Low amplitude on ETA 2638 - mainspring?

[patchwerk]

I am servicing a 2638 from the 60s. I have the done the full clean and lubricate (epilame, train, escapement, braking grease, balance caps). I cleaned and pithed the train twice. The movement is in good condition and the train wheels spin very freely, and the balance is spins freely as well. Yet I am only get 180 degrees amplitude. This is without the calendar works engaging as well. The endshake seems fine and I even made an aluminum foil shim to add just a little but it didnt help. I feel like I have eliminated many options but still cant figure it out. My question is could it be the alloy mainspring? It still has it shape, but is it possible for alloy mainspring to go weak without looking like a set spiraled steel mainspring? I know it cant hurt to change it but could this be the culprit? 

Thanks.

[jdrichard]

40 minutes ago, patchwerk said:

I am servicing a 2638 from the 60s. I have the done the full clean and lubricate (epilame, train, escapement, braking grease, balance caps). I cleaned and pithed the train twice. The movement is in good condition and the train wheels spin very freely, and the balance is spins freely as well. Yet I am only get 180 degrees amplitude. This is without the calendar works engaging as well. The endshake seems fine and I even made an aluminum foil shim to add just a little but it didnt help. I feel like I have eliminated many options but still cant figure it out. My question is could it be the alloy mainspring? It still has it shape, but is it possible for alloy mainspring to go weak without looking like a set spiraled steel mainspring? I know it cant hurt to change it but could this be the culprit? 

Thanks.

Is the hairspring really well cleaned? I had the same issue with a Rolex Oyster and had to clean the hairspring twice with lighter fluid to get the smallest amount of oil off the hairspring. After that the amplitude doubled:)

[patchwerk]

I have cleaned it L&R cleaning and rinse but could give it another wash.

[nickelsilver]

Amplitude that low is almost certainly not the mainspring. Well, supposing it has the correct mainspring- if a thinner one was substituted in the past that can zap quite a few degrees. It should have a 1.20 x 0.125 x400 spring. If yours measures up close to that (especially the thickness) you have other issues.

 

Even with machine cleaning it's worthwhile to peg out jewels, particularly the pallet fork jewels. I've skipped that step and scratched my head why a spotless watch was missing 40 degrees of amplitude, peg the holes, and voila.

 

If the barrel is worn, or its holes in the bridge and plate are worn, it can tilt and rub, robbing quite a bit of power. I can't remember on this one but a number of ETA pieces from that period have a sort of hourglass shaped bushing for the lower 3rd wheel pivot that can wear quite a bit. This will suck out a lot of power too. Sometimes under high magnification you find funny white deposits in the fork slot as well as muck on the roller jewel, that can eat 40-50 degrees easily. Use freshly cut pegwood to clean there.

[patchwerk]

I cleaned the balance assembly again and made sure it was dry with no coils touching. Same result.

I did do a round of pegging during precleaning and the pallet fork moves freely (falls up and down when I tilt the watch). Top barrel bridge bushing is not worn and the barrel doesn't wobble. The bottom one is jeweled. I will do another inspection of the train and jewels.

I am still interested in whether alloy mainspring can become weak. I feel I have seen several watchguy.co.uk jobs where he changes the alloy mainspring for whatever reason.

[Nucejoe]

1 hour ago, patchwerk said:

My question is could it be the alloy mainspring? It still has it shape, but is it possible for alloy mainspring to go weak without looking like a set spiraled steel mainspring? 

Yes, mainspring normally weakens with use whatever the alloy, the old black carbon steel ones are worst, specially if this is the same one that left the company.

Do your own test, turn the barrel to suppliment power to get a feel of how things are. You can also full wind the barrel through ratchet screw and onbserve the amplitude.

 

[nickelsilver]

Modern alloy springs really don't weaken with age. They might develop wear marks which could increase friction as it uncoils, but its modulus of elasticity doesn't change. What can happen with steel springs is they can experience plastic deformation, "take a set", where the coil shape when out of the barrel is clearly smaller than when new. This will reduce its power somewhat, but actually isn't the huge problem many think.

 

The main reason many watchmakers change the spring each service is to avoid broken springs, and for many calibers they are readily available and it's a simple thing to change and eliminate that potential problem (for many calibers they might change the whole barrel). But, I do quite a bit of work for a big, old brand, on their vintage pieces, and 9/10 we don't change the spring This is their decision, and these jobs are often quoted at 10, 20, 30 hours- so the cost of a mainspring isn't a concern when dealing with those kinds of repair prices. Often the watch  clearly has its original spring, and I'm talking about 40-70 year old watches.. I think in 6 years we've had one watch come back for a broken spring. I also do work for a shop that sees a lot of vintage chronographs and other really nice pieces, with him also we change the spring when it's necessary, but not as a matter of course except newer stuff like ETA current calibers.

[jdrichard]

1 hour ago, patchwerk said:

I have cleaned it L&R cleaning and rinse but could give it another wash.

and if you have a microscope, have a look.

[patchwerk]

1 minute ago, jdrichard said:

and if you have a microscope, have a look.

Was literally doing that when you posted that.
Hairspring looks perfect. Jewel holes seem good at inspection but I will take the train apart and inspect each one next.

[jdrichard]

1 hour ago, patchwerk said:

Was literally doing that when you posted that.
Hairspring looks perfect. Jewel holes seem good at inspection but I will take the train apart and inspect each one next.

and check the snap on the palate fork

[JohnR725]

4 hours ago, patchwerk said:

180 degrees amplitude

How do you know you're getting 180°? I'm assuming you're using some sort of timing app or timing machine? I don't suppose we can have an image?

