Timegrapher show -220 watch runs +1 hour per day fast

[SpringMangler]

Bought an old Oris off eBay which has an AS 1863 movement. The movement in the watch was shot so picked up a very good condition replacement movement.

Prior to service it was running at -16 secs with a beat error of 0.4, so acceptable. 

Serviced the watch and after service it was still running slightly slow (-10), beat error around .4 but with better amplitude.

I put the dial and hands on and let the watch sit for around 15 hours. In that time the watch gained around 50 minutes. Back on the timegrapher but with the same result.

I demagnetised the watch again, centred the rate adjustment and the timegrapher shows -220 secs per day, beat error and amp the same as before.

The watch is still running at around 4 minutes per hour fast.

Confused. What does the timegrapher actually measure off? Could the balance be catching but if so should that not reflect in the timing?

[ColinC]

When you let the watch sit around - what position is it in?  Is the difference dial up vs dial down?

 

[nevenbekriev]

Put the watch on the timegrapher, adjust the rate as possible closer to 0 s/day, do not change the position and observe the second hand. Is it making one revoluton exactly per minute, acording to another trusted timer/watch?

Edited November 10, 2023 by nevenbekriev

[Nucejoe]

 Are you sure you have the right balance-cock assembly on your watch ?      Balance out of AS 1717  ( a 18000 beater ) fits in AS 1863.

 

 

[SpringMangler]

Thanks for the replies.

It's running the same both face up and face down on the timegrapher. I left it running face up but placed it face down about 90 minutes ago and it's already gained 4 minutes.

I'll try the adjustment to zero and time it with a stop watch.

The balance is the one that came with the movement, although that may not be the correct balance. Assumption is always dangerous! I have a balance from an AS 1717 so will give that a try.

EDIT - Tried the timing test with the existing balance and one revolution of the second hand is taking 56 seconds. Time to try the other balance.

 

Edited November 10, 2023 by SpringMangler

[Nucejoe]

24 minutes ago, SpringMangler said:

 I have a balance from an AS 1717 so will give that a try.

  AS 1717 is 18000 beater and AS 1863 beats 21600 bph , so if you have a 1717 in a 21600 movement , it will run slow.

Otherwise its a serious hairspring issue or loose canon pinion and the like.

Rgds

 

 

Edited November 10, 2023 by Nucejoe

[SpringMangler]

4 minutes ago, Nucejoe said:

  AS 1717 is 18000 beater and AS 1863 beats 21600 bph , so if you have a 1717 in a 21600 movement , it will run slow.

Otherwise its a serious hairspring issue or loose canon pinion and the like.

Rgds

 

 

Ah, sorry I misunderstood. I though you meant they were interchangeable. 

[nickelsilver]

What frequency is your timing machine saying you are running at?

 

I believe the balance are interchangeable, in the sense they will tick in either movement. But if you have a 21,600 balance in a watch that needs 18,000, you are getting an extra tick each second. 12 seconds extra per minute. You observed the minute hand making a turn in 56 seconds, so the math doesn't seem to work-except, often the higher beat watch will have 20 or so teeth in the escape wheel, and the other gearing has changed.

 

On that note, for a "family" of calibers, often the center distances didn't change even though there were gearing changes to allow raising the frequency. In some cases the gearing can get mixed up- so maybe you have a 4th wheel meant for an 18,000 beat watch but a 21,600 balance, or something similar; it can get a bit complicated.

Edited November 10, 2023 by nickelsilver

[SpringMangler]

 

It is showing as 21,600 so the balance appears correct. 

However, I think you may have hit the nail on the head with the gearing. I've swapped the 4th wheel out for the one off the old movement. Initial timing of the seconds hand appears to give a 60 sec revolution but I'll let it run for a few hours to see how it goes.

Many thanks for the help and saving what bit of hair I've got left. 

[Nucejoe]

14 minutes ago, SpringMangler said:

Many thanks for the help and saving what bit of hair I've got left. 

Haha, Do you remember the emoticon wrt forum use to have of the guy plucking out his own hair , I miss the guy. 

Balances are interchangable , balance completes are not. Gears of the train too are for different beats.

We should have a thread on Dr ranfft's site, tons of info is packed in it. 

Rgds

 

Edited November 10, 2023 by Nucejoe

[nevenbekriev]

It was clear from the beginning that the balance is not ‘wrong’ and it is something about the ratio of the train.

