PDA

View Full Version : Verge adjustment question



DC Kelley
12-03-2007, 11:23 PM
When replacing the verge on an American clock what logic/process should I use to determine where to bend the crutch wire to fit to the pendulum? I.e. how far down is best?

If the old verge is there there I have always tried to follow that as a lead, but "higher" on the pendulum seems to give a wider swing in some cases. But this also requires more work to be done (hence lowering the Q of the pendulum) so I suspect it will cost me in time keeping accuracy. Lower, taken to extremes, gives poor lock and produces a weak swing that is more susceptible to being stopping when the load on the train changes (like when lifting a 1/2 hour strike).

I don't know enough to follow the lifting angles and faces of the pallets to know where the "optimum" point for this should be in this, so some good practical advice is sought.

harold bain
12-03-2007, 11:41 PM
DC, I don't think there really is any hard and fast rule to follow. I try to go with what the manufacturer used originally, if you still have the original, or if you don't, try to find a similar clock for guidance.

Scottie-TX
12-04-2007, 12:39 AM
higher" on the pendulum seems to give a wider swing in some cases. But this also requires more work to be done (hence lowering the Q of the pendulum) so I suspect it will cost me in time keeping accuracy. Lower, taken to extremes, gives poor lock and produces a weak swing that is more susceptible to being stopping when the load on the train changes (like when lifting a 1/2 hour strike).
I'm just takin' notes.
I never knew alla that about crutches.

Ralph
12-04-2007, 11:06 AM
Saunier discusses crutch length in his Treatise. In general, the crutch is 1/3 to 1/4 the length of the pendulum, except in the seconds pendulum where it might be 1/5th the length.

There are a number of considerations, he discusses, but, generally, it comes down to the ROT.

Ralph

eskmill
12-04-2007, 11:48 AM
I think DC Kelley's concern of how the ratio of the pendulum length to crutch length affects "Q" is interesting.

Although Saunier suggests a rule-of-thumb for an optimal crutch length, I think Kelly might find the writing of Rawlings and more recently, what Woodward have written on pendulum oscillators and escapements. Then, I believe he will find that the simple American made Connecticut clock with it's recoil escapement is a highly complex machine. :o

LaBounty
12-04-2007, 01:20 PM
John Jarvis has a series of articles in BHI's "Horological Journal" titled "The role of a Crutch in Pendulum Clocks". The series starts in the October, 2007 issue and concludes in the December, 2007 issue. I'm still studying it :).

Hope that helps!

shutterbug
12-04-2007, 10:29 PM
Most of the crutches I've seen are terminated before exceeding the length of the movement. I have a post here about a long case Japanese clock that I had to shorten the crutch by 3 inches to get enough swing to unlock reliably. I also discovered that the location of the verge is important and that more lock is not a good thing. This was a recoil type and is finally running reliably. If I were you, I'd try for a longer length that can be shortened without replacing it. Then experiment to see how it runs. Pay close attention to overswing.

Scottie-TX
12-05-2007, 03:59 AM
BUGGMAN, that does seem to be true of spring powered movements, but VERY unlike weight powered movements that ALWAYS have a crutch that extends an inch or so below plates REGARDLESS of pendulum length.
Reckon what that's all about, LAB? Why spring powered crutches would vary so drastically in length compared to weight driven ones?
I'd LOVE to read that treatise LAB, but in lieu of that perhaps you could sort of recap - give us a thumbnail version of JARVIS' thoughts on crutches.

R. Croswell
12-05-2007, 10:19 AM
I to would love to read that treatise about the crutch and pendulum. I think the relationship depends to a large extent on the position of the anchor (verge) pivot point relative to the suspension point of the pendulum. If the anchor pivot point and the pendulum suspension point are in the same plane, as is sometimes found in better weight powered clocks, the crutch and pendulum would seem to trace an identical arc, so the length of the crutch would appear to be of little importance. In many American shelf clocks the pendulum is suspended below the hour hand tube while the escape wheel and anchor are near the top of the clock. I would guess that the length of the crutch would be much more important in this case.

That is, when the anchor pivot and suspension point are not the same, then as the position of the crutch loop approaches the pendulum suspension point, the motion of the crutch approaches zero, but the motion of the free swinging pendulum is the same. When the anchor point and the suspension point are the same (in the same plane) the crutch will always move the same amount as the pendulum at the point of the loop.

Now I have no idea how to determine the optimum length of the crutch other than trial and error when the anchor point and suspension point are not in the same plane. Maybe someone can add to this.