 

4 hours ago, patchwerk said:

My question is could it be the alloy mainspring? It still has it shape, but is it possible for alloy mainspring to go weak without looking like a set spiraled steel mainspring? I know it cant hurt to change it but could this be the culprit? 

Yes with a mainspring anything is possible. But as I'm typically working on vintage watches I run into set mainsprings typically blued steel that are set. But even a set mainspring wound up tight still has enough energy to run the watch. It just won't run the watch for any length of time. So it's extremely unlikely your mainspring is the issue. Then also unlikely to find a set modern mainspring.

 

4 hours ago, patchwerk said:

I have cleaned it L&R cleaning and rinse but could give it another wash.

How do curiosity how long do you leave it in the cleaning machine and which cleaning fluid are you using?

3 hours ago, patchwerk said:

I feel I have seen several watchguy.co.uk jobs where he changes the alloy mainspring for whatever reason.

Depending upon the watchmaker determines whether the spring gets changed. The previous watchmakers that we had wouldn't even touch a watch without a new mainspring. Apparently they were taught in school that you cannot guarantee a repair unless the mainspring is new. But if the mainspring is white it looks right and it runs the watch it's fine. On the other hand usually factories that have infinite supplies of mainsprings in their service centers always change the mainspring just because and usually dull change the entire barrel assembly because they have a infinite supply of those.

But no matter what your mainspring still should have enough power to power the watch at more the hundred and 80° it would have to be so soft that you can literally bend it like a piece of lead then it definitely would have a problem.

If you take the balance wheel out and take the pallet fork out and wind the watch really fast in other words you can find faster than the watch can let the energy escape with no fork of course it doesn't seem to accumulate power does it continue to spin the train for a while? Then of course when it reaches the end do you see the back's been? Or for that matter when you take the balance wheel out do not let the power off and take the four Does the train spin down if it does your mainspring is fine but I think your mainspring is fine anyway

1 hour ago, jdrichard said:

and check the snap on the palate fork

Oh and of course before letting the power down by taking a pallet fork out this is also an excellent test but you should have done this already when you are lubricating the escapement. In other words when you wound up the watch the first time you go to lubricate the escapement you typically push on the pallet fork and verify that it has a nice snap as you're putting a lubrication on.

 

[patchwerk]

I am using watch-o-scope and verified with a slomo video on my phone

I am using L&R extra fine cleaning and rinse solution at 15-30 then 3 minutes. I pegged the jewels again and polished the pallet fork pivots to make it more more freely. The fork does snap but the little sound it makes isnt as loud as I am used to in other watches. I also polished the barrel holes to allow freer movement of the barrel arbor

When the train is free spun with pressure on the barrel it spins down very slowly but does not recoil. In fact tapping the movement when it stops makes it spin a little more in the same direction. I suppose this tells me there is friction in the train somewhere. I have inspected the train pivots and jewels under high magnification but I guess one could be bent so slightly I wouldn't see it. I think the next step is to put the wheels one by one in the movement and see how they freely individually.

[JohnR725]

That really is a super perfect looking waveform on the timing software? I assume it's dial down? Why don't you turn the watch to crown down and see what it looks like. Then maybe even dial up that might be interesting to see also. It's always nice to do multiposition timing when you're having an issue because it sometimes helps to show things

The reason I asked about the cleaning fluid is often times people don't realize it only takes about four or five minutes in the cleaner and anything more is usually bad. The rinse is harmless and be in there much longer.

1 hour ago, patchwerk said:

I think the next step is to put the wheels one by one in the movement and see how they freely individually

Another option would be especially after I went searching online for a picture of the watch. It's an automatic isn't it? I'm guessing the plates especially for the gear train might be then perhaps easy to bend. Take the balance in pallet fork out and the mainspring barrel but leave the rest of the train in. Then push on it like the center wheel how does spin how does it feel seem to spend nice doesn't have resistance? Then I usually find a clean oiler works nicely for this go in and lift up each the wheels or move them around see what the end shake feels like.

 

 

[rodabod]

I was about to reply, but it would be an echo of what nickelsilver said above.

What is important to appreciate is that wheel train does not mesh under load in exactly the same when as unloaded, when there is side shake. So, basically, what I'm saying is that the side-shake really kicks in when there is force running through the train, and this can lead to poor meshing if there is indeed wear. 

It's easier to test on clocks, but when you get a suspect wheel, you can run it standalone in the train with one other adjacent wheel as a single meshed pair. Sometimes by holding one stationary and rocking the other back and forth you can see excessive side-shake. Or if you allow one to slowly run against your finger to create a load, you can rotate the other to feel if it turns smoothly. A bad depth will sometimes cause a lumpy feeling.*

*Your fault may not be due to side-shake, but thought it was worth mentioning some of the methods of diagnosing it.

[patchwerk]

So I changed out the mainspring and amplitude is 280 now. The mainspring that was in there was .06mm thick...so half as thick as it should have been. I cant imagine why anyone would put one like that in there.

Edited December 16, 2020 by patchwerk

[nickelsilver]

Haha

37 minutes ago, patchwerk said:

So I changed out the mainspring and amplitude is 280 now. The mainspring that was in there was .06mm thick...so half as thick as it should have been. I cant imagine why anyone would put one like that in there.

Haha, oh, wow. 0.06 is like the smallest watch movements. Doubling that and getting correct amplitude, good job. Good job.

[patchwerk]

I should have checked its thickness first thing and just assumed it was right because it looked like a automatic mainspring in good shape.

[nickelsilver]

2 minutes ago, patchwerk said:

I should have checked its thickness first thing and just assumed it was right because it looked like a automatic mainspring in good shape.

Hahaha, the number of times I've been ****** because something should have been. Welcome to the club.

[JohnR725]

Yes the constant amusing problem of whoever worked on the watch before us why the heck did they do that?