When just the balance is the ‘wrong’ one, then the movement will gain or loose more than 4 hours /day:

21600/18000=1.2

24x1.2=28.8       28.8-24=4.8h/day

24/1.2=20           24-20=4h/day

Now, that You had changed the 4th wheel, we have correct ratio to the second hand. Hopefully all will be OK, but still the ratio to the 5th wheel has to be checked.

Changing the wheels between one family of calibers with different BPH not only leads to wrong train ratio, but to incorrect gearing, as the wheels have different modules or diameters. That’s why, the wheels must go together as a set and not to be changed between sets with different BPH. But, as You see, sometimes the movements work without stoppings, even with wheels from different sets, despite the wrong gearing.

Edited November 10, 2023 by nevenbekriev

[JohnR725]

1 hour ago, nickelsilver said:

What frequency is your timing machine saying you are running at?

4 hours ago, SpringMangler said:

What does the timegrapher actually measure off

Mechanical watches can be quite interesting at times. The timing machine is listening to the sounds of the escapement controlled by the oscillation of the balance wheel. The timing machine does not care about the hands or anything else it only cares about picking up the sound of the ticking watch. So you can have a absolutely perfect timekeeping on the timing machine watch that can have other issues as you've now discovered.

26 minutes ago, SpringMangler said:

However, I think you may have hit the nail on the head with the gearing. I've swapped the 4th wheel out for the one off the old movement. Initial timing of the seconds hand appears to give a 60 sec revolution but I'll let it run for a few hours to see how it goes.

So balance wheel oscillates at a frequency. The gear train of the watch is to reduce the frequency down so that the hands revolve at the correct rate.

When I was a young student in school working on an Omega watch I had your exact same problem. We ordered a replacement wheel and the watch did not keep time.

This comes up for unsuspecting Rolex collectors the exact same thing.

Your ending up with the same problem here is a series of watches that were made in differing frequencies. Where the base caliber is 18,000 and later versions basically identical watches came in other frequencies. So the difference would be the balance wheel but oscillates at its correct frequency and just a couple of the gears in the gear train that a change the ratio. So eager to assemble the Rolex for instance will mix-and-match and then you have a serious problem.

So timing machine can be very happy with the rate of oscillation of the balance wheel. But you do have to have the right gear train ratio to go with your oscillation timing machine cannot measure that.

I suspect if you take out the offending wheel and compare it with the gear that should be in there and look at the number of teeth on the gear and/or pinion You will observe that they do not have the same number of teeth.

[mbwatch]

On 11/10/2023 at 9:14 AM, nickelsilver said:

In some cases the gearing can get mixed up- so maybe you have a 4th wheel meant for an 18,000 beat watch but a 21,600 balance, or something similar; it can get a bit complicated.

Glad I found this. I have an absolutely cursed Lorsa P72a (Ranfft lists as 21600, no mention of 18000) that has been a "learning machine" for me. Everything that could go wrong has gone wrong; I got a balance complete delivered from Italy, pallet fork horns were bent, host of other problem all solved until it ran perfectly on time at 280deg, 21600. Then last night I put the seconds hand on it and it makes a perfect 50 second revolution.

I have a donor I didn't use for anything and will start by checking the 4th wheel teeth count in hopes that one is 18000. It's very clear I put in a 21600 balance complete.

[Neverenoughwatches]

33 minutes ago, mbwatch said:

Glad I found this. I have an absolutely cursed Lorsa P72a (Ranfft lists as 21600, no mention of 18000) that has been a "learning machine" for me. Everything that could go wrong has gone wrong; I got a balance complete delivered from Italy, pallet fork horns were bent, host of other problem all solved until it ran perfectly on time at 280deg, 21600. Then last night I put the seconds hand on it and it makes a perfect 50 second revolution.

I have a donor I didn't use for anything and will start by checking the 4th wheel teeth count in hopes that one is 18000. It's very clear I put in a 21600 balance complete.

21600÷18000=1.2   60sec÷50sec =1.2   sound like the train is for an 18000 balance.

12 minutes ago, Neverenoughwatches said:

21600÷18000=1.2   60sec÷50sec =1.2   sound like the train is for an 18000 balance.

P72 runs at 18000

[mbwatch]

2 hours ago, Neverenoughwatches said:

P72 runs at 18000

I see that now in Ranfft's archives. The other difference illustrated there was that P72a has a mobile stud carrier while P72 is fixed. Mine had a mobile carrier to begin with so I assumed P72a (even though it says P72 under the balance). Same mobile stud carrier on the other parts movement I had on hand (a P75 with calendar). I did not notice the difference in beat there before.

