TIMEKEEPING PROBLEM: Impact of changing the platform escape wheel on a carriage clock

[zygo]

Here's my first proper post, copied from the one I made on the NAWCC forum last week, so far unanswered.

I am working on an antique French carriage timepiece. It's a good make (Hollingue Freres who produced for Drocourt), with lever platform by Joseph Soldano (who also supplied Dent), all freshly overhauled. It runs with a good amplitude at the usual 18000bph. But it gains time massively - about three minutes each hour (that's over an hour per day!).

The wheels are all perfect with no damaged or missing teeth and the silvered Soldano platform is almost certainly the original. Now, the escape wheel has 15 teeth and its pinion has 8 leaves - all pretty common. It looks original but in the absence of any other explanation, I'm wondering if it's a replacement and if the original had 15 teeth and seven leaves. If I were to change the platform for one with an escape wheel with 7 leaves, it would obviously run slower while maintaining the 18000bph, but how much slower? Is it a simple math question of seven-eighths as fast as the current one with 8 leaves?

63mins x 7/8 = 55mins would mean it's far too slow.

What am I missing here? @nickelsilver

[JohnR725]

Are you open to other things that would make your timepiece go faster.

Does it have a bimetallic balance wheel for instance? Often times they get squeezed they're no longer round the arms get bent in that would speed you up. Also when you're looking at your balance we'll make sure you have even pairs of timing screws. It's amazing how fast you'll go if you're missing one of the timing screws so there have to be opposite pairs in other words if there's a screw on one side there should be a screw on the other side.

If the hairspring is magnetized has oil on it or anything that causes it to be stuck together. Or for that matter it's touching anything. Also how was the balance wheel removed from the bridge oftentimes they're unpinned as do not do not have a stud and if that happens somebody will shorten their spring.

 

[nevenbekriev]

John, I believe that @zygo has checked the beat rate with timegrapher and has adjusted to exactly 18000, but still the clock goes faster.

@zygo, please count the teeth of the wheels and pinions in the train. But the most pobable answer here is:

18000 X 24 / 25 = 17280, which is another well used in various movements beat rate. May be somebody replaced the hairspring and vibrated it for 18000 and wasn't clever enough to count the teeth...

Edited May 12, 2024 by nevenbekriev

[zygo]

Fast responses - thank you.

Yes, I mentioned the 18000bph to illustrate that it's not a matter of touching coils, over banking, shortened hairspring etc. I even took off the platform and temporarily fitted a another NOS one also with 8 leaves and beating 18000 but the colossal gain is the same.

I'm resigned to buying and fitting a new Gorge platform plus a 7 leaf pinion escape wheel but only if I think it will work; otherwise it's a waste of £200 and a couple of hours work. Hence the underlying question is: all other things being equal, what would the timekeeping effect be? Is it a simple 7/8ths as fast? Logic tells me it is but can it be that linear?

[nickelsilver]

Let's say, like in a watch, you have  a 70 tooth 4th wheel, which needs to make a turn in one minute, with an 18,000 beat escapement. 15 tooth escape wheel as usual, and a 7 tooth pinion on it.

 

70 x 15 x 2 /  = 300  That's 300 beats, and there are 5 beats per second, so divide and you get 60s, one minute- all is well. If we do 70 x 15 x 2 / 8 that equals 262.5. With an 18K balance, which will let the escape wheel move one turn in 6 seconds, the 4th wheel will have made like 1.14 turns. So the time display is running fast. But this comes to like 7.5 seconds over per minute, which is about double your gain (maybe I'm wrong- I'm mathing on my phone while writing on my phone ) but pretty sure the gain is more than you are observing.

 

A possible culprit is that crown/4th wheel isn't running true, and a tooth is slipping past a pinion leaf once per revolution. I've seen this. If the platform is original and it's a nice piece they are often pinned, so the depthing isn't an issue. If the platform isn't pinned, check the depthing. And at any rate check the crown wheel between centers and make sure the teeth are running true.

Edited May 13, 2024 by nickelsilver

[oldhippy]

It sounds to me like it's a hairspring problem. Are you sure it is the correct one. As already explained teeth and pinion count are correct.    

[zygo]

Thanks for the thoughts. I didn't think the hairspring per se was the problem: 18000bph is by far the most common rate, it looks nice and evenly coiled and even fitting a new unused platform gives the same result. All the teeth on the wheels, including the contrate, are fine - so there's no skipping. A tooth count may solve it:

Wheel          teeth          leaves

Centre         64

Interm         80               12

3rd              64                8

contrate     70               10

scape         15                 8

This indicates a required beat of 16800bph which is a rate I've never encountered before but might be common to some watches (about which I know little). So despite the hairspring looking perfect visually, it should be longer. It's anchored at the very end so I can't do much about that. I will try adding a couple of balance screws to slow it down as unless other members have a better idea, the only other option would be to cut and mount a new contrate wheel with 75 teeth - a bit odd !

 

[oldhippy]

Odd teeth and pinion count do pop up. As you have said timing screws would be the place to start, you don't want to cut a contrate wheel that would be the last resort. It is more than likely the hair spring broke at some time. How about a nice photo of the troublesome clock.  

[nevenbekriev]

Ok, with the parts You have, the easiest thing to do is vibrate a new hairspring, of cousre for 16800 BPH. A couple of balance screws will be not enough, at least 3-4 aditional couples will be needed. And this if You have that many free holes. And cutting 75 teeth table for the contrate wheel is the other reasonable option

Different things may have happened to the clock - the contrate wheel may have be replaced from another modification, the platform may not be original, the hairspring may have been replaced and vibrated for 18000 or the entire balance may have been replaced wit the wrong one...

Edited May 14, 2024 by nevenbekriev

[zygo]

Great help for me - thank you all. If the hairspring has been shortened, it must have have even tighter coils than it does now and they all look utterly concentric but nevertheless I'm not leaning further towards that explanation. For me, vibrating a new hairspring would be far from easy - I wouldn't know where to start.And if a new contrate wheel has been cut and mounted, it's been done very skilfully, leaving no obvious telltale signs. Cutting and crossing out contrate wheels takes skill, too, so a bit of a last resort. and why would the he have cut 70 teeth if the existing one had 75. The platform certainly looks original to the clock but I cannot rule out the possibility that it has been taken from a similar period one (with a different tooth count). It was produced by Soldano (who supplied Dent) and the movement is by Hollingue Freres (who worked for Drocourt) so it's a nice small one (not a miniature) but I have to say that it does look like it has been worked on by someone with a hammer and centre punch instead of a bushing tool. The history isn't know - the current owner has never seen it working.

I completely dismantled it again at the weekend to make the tooth count but when I've figured it out, I'll take some photos. But for now, I think I've reached the end of the line.