HELP!!! Screwed up my 2892-A2

[adiorio110]

 

To All

Ok I purchase a clone 2892 took it apart and serviced it with good results. So I felt brave enough to work on my Cartier Pasha with a 2892-A2.. Apparently I I screwed up  Timegrapher all over the place ...I used it on several movements and always had better results than when I started..Of course they were always the cheap Chinese movements. 

 

Here is what I am up against. I took it apart and examined but did not notice anything wrong From what I read it is most likely the escape wheel or the 4th wheel. But I am not sure if I am analyzing  this right.I enclosed pictures of what I am seeing ... Sometimes the movements can get even more extreme than the pictures I uploaded  

 

 

 

 

[JohnR725]

Definitely an interesting waveform! Can you change from the current to the B graph? In the B graph mode will display 192 seconds which will make it easier to look for patterns if there are any. Then I'm a big fan of timing in different positions to see if the problem occurs in all the other positions?

[rodabod]

Obvious and simple things to check when the rate varies like this (plus it’s a newer movement so seems less likely to have an eccentric wheel):

- magnetism

- sticky hairspring coil, or coil fouling part of the movement. 

[adiorio110]

4 hours ago, JohnR725 said:

Definitely an interesting waveform! Can you change from the current to the B graph? In the B graph mode will display 192 seconds which will make it easier to look for patterns if there are any. Then I'm a big fan of timing in different positions to see if the problem occurs in all the other positions?

Hi John

 

Thanks for reply

This three shots are in B mode 

Face down 

Face up

Crown Right

Same type results

I am tearing it apart to examine a little closer I obviously missed something.

[adiorio110]

2 hours ago, rodabod said:

Obvious and simple things to check when the rate varies like this (plus it’s a newer movement so seems less likely to have an eccentric wheel):

- magnetism

- sticky hairspring coil, or coil fouling part of the movement. 

Hi Rodabod

Not magnetized and coil spring looks fine .. 

[Nucejoe]

Hi, did you remove the staff jewels, is there any chance of lower and upper jewels having got mixed up. 

If there is the slightest wobbling in  the wheel , it may be touching something between 140 and 280 travel.

Regards 

 

[adiorio110]

Nucejoe

Only removed one at the time no way they got mixed up. I just finished tearing it down again and all jewels look good pivots look good Almost finished putting it back together. Lets see what happens 

[Pip]

Nucejoe
Only removed one at the time no way they got mixed up. I just finished tearing it down again and all jewels look good pivots look good Almost finished putting it back together. Lets see what happens 

Well my tech knowledge isn’t good enough to help but I’m watching with interest and hope you manage to get it sorted.


Sent from my iPad using Tapatalk

[adiorio110]

52 minutes ago, Pip said:

Well my tech knowledge isn’t good enough to help but I’m watching with interest and hope you manage to get it sorted.


Sent from my iPad using Tapatalk

Were in the same boat ...Still not working right can not figure what I am doing wrong ...The only variable is I used 9415 on pallets instead on 9010..could that b the problem? 

 

I am leaning towards on of the drive train wheels is screwed up 

[adiorio110]

Its weird because I can not see anything that looks off ..No wobble in any of gears ..balance wheels looks good to me ...I suppose it could be so slight it is not noticeable without proper measuring devices 

[JohnR725]

What we need access to is a different timing machine? One thing that can cause timing fluctuations is change in amplitude because of issues with the gear train or basically fluctuations of power through the gear train. There timing machines that will run time plots for both the rate an amplitude which makes it really easy to see if you have a gear problem. This is where was hoping with the longer display we can possibly see a repeating pattern and figure out if there's a particular wheel that's the problem. This is one of the times where paper tape was nice because you to run on nice long piece of paper and look for things.

Then the balance jewel assembly? When you disassembled are you 100% sure that you put the Cap stones back correctly? One side is curved one side is flat it's really easy to sometimes flip them over. If you look at them like under a florescent light you can look at the reflection on the top and see if the reflection looks flat or whether it's curve the way it's supposed to be because the curved side goes out the flat side goes down.

When you have it on the timing machine leave it in one position and watch the amplitude see if it's going up and down and how far does it fluctuate up and down? Then if it does fluctuate see if the fluctuation occurs at a regular interval.

 

[adiorio110]

3 minutes ago, JohnR725 said:

What we need access to is a different timing machine? One thing that can cause timing fluctuations is change in amplitude because of issues with the gear train or basically fluctuations of power through the gear train. There timing machines that will run time plots for both the rate an amplitude which makes it really easy to see if you have a gear problem. This is where was hoping with the longer display we can possibly see a repeating pattern and figure out if there's a particular wheel that's the problem. This is one of the times where paper tape was nice because you to run on nice long piece of paper and look for things.

Then the balance jewel assembly? When you disassembled are you 100% sure that you put the Cap stones back correctly? One side is curved one side is flat it's really easy to sometimes flip them over. If you look at them like under a florescent light you can look at the reflection on the top and see if the reflection looks flat or whether it's curve the way it's supposed to be because the curved side goes out the flat side goes down.