Bob C

LaBounty
12-05-2007, 10:46 AM
Hey Scottie-

As Ralph suggests from Saunier, the crutch length is proportionate to the pendulum length. Spring driven movements will generally have a short pendulum and short crutch while weight driven movements will generally have a longer pendulum and longer crutch. There are certainly exceptions, like the spring-driven long pendulum mvt. in the ST "World" or the electrically wound/weight-driven mantel clock from the American Clock Co. but I believe the proportions hold.

And I'd be happy to give a synopsis of Jarvis' article. So far, just having skimmed it, I believe he focuses more on the effect of the mass of the crutch rather than the length. But I have a ways to go in digesting the material.

Regards,

eskmill
12-05-2007, 12:39 PM
Croswell is correct in his statement: "the relationship depends to a large extent on the position of the anchor (verge) pivot point relative to the suspension point of the pendulum. If the anchor pivot point and the pendulum suspension point are in the same plane, as is sometimes found in better weight powered clocks, the crutch and pendulum would seem to trace an identical arc, so the length of the crutch would appear to be of little importance. "

But the "suspension point" of the pendulum being aligned with the verge pivot holds only for knife-edge or Cuckoo suspensions. :o

For more conventional pendulum suspensions, the "center of flexiture" of the suspension spring should be in line with the verge pivot so that the pendulum and crutch move in the same arc.

Not every clock has a short stubby suspension spring. Those designed along the lines of the traditional English pattern often have very long suspension springs. Determining the "center of flexiture" of these long suspensions would depend on the spring material and the geometry of the pendulum arc; a puzzle for the math professor.

I think it could be safely stated that the middle of the pendulum suspension device is usually positioned to be aligned with the verge pivot. This to assure that the pendulum and its driving crutch move in the same arc so as to minimize friction between the crutch loop and the pendulum rod.

I note that many sketches of clock pendulum escapements appear to assume that the suspension spring will bend at the point where the spring is clamped.

More: Take a close look at the photo in another post asking opinion on the suspension spring for a Pin-Wheel movement. It appears to me that the overall design of the pendulum suspension is flawed.

JMO Les

LaBounty
12-06-2007, 12:41 PM
A short synopsis of John Jarvis' "The role of a Crutch in Pendulum Clocks"...

Mr. Jarvis uses a constant crutch length of roughly 1/5 the length of the pendulum (PL = 54 cm, CL = 10 cm) but varies the mass of the crutch and notes the effect on the period of the pendulum. His conclusion is that, as the mass of the crutch is increased, the period of the pendulum is decreased.

In his mathematical model, he treats the crutch system like a pendulum, accounts for air resistance, includes the moment of inertia for each gear in the train, and takes into account the forces acting on a deadbeat escapement.

He references a prior article by Richard Stephen "The Effect of the Crutch on the Dynamics of the Pendulum", HJ, 144, pp. 201 - 203 and HJ, 144, pp. 255 - 257. My old periodicals are still packed so, if someone else has the mentioned article, it would be nice to know if Mr. Stephen varies the crutch length.

Hope that helps!

Ansomnia
12-06-2007, 05:24 PM
I would like to put forward another view on this topic.

I believe the workings of the clock movements we are discussing are all roughly based on theory of the "simple pendulum".

The simple pendulum theory assumes the connection between the center of gravity of the mass at the end of the pendulum (the bob) and its point of attachment (verge) to be rigid and of zero mass.

In our discussion, one exception to this simple pendulum setup may be the instance of shelf clocks mentioned by Bob Croswell where the base of swing of the pendulum (i.e. suspension) and the centre of swing of the verge (anchor pivot) are quite different.

So it would seem to me that in most cases, the crutch is really just designed to compensate for the non-rigid nature of the suspension. The suspension itself is a modification to the natural period of the pendulum in use in order for it to be "in beat" to keep time according to our units of time measurement.

If I am correct then the length of the crutch should be no more than necessary to work with the design of the suspension - meaning its length depends on the length of the suspension. To reduce the effect of mass and air resistance, the crutch should be as light and as short as possible. The crutch is simply designed to make the pendulum assembly rigid.

Having said this, I would also say that you should stay with the length of crutch already used by your clock because it was designed to work with the mass of the bob and the type of suspension the clock originally came with. One should not change the design of an existing crutch unless it is not working properly (for instance, if someone replaced the bob with the wrong one).

Other than this, as inferred by LAB, if you use a heavier crutch you will effectively move the centre of gravity of the pendulum up its length and reduce the period of the pendulum… and vice versa.


Michael
added reference to LAB's comments

DC Kelley
12-07-2007, 01:32 AM
The critical issue as I see it is understanding that the pivot point of the verge and the effective pivot point of the pendulum ARE NOT in the same plane (x or y) and this creates a ratio of some sort where the "force-length" of the verge pressing on the pendulum (ignoring mass and friction for a moment) do not induce the same movement in the pendulum.

