Good day, fellow addicts... I recently acquired a 1950's Smiths Enfield long-case clock at auction. It looked immaculate, but needed nine (9) bushings, including one end of the escape wheel. After rebuild and on test, it was running fast - about 45 seconds per hour. I hadn't tested it before re-bushing so I don't know anything about its time-keeping before I worked on it. The pallet cock is not adjustable; I didn't move the pallets, and I didn't remove or move the back pallet cock. The pallets required only minimal polishing, do I don't believe I changed the geometry of the pallets in relation to the escape wheel, except for bushing one of the EW pivots. So - on test, it's running fast. Pendulum adjustment was untouched since I acquired it, and the marks on the wooden pendulum rod suggested that the pendulum bob had been in that position for many years. However, on lowering the pendulum bob, I could see that it had been in other positions in its lifetime. But even after lowering the pendulum to its lowest possible position, it was still running fast. I've only managed to get it regulated by adding a magnetic weight to the back of the (very light) bob. I've read up a bit on deadbeat escapements and I understand that insufficient depth will cause the escapement to behave as a recoil. I can see some recoil, but I'm unsure as to whether what I have is actually a half-deadbeat where recoil on one pallet is to be expected: Should I be looking at adjusting the back pallet cock to lower both pallets? Simon
Yes, I would try lowering the pallet arbor first, and see what happens. Lowering an individual pallet may be necessary but that will make the pallet tip spacing decrease. Check E- wheel's pinion mesh to make sure the new bushing is not positioned low. Nice video, WIllie X
Yes. Sometimes only one pallet cock is movable. Can we see pictures of the pallet cocks? and perhaps the whole clock and movement? If the escapement was working correctly before you installed bushings but now it doesn't, I would question if perhaps the bushings were not installed precisely centered on where the original hole was. Don't worry about the time keeping until you get the escapement operating as a deadbeat. Sounds like an interesting clock. RC
The pallets don't look right to me. The entrance pallet is locking as it should, but still recoils. The exit pallet is definitely landing on the impulse face, so that's causing some recoil there. But, if the pallets are made correctly, there should be no recoil, and the fact that the entry pallet is recoiling makes me think that someone may have transferred the pallets from another clock. Also, the pallet on the entry side looks like it has a weird bend at the end. Maybe a home made attempt?
I don't know. I thought I have seen the teeth landing on the impulse faces of both pallets. That's where the recoil comes from. Uhralt
On the entrance pallet there's a reflection that makes it look like it's hitting the impulse face, but I don't think it is If you stop the video right at the locking point it becomes a bit more clear. I also ran it it slow motion, and think it's getting a little lock. That reflection is odd, and makes it look like it's missing the pallet. Peanuts: if you could do a manual check for us, you could clear that part of the escapement for us. Just move the crutch slowly by hand, and see if the tooth hits the dead face or the impulse face.
Thanks all. You've confirmed some of my suspicions, and I'll do some more investigation tomorrow. Sorry about the reflection in one of the videos; it does hinder the observation of where exactly that pallet lands. I've found a couple of other videos that I shot earlier, before the clock was properly in beat, but maybe they show what's going on more clearly (I think between them they show both pallets, but working upside down, using a pencil camera with a mirror on the end, hand-held, while trying to start the video with the other hand, gets confusing!): In the meantime, some pictures I've already taken: The whole clock: The movement front and back (before cleaning): The anchor: The pallet cock, showing no adjustment is possible at that end: The amount of pendulum adjustment available when I acquired the clock: And, just for fun, the original price tag that was given to the retailer to put on the clock (but which had never been filled in, so I don't know how much it cost): I also have the guarantee, shop setting-up instructions and customer instruction booklet! Simon
That shows that I was seeing it wrong. Both sides are hitting the impulse face. After you adjust the height of the anchor, show us another video please. Good pictures BTW. I don't see what looked like a bend in the pallet there.
