Is it possible to modify a balance assembly?

[LazyTimegrapher]

As I am learning more about the inner mechanics of watches, I have started thinking about various ways of making customized watches by starting with an assembled watch and modifying various parts of it.

It seems fairly straightforward at least in principle to make custom dials. Making custom watch hands also seems conceptually simple, but perhaps challenging to fabricate parts with sufficiently fine tolerances. 

Making watch cases and bracelets seems a bit more challenging from a practical perspective, but conceptually seems fairly simple.

My main question is whether it’s feasible or possible to make significant modifications to mechanical watch movements. In particular I really like the ETA 2824 and the ETA 2892, but I would be curious to experiment with different types balance assemblies. 

I was curious if anyone has tried to replace the regulator style assembly with a free sprung  balance in one of these movements and if you had any luck doing this.

 

One possible path for doing this would be to take the balance assembly from a Powermatic 80 or other Swatch movement based off of the 2824 and try to put it into a regular 2824. I’m not sure if that would work or not. Maybe the gear train needs to be matched to the balance assembly or maybe there’s some other factor I’m not considering.

 

If no one has any experience trying this, I may just buy some cheap watches and mess around with this sort of thing and see how it goes.

 

Anyway, any thoughts or ideas would be appreciated. Thanks!

[JohnR725]

2 hours ago, LazyTimegrapher said:

Powermatic 80

I'm curious as to where you're going to buy replacement components for this watch from? As it's a Swatch group proprietary movement I don't think there's any replacement parts anywhere unless you're working a Swatch group service center where they have an endless supply of spare parts.

2 hours ago, LazyTimegrapher said:

based off of the 2824

The problem with things based off of is what exactly has been changed and because parts are no longer available parts cross references to figure out exactly what is been changed become nonexistent.

It depends upon how far you want to go into this. A lot of times students that go to the modern school of watch repair where they spend or at least used to spend a lot of time making things. I think they dropped the making part because are having a problem. The problem was that when they spent all his time teaching the students how to make things then they didn't want to go work in the service centers instead they would launch their own watch company making and/or modifying things. So very common watch to start with is the 6497 or 6498 and they would modify the plates. So you'd make use the wheels modify the plates end up with a really pretty looking pocket watch.

Then the problem with modifying the balance wheel is you would need a hairspring and that would be a challenge to get at least today unless you want to buy a few thousand of them than that's not a problem.

 

 

[LazyTimegrapher]

Thanks for the reply. That’s very interesting information. I don’t work in watch repair. I’ve just slowly become increasingly interested in the details of how watches work and started wondering how far one can go into tinkering with them. 

So I’m not in the watch business. I’m just someone who has a modest watch collection and am at the point where instead of going the route of saving up for increasingly expensive watches, I’d rather start tinkering with cheaper watches on my own and make my own unique probably fairly quirky watches. I’m a mechanical engineer for my day job so I do like thinking about how things work, how to take them apart, and put them back together (always the hard part).

 

I just figured that it’s possible to buy a 2824 for $150-300 or a Chinese copy for $50 and started wondering how hard it’d be to make different modified movements with different types of balance assemblies. Rolex and omega seem to use free spring balances so I wondered if it’d be possible to somehow take a balance/hairspring from one watch and put it in a different one and make it work.

 

Maybe it’s an impractical plan but I may give it a try since it seems fun even if it doesn’t work.

It could be cool to make a fancy engraved/ skeletonized 6897, but the engineer in me wants to also mess with the mechanical part of the gear train and balance assembly too.

 

when you say watch schools used to teach people how to make things, do you mean that it’s possible to fabricate your own gears and balance wheels and things like that out of brass just using jewelers saws and drills? If so that sounds like it’d be really fun to attempt.

[Nucejoe]

I am self thought watch destroyer. 

Though any watch balance wheel can be matched with hairsprings of any strength, and  I have built a few balance completes, for 

Movements of unknown calib.