So I bought the wrong balance, and while I was certain I remembered the ebay listing saying P62/62A, P72/72A I now look back and it definitely says just P62A/P72A. What a mess. I was so happy to see this watch tick after all the issues I overcame, and now I need to decide if the existing sunk cost is worth another $30 and a month wait for the right balance. Boo.

Additionally, there are variants for MT Antichoc (which is what I need) and Incabloc. The former seems harder to come by.

(Edited - correcting the listing as P62A/P72A)

Edited January 8 by mbwatch
correcting the listing as P62A/P72A

[Neverenoughwatches]

8 minutes ago, mbwatch said:

I see that now in Ranfft's archives. The other difference illustrated there was that P72a has a mobile stud carrier while P72 is fixed. Mine had a mobile carrier to begin with so I assumed P72a (even though it says P72 under the balance). Same mobile stud carrier on the other parts movement I had on hand (a P75 with calendar). I did not notice the difference in beat there before.

So I bought the wrong balance, and while I was certain I remembered the ebay listing saying P62/62A, P72/72A I now look back and it definitely says just P62/P72. What a mess. I was so happy to see this watch tick after all the issues I overcame, and now I need to decide if the existing sunk cost is worth another $30 and a month wait for the right balance. Boo.

I'm a bit confused by your last statement. If the listing now says p62/p72 that would be the 18k bph, but you have a 21.6k balance. 

16 minutes ago, mbwatch said:

I see that now in Ranfft's archives. The other difference illustrated there was that P72a has a mobile stud carrier while P72 is fixed. Mine had a mobile carrier to begin with so I assumed P72a (even though it says P72 under the balance). Same mobile stud carrier on the other parts movement I had on hand (a P75 with calendar). I did not notice the difference in beat there before.

So I bought the wrong balance, and while I was certain I remembered the ebay listing saying P62/62A, P72/72A I now look back and it definitely says just P62/P72. What a mess. I was so happy to see this watch tick after all the issues I overcame, and now I need to decide if the existing sunk cost is worth another $30 and a month wait for the right balance. Boo.

I think i would go with that the movement is p72 that runs at 18k. Your timegrapher has already indicated that the new balance is vibrated to 21.6k . Since the hands are moving at 1.2x , the train must be geared to run correctly at 18k . Can you varify it with a tooth count of the gears. 

[mbwatch]

22 minutes ago, Neverenoughwatches said:

I'm a bit confused by your last statement. If the listing now says p62/p72 that would be the 18k bph, but you have a 21.6k balance. 

Hastily typed, and missing the "A". The listing was P62A/P72A. Apologies for the confusion.

[Neverenoughwatches]

9 minutes ago, mbwatch said:

Hastily typed, and missing the "A". The listing was P62A/P72A. Apologies for the confusion.

Ah ok, so it seems the balance is incorrect and too fast.  Out of curiosity if you set your timegrapher manually to 18000 what time rate would you get ? Surely out of the tg range as it would be 240 minutes fast.

58 minutes ago, mbwatch said:

The other difference illustrated there was that P72a has a mobile stud carrier while P72 is fixed. Mine had a mobile carrier to begin with so I assumed P72a (even though it says P72 under the balance). Same mobile stud carrier on the other parts movement I had on hand (a P75 with calendar). I

The carrier might have just have been a modification if it fits both calibers. Either fitted in the factory as an alternative to the fixed stud, or maybe the balance complete was changed afterwards by a repairman, maybe all that was available at the time.

[mbwatch]

46 minutes ago, Neverenoughwatches said:

Ah ok, so it seems the balance is incorrect and too fast.  Out of curiosity if you set your timegrapher manually to 18000 what time rate would you get ? Surely out of the tg range as it would be 240 minutes fast.

I'll test this later. I'm am pretty confident if it can read it, it will be bang on 240 min fast. The movement is running perfectly +2s, 0.0, 280deg, albeit in an alternate world where one minute is only 50 seconds long. I use the open source tg timing software rather than a Weishi, and it is pretty flexible in what it's willing to calculate and display.

I did count the teeth on my spare P72 escape & 4th wheels. They're 15 & 60 respectively, same as on my too fast movement. So I won't be swapping in train parts I happen to have on hand. It will need to be a different balance if I do pursue this.

[Neverenoughwatches]

2 minutes ago, mbwatch said:

I'll test this later. I'm am pretty confident if it can read it, it will be bang on 240 min fast. The movement is running perfectly +2s, 0.0, 280deg, albeit in an alternate world where one minute is only 50 seconds long. I use the open source tg timing software rather than a Weishi, and it is pretty flexible in what it's willing to calculate and display.