Hey, that puppy has plenty of adjustment. The front adjuster appears to be in the full up pisition! All you need is around 6 - 8 thou downward at the pallet. It doesn't matter if the arbor is not level as long as there is no binding. WIllie X
Willie, that photo of the front plate was taken at a bit of an angle. This one (again, before the strip-down - look at the poop on that ew pivot) shows a more accurate representation of the position of that front adjuster: But still room to adjust it downwards. I'll try that, as shutterbug suggested, and take some more video. And will try to check the accuracy of the placement of the ew bush. And now I look at it more closely - someone has attacked that screw before. Not me - honest. I'd formed an opinion that this clock had never been touched before but I'm clearly mistaken. Maybe they messed with the pallets too? I don't have a clear photo of those screws. I'll take a look tomorrow (nearly bedtime this side of the pond...). Just so I understand - am I looking at a deadbeat, or a half-deadbeat? Simon
What Willie said. There is nothing "half" about a half deadbeat. Both setup the same way, both have dead faces, both have impulse faces. The half deadbeat is really just a deadbeat that doesn't quite have perfect shape and may have a slight bit of recoil on the dead face where a true deadbeat will have zero recoil when properly adjusted. RC
This is how it's working now: Does this look OK? Any lower and it won't run at all. The pictures and videos were taken using this USB camera, which connects to a tablet, smartphone or computer: Focuses from infinity to about 3mm (1/8 inch), has illumination built in, and comes with a mirror: Yep, hardly any shift at all got it back to where it is in the videos. The pendulum bob is now back at the original position, and the regulation is spot-on at 3600 BPH. Thanks for that clarification. I'm very grateful for the help I got on this issue from everyone. Simon
I think you are going in the right direction but are not quite there. The first video seems to show that the teeth now land at the edge of the impulse face and a slight recoil causes them to touch the stop face. I think that the lock is not secure enough and the clock will run correctly only for a relative short time. Maybe adjusting the pallet down a very small amount will solve the problem. Perhaps this is only my impression from the video and when you look from a different angle you may be able to see if you have secure lock or not. Uhralt
The escape wheel teeth are still not locking on the dead face. It will never run right until you get it locking on the dead face. If you can't get the verge any closer without it locking up the escapement then there is an issue with the pallet spacing. In this type of verge it isn't unusual to find that someone has "messed with" the position of the pallet nibs, or even bent the verge. It may be time to consult the literature such as Practical Clock Escapements, by Lauire Penman, or similar references that describe in detail the geometry of and how to adjust Vulliamy pallet deadbeat escapement. Penman illustrates this quite well. RC
OK, you have a piece of lint on one of your e-wheel teeth there! If you're at the 'lock-up' point. Looks like your pallets are two wide. You will need to push one of the pallet bits out a bit (toward the e-wheel) to narrow the spacing. Usually the pallet heights are not exactly the same, just move the one that will make them more even. They don't have to extend exactly the same and you will rarely have to move both. After this you will probably need to raise the pallet arbor back upward a bit. WIllie X
You're right; the screw slots for both pallets show signs of previous adjustment. I'll do that over the weekend and report back. I'm determined to get this right. Thanks all! Simon
This always looks simple but it often takes many trials to get it back to exactly where it needs to be. Adjusting just one pallet will reduce 'overshooting the mark'. This can be one of the main problems. One of my tricks is to loosen and then lightly snug down the pallet (you plan to adjust) to a point where it can still be moved. This way you can easily make any number of adjustments, using a hemostat or similar tool, and you only have to remove the pallet assembly twice. Willie X
I may have got this right first time through dumb luck. The pallet adjustment had clearly been messed with previously: Not only are the screw slots chewed up, one of the original screws was replaced. I decided to adjust only the pallet that had two original screws, in case the thread in the anchor where the replacement screw had been used had been compromised. Good idea! I would like to have used this method but I didn't have a suitably angled tool. I'll be looking to add one to my toolbox... My first adjustment was by .25mm (1/100th of an inch). I then had to move the adjustable pallet cock up, as you predicted, to stop the escapement from locking. This was the result: I can't see any recoil; the pendulum amplitude is greater, and the position of the adjustable pallet cock looks centralised instead of lower than central. Am I done? Or should I be thinking of a little more? Simon
You have lock now on both pallets but the lock shown in the first video is very small. The clock will run this way and keep time but personally I would make the small lock a bit more secure by lowering the pallet a very small amount. Then you will be good! Uhralt
A little deeper might be good but that's probably OK as is. You have lock but I can't tell exactly how much. When you look at it, with good magnification, slowly move the crutch back and forth. The tip of each tooth should fall clearly onto the dead face by about 2 or 3 thou, make sure you check it for every tooth. A slight discrepancy with just one tooth can spell trouble when the locking is marginal. WIllie X
I'm thinking it may be fine the way it is too. There's more lock on the entrance than on the exit, but I believe it will run fine that way. No recoil now! Good job!
Looks like it could use a hair more lock, but that won't make it run any better, so if it is happy I would leave it as is. Check it in a year or so and make sure there is still no recoil. RC
I would lower the arbor adjustment just a tad and see what happens. I think I'm seeing a good bit more drop onto the entrance pallet than the other. Wonder if you could move that fancy little scope out a bit, enough to see both pallets at the same time. WIllie X
I did try that, but any lower and it locks up, so I'd have to move the pallet some more. The clock is back in its case in our hallway now, so I think I'm going to run it as it is for a while and see if anything changes. I've observed a full revolution of the EW several times, and I can't see that any tooth behaves any different from all the rest. I've also observed the trace on a beat monitor application on my smartphone, and again there doesn't seem to be any variation throughout several revolutions. I can't find a viewpoint that allows both pallets to be observed square-on at the same time. I suppose if I had a couple of small mirrors that I could position either side of the arbor it could be done. I'll see about fabricating something like that for future use. Thanks again to all who contributed. That's the first time I've had to adjust a dead-beat and without your support I know I'd have struggled. Simon