I think R Croswel is on the right track that it relates to the the plane they are in. In the case of this clock, the verge point is below and to the side, creating some mathematical relationship that is beyond me to express at present. If they were in the same line/plane (as in either above or below in a line), then you probably just have some multiplier effect to deal with.

I may look dumb by saying this but here goes... it seems to me that if the angle of movement is constant (say 2 deg for each side off the pallet faces, a one degrees for lock or 5 degs of rotation or so). This motion is translated to the verge by the ratio of the distance of the pallets to the pivot and the distance from the pivot to the bend in the crutch, usually a gain. If the distance is too short the force on the pendulum will be excessive (you would see some over swing? because it is pushing "too hard" for "too short" a distance). If the distance is too far the force on the pendulum will inadequate to cover the full range. Perhaps it wont have energy to clear the next pallet? This is also a rule of parallel axis theorem of forces going on here when the verge point is offset from the pendulum point. All of this should in more or less independent of the free oscillation period of the pendulum, and that is where is gets confusing in my mind. I cant make it all work, thankfully most clocks seem tolerant of a range of values.

Many references I need to look up next...
Ralph mentions
...Saunier discusses crutch length in his Treatise

Eckmill mentions
...Rawlings and more recently, what Woodward have written on pendulum oscillators

LaBounty metions
...John Jarvis has a series of articles in BHI's "Horological Journal"

Laurie Penman (who I find to be full of practical advice on such topics) skips over bent strip verges as "cheap" and never mentions to verge length issue.

Every time I ask a question like this I end up with loads more insight and places to learn that I ever knew existed, many thanks to all for my continuing education.

And by the way this all came up because the clock had an obviously poorly replaced verge on it, so I did not (do not) trust the setting I originally found on it.

Scottie-TX
12-07-2007, 02:24 AM
When replacing the verge on an American clock what logic/process should I use to determine where to bend the crutch wire to fit to the pendulum? I.e. how far down is best?
Ah; I dunno.
Yeah, I too enjoy the minutae - the esoteria of these finer points of horology.
Thanks LAB for th' recap. Yeah. Understand. Heavier crutch - bigger load - a - "DUH"?
Don't get me wrong. I'm guilty. Some of the topics I've started attest to that - that I sometimes belabor a point that is really not all that important or decisive. . . . . . . ... and I'm not saying this is one of those but:
(. . . . . . . and we all know the meaning of "but")
BUT: If my store regulator INGERHAM is missin' it's crutch - you did seek a logical process to determine the length - if my INGERHAM was missing the crutch, I'd post a pixture of it here and ask about the length it should be. BONG would have one and tell me it should be four inches long.
WORST case scenario: ( very rare here) "We've never seen one!"
I GUESS and bend it three and a half or four anda half.
Now HOW much difference ya reckon a half inch error's gonna make on this el-cheapo INGERHAM recoil regallator?

harold bain
12-07-2007, 10:30 AM
It relates to levers and fulcrums to some extent as well. The further down the pendulum the crutch is, the more power it takes to move the pendulum, which can be critical with a heavy pendulum

R. Croswell
12-07-2007, 12:20 PM
Scottie,
While we are into “the minutae - the esoteria of these finer points of horology”, I’ll throw another largely irrelevant (from the practical standpoint) variable into the equation. As we know, there must be a small gap between the crutch loop and the pendulum leader, As the loop is moved closer to the pendulum suspension point (crutch shortened), the distance traversed by the loop becomes less and less, but the gap between the loop and the leader remains the same. So the gap becomes a slightly greater percentage of the total distance traversed as the crutch is shortened. I seriously doubt that this would have any observable effect on the clocks operation unless the gap was excessive to be begin with.

Seeing that your old Ingraham is missing its crutch, it might be interesting if you could set it up with a very long leader, and make a crutch from several telescoping brass tubes (from the hobby shop) such that you could lengthen and shorten the crutch at will without changing the crutch mass (weight). That clock should swing slow enough that air resistance should be negligible on the crutch. Would be interesting to see how it affects the running of the clock. Just a thought if you can find an old verge and crutch to sacrifice, and enough old time to do it.

Bob C.

Ansomnia
12-07-2007, 12:48 PM
It relates to levers and fulcrums to some extent as well. The further down the pendulum the crutch is, the more power it takes to move the pendulum, which can be critical with a heavy pendulum

Harold, I think the crutch should be considered an extension of the pendulum as far as impulse from the escapement is concerned. So the crutch length would only be a factor as added weight and not so much as leverage since that is determined by the centre of gravity of the pendulum which should weigh much more than the crutch.


Michael