Why is it? Poising Question

[D N Ravenna]

I did a search through the archived files, but could not find an answer. So here is my question.

Why does a balance wheel need re-poising? Note that I say "re". In the PW's creation, the balance wheel had to have been poised prior to its being sold to the first customer.

So why is that some 90 years later, when I place the balance wheel on the poisting tool, I need to add or remove weight, and sometimes by a lot!

I can understand out of roundness (perhaps creep of the metal over time or abuse), but in adjusted and unadjusted watches, I have been surprised by the fact that they are out-of-balance.

Anybody have any clues/insights on this?

Thanks in advance,

Dan Ravenna

[D N Ravenna]

I did a search through the archived files, but could not find an answer. So here is my question.

Why does a balance wheel need re-poising? Note that I say "re". In the PW's creation, the balance wheel had to have been poised prior to its being sold to the first customer.

So why is that some 90 years later, when I place the balance wheel on the poisting tool, I need to add or remove weight, and sometimes by a lot!

I can understand out of roundness (perhaps creep of the metal over time or abuse), but in adjusted and unadjusted watches, I have been surprised by the fact that they are out-of-balance.

Anybody have any clues/insights on this?

Thanks in advance,

Dan Ravenna

[GandalfPC]

You shouldn't have to add or remove weight normally - it should only be after you have confirmed that it is absolutly required.

First use a pair of balance calipers to confirm that it is true in the round and flat - and to confirm that the staff isn't bent. Rough handling may have bent it so this is a fairly normal repair.

If it is still out of poise, check to see that the meantime screws are both adjusted by the same amount - if one is screwed in more than the other they will throw it out of balance.

If still out, check to see if any balance screw washers have been added by a previous watchmaker.

Finally it is time to add/remove weight - I have not needed to do this yet - be sure not to change the overall weight though, add to one side, remove from the other.

A balance will only need to be adjusted by adding/removing weight if (1) it was made incorrectly, or (2) it was adjusted by a previous watchmaker incorrectly.

Also, such tasks as replacing a staff or a roller jewel might take it out of poise I imagine...

As far as I know time alone will not cause it to go out of balance. (I could be wrong...)

Jason Goldman
Apprentice Watchmaker

[This message was edited by GandalfPC on April 06, 2003 at 22:13.]

[D N Ravenna]

Thanks for the response. I was just wondering out loud.

Most of the times, I imagine the out of poise problem is due either to re-staffing or putting in a new jewel. But the world is nuts I tell you. I actually have a couple PWs where the weights do not match up (like a larger one on one side and a smaller one on the other).

That is what led me to wonder, why?

:confused:

Dan

[GandalfPC]

Probably someone trying to adjust the timing added more weight by replacing a screw with a larger one, without realizing it had to be in balance... (or too many cocktails at lunch :smile:)

Jason Goldman
Apprentice Watchmaker

[Steve Maddox]

For a number of years, I've made a particular study of the subject of positional adjusting, static poising, dynamic poising, etc., and while the opinions on the subject vary, it's generally considered that the more perfectly poised a balance wheel is, the better will be the resulting rates in all positions.

If a balance is intentionally thrown out of poise to correct an error in one particular position (or more specifically, in two opposite vertical positions), the the result will be that what reduced the errors when the balance amplitude is high, will increase them when the balance amplitude is low. The "break even" point is 1.25 turns of total balance motion (225 degrees in each direction from the natural position of rest).

Let's say that a given watch, when fully wound, presents a good rate (+/- 0 sec/day) in the stem-up position, but gains 10 seconds per day when stem-left, and loses 10 seconds per day when stem-right. In such a case, it would be possible to alter the poise of the balance so that the 10 second gain and loss could be brought to 0 (or very near it), so long as the balance motion remained greater than 225 degrees.

Unfortunately, the reason that isn't necessarily a good idea is that a poise alteration which decreases positional errors when the balance motion is above 225 degrees, will have exactly the opposite effect when the motion falls below that. The original errors that had been +10 and -10 seconds, would increase to +20 and -20 seconds! In short, for best positional rates under all possible conditions, the balance wheel should be as perfectly poised as possible.

Now, most particularly in watches having split compensation balance wheels, more esoteric problems arise. For one thing, no matter how carefully a split compensation balance wheel is made, the thickness of the brass and steel will never be perfectly uniform all the way around the wheel. In other words, one side or the other will have a little more brass and a little less steel, and the result will be that those sides will react differently to changes of temperature. One side will always move in or out just a bit more than the other, with the result being that the disruption of balance poise will be significant. That's why in adjusting and observatory rating, watches were never tested simultaneously for positional error and temperature compensation.

An additional problem with split compensation balances is that the two halves of the rim always display slightly different centrifugal characteristics. In other words, when the balance is in rotating, the weaker side of the rim (and there always is one) will be pulled out a little more than the stiffer side, with again the result that the poise is effectively altered. The realization of this resulted in the development of "dynamic poising," which essentially means that the wheel is poised while it's spinning, rather than when it's merely resting on the jaws of a poising tool. While dynamic poising is an interesting subject, it's actual benefits are somewhat dubious, and most ordinary watchmakers (including this one) don't have the necessary tools to accomplish it.

In any event, as to the original question, there are several reasons that balance wheels today often require poising. One of the main reasons is rough handling, which results in bending of the balance rim (even an imperceptible amount will cause significant poise errors). Other reasons can include loose balance screws, badly adjusted meantime screws, bent balance pivots, or the replacement of a rivited staff.

The reason the balance poise is usually disturbed when new staffs are riveted in place, is that regardless of how careful a watchmaker is, the riveted portion of staff seldom expands evenly in all directions. Obviously, if the rivet expands more on one side than the other (due to inconsistency of the metal, slight irregularities in the stake, etc.), the result will be that the center of the new staff will be slightly eccentric with the rim of the wheel. Also, if the rivet is expanded too much, it can actually stretch the hole in the balance. When that happens, the metal on one side will always shift a little more than on the other, with the result being that the spoke arm bends (like a bananna) toward one side.

It's useless trying to poise a balance with bent pivots. Before any poising is attempted, it's essential that the pivots be perfectly straight, and that the balance rim be as round, level, and well-centered as possible. Obviously, blalance calipers, poising tools, balance wrenches, pivot straighteners, etc., are