Tid-Bit 7 - Tipping Escape-Wheel Teeth
Published in the October 2010 NAWCC Bulletin starting on page 574
The even ticking of a clock. Quietly telling you of how great a job you did restoring the mechanism - a joy to hear. But, what if the beat is not even? Or, more aggravating, what if it is mostly even, but sometimes seems to shift a bit - to have an intermittent tick and a tock that are just not right. Welcome to the world of escape wheels - and, more specifically, the world of long and short escape-wheel teeth.
Let=s start out with this underlying assumption behind us: You have already done all you can to get the clock in beat - and, are satisfied that all is as right as you can make it. If the beat seems to periodically change you have already tried letting the mechanism spin very slowly without the anchor in place - looking for bent arbors or pivots, or fouled teeth.
If your mechanism
....Seems to have an even beat, except for one or two beats each rotation of the escape wheel, or,
....You have a clock that does pretty well for a while, but it seems that, just before it stops, the beat becomes erratic,
I am hopeful that this Technical Tid-Bit will help you understand what it is that is frustrating you, and also perhaps help you understand what you would do to Amake it all better@!
When a pendulum is swinging with a lot of over-swing (more swing than necessary to just release the pallets), the beat tends to be more even than when the pendulum is swinging just far enough to release the pallets. I have had a number of chances to work on escape-wheel teeth because of the number of long duration Viennese clocks I have had the chance to work on. In general I find that longer duration clocks operate a lot closer to the minimum weight required to swing the pendulum just far enough to release the pallets.
Why are long duration clocks more prone to having problems when escape-wheel teeth have been damaged? Let=s think through how much weight is typically used to run long duration clocks. It is not uncommon to find a 3 pound weight driving the trains on a week-duration Vienna Regulator. OK, now, assuming a years duration clock is as efficient as the week duration, it is only logical to say the year runner should have a weight that is 52 times as heavy as the week runner. But, well, 52 times 3 is 156 pounds. Yes, I have seen British years runners with weights in the 40 pound range. But, a Viennese years runner, with its exceptional pivot, gear and escapement work, typically requires about 16 pounds. In fact, I am currently finishing up work on a clock that runs 16 months on 16 pounds, and I have a 3 month runner in my collection that runs on 3 pounds. Almost sounds like a pound is the minimum for each months duration to me.
Running on such little weight typically results in a very narrow pendulum swing - people often ask if my long duration clocks are even running!
This is why, when I am setting up a clock, be it an Anniversary clock, or a conventional pendulum clock, I always try to set the pendulum swinging (or rotating for an anniversary or other torsional clock) just enough to barely Atick over@. This amplifies the impact of any Aout of beat@ issues.
But what about when you have your mechanism in beat, to the best of your ability, and it is still sometimes wanders off for a beat or two? Usually when I get a call or an e-mail about this I find that the repair person has already tried to rectify the problem - usually by Astraightening@ the Aobviously bent@ escape-wheel teeth. And, while the clock may have run before, well, now it doesn=t seem to want to any more.
In general (yes, I realize that a statement like this just begs for contradiction) bent escape-wheel teeth will result in an escapement hanging up - a pallet fouling or Arunning into@ a tooth when it is supposed to just clear it. Yes, a bent tooth does end up just a bit shorter than the other teeth, but this is a very small change in length (has to do with the cosine and the tangent of the angle by which the tooth is bent from its correct position - and, since I do not want to remind any of you how much fun geometry was, let=s just say it is a very small change in length). But, in general, a bent tooth does not result in an uneven beat. That takes a tooth that is either too long or too short. Funny thing is, often when a repair person tries to straighten a tooth they end up making it longer, and now it is really out of beat. Or they break off the end of the tooth, but we don=t really want to go there.
A clock that wanders out of beat has a tooth or two (or three or four, well you get the idea) that is not real close to the length of the other teeth. Let=s think about this for a second to see why this should cause an uneven beat. As the pendulum swings it moves a pallet out of the way of an escape-wheel tooth. But, let=s say a tooth is shorter than most - the pallet will Arelease@ the tooth sooner in its movement than with the rest of the teeth. Because the pallet releases an escape-wheel tooth sooner in the pendulums swing the tick is shorter. Likewise, it the tooth is longer, the pendulum has to swing farther, and the tock is longer.
I think now is the time for a word or two of caution. The subject of this Tech Tid Bit, ATipping Teeth@ is something that requires a lot of care and a lathe with a cross slide. Yes, I have talked to people who have successfully tipped escape wheels with a power drill and a file. But I have also had calls asking if I had a spare escape wheel for a Gustav Becker anniversary clock, amongst others.
Why is it important that the beat be even? In many clocks it isn=t. Especially if they have a weight big enough to provide a lot of over-swing. Then it just doesn=t matter. But, as we take things to the extreme (think long duration which usually translates to minimal over-swing) an overly long escape-wheel tooth will shut it down. And a short one will not allow the mechanism to run as well as it could, and to stop sooner than it might otherwise.