When you have it on the timing machine leave it in one position and watch the amplitude see if it's going up and down and how far does it fluctuate up and down? Then if it does fluctuate see if the fluctuation occurs at a regular interval.

 

John

Yes I I rebuilt this watch three times now and same results I am sure I installed cap jewels correctly...

This watch is close to twenty years old but I rarely  wore it ...I did all ready order a new mainspring with barrel and arbor ...The price of getting mainspring alone was negligible.

  According to this article it sounds like it would be the fourth wheel ...Which is the seconds wheel right? I get my parts confused sometimes lol..

 

https://adjustingvintagewatches.com/rate-stability-wavy-traces/

 

[adiorio110]

1 hour ago, JohnR725 said:

What we need access to is a different timing machine? One thing that can cause timing fluctuations is change in amplitude because of issues with the gear train or basically fluctuations of power through the gear train. There timing machines that will run time plots for both the rate an amplitude which makes it really easy to see if you have a gear problem. This is where was hoping with the longer display we can possibly see a repeating pattern and figure out if there's a particular wheel that's the problem. This is one of the times where paper tape was nice because you to run on nice long piece of paper and look for things.

Then the balance jewel assembly? When you disassembled are you 100% sure that you put the Cap stones back correctly? One side is curved one side is flat it's really easy to sometimes flip them over. If you look at them like under a florescent light you can look at the reflection on the top and see if the reflection looks flat or whether it's curve the way it's supposed to be because the curved side goes out the flat side goes down.

When you have it on the timing machine leave it in one position and watch the amplitude see if it's going up and down and how far does it fluctuate up and down? Then if it does fluctuate see if the fluctuation occurs at a regular interval.

 

John

Amp stayed about the same but very low. Look how wacky this chart is 

 

I upload a picture of my 6498  just to show I can dial it in if it lets me.

Edited February 24, 2019 by adiorio110

[JohnR725]

If this was a pocket watch with a little secondhand then that wheel is the fourth wheel. Typically it would go around once per minute indicating the seconds but on a modern watch conceivably it could go at a different rate.

Gear trains are interesting in that they have power fluctuations. Modern watches the power fluctuations are less because they understand things and have nifty test equipment to make sure they're less versus older watches that existed before timing machines and nifty computer simulations. So gear train fluctuations typically don't show up on the timing machine which is usually averaging over time anyway which is the average out all the short-term fluctuations. The only time you might see something is if you're having a problem and when the amplitude is low problems can show up as everything gets magnified.

At work I get spoiled the witschi machine I have has a full automatic microphone I time everything in six positions. Occasionally on American pocket watches I will get suspicious and look at the numbers on the last page then I run the whole thing again. On the watches I'm suspicious of the numeric page the numbers are different. A lot of times even between dial-up and dial down which should be the same there's enough time between timing that the numbers are different. But a nice characteristic of a watch is it tends to average out things like this and it's not noticed usually.

Then there is a way to see gear train fluctuations if your timing machine has a time plot feature. The last time I did a firmware update on the machine at work I now have a nifty time plot feature. It's interesting in that it auto scales to whatever the input is so the graphical displays might look similar but you have to look at the numeric scale on the left-hand side to see what they mean.

So to understand the time plot feature I'm attaching images. You'll notice the Rolex appears to have a problem but timekeeping wise it still looks fine. The Hamilton hasn't been serviced in I have no idea when this is the way I got it when I purchased it and it also looks outstanding for timekeeping even with the fluctuations. Both the Rolex and Hamilton were time when fully wound up. Then the seven jewel laco timed at the end of 24 hours which is why the amplitude is low looks really bad.

[adiorio110]

2 hours ago, JohnR725 said:

What we need access to is a different timing machine? One thing that can cause timing fluctuations is change in amplitude because of issues with the gear train or basically fluctuations of power through the gear train. There timing machines that will run time plots for both the rate an amplitude which makes it really easy to see if you have a gear problem. This is where was hoping with the longer display we can possibly see a repeating pattern and figure out if there's a particular wheel that's the problem. This is one of the times where paper tape was nice because you to run on nice long piece of paper and look for things.

Then the balance jewel assembly? When you disassembled are you 100% sure that you put the Cap stones back correctly? One side is curved one side is flat it's really easy to sometimes flip them over. If you look at them like under a florescent light you can look at the reflection on the top and see if the reflection looks flat or whether it's curve the way it's supposed to be because the curved side goes out the flat side goes down.

When you have it on the timing machine leave it in one position and watch the amplitude see if it's going up and down and how far does it fluctuate up and down? Then if it does fluctuate see if the fluctuation occurs at a regular interval.

 

My  Chinese wieshi is not quite a Witschi LOL ... 

 

I do not have access to a better unit...Is there any one thing in particular you would concentrate on ?  ..I already ordered a mainspring and barrel . If I have to replace one part at the time that is what I will do. Could always use the extra parts for another project.

