember any thing like this? Then did you know that Rolex at least in Geneva as I visited their service center replace all the screws in the watch every time they service the watch. then why did they do that because they use power screwdrivers and tighten those things down as tight as you can get them which has a problem. How many times can you tighten screw down really tight before the hints break off. So they replace the screws every single time. So personally I don't think they have to be so tight that you're in danger snapping head off that's too tight in my opinion. but the screws definitely shouldn't be falling out either which I've occasionally see and where people just don't tighten their screws tight enough.   Horologica Times -- May 2004 From the Workshop witschi time plot.pdf
    • I guess it is a possibility, but the train wheel bridge was pressed down all the way so I'm not so sure. Then again, the screws weren't tightened at all (imagine being screwed down all the way but using a piece of Rodico instead of a screwdriver). Anyway, the movement is now fully stripped so we'll have to see once I've assembled it again. BTW I found the post where @nickelsilver wrote about tightening screws: As he writes: "In school, if your screws aren't tight, like you think they might snap, you get your movement tossed in the sawdust box!" I'm really curious to know why it is so important to tighten the screws that hard. I usually stop when it feels like there is no chance the screw can start to unscrew itself. Also, screwing down that hard requires perfectly dressed and perfectly sized screwdrivers to avoid slippage and/or damaging the screw slot.
    • This place has them, cheaper than I saw on ebay and they appear to be a legitimate supplier: https://maddisonsofdurham.co.uk/watch-parts/capacitors/seiko-batteries-capacitors/seiko-capacitor-kinetic-30235mz-tc920s-5m42-5m22-5m23-battery-3023-5mz-3023-5my/?gad_source=1&gclid=CjwKCAjw7-SvBhB6EiwAwYdCASmviGb9G2ZGW3CtcUZBkNgglcgfPKoqnpOzrzruiPtm69f6DX7UGhoCKl4QAvD_BwE There is also a note at the bottom of the page about them being a newer type, with a slightly different part number.  
    • Are we ignoring  that another watch on tg showed similar rate fluctuation. Did I miss any conclusion made on this sigificant  point ?    Springs have high fatigue threshold,   meaning angle of  bend/distortion  has to be very sharp to cause material fatigue .  An evidence to this point is the bend we form at end of terminal curve or the bend at the collet,  are we causing fatigue there? definately not.       
    • Yes, I do have the Seitz pivot gauge which is worth a small fortune these days. Got mine for about £200 which I thought to be crazy expensive at the time, but I've now seen asking prices over twice that. And no, I do not have the scaled pin gauge, but it would be convenient. Let me know if you find them! Anyway, the Seitz pivot gauge is in my opinion not really necessary if you have the JKA Feintaster. It can measure even the very small pivots w/o making any dents. One will have to be a lot more careful when using the Bergeon micrometres.
×
×
  • Create New...