With balance wheels of unknown calibs

Hairsprings of unknown strength.

 The finished product can be regulated to exact beat needed  to show right time with the train at hand, as long as in static position, thats on bench, but runs inaccurate in dynamic ,on wrist, in pocket.

Always a pain to get an acceptable avg rate of all positions in dynamic state.

 Several  prametres get in the picture in balance rim  design and alloys are     the main     in hairspring  technology ,whilst one can't  care less what material or alloy the pendulum of wall clocks is made of.

Rgds

 

 

 

 

 

 

Edited August 15, 2024 by Nucejoe
Spell

[LazyTimegrapher]

Thanks very much for sharing your experience. That’s very helpful information. I think it would be fun to follow your path of using trail and error to play around with different movement designs.

 

Since you mentioned clocks, i actually had the idea that maybe I should start by making my own clocks by hand. I’d imagine one could be made out of soft metal and wood and it’d be easier to see how it all fits together and how things work since clocks are larger.

 

why is it that the main technology is alloyed materials for springs and balances? What’s the drawbacks of metals like brass or steel for balance assemblies?

 

thanks again for the information!

[nickelsilver]

5 hours ago, LazyTimegrapher said:

 

when you say watch schools used to teach people how to make things, do you mean that it’s possible to fabricate your own gears and balance wheels and things like that out of brass just using jewelers saws and drills? If so that sounds like it’d be really fun to attempt.

Most schools won't teach how to cut gears, or if they do, just the basics. You do learn how to saw and file to a high level, and how to make turned parts on a lathe, like winding stems and balance staffs. To cut gears you need some special machinery; lathes can be adapted to do it, but small watchmaker lathes are far from ideal. You also need the profile cutters to form the teeth.

1 hour ago, LazyTimegrapher said:

 

why is it that the main technology is alloyed materials for springs and balances? What’s the drawbacks of metals like brass or steel for balance assemblies?

 

 

Once the hairspring was invented, two big issues became the focus. Isochronism- the stability of rate over high and low amplitudes, and the effects of temperature.

 

Up to the 1930 or so, the best option for compensating the effects of temperature (gaining rate in colder temp, losing rate in hotter) was a bimetallic balance of roughly 2/3 brass and 1/3 steel, in various configurations but usually with the rim cut at the cross arm on both sides. As the balance heats, the sections of rim contract, as it cools, they expand, thus correcting for the change in elasticity of the hairspring (which was made from steel).

 

Charles Guillaume invented Invar in the early 1900s, which is an alloy that doesn't change length with temperature change, then Elinvar, which doesn't change elasticity with temperature change. It was a pretty big deal, netting him the Nobel Prize. Now a watch could be made to rival the best bimetallic balances using a monometallic uncut balance coupled with an Elinvar hairspring. Over the years further developments were made, in particular making the hairspring alloy virtually non-magnetic (another big deal).

 

You can make a balance from a number of materials and get good results using a "compensated" hairspring, but most high quality balances are made from beryllium copper.

 

If you want to make a balance that uses a regulator into a freesprung, you will need to add some sort of adjustable masses to the balance rim. This will cause the balance to be too heavy for its hairspring- but you can also remove material from the rim to compensate for the added mass. It's pretty high-level watchmaking work, not something to start with.

[Nucejoe]

2 hours ago, LazyTimegrapher said:

why is it that the main technology is alloyed materials for springs and balances?

 Balance rim  expands & contracts with heat, which in turn affects angular inertia of the rim,  likewise do hairsprings, so for the industry, the idea of self compensation in design and alloy  was worth investing in ,  similarly springness of hairsprings being a function of heat, affects the oscillation frequency. Despite improvements achieved, we still  see quality watches  being adjusted for temperature.

Some less magnetizable metals can also be alloyed with non metals. 

Clocks are our good moderator's expertise  "oldhippy"  whom we call OH.

Rgds

I see Nickelsilver beat me by four minutes and tons of knowledge. 