I decided to write this article when I was faced with tipping an escape-wheel from a truly magnificent mechanism. As you can see in Figure 8, the workmanship (not sure how to make that term asexual) required to make this escape-wheel is phenomenal. Think I wanted to tip the teeth? Think again. But, the mechanism did not want to run. Yes, I probably could have put a bigger weight on the mechanism and overcome the problem. But, being a 6 month runner, and being convinced I could tip the teeth without destroying the wheel, I decided to go forward.
Now for another word of caution. I use a cross slide to hold a cutting tool to cut the tips of the teeth. If you have every used a cross slide you know that there are knobs (or levers if you have a lever-actuated cross slide) you twist to move the cutting tool back and forth and closer and farther from what you are cutting. I make myself practice with both knobs, making sure that I know which direction to twist the knob to move the tool in and out, and back and forth. If you inadvertently go the wrong way and cut too quickly I pretty much assure you that you will bend teeth and potentially trash your escape wheel.
I don=t recommend tipping teeth with a lever-actuated cross slide. I really like them for some operations, but not one as delicate as tipping teeth. And, I don’t recommend setting up to tip teeth with a stone – a properly-sharpened graver will cut extremely fine bits of brass off of the teeth, and will only be cutting right at the tip – if a stone is used it will be cutting across the entire tip of the tooth, significantly increasing the risk of bending teeth.
The first step is to mount the escape wheel in your lathe. Typically I find that holding the escape-wheel arbor in a collet is accurate enough. That is assuming the run out on the exposed end of the arbor (the amount that the exposed pivot moves as the gear is spun in the lathe) is minimal. Typically I chuck up the gear in the lathe, check it for run-out, if not acceptable I then loosen the collet, rotate the arbor in the collet a quarter turn, re-tighten, check run-out, and repeat until I get the least run-out possible. Then I will put either a steady rest on the exposed arbor or use the tail-stock with an inverted cone (with a hole in the center where the pivot can rotate without touching the cone) to hold the arbor as centered and as steady as possible. Remember to put a small drop of lubricant on the jaws of the steady rest or the inside of the inverted cone.
If I find that I can not hold the arbor in a collet because the exposed end is not steady (caused by an arbor that was bent when it was made, too short of an arbor, or a tapered arbor that can not be readily chucked up), or I have noted an offset pivot on one or both ends of the arbor, I will turn the escape wheel between centers with a drive-dog on the escape-wheel arbor. I will have to write an article on turning between centers one of these days.
Do not chuck up on a pivot to spin the escape wheel. This will only give you a chance to practice repivoting.
My typical set-up is shown in Figure 1.
Read MoreThe even ticking of a clock. Quietly telling you of how great a job you did restoring the mechanism - a joy to hear. But, what if the beat is not even? Or, more aggravating, what if it is mostly even, but sometimes seems to shift a bit - to have an intermittent tick and a tock that are just not right. Welcome to the world of escape wheels - and, more specifically, the world of long and short escape-wheel teeth.
Let=s start out with this underlying assumption behind us: You have already done all you can to get the clock in beat - and, are satisfied that all is as right as you can make it. If the beat seems to periodically change you have already tried letting the mechanism spin very slowly without the anchor in place - looking for bent arbors or pivots, or fouled teeth.
If your mechanism
....Seems to have an even beat, except for one or two beats each rotation of the escape wheel, or,
....You have a clock that does pretty well for a while, but it seems that, just before it stops, the beat becomes erratic,
I am hopeful that this Technical Tid-Bit will help you understand what it is that is frustrating you, and also perhaps help you understand what you would do to Amake it all better@!
When a pendulum is swinging with a lot of over-swing (more swing than necessary to just release the pallets), the beat tends to be more even than when the pendulum is swinging just far enough to release the pallets. I have had a number of chances to work on escape-wheel teeth because of the number of long duration Viennese clocks I have had the chance to work on. In general I find that longer duration clocks operate a lot closer to the minimum weight required to swing the pendulum just far enough to release the pallets.
Why are long duration clocks more prone to having problems when escape-wheel teeth have been damaged? Let=s think through how much weight is typically used to run long duration clocks. It is not uncommon to find a 3 pound weight driving the trains on a week-duration Vienna Regulator. OK, now, assuming a years duration clock is as efficient as the week duration, it is only logical to say the year runner should have a weight that is 52 times as heavy as the week runner. But, well, 52 times 3 is 156 pounds. Yes, I have seen British years runners with weights in the 40 pound range. But, a Viennese years runner, with its exceptional pivot, gear and escapement work, typically requires about 16 pounds. In fact, I am currently finishing up work on a clock that runs 16 months on 16 pounds, and I have a 3 month runner in my collection that runs on 3 pounds. Almost sounds like a pound is the minimum for each months duration to me.
Running on such little weight typically results in a very narrow pendulum swing - people often ask if my long duration clocks are even running!
This is why, when I am setting up a clock, be it an Anniversary clock, or a conventional pendulum clock, I always try to set the pendulum swinging (or rotating for an anniversary or other torsional clock) just enough to barely Atick over@. This amplifies the impact of any Aout of beat@ issues.
