Setting Beat on Kern #1406F Miniature, and other dilemmas

[hannmade95]

After finally getting some slightly better tools than my eyes and ears I have been recording my clock's escapement via a timetrax amplifier outputting to software on my pc to analyze the wave form. And I cannot for the life of me get the beat error to be less than 0.150 s. The error changes a bit as the clock settles into its beat but its generally somewhere between 0.100 s and 0.200 s off from the counterclockwise to the clockwise rotation of the pendulum. I haven't found anywhere what the acceptable beat rate error is on one of these very slow moving clocks. My limited experience thus far has been with wristwatches so this much slower moving machine has been somewhat frustrating to me. I had cleaned it initially because it was not running, oiled it, and set the beat the best I could by ear and visually and gave it back to my coworker after having it run for two weeks straight at my home. Upon receiving it he set it up and could not get it to run more than an hour or so. I took it back and began the process again, this time taking the time to replace the mainspring as it was admittedly worn out ( not even twice the size of the barrel when removed). I've now replaced the mainspring and with my new equipment I set it in beat (probably alot better than I originally did by ear). Other than the slight beat error I believe it will run fine but will let it sit on my mantel a while prior to giving it back. Does anyone have any suggestions as to what may have been stopping it besides beat error or a worn out spring? I should also mention I replaced the suspension spring to the correct length it it accurately tells time. How on earth do people work on these things and know that it will run a whole year with no problem? Also one other little thing I noticed while looking at the wave form I recorded is that the escape wheel moves slightly right before actually dropping a tooth on the clockwise rotation of the pendulum. Is this normal? It's only in that direction. Any help would be wonderful. This may be a bit of an information dump but I would really like to get this working for my coworker as it was his mother's and he's never seen it run even in childhood.

 

[KurtinSA]

As for the beat error, I think that too much information can "destroy" the nature of the clock! If I'm understanding things correctly, the issue might be that the relationship between the fork and the anchor pin as well as the position of the anchor pin when the fork begins its push might be different going in one direction versus the other. You might get an uneven push in that case.

It is maddening to have a clock run in one place and then not run in another. When I first started working on them, I loathed the idea that I was going to have to touch the pendulum adjuster in fear that the clock would then stop. It doesn't take much at all to stop them, given their lack of excess power.

My goal continues to be to get a clock to run for a year, but I'm beginning to soften my position. I don't think they will run that way now they are so many years old. They might have done that right out of the factory, but it's hard to believe they will ever do that again. Typically, people will just wind them twice a year or at the times of the clock changes in the spring and in the fall. Twice a year is not so bad!!

As for the escape moving in reverse, it could be that the lock face profiles have been modified over the years and the tooth is sensitive enough to respond to that. Again, if it doesn't stop the clock, maybe it's OK. It might be a very slight drag on power.

Kurt

 

[hannmade95]

Sorry I may have communicated something wrong. The escape is not moving in reverse, it just kind moves forward a little right before the anchor has fully moved up and allows the tooth to drop in, right on the little bevel that is on the anchor/pallet. I don't think its causing a problem but I also don't know enough to tell you haha. I'm glad I'm not the only one that has gotten a clock running on their bench and then it stop on someone else. I might offer to set it up for him. I haven't checked to see if I can but I will attempt to attach a short video of the escapement, its a hard angle to film.

 

[hannmade95]

Like in this diagram, when "A" is contacting "C" as "C" rises up the tooth jumps forward just a bit on that beveled part of the pallet fork before fully dropping and allowing the tooth near "B" to strike that fork. It consistently does it on every took at the exact same time every time. It may be normal I'm not sure. It is audible though. Just a slight "little tok" before the real tok.

 

[KurtinSA]

I don't see anything wrong with that escapement. I'm not there, so it's difficult to see the details.

Kurt

 

[hannmade95]

I don't see anything wrong with that escapement. I'm not there, so it's difficult to see the details.

Kurt

If you don't see anything wrong then I don't either haha. I appreciate it. Like I said I don't think its an issue but it was just something I noticed. I will say that the beat amplifier going to my computer so I could pin point the exact moments of the beats and use it to tell me the time in between was so much more precise than the other methods I'd used based on what other folks do. Don't get me wrong, the manual method of marking when the escapement ticks and where the overswing goes to worked, but after hooking it up to the computer I got it alot closer. And the best part Is that I don't have to wait a whole day to adjust the speed of the pendulum. I hope I have it running great now, I bought my own junk clock for ten dollars for me to play with, doesn't seem as nice though.... its a Herr and the horolovar book doesn't really have much information on it.

 

[KurtinSA]

I'm sure you're aware that you can use a stop watch to time the speed of the pendulum after just a few minutes of running. 8 beats in 60 seconds for larger sized clocks. It's difficult to be extremely precise of watching the pendulum and stopping the timer...but do it for 2 or 3 minutes and divide things out to help eliminate some of the error. I think most spring-driven clocks are going to have errors in the time keeping. Probably only gravity-driven clocks have a better chance of being more precise over time...at least the applied power should be relatively constant.

Kurt

 

[hannmade95]

I'm sure you're aware that you can use a stop watch to time the speed of the pendulum after just a few minutes of running. 8 beats in 60 seconds for larger sized clocks. It's difficult to be extremely precise of watching the pendulum and stopping the timer...but do it for 2 or 3 minutes and divide things out to help eliminate some of the error. I think most spring-driven clocks are going to have errors in the time keeping. Probably only gravity-driven clocks have a better chance of being more precise over time...at least the applied power should be relatively constant.

Kurt

You're right, I can't make it perfect. And yes sir I was aware that's how I was doing it before. Unfortunately I'm a bit of a perfectionist though. Trying to get over that part of me haha. I really appreciate the help and the conversations Kurt, I look forward to speaking with you again.

 

[Wayne A]

The escapement operation looks ok. Trying to figure out what the error you perceived and perhaps the actual impulse is what your talking about. Can't see enough to comment on beat or overswing.

Wayne

 

[hannmade95]

The escapement operation looks ok. Trying to figure out what the error you perceived and perhaps the actual impulse is what your talking about. Can't see enough to comment on beat or overswing.

Wayne

My main worry was that the difference between the beats were too much. I'm used to watches and the beat error is measured in milliseconds but those movements beat tens of thousands of times per hour. I don't thing its has to be as precise with these clocks is what I'm figuring out. The clock is working fine right now I was basically just asking how close the pendulum rotations from left to right (each beat as the teeth strike the pallet forks) had to be. I've got them within a tenth of a second apart right now. And I apologize if what I'm saying is confusing, I know what I'm trying to say and I understand how the movement works and how to work on it, I just don't quite know the right lingo that everyone else uses.

 

[Wayne A]

I see, well a clock beating 600 beats an hour is a bit different from a watch buzzing along at 18k+. There is normally some small variation in beat on these clocks often caused by the ew teeth. The way I look at beat error is with a degree wheel under the pendulum, normally try to get the beat within a couple of degrees in each direction.

Wayne

 

[tracerjack]

The double tick you hear is correct and a sign the escapement is functioning properly. Soft tick as tooth begins slide over impulse face to “kick” the fork, louder tick from another tooth hitting a dead face. The difference in volume from one top rotation to the other I would think to be related to entrance and exit hits. Coming from the world of watches, I think you will need to adjust your expectations of time keeping on these movements. They are adequate time keepers, have the potential to be good time keepers, but lack any means of refined adjustment. I don’t consider that wheel on the pendulum to be “fine” by any stretch of the imagination. As to running in one place and stopping in another, that describes these perfectly. I’ve had one go out of beat because I bumped the pendulum lifting the glass dome.