Mainspring dimension influence?

[bsoderling]

Hi all,

 

I have understood that mainsprings are generally defined by their dimensions (height, thickness, length).

 

It seems (and is logical) that thickness is a vital parameter for the ”power” going into the wheel train. I often see recommendations to ex. Step up thickness if a movement is showing poor amplitude and all other influences have been exhausted.

 

But how influential is ex. Height ?

 

The thing is that I have collected small batches of springs on auction sites when there’s been opportunity. Once in a while I have a match with Ranfft specs but every so often the thickness is off by one or two tenths of a mm but it would fit in the barrel and thickness is correct.

 

I presume a higher spring will provide more power, all other parameters equal?

 

But what’s the sensitivity here?

 

Can I presume it’s considerably less than the thickness?

 

Trial and error is one way here but your feedback and experience is appreciated.

 

 

 

 

Sent from my iPad using Tapatalk

[jdm]

6 minutes ago, bsoderling said:

I often see recommendations to ex. Step up thickness if a movement is showing poor amplitude and all other influences have been exhausted.

Where exactly you have seen that recommendation?

[bsoderling]

”Often” was probably an exagerration...and I can’t provide any exactness either.. :-)

It’s more something I picked up as one of the tools in the investigation toolbox.

I recall one instance where I did the opposite (reduced thickness) in an old rebanking movement that I could only find a ”modern”
m/s for (that was supposedly ”stronger” at the same thickness).

Are you hinting that I’m out in the woods here... ? :-)




Sent from my iPad using Tapatalk

[Nucejoe]

If by power you mean torque, given all else the same, thicker MS supplies stronger torque, however, power is the rate at which energy is spent, which has to do with the escapement. 

In second part of your question you bring in height as a parameter, then torque is actually a function of the cross section of the spring, that is   (thickness)×(height),  again given all else remains the same.

Spring production follows industries standards, which designers are to go by and is a common way of commnuication for watch repair as well, going by what is produced by standards.

[bsoderling]

I’m sure you are correct when it comes to torque vs. Power.

I was probably thinking more along the lines of ”force” which I presume converts over to torque when the force is applied over an arm of certain length? I think Nm (Newtonmeter) was the unit we used in physics class?

So the height of the m/s spring has the same ”weight” as the thickness in these equations?

That’s interesting and sort of indicates one should be just as careful with height as thickness.

I do understand that the pro’s go by the standards and manufacturer recommendations but for me as the happy amateur just fiddling with oldtimers I buy for pocket money and try to get running and understanding what the various influences are, it’s a bit different. And why I post these questions.

When it comes to the long term aspects, I was under the impression that m/s one can get today are made of different alloys with other mech properties than the ones I usually find (broken) in my old watches. How is this taken into account (if my assumption is correct)?







Sent from my iPad using Tapatalk

[Nucejoe]

Same question has been on my mind, unless aftermarket spring is specially made for the caliber at hand, I have just been going with the closest one I can find.

[rodabod]

I’ve covered this before here, so please search, but the height of the spring has a linear relationship with the force it produces. However force is related to the cube of the thickness. That’s why it is more critical in terms of force (torque developed). In terms of height, it needs to fit the barrel snugly. 

[nickelsilver]

Yes, like rodabod says, height has a small influence and thickness a comparatively massive one. In a pinch I will put say a 1.20 height spring where a 1.30 is called for, same thickness. Amplitude doesn't suffer. But put a 0.11 thick spring where 0.10 is called for and you might run into rebanking. You can go down 0.10mm in height but can't go up- there might be 0.05mm height clearance designed in, the higher spring would bind.

[JohnR725]

one of the problems with the discussion is this is a spring. the spring part is really important for this discussion. If your mainspring has lost its spring characteristics it's basically worthless and it doesn't matter what any of the rest the dimensions are it has to have a spring characteristic. Note I refused to look up the definition of a spring for you. Then some mainsprings have really interesting back curves and these interesting back curves help to equalize out the force. the basic meaning of this is the end up with a really nice even power throughout the running of the watch. Blue Springs are typically older older Springs can set especially the blued steel Springs.

