Pinwheel escapement problem

[brianw]

I’m trying to fix a pinwheel deadbeat escapement which has all the pins on one side and pallets of differing lengths. I’ve had to regrind the pallet faces which had become curved and have closed the increased gap to the correct distance. The problem I have is that the clock runs for a while and then stops. There doesn’t appear to be anything wrong with the train. I’ve check it very carefully.
The only thing that seems very odd is that the drop on the top pallet is far greater than the drop on the bottom pallet. Also, the upper surface of the top pallet does not seem concentric with the pallet arbour as it should be. This means that there is a very slight recoil on the subsidiary arc of this pallet. I was thinking of adding some metal to this pallet and file it to be concentric and to reduce the drop. This would mean that the top pallet would be thicker than the bottom one, but I can’t see that that would matter.
I suspect that these pallets have been filed back by someone who didn’t really understand how they work, but I can’t be sure.
My question is this. Has anyone in this group come across an escapement like this where one of the pallets is thicker than the other? If the answer is yes, I will go ahead with the alterations. If the answer is no I am open to suggestions as to what to do.

[brianw]

I found out that it is indeed OK to have different thicknesses of pallet to ensure that the drop is the same on both pallets.
I made up a sliver of gauge plate, filed to size, hardened, soldered to pallet and ground to shape. Not an easy job, but it has fixed the problem.
The pallets are now operating correctly, and the clock is ticking away as it should.
This was a tricky one for me to diagnose and get the geometry right but all the more satisfying now that it is working.

[Dell]

It’s nice when we get a problem clock sorted.

Dell

[Author]

Posts