Erratic behaviour crown down

[aac58]

Hi again!

I'm working on a Seiko 7005-7012 that I bought non-running. On disassembling I inmediatly discovered that the bottom escape wheel pivot was broken, so I replaced it. Then when testing the motion side on the timegrapher the machine could not read anything clear, it was all a nightmare of dots on the screen.

As the escape wheel was faulty, I suspected of the pallet fork, so I also change that, and yes, now everithing is ok on the timegrapher. Everything except...

This is what I have:

So I'm very happy in most positions, but the data in position crown down/9 up is awful! And the lines are not as clean. I have the watch on the timegrapher while I'm writing this and now it says the amplitude is 145 and the beat error is changing, sometimes 4.4, or 2.7, or ...

Any clue about what can be the most probable cause? Maybe I'll try another balance? I thinh everything is really clean, except maybe the pallet fork, as I took a spare whitout cleaning it and I do not know it it was cleaned by me before.

As always, thank you for your help.

[Nucejoe]

If you have access to high magnification, I would check both pivot ends, rounded, burred...

  Given enough end shakes, staff rides on pivot ends. Sometimes I  get by with just a reduce of end shake. 

Best

[jdrichard]

I had the same problem on a pocket watch. Turned out to be a loose impulse jewel. Re-shellaced it and all was good.


Sent from my iPhone using Tapatalk Pro

[aac58]

Thank you both for your answers.

As my skills are so limited I cannot reshellac an impulse jewel. I have now another balance on the ultrasonic right now with its incabloc disassembled too, so I will try this one. Anyway, I've tested this new balance on a 7009 I have for tests and this one also have some problems in some positions, we'll see how it behaves after the clean and oiling of the cap jewel.

[jdrichard]

Thank you both for your answers.
As my skills are so limited I cannot reshellac an impulse jewel. I have now another balance on the ultrasonic right now with its incabloc disassembled too, so I will try this one. Anyway, I've tested this new balance on a 7009 I have for tests and this one also have some problems in some positions, we'll see how it behaves after the clean and oiling of the cap jewel.

Just grab the impulse jewel carefully with tweezers and see if it moves at all.


Sent from my iPhone using Tapatalk Pro

[JohnR725]

Does your timing machine show a graphical display you can show us?

Besides swapping escapement components what other servicing have you done to this watch? Plus what sort of lubrication are you using especially on the escapement components.

[aac58]

I cleaned the hole thing, and oiled it using 9010 and hp1300. On the exit stone of the first pallet I put a drop of 9010, I've seen others using it although I now it's not the best there. I didn't oiled the second pallet.

I have no photos of the timegrapher screen, and as I do not have a permanent desk I have uninstaled it. I'll take some photos next time I check the watch. But the data I gave is about what you can see there, the lines are very clean and nice in all positions except crown down, where they are quite messy. Lift angle was established at 53.0. I do not think there is any other relevant information you can see on the screen, is there?

[AndyHull]

Possibly *something*, most likely a jewel,  is worn ovoid (egg shaped), or partially gummed up,  so things are fine until a pivot drops in to the worn part of the ovoid hole, or sticks on the gunk,  whereupon the relative geometry of the gears change, as pivot ends are no longer parallel, or running true and free.

Clean and check the balance jewels, the fork jewels and the escape wheel jewels, under high magnification, also check the end shake of these components. Clean and re-check their pivots too. 

Also check the hairspring is completely flat, as if not, it may be able to foul on something, but only in certain positions.

[JohnR725]

22 hours ago, aac58 said:

I cleaned the hole thing, and oiled it using 9010 and hp1300. On the exit stone of the first pallet I put a drop of 9010, I've seen others using it although I now it's not the best there. I didn't oiled the second pallet.

I have no photos of the timegrapher screen, and as I do not have a permanent desk I have uninstaled it. I'll take some photos next time I check the watch. But the data I gave is about what you can see there, the lines are very clean and nice in all positions except crown down, where they are quite messy. Lift angle was established at 53.0. I do not think there is any other relevant information you can see on the screen, is there?

Using 9010 on the escapement is fine it's what they used to use.  Did you put one drop on the exit stone only? 

The numbers up above in a spreadsheet did they come from a timing app or a timing machine?

The only reason I want a photograph of the timing machine is sometimes perfect lines to one person may not be perfect to someone else.

Your amplitudes appear to be low even for Seiko especially if the watch has been fully wound up?

 

[aac58]

So, I let it stop from fully wind to fully unwind, mostly 9 up, and now I've wound it fully and have it on the TG. Here is what I'm seeing:

It is near what I had previously. Once thing to mention is that it takes quite some time to stabilize after moving from one position to another, maybe too much.

I agree that the amplitude could be better, but honestly this is about the best result I've got so far. Sometimes I see up to 225º but for short periods of time. Figures change on screen more than I'd like.

I could start testing things like changing the balance, but I'd love to have a reason before I do it. For example, if I suspect the hair spring can be touching something only in that position, that would be a good reason to check another balance.

What do you guys think is the best rasonable way to proceed?

EDIT: BTW I don't beleive in miracles, but after a few minutes on the TG 6 UP lines are getting cleaner (and amplitude lower) Not sure if it is reliable.

EDIT AGAIN: No it's not a miracle, it was clean only for a few minutes.

Edited October 19, 2019 by aac58

[rodabod]

A couple of simple tests:

can you grab the balance rim and lift and drop the wheel to see if the end shake appears excessive? Even if you don’t know what this should look like, start doing it now as it’s a common, trivial examination. Be careful when doing this with non-antishock balance jewel settings. 

Secondly, hold the watch in the questionable position, and under lots of light and magnification, observe the balance wheel and hairspring: does it rub anywhere?

[aac58]

When I was looking the balance under the USB scope I noticed that the balance was swinging a bit, not as much as in the 6119 I previously posted, but it wasn't moving as it should. I think I have lots of 700x balances in poor conditions!

As I said before, I was cleaning and oiling another balance for this movement. It has given me some problems to clean and oil properly, but finally I've instaled it in this movement, and the results are much better so far. I'll let it run out of the case for a while before installing the calendar works, dial and hands and put it into its case.