Pin Pallet Adjustments

[eclecticbeat]

Hi All and Happy New Year,

This is only my second attempt at a torsion clock service and all beginner's luck has vanished. The first one was a relative's K und O Standard from the 50's. Just from reading the threads here and Terwilliger's Repair Guide, it went without a hitch and runs great! After performing all the normal tasks with this second one - I'm stymied.

Details:

Pin pallet, lantern pinions (c. 1913-1915 per JH)
Plate #1041
Pendulum #38 (has some lead inside I'm not sure should be there - easily removable but for now, it's in there)
Inside front plate, inside rear plate and bottom of pendulum are stamped 17001
Bracket/Saddle very similar to a #1
New Horolovar mainspring and suspension spring (.0038")
New Single tine fork (no binding I can detect)
All is clean/lubed and it powers up with a 1/4 turn of winding arbor
End shake feels good and there's no evidence of bent pivots
Beat is equal but not much overswing
Pallets look a bit tapered to me; wear marks on EW teeth seem to support that


Problem:
It just barely runs. 180~190 total rotation and I calculate it's losing about 1.5 hours in 24. I suspect the pin pallets but don't know if there's anything I can do with them.
According to 'The Book', the anchor pallets are not performing well at all. Drop is unequal and the pins are not drawn all the way to the rim of the wheel.
Measuring drop as the distance before a tooth falls onto a pin - the exit pallet drop is greater than the entrance drop.
The exit pin goes the deepest onto a tooth but only about half way down the tooth; entrance pin - not quite as far.

Are there any adjustments or did I miss something causing the pallets to act this way?


Couple of pics from the auction

 

[shutterbug]

The pallets and ew can't really be compared to a typical escapement, and are rarely an issue in 400 day clocks. One quandary I see though - you say it has poor over-swing, but then say it has poor rotation. Normally it's going to be one or the other, not both. Rotation is increased by raising the fork, but that action tends to decrease the over-swing which is essential for the clock to run. "In beat" is when both extremes of the pendulum over-swing the same amount. So, maybe the fork is the place to look first. Remember that small adjustments make large changes. Also check that your fork has a paper thick gap between the pin and either edge. Losing time is probably due to the pendulum/suspension spring differences. A thicker spring will speed it up, as will a lighter pendulum. Maybe you could post a pic of the pendulum, so we can see if it's the right one for the clock

 

[eclecticbeat]

Hey shutter,

It's been running almost a day now and is over an hour slow. The total pendulum rotation is right at 180 degrees. The overswing is 40 degrees on either side.
Kind of cool (to me) but like watching paint dry compared to that K und O. It had 360 total and 172 of it was overswing (86 either side).
I know exactly want you mean about 'small changes'. That one had a set screw holding the saddle in place - just tightening it after setting the beat could alter the beat 5 degrees. I had to set it up so the rotation of tightening the screw would bring both sides perfectly equal. Slightest tap on the tapered pin could alter it 20 degrees.

I've had this fork at different heights but it wouldn't run at all for me unless it was dead straight in the loop. The anchor would just 'snap' back and forth a few times then quit.

Only info on the correct pendulum for this clock I could find was on this thread. The clock there is very, very similar but the pendulum on it might not be correct.

This pendulum was pretty banged up when I got it. The flat top was bowed-in and the regulation rod broken. I straighten it out and made a new rod. There's about a 1/4" worth of thin steel rings that go around the inside edge and two thick support plates under the flat top. The lead was poured in at some time and the imprint on the reverse and sides tells me where it was. I just don't know when it was put there; would the factory do it, to get it to just the 'right' weight they wanted?

When the flat top was damaged I can only guess the lead broke into these three pieces. Some wadded-up paper between them and the bottom cover was holding them in place, so I just did the same when I put it back together. Since I took the pictures I put the bottom cover back on but left the lead out to see if that makes any difference.
I'll check it in the morning.

 

[ivancooke]

The lead was added to the pendulum to make the clock go slower. Years ago there wasn't the same selection of suspension springs we have now, so more or less weight at the pendulum was the only option to get the clock within close adjustment.
The steel rings are also a type of adjustment.
Add rings to make it go slower.
Remove rings to make it go faster.



Ivan.

 

[eclecticbeat]

All I attempted last night was to remove the lead (and reset the beat just to see if I could repeat it). The clock was still running in the morning. Total swing improved to 195, overswing still at 40 but pallet lock had all but disappeared.

The exit was sometimes locking on the lifting face and entrance always on the corner of a tooth. I'd think it would have stopped running in minutes like that but it was still going 9 hours later

"In beat" is when both extremes of the pendulum over-swing the same amount. So, maybe the fork is the place to look first.

Why does more overswing result in less total rotation and vice-versa?

