Running Oh So Fast

[RyMoeller]

This one is a real mystery so I thought I'd toss it up here in case someone else has come across this problem.

I've reassembled the Speedmaster which was lacking parts (see my earlier inquiry) and have it up and running again.  It's a long story, but amazingly, after missing a bundle of parts and being terribly neglected it's ticking away happily and the chronograph works a treat- but there is one BIG problem.  The watch runs about two seconds per minute fast.  That adds up pretty quickly.  The timegrapher trace isn't too clean but it is consistent and it shows the watch running much better than that (note the image only shows about fifteen seconds of testing on the timegrapher- rest assured, I've had it on there a lot longer than that and have pretty much the same results in all positions)

So why does the timegrapher show the watch running less than a minute off per day but the real world results are so much different?  Well there are a few possible problem areas.  The escapement may be the problem but the pallet and balance (complete) are NOS replacements from Omega.  The escape wheel may be an issue- perhaps it's "slipping" and the pallet isn't locking consitantly.  I haven't seen this on the timegrapher though nor have I seen it while watching the escapement under the microscope.

Still, there are some obvious issues in the movement.  Whoever disassembled it before I got it made a real mess of things.  For example the Delrin wheel on the coupling clutch was vandalized terribly by a wayward screwdriver.  It took quite a while to get the gear teeth back in order  (the "before" picture is below).  I secured a NOS replacement just in case the movement didn't accept the repaired coupling clutch.

Despite this damage, I don't think the coupling clutch is the problem.  The fourth wheel is positioned between the coupling clutch and the escape wheel.  It's regulated by the escape wheel and the running seconds hand is attached to the fourth wheel.  When I time how long it takes for the running seconds hand to make a full revolution using the stopwatch built into my iPhone, I come up with 58.02 seconds (give or take a few tenths).

I'm thinking the problem lies with the fourth wheel. The Lemania 5012 chronograph movement (Tissot 2920) runs at a slower 21600 BPM.  The replacement of the fourth wheel in this movement (a Lemania 5100, 28800 BPM derivative) with one manufactured for the 5012 movement would result in the watch running fast.  It's questionable whether the gear teeth would mesh properly though.  Right now it's just a guess.  I may need to tear the movement back down to check the gear ratios for each of the wheels in the train to confirm they're the right wheels for the  movement but before I do that I was hoping someone might have solved a mystery such as this before and have some good advice to offer!  Thanks in advance for the comments as they are always helpful! 

Here's a few pics of the watch as it came together- just because posts with pictures are always more interesting.  

Before service begins...

Why you should periodically replace your spring bars-

Omega gaskets self destruct (turn into tar) if they aren't replaced for forty years...

This movement is plastic fantastic but I love it.  Other than not keeping time at all, it's the easiest chronograph movement I've worked on.

Look at all those scratches- previous watchmaker's tool of choice- screwdriver or machete?

It is a pretty beast though. 

[Delgetti]

I think it's a gear ratio issue. A friend had a similar problem on a rolex 1570. He bought a spare part (in his case second wheel) on the internet. But the seller mixed parts of 1570 and 1530 which are very similar but have different numbers of teeth. So the wrong wheel fitted but established a wrong ratio. 

 

I would count the teeth on the wheels you replaced and compare that with the originals. 

[JohnR725]

Somewhere in the discussion group within the last year or so this came up with another watch. So the balance wheel oscillates at its proper frequency the timing machine shows that. The gear ratio has to correspond to the rate of the balance wheel as pointed out above. If the gear ratio is wrong the hands will move at the wrong rate. 

I've seen this occur with several watches they came in several different frequencies. So basically physically identical watches a lot of parts would interchange except gear train they be some differences there for the gear ratio and the different balance wheels. This met when you're ordering replacement gear train gears you had the make sure you ordered the correct wheels or you'd get exactly what you have strange timekeeping.

 

[clockboy]

Yes agree with the  previous two posts it is most likely the wrong gear ratio. Also noticed one side of the escape is not good.If it was me I would check the replacement wheels and then re-clean and lubricate the balance etc. 

[RyMoeller]

Thanks for the responses.  It's nice to see a consensus as it gives me confidence with what to do next.  I noted also the dirty read on the timegrapher as @clockboy pointed out and will see if I can clean that up a bit. 

Apologies for not responding earlier- my wife's Mini Cooper blew up and the repair has kept me quite busy and away from the desk!

[RyMoeller]

Well the replacement fourth wheel finally arrived so I was able to compare it against the fourth wheel that was in the movement. Now I know the wheel I purchased is correct for the 1045 movement because it was still in it's original package (Omega Part #1045-1243) but to the naked eye (well, with the help of a loupe) both wheels looked exactly alike. This caused me a bit of concern since I was pretty darn sure the fourth wheel was the problem area. Counting gear teeth under the microscope the old wheel came in at 83 teeth and the new wheel at 84! Now we're getting somewhere!

So I swapped out the wheels, reassembled the movement and let it run for a day- it's now just a minute fast per day and I should be able to regulate that out.

Looking at the wheel to pinion tooth ratio I came up with the following- which obviously isn't correct so maybe someone can check my math! The movement beats at 28,800 bph.  That's 8 beats per second.  The escape wheel has 20 teeth on the wheel and 7 on the pinion.  The correct fourth wheel has 84 teeth.  If I figure there are 480 beats per minute (8 x 60 = 480), then divide that by the twenty teeth of the escape wheel (480/20 = 24), multiply the quotient by the number of teeth on the escape wheel pinion (24 x 7 = 168), and lastly divide that by the number of teeth on the fourth wheel (168/84 = 2) I end up with two revolutions of the fourth wheel for every 480 beats; clearly this isn't correct.  The correct answer should be one revolution of the fourth wheel for every 480 beats.  Somewhere I goofed up.  If I run the same calculation using the old fourth wheel (83 teeth) I end up with 2.024096 revolutions per minute- which is clearly faster.

At any rate the problem has been solved and the watch is finished.  Thanks again for the input everyone.

[rodabod]

To do a train count (apologies if you know this already), you just need to know the tooth and pinion counts of each wheel from the centre wheel onwards. From there you can calculate how many beats are required to make the centre wheel turn once (ie. one hour).

 

 

[Delgetti]

You calculated 24 revolutions of the escape wheel per minute which is the error in my point of view. Looking at a working escapement you will find that it needs 2 beats to let the escape wheels move 1 tooth further. So the ew makes 12 revs per minute and your further calculation fits. 

[Nucejoe]

Hi RyMoeller, A mystery indeed. I thanks for sharing.

Congrat.  

 

 

[nickelsilver]

To calculate the train, you can start from the center wheel for beats per hour or in your case the 4th wheel for beats per minute (if the watch has a 4th wheel that turns once per minute, usually but not always the case- of course it is for a chrono), as you like. From center wheel you multiply all the wheels, CW x 3rd x 4th x escape, x 2 (for the two beats for a tooth to "escape"), and divide that by the driven pinions multiplied together, so 3rd x 4th x escape. That gives the beats per hour.

In your case it's from the 4th, so 84 x 20 x 2= 3360 divided by 7=480 that's the beats per minute, or 8 per second, or 28,800 per hour.