Edited August 15, 2024 by Nucejoe
Adding less knowledge

[VWatchie]

Well written and very interesting post @nickelsilver!

[lucgyver]

you could do like i did.

buy a pile of watches or just movements that has plenty of similar ones and some that look like there is some life in and see how far you can get.

i have learned a lot about interchangebility already.

watches are very simple, you have your power (the spring) a way to release that power in a controled way (the balance asembly) and a way to convert that controled release of power in usefull work (the gears).

 

[LazyTimegrapher]

7 hours ago, nickelsilver said:

Most schools won't teach how to cut gears, or if they do, just the basics. You do learn how to saw and file to a high level, and how to make turned parts on a lathe, like winding stems and balance staffs. To cut gears you need some special machinery; lathes can be adapted to do it, but small watchmaker lathes are far from ideal. You also need the profile cutters to form the teeth.

Once the hairspring was invented, two big issues became the focus. Isochronism- the stability of rate over high and low amplitudes, and the effects of temperature.

 

Up to the 1930 or so, the best option for compensating the effects of temperature (gaining rate in colder temp, losing rate in hotter) was a bimetallic balance of roughly 2/3 brass and 1/3 steel, in various configurations but usually with the rim cut at the cross arm on both sides. As the balance heats, the sections of rim contract, as it cools, they expand, thus correcting for the change in elasticity of the hairspring (which was made from steel).

 

Charles Guillaume invented Invar in the early 1900s, which is an alloy that doesn't change length with temperature change, then Elinvar, which doesn't change elasticity with temperature change. It was a pretty big deal, netting him the Nobel Prize. Now a watch could be made to rival the best bimetallic balances using a monometallic uncut balance coupled with an Elinvar hairspring. Over the years further developments were made, in particular making the hairspring alloy virtually non-magnetic (another big deal).

 

You can make a balance from a number of materials and get good results using a "compensated" hairspring, but most high quality balances are made from beryllium copper.

 

If you want to make a balance that uses a regulator into a freesprung, you will need to add some sort of adjustable masses to the balance rim. This will cause the balance to be too heavy for its hairspring- but you can also remove material from the rim to compensate for the added mass. It's pretty high-level watchmaking work, not something to start with.

This is all great information. Thanks very much!

 

It’s interesting to know what it’s possible to make gears yourself. It could be interesting to try making gears to understand what’s involved in that. My wife has done some jewelry making courses which are very manual in the methods they teach. Forming metal into rings or other jewelry and then making prongs to hold stones can be done entirely by hand but it’s very tedious work with fine saws and files and delicate tools. It seems like making gears would be similar to that in some ways.

 

I like the idea of directly modifying a regulated balance into a free sprung balance by directly modifying the parts themselves. Given all the advanced materials science that goes into making these things, that sounds like the best path forward since making these materials or buying them directly doesn’t seem feasible.

 

Also, I’m more interested in the mechanical engineering aspects of the movement rather than the material science of it. The fact that the true innovations that enabled highly precise watches was materials science rather than mechanical engineering is a good lesson for me to really internalize. While the first few steps of innovation for making a watch were mechanical innovations ( pendulum clocks and then rotational inertia wheels with long hairsprings to miniaturize clocks and convert them into watches) the next step for future improvements came from a different discipline  altogether. At the highest levels of engineering, it seems like multidisciplinary teams tend to be what’s required to create the best technologies. This is something I need to constantly remind myself of.

 

One idea I had was that I could probably use the material from the spokes of the balance wheel to make threaded rods and then take other material from a balance wheel to make screws to put on the rods. And then bore out holes in another balance wheel to put the rods into. That way I could use material from spare balance wheels to create a modified balance wheel with adjustable inertia and have it be of the correct material. As you say if need to remove material to compensate for the weight I’m adding with the rods and screws. 
 

I suppose I’d also have to figure out how to setup the hairspring attachment point once I remove the regulator arm.