I did count the teeth on my spare P72 escape & 4th wheels. They're 15 & 60 respectively, same as on my too fast movement. So I won't be swapping in train parts I happen to have on hand. It will need to be a different balance if I do pursue this.

Thats a bugger, its certainly not something you can adjust with the quicker balance. I'm not sure of complication of changing the gearing, might be just a 72 tooth fourth wheel but could be all of needs changing.

[mbwatch]

32 minutes ago, Neverenoughwatches said:

might be just a 72 tooth fourth wheel but could be all of needs changing

There are a lot of 72A junk movements available, mostly without a balance, even in my own country. So replacing the whole train at not too much more cost is feasible. I wouldn't take a chance on ordering a 4th wheel alone, in case it's mislabeled. After that I would have a nicely hot rodded old Waltham with a perfect dial and higher bps.

[JohnR725]

15 hours ago, mbwatch said:

Lorsa P72a (Ranfft lists as 21600, no mention of 18000)

this appeared to be a hard movement to track down but I had a suspicion of something and found the verification here

now with the proper name that I can look up we find something interesting? This is a problem that comes up commonly with Rolex for instance where and I haven't even looked at the parts I just snipped out the listing were often times things refer back to the base caliber without the letter a perhaps the case of Rolex a lot of the movements will go back to the base caliber run it 18,000 but the other ones are running at higher frequency and of course people start mixing and matching and not look at the exact parts and then you get stuff like this balance wheel one frequency watch another or worse yet mixing components the gear train is now at some bizarre frequency

so for my listing above it looks like that letter a is really important if you what your balance to oscillate at the correct frequency. Then 72 isn't even the base caliber it looks like the 62 is the base caliber. A case for balance wheel cross references like this

 

 

 

[Neverenoughwatches]

3 hours ago, JohnR725 said:

this appeared to be a hard movement to track down but I had a suspicion of something and found the verification here

now with the proper name that I can look up we find something interesting? This is a problem that comes up commonly with Rolex for instance where and I haven't even looked at the parts I just snipped out the listing were often times things refer back to the base caliber without the letter a perhaps the case of Rolex a lot of the movements will go back to the base caliber run it 18,000 but the other ones are running at higher frequency and of course people start mixing and matching and not look at the exact parts and then you get stuff like this balance wheel one frequency watch another or worse yet mixing components the gear train is now at some bizarre frequency

so for my listing above it looks like that letter a is really important if you what your balance to oscillate at the correct frequency. Then 72 isn't even the base caliber it looks like the 62 is the base caliber. A case for balance wheel cross references like this

 

 

 

Mbwatch didnt say why the balance that came in the watch needed changing. Seeing as there are a lot more variations for movements that run at 21600bph that one was probably incorrectly fitted at some point.

[JohnR725]

6 minutes ago, Neverenoughwatches said:

Mbwatch didnt say why the balance that came in the watch needed changing.

I think he was hinting without saying

18 hours ago, mbwatch said:

Everything that could go wrong has gone wrong; I got a balance complete delivered from Italy,

the first part of what I quoted above the says about everything went wrong and he got a balance complete from Italy. This means that an accident probably occurred.

18 hours ago, mbwatch said:

I have an absolutely cursed Lorsa P72a (Ranfft lists as 21600, no mention of 18000) that has been a "learning machine" for me. Everything that could go wrong has gone wrong; I got a balance complete delivered from Italy, pallet fork horns were bent, host of other problem all solved until it ran perfectly on time at 280deg, 21600. Then last night I put the seconds hand on it and it makes a perfect 50 second revolution.

I have a donor I didn't use for anything and will start by checking the 4th wheel teeth count in hopes that one is 18000. It's very clear I put in a 21600 balance complete.

I know it's a silly question but are you sure that it is a P72 a? Because if that's what you have it's supposed to be running at 21,000.

14 hours ago, mbwatch said:

spare P72 escape & 4th wheels

there is a number again but where's the letter a? This as you're finding out what happens when you start mixing and matching and their variations of the watch frequency the balance wheel oscillate at its natural frequency and the gear train will revolve at its gear ratio and have somebody's been mixing and matching then things get really interesting as you've discovered.

[Neverenoughwatches]

2 hours ago, JohnR725 said:

know it's a silly question but are you sure that it is a P72 a? Because if that's what you have it's supposed to be running at 21,000.

Further up back in post mb tells us that it was an assumption based on the balance that it came with. P72 was found under the balance.