But what about when you have your mechanism in beat, to the best of your ability, and it is still sometimes wanders off for a beat or two? Usually when I get a call or an e-mail about this I find that the repair person has already tried to rectify the problem - usually by Astraightening@ the Aobviously bent@ escape-wheel teeth. And, while the clock may have run before, well, now it doesn=t seem to want to any more.
In general (yes, I realize that a statement like this just begs for contradiction) bent escape-wheel teeth will result in an escapement hanging up - a pallet fouling or Arunning into@ a tooth when it is supposed to just clear it. Yes, a bent tooth does end up just a bit shorter than the other teeth, but this is a very small change in length (has to do with the cosine and the tangent of the angle by which the tooth is bent from its correct position - and, since I do not want to remind any of you how much fun geometry was, let=s just say it is a very small change in length). But, in general, a bent tooth does not result in an uneven beat. That takes a tooth that is either too long or too short. Funny thing is, often when a repair person tries to straighten a tooth they end up making it longer, and now it is really out of beat. Or they break off the end of the tooth, but we don=t really want to go there.
A clock that wanders out of beat has a tooth or two (or three or four, well you get the idea) that is not real close to the length of the other teeth. Let=s think about this for a second to see why this should cause an uneven beat. As the pendulum swings it moves a pallet out of the way of an escape-wheel tooth. But, let=s say a tooth is shorter than most - the pallet will Arelease@ the tooth sooner in its movement than with the rest of the teeth. Because the pallet releases an escape-wheel tooth sooner in the pendulums swing the tick is shorter. Likewise, it the tooth is longer, the pendulum has to swing farther, and the tock is longer.
I think now is the time for a word or two of caution. The subject of this Tech Tid Bit, ATipping Teeth@ is something that requires a lot of care and a lathe with a cross slide. Yes, I have talked to people who have successfully tipped escape wheels with a power drill and a file. But I have also had calls asking if I had a spare escape wheel for a Gustav Becker anniversary clock, amongst others.
Why is it important that the beat be even? In many clocks it isn=t. Especially if they have a weight big enough to provide a lot of over-swing. Then it just doesn=t matter. But, as we take things to the extreme (think long duration which usually translates to minimal over-swing) an overly long escape-wheel tooth will shut it down. And a short one will not allow the mechanism to run as well as it could, and to stop sooner than it might otherwise.
I decided to write this article when I was faced with tipping an escape-wheel from a truly magnificent mechanism. As you can see in Figure 8, the workmanship (not sure how to make that term asexual) required to make this escape-wheel is phenomenal. Think I wanted to tip the teeth? Think again. But, the mechanism did not want to run. Yes, I probably could have put a bigger weight on the mechanism and overcome the problem. But, being a 6 month runner, and being convinced I could tip the teeth without destroying the wheel, I decided to go forward.
Now for another word of caution. I use a cross slide to hold a cutting tool to cut the tips of the teeth. If you have every used a cross slide you know that there are knobs (or levers if you have a lever-actuated cross slide) you twist to move the cutting tool back and forth and closer and farther from what you are cutting. I make myself practice with both knobs, making sure that I know which direction to twist the knob to move the tool in and out, and back and forth. If you inadvertently go the wrong way and cut too quickly I pretty much assure you that you will bend teeth and potentially trash your escape wheel.
I don=t recommend tipping teeth with a lever-actuated cross slide. I really like them for some operations, but not one as delicate as tipping teeth. And, I don’t recommend setting up to tip teeth with a stone – a properly-sharpened graver will cut extremely fine bits of brass off of the teeth, and will only be cutting right at the tip – if a stone is used it will be cutting across the entire tip of the tooth, significantly increasing the risk of bending teeth.
The first step is to mount the escape wheel in your lathe. Typically I find that holding the escape-wheel arbor in a collet is accurate enough. That is assuming the run out on the exposed end of the arbor (the amount that the exposed pivot moves as the gear is spun in the lathe) is minimal. Typically I chuck up the gear in the lathe, check it for run-out, if not acceptable I then loosen the collet, rotate the arbor in the collet a quarter turn, re-tighten, check run-out, and repeat until I get the least run-out possible. Then I will put either a steady rest on the exposed arbor or use the tail-stock with an inverted cone (with a hole in the center where the pivot can rotate without touching the cone) to hold the arbor as centered and as steady as possible. Remember to put a small drop of lubricant on the jaws of the steady rest or the inside of the inverted cone.
If I find that I can not hold the arbor in a collet because the exposed end is not steady (caused by an arbor that was bent when it was made, too short of an arbor, or a tapered arbor that can not be readily chucked up), or I have noted an offset pivot on one or both ends of the arbor, I will turn the escape wheel between centers with a drive-dog on the escape-wheel arbor. I will have to write an article on turning between centers one of these days.
Do not chuck up on a pivot to spin the escape wheel. This will only give you a chance to practice repivoting.
My typical set-up is shown in Figure 1.