 

I don’t know if or when I will actually attempt this as a project, but just understanding what would be involved and think about how it could be done it a lot of fun in and of itself.

 

i really appreciate everyone’s input and information on this topic. It’s a very interesting topic.

7 hours ago, lucgyver said:

you could do like i did.

buy a pile of watches or just movements that has plenty of similar ones and some that look like there is some life in and see how far you can get.

i have learned a lot about interchangebility already.

watches are very simple, you have your power (the spring) a way to release that power in a controled way (the balance asembly) and a way to convert that controled release of power in usefull work (the gears).

 

I think this is the mostly likely path for me to start on this. To just accumulate enough spare parts to be willing to break things while experimenting with movements.

 

i also think starting with clocks would probably make things easier since they’re larger, don’t move around, and just seem easier to work with generally.

[rjenkinsgb]

I'd think the free balance mod could be quite easy on a movement that already has adjusting screws on the balance wheel?

If you remove the regulator arm, the hairspring will have a slightly longer effective length, all the way to the anchor stud, so the watch will run slow.

That can be compensated for by filing or grinding small amounts from an opposite pair of weights (or removing shims), then the fine adjustment made using the existing screws.

eg. from a quick look on ebay - a pocket watch movement and a wristwatch movement. The pocket watch has clearly visible adjusting screws; I think the other has those also, though the image is not so detailed. There are many others, if you look as mechanical watch movements on ebay, including some ETA ones.

 

[LazyTimegrapher]

2 hours ago, rjenkinsgb said:

I'd think the free balance mod could be quite easy on a movement that already has adjusting screws on the balance wheel?

If you remove the regulator arm, the hairspring will have a slightly longer effective length, all the way to the anchor stud, so the watch will run slow.

That can be compensated for by filing or grinding small amounts from an opposite pair of weights (or removing shims), then the fine adjustment made using the existing screws.

eg. from a quick look on ebay - a pocket watch movement and a wristwatch movement. The pocket watch has clearly visible adjusting screws; I think the other has those also, though the image is not so detailed. There are many others, if you look as mechanical watch movements on ebay, including some ETA ones.

 

This is also a good idea. If the balance already has screws that would make it easier to adjust. It sounds like there are several paths forward if I decide to attempt these sorts of movement modifications.

[JohnR725]

5 hours ago, LazyTimegrapher said:

i also think starting with clocks would probably make things easier since they’re larger, don’t move around, and just seem easier to work with generally.

Clocks definitely open up all sorts of possibilities and solve a lot of problems. Like you don't need a balance wheel anymore you can have a pendulum.

Many years ago in the national Association and watch and clock collectors magazine I saw an article of somebody who made a clock. Now making a clock for them is not that unusual Each year they have a craft contest and usually somebody each year makes a clock or two or three. But that particular magazine somebody filed all the gears. Personally I think that's insane but the person made his entire clock and it was truly a thing of beauty. Although he did have a bad choice for the plates use some sort of Super tough bronze that cut his pivots off. So add the disassemble o'clock and put bushings in that were more friendly

If you're still interested in watches why don't you get the book watchmaking by George Daniel's he tells how to make a watch and all of the components so you can see how to do that.

Then here is a YouTube channel that will help you get started with your watch and or clockmaking

https://www.youtube.com/@Clickspring

[LazyTimegrapher]

2 minutes ago, JohnR725 said:

Clocks definitely open up all sorts of possibilities and solve a lot of problems. Like you don't need a balance wheel anymore you can have a pendulum.

Many years ago in the national Association and watch and clock collectors magazine I saw an article of somebody who made a clock. Now making a clock for them is not that unusual Each year they have a craft contest and usually somebody each year makes a clock or two or three. But that particular magazine somebody filed all the gears. Personally I think that's insane but the person made his entire clock and it was truly a thing of beauty. Although he did have a bad choice for the plates use some sort of Super tough bronze that cut his pivots off. So add the disassemble o'clock and put bushings in that were more friendly

If you're still interested in watches why don't you get the book watchmaking by George Daniel's he tells how to make a watch and all of the components so you can see how to do that.