I checked fork again and I think that's it (or at least a likely suspect). I just don't know what to do about it. When the anchor is in the center position the fork is perfectly free with no binding. I didn't think to check it at the extreme ends of motion before. Tilt anchor to the exit side and there is a little bit of binding but on the entrance side - lots of binding. I adjusted the fork height several more times but each time - all that accomplished was to have it bind in the center and the ends. I've managed to get it back where it doesn't bind in the center but either end is still a problem. The darn thing is acting like a brake severely arresting the entrance pin. What's the cure? spread the loop?

Hi Ivan,
Thanks

 

[ivancooke]

I have seen clocks with this pin pallet movement and with a loop on the anchor pin before. They both had the suspension fork fitted level with the loop.
The fork needs to be a lot wider than the normal one, to straddle each side of the loop.
The photo below shows a similar clock to yours, where I fitted a single pin fork, with the pin through the loop.
It didn't work very well and kept catching, so I widened a two pin fork,to fit round the loop, and it worked fine.


Ivan.

 

[John Hubby]

First comment is that the pendulum with your clock is the correct one, but should NOT have any lead in the disc. Removing all the lead will definitely allow the clock to run faster and with greater rotation of the pendulum. These were constructed with a thin brass shell, one or two steel discs with slots for the weight guide pins to fit inside the shell, and then the weight rings included for adjusting rate. The bottom cover of the disc originally would be like the photo I've posted below.

All I attempted last night was to remove the lead (and reset the beat just to see if I could repeat it). The clock was still running in the morning. Total swing improved to 195, overswing still at 40 but pallet lock had all but disappeared.

The exit was sometimes locking on the lifting face and entrance always on the corner of a tooth. I'd think it would have stopped running in minutes like that but it was still going 9 hours later

What this immediately says to me is that lock and drop on your clock is not set properly. On these clocks, the lock and drop are set by moving the upper suspension bracket that also serves as the anchor arbor pivot plate. This can easily be put out of adjustment if you disassemble the clock and don't mark exactly where this part is situated before removing it. From what you say, this part needs to be slightly lowered to allow the pallet pins to lock on the locking surface of the escapemet wheel teeth instead of on the impulse surface or just on the corner. That will increase lock for both pallet pins. To adjust drop may require moving the bracket slightly to the left or right depending on how much you have on the entrance and exit pallets.

Why does more overswing result in less total rotation and vice-versa?

This is basically the physics of conservation of energy. When you have less overswing it is because the pendulum has to turn further between locks to unlock one escape tooth and lock on the next one. When you raise the fork, that is what happens. It also means the amount of impulse energy transmitted to the pendulum is greater because of the longer time the pin pallet is being pushed by the impulse surface of each tooth, giving more rotation. Lowering the fork has to opposite effect, overswing is increased but rotation is decreased due to less energy being transmitted to the pendulum.

I checked fork again and I think that's it (or at least a likely suspect). I just don't know what to do about it. When the anchor is in the center position the fork is perfectly free with no binding. I didn't think to check it at the extreme ends of motion before. Tilt anchor to the exit side and there is a little bit of binding but on the entrance side - lots of binding. I adjusted the fork height several more times but each time - all that accomplished was to have it bind in the center and the ends. I've managed to get it back where it doesn't bind in the center but either end is still a problem. The darn thing is acting like a brake severely arresting the entrance pin. What's the cure? spread the loop?

As Ivan mentions this type of anchor pin should NOT use a single pin "fork", because there will be inevitable interference either at the ends of oscillation or at the center or both. The fork should have wide tines that allow the loop at the end of the anchor pin to easily slide up and down as the fork turns from right to left and back. See the second photo posted below. NOTE: With this design there if no benefit from raising or lowering the fork, since the contact point on the anchor pin doesn't change.

 

[eclecticbeat]

Thanks Ivan and John,
Oh, Ok - the fork goes on the outside of the loop
Ivan's Badische restoration was so beautiful it never crossed my mind the fork was also incorrect.
The profile of the correct fork looks pretty common; any recommendations on a modern replacement part suitable for modification?

Picture of pendulum bottom cover is posted below. Does the number help narrow down who made this clock and when any further?

The pallets and ew can't really be compared to a typical escapement, and are rarely an issue in 400 day clocks.

A typical Graham or any escapement? Seems like a half/deadbeat to me, except it uses pins instead of flat sheet metal.

 

[shutterbug]

A typical Graham or any escapement? Seems like a half/deadbeat to me, except it uses pins instead of flat sheet metal.

One of the things most clock smiths find frustrating about 400 day clocks is the way the escapement works. In principle, they work the same way ... but when people start altering the pallets they usually get in way over their heads. The 400 Day Guide has excellent information on proper setting when there is an issue, but like I said before .... most of the time it's going to be something else. If yours is out of adjustment, chances are someone else got to it first