 

[bsoderling]

Hi guys and thanks for all feedback.

I guess we can at least make one safe conclusion; this is complicated stuff!

An idea that popped up in my head that I wanted to try with you...

Wouldn’t it be possible to design and build a fairly simple device to actually measure the torque in a barrel mounted m/s ?

If the barrel is fixed mounted and a bar with some defined length is mounted in the hole normally used for the ratchet wheel, one could maybe measure the counterforce at the end of the bar with some relatively sensitive weighing scale with a reasonable winding of the spring.

But maybe the torque varies to much with winding?

Johns comment about back curves in newer springs seem to indicate this?




Sent from my iPad using Tapatalk

[nickelsilver]

They have electronic devices to measure the power curve of the spring in the barrel, but back in the day they used manual machines. They are simple in concept but would be a lot of work to make one. The main component of the machine is a steel spring, somewhere around 1mm thick, about 35mm high and about 80cm-1m long. It is fixed at one end, and has calibration marks along its length to show the force needed to move it about 6cm. A holder for a pen can be fixed anywhere along the bar. The barrel is held by a split chuck, and the arbor in another. Through gearing, turning a crank, you can wind up the spring and then release the power; while doing that a bed moves along in front of the spring- this has paper on it. Also while winding a string wraps around a drum and is connected to the pen holder... so you make a graphic chart of both the winding power curve and unwinding curve.

 

I can't find a pic of one but I have used them in the past, really neat machines.

 

The graph looks like the pic- though this is just showing one of the two, there is a difference between the winding up graph and winding down.

Edited September 11, 2020 by nickelsilver

[rodabod]

I’m more concerned with the torque developed through a fusee and it’s interaction with old springs and their “set up” so can use a torque meter to measure this on clocks via the winding arbor (on the fusee). 
 

I’ve thought about this for watches, and what would possibly make it quite simple would be a screwdriver with built in torque gauge, such that you just turn the click wheel screw. 

Edited September 11, 2020 by rodabod

[Nucejoe]

As can be seen on the graph above, there is no mention of power, your axis are torque versus wind. The area under the curve gives the reserved energy.

Hook's law states that spring's dispalcement scales linearly with the force exerted. Put this in a polar coordinate, you get a curve.  

You don't exactly get the lihears scaling in a coiled mainspring.

rodabod's tool are on sale at sears and tool stores and is called torque wrench, but only for regular bolt and nuts, mainly auto mechanic use.

[clockboy]

It is an interesting topic for sure. However replacing a vintage watch mainspring with an exact sized replacement is not always possible. I have found providing you don't go a long way off the original spec and the movement has been serviced it runs just fine. However two issues to be observed. Going above the original height is risky as binding on the barrel lid can happen also going way below the original height means the arbour hook its not naturally in the centre of the spring. The length will alters its power-reserve. I have GR resources book which is really handy when faced with this dilemma. Unfortunately this is now out of print, but might be available on-line.

[Nucejoe]

Take another look at shape of the curve, what dose it resemble?   .......     Modern mainsprings. 

Mainspring old or modern is coiled inside the barrel "you polar coordinates" , take the spring out you have it in a cartesian coordinate, shape of the curve= modern mainspring.

 A good portion of the curve is a straight line, corresponding to a linear torque supply.

Mainsprings production is based on a standardize chart, you just use one closest to what do the job.

[bsoderling]

Lots of interesting info for the hobbyist here!

The thought of a ”mini torque wrench” is interesting. As far as I recall the (big) ones you use on your car etc. are designed to have you set the desired tightening torque on a scale and the wrench will slip as you tighten the bolt/nut and reach the desired torque.

And then I guess you would run into those movements with a left-threaded screw on the ratchet wheel, making it hard to wind via that screw...

At the end of the day it sounds like a bit of trial and error may still be where I end up. The feedback here has definitely provided some guidance on where the influences and limitations are.




Sent from my iPad using Tapatalk