Then here is a YouTube channel that will help you get started with your watch and or clockmaking

https://www.youtube.com/@Clickspring

I would agree that making every piece yourself is a very impractical way to obtain a working clock or watch. I think my objective would more be to get a deeper intuitive understanding of how these mechanisms work and I think going through the process of making each aspect of it would really help me understand them better.

 

i do plan on buying some of the classic books on watchmaking to learn from. The Daniel’s one sounds like a good book to buy.

 

If I end up attempting this, I’ll post my results on this forum for others to see and possibly learn from.

 

one other question I had was aside from the regulator arm in the ETA 2824 and the free sprung in balance in a Rolex 3235 are there any other very different methods for making a balance assembly? I’ve seen combinations of variable interia balance wheels and regulator arms on some watches. I suppose in principle either you change the springs length or the balances inertia so there aren’t that many options, but I wondered if there were any other types of balances that have been used in watches that I don’t know about.

[Neverenoughwatches]

17 hours ago, Nucejoe said:

I am self thought watch destroyer. 

Newly recruited X Men character  " The Watch Destroyer " I look forward to your latest Marvel movie Joe.  

13 hours ago, lucgyver said:

watches are very simple

thats all i can come up with

[Nucejoe]

6 hours ago, LazyTimegrapher said:

It’s interesting to know what it’s possible to make gears yourself. It could be interesting to try making gears to understand what’s involved in that. 

 Manufacturers are equiped with batch gears cutters, I guess its  all programmed on CNC machines.

Considering your interests we ought to put you in contact with @nevenbekriev, he can make the tools and parts he needs, just recently I asked him if he can make escapewheels, I be darn, he posted a vid showing how he does make them two days later with some rather basic machines he got,  he did saw and filing manually.

You can find vids of such industrial production machines on youtube or machine makers adverts.

 

 

 

 

 

 

[LazyTimegrapher]

2 minutes ago, Nucejoe said:

 Manufacturers are equiped with batch gears cutters, I guess its  all programmed on CNC machines.

Considering your interests we ought to put you in contact with @nevenbekriev, he can make the tools and parts he needs, just recently I asked him if he can make escapewheels, I be darn, he posted a vid showing how he does make them two days later with some rather basic machines he got,  he did saw and filing manually.

You can find vids of such industrial production machines on youtube or machine makers adverts.

 

 

 

 

 

 

That sounds super interesting and like it’s be a fun thing to try. Depending on how expensive the tools and machines are it may be outside my price range for now but it is something I’d be interested to try doing in the future. Does he have a YouTube channel? What’s it called or where can I find his videos?

[Nucejoe]

1 hour ago, Neverenoughwatches said:

Newly recruited X Men character  " The Watch Destroyer " I look forward to your latest Marvel movie Joe.  

Haha , well I done my share of ruining  watches,

called it part of my learning process, not to feel  shame.

Rgds

2 minutes ago, LazyTimegrapher said:

That sounds super interesting and like it’s be a fun thing to try. Depending on how expensive the tools and machines are it may be outside my price range for now but it is something I’d be interested to try doing in the future. Does he have a YouTube channel? What’s it called or where can I find his videos?

Not sure but if memory serves me he does post on youtube . Just notify him that he is got a msg on the forum.

Born a craftsman. 

 

 

[tomh207]

1 hour ago, LazyTimegrapher said:

That sounds super interesting and like it’s be a fun thing to try. Depending on how expensive the tools and machines are it may be outside my price range for now but it is something I’d be interested to try doing in the future. Does he have a YouTube channel? What’s it called or where can I find his videos?

Here you go

 

Tom

[LazyTimegrapher]

16 hours ago, tomh207 said:

Here you go

 

Tom

That’s amazing. It’d be really interesting to be able to work with that machine. Thanks for the link.