If you work on vintage watches long enough, it won’t take long before you run into a mainspring that’s either completely the wrong size or the mainspring is too weak and doesn’t provide enough power to the escapement resulting in low amplitude.

Whatever the case is if you are going to be in this game you’re going to need to be able to figure out for yourself what size mainspring that barrel should have in it. And that is what we are going to be learning about today.

Troubleshooting Watch Amplitude Problems

Never Trust the Originality of a Watch Movement

Look ever since Covid started there has been a boom in watch repair interest, and you’re going to buy a watch in which the movement has been worked on by someone who knows less about watch repair than you.

Now,  that’s not a knock against you but I want you to think about that for a second, OK?  Home watchmakers sell their movements that they can’t get to run on eBay.  So you can never trust that things are original or correct in the movements that you buy.

  If you are working on a Pocket watch, it literally may have been worked on 5 or 6 times before it got to your bench.

Problems associated with bad mainsprings

Now there are tons of scenarios and reasons that an amateur or even a working watchmaker may have put an incorrect mainspring in a barrel, and there are all kinds of problems that can be caused by it.

If it’s too short, it may have a shortened power reserve which can affect your daily rate

With springs that are too strong, the amplitude may be too high and you may get some knocking which can cause damage

If too weak the amplitude will be low.

Too wide and it will not fit in the barrel.

Too narrow may the spring to twist a bit inside the barrel and cause wear.

Mainspring Theory

I’m going to bet that the majority of you watching this video assume that mainsprings sizes are absolute. Well if you are working on modern movements you probably should stick with what the manufacturer is calling for.

But, it’s important for you to know, that you do have some leeway when sizing a mainspring in a vintage watch and do not have to follow exact factory sizing in most cases.

One example might be a 7-jewel watch suffering from low amplitude because there is some wear in the un-jeweled pivot holes. Now, you could certainly fix that problem by re-bushing the holes, but what if that’s not an option because of your current skill level or you have the skill but the value of the movement doesn’t warrant that much work to be put into it?

Well, another thing you can do is just increase the strength of the mainspring,  increasing the available power coming from the barrel to compensate for the wear in the movement that’s eating up your power before it gets to the escapement.

How many times have you heard me say that 70 % of mainsprings power is lost by the time it gets to the escapement?

Increasing the power of the mainspring means there will be more available at the pallet fork, plain and simple

MainSpring Sizing Theory

If you were to look into mainspring sizing theory, DeCarle uses what’s known as the Rule of Thirds, which simply says that if you took a barrel and split it in half,  1/3 arbor (half the actual arbor), 1/3 unwound mainspring and 1/3 open space. As you can see in the diagram, the unwound spring is slightly less than the 1/3rd.

In George Daniel’s book “Watchmaking” he says that the spring should take up exactly half the area between the arbor and the barrel wall.

And in the Swiss “Theory of Horology”, says that The mainspring should ALWAYs occupy 50 % of the open space in the barrel no matter if it’s at rest or fully wound.

In the  Generale Ressorts catalog, there are around 150 different springs for 11mm barrels alone  which range in thickness from 0.095mm to 0.160mm and lengths from 300mm to 460mm 

So why all the differences of opinion? What this should be telling you is that there is no exact science, and you have some leeway when fitting a mainspring as long as you stay within certain parameters.

The Mainspring Fitting Formula

Ok so now I am going to share how I learned to size mainsprings and we are going to compare this formula to a couple of known mainspring sizes.

The formula is simple.

To calculate the Strength of the mainspring, you measure the inside dimension of the barrel and divide it by 87.

The length of the mainspring is easy. You use the Inside dimension and multiply it by 30 for American movements or 35 for Swiss movements.

To get the width, you measure inside the barrel, from the bottom to the top edge, then subtract the thickness of the lid plus another 0.1 mm for clearance.

Now, let’s test this formula against a couple of known mainspring sizes.

Comparing the formula to an ETA 2824

We will use the barrel in an ETA 2824

THE I.D of the barrel is 11mm, so 11mm divided by 87 equals 0.126 so that’s our strength or thickness

Then we multiply the Barrel ID by 35 to get our length which is 385 mm

To get the width or height of the mainspring, measure from the bottom of the barrel to the edge 1.64 mm.  Then subtract 0.1 mm for clearance and subtract the thickness of the lid, which is 0.31

  In this case, we came up with 1.23   x .126 x 385.    

ETA specs are 1.23 x 0.125 x 400

Comparing the Formula to an American Pocketwatch

Let’s do another. This time an American Pocket Watch the Elgin Model 315

I can look up the part number in my original factory parts book.

And cross reference that part number in my Swigert catalog to get the 2.05 x .196 x 508

When I measure the barrel I get an inside dimension of 15.50 mm and if I divide that by 87 I get a strength of .179 which is off by less than 200 of a mm.

If I multiply the Barrel ID by 30 I get a length of 547 mm which is almost 40mm longer than the factory spec but it would still easily fit into this big barrel that has an inside circumference of about 48 ½ mm, so there’s plenty of room for another coil

When I measure the barrel depth I get 2.59 mm minus .1 mm for clearance and minus the thickness of the lid .5 mm leaves me 1.9 mm which is 15 hundreds of a mm smaller in width than the factory 2.05 so that would work.

What’s the takeaway

So what’s the takeaway here? Well if you are having a problem with low amplitude, you have to start troubleshooting at the power source. You have to verify that the mainspring is correct or at least that it’s not too weak. Only then, once that’s verified,  can you work your way through the power train to the escapement. And if you start checking your mainspring size, it will make you a better watchmaker