What pisses me off is that I worked on disassembling and assembling a several different pocket watch movement and watch movements plus several  Chinese knockoffs movements   including a clone 2892....They all went off pretty well ...This movement was from my Cartier Pasha I bought new about 20 years ago it keep pretty good time I just figured why not freshen it up ...and instead I screw it up . I 'll get it working again no matter what and chalk it up to a lesson learned

 

 

 

[Nucejoe]

Like rodabod said simple things, like contaminated  HS or .......

So lets find the cause to fix it.

If it run 32 hrs on full manual wind, I see no reason to replace the original high quality MS with some after market which even the supplier dosn,t know who made.

I learned a golden rule from a wacthmaker, when it gets you tired, put the movement in clean container and come back when you have rested .   

Regards 

 

[adiorio110]

1 hour ago, Nucejoe said:

Like rodabod said simple things, like contaminated  HS or .......

So lets find the cause to fix it.

If it run 32 hrs on full manual wind, I see no reason to replace the original high quality MS with some after market which even the supplier dosn,t know who made.

I learned a golden rule from a wacthmaker, when it gets you tired, put the movement in clean container and come back when you have rested .   

Regards 

 

Nucejoe

Thats good advice ...I have already put it aside twice...tried rebuilding it from scratch slowly going over every aspect...I am at a loss for any solutions ..my only recourse is to replace parts in drive train one by one till I come across the culprit ... no one really seems to sure exactly where the fault lies from the info I can supply with my limited experience and testing equipment. I had already ordered the mainspring prior to knowing about the problem as a backup in case I messed up greasing the barrel (which I did not attempt)  ..worst case I'll know where to look first I ever see a similar pattern.

I'll fully wind watch tonight and see how long it takes to wind down but I am pretty sure that is not the problem as it worked fine before the tear down. 

 

 

[JohnR725]

2 hours ago, adiorio110 said:

but I am pretty sure that is not the problem as it worked fine before the tear down. 

When you say the watch worked fine before the tear down Did you put it on the timing machine? Then you purchase the watch 20 years ago when was the last time it was serviced?

I was rereading everything did you replace the mainspring or barrel complete?

[adiorio110]

8 hours ago, JohnR725 said:

When you say the watch worked fine before the tear down Did you put it on the timing machine? Then you purchase the watch 20 years ago when was the last time it was serviced?

I was rereading everything did you replace the mainspring or barrel complete?

John

Yes I put it on time machine ... I lost the hand written measurements I took don't remember exactly what readings were but they where consistent and not running all over the place Line was running relatively straight. 

I did not replace mainspring or barrel ..Did not even service it. I had ordered a new one and still waiting for it to be delivered.  Just in case I screwed up servicing mainspring .. I just purchased mainspring winders so I did not have the ability to practice mainspring service ..so I left well enough alone.

[adiorio110]

9 hours ago, JohnR725 said:

When you say the watch worked fine before the tear down Did you put it on the timing machine? Then you purchase the watch 20 years ago when was the last time it was serviced?

I was rereading everything did you replace the mainspring or barrel complete?

John

Yes I put it on time machine ... I lost the hand written measurements I took don't remember exactly what readings were but they where consistent and not running all over the place Line was running relatively straight. 

I did not replace mainspring or barrel ..Did not even service it. I had ordered a new one and still waiting for it to be delivered.  Just in case I screwed up servicing mainspring .. I just purchased mainspring winders so I did not have the ability to practice mainspring service ..so I left well enough alone.

[adiorio110]

BTW it was hardly ever wore and never serviced

[Nucejoe]

Hope your next reading be satisfactory. Enough skilled folks here to gurantee excellent results.

More info may help. So if you will, check side and end shake on all arbors specially staff.

I suggest posting pix of the piece at different stages of strip down, someone may see what you have not noticed, video of runing parts, balance HS, fork pallets so on.

Regards 

[adiorio110]

5 hours ago, Nucejoe said:

Hope your next reading be satisfactory. Enough skilled folks here to gurantee excellent results.

More info may help. So if you will, check side and end shake on all arbors specially staff.

I suggest posting pix of the piece at different stages of strip down, someone may see what you have not noticed, video of runing parts, balance HS, fork pallets so on.

Regards 

Will do ..Waiting for delivery of mainspring and barrel before I make another attempt. So at least we can rule that out... 

[rodabod]

Unless you are also seeing fluctuations in amplitude which correspond with the fluctuations in rate, then I would not initially suspect that it is an issue with power delivery (mainspring and barrel). 

If you are noticing near-instantaneous changes in rate then I would suspect it is an issue with the oscillator. 

[matabog]

Take out the balance and escapement and see how the wheels are freely turning.

Then put the pallets and see how they snap from one side to the other. Se how they interract with the escape wheel.

Check the pallets not to be loosened or anything...

Check the pallets back-end - the one that interracts with the balance impulse pin - it must look nice and polished.

Then check the impulse pin not to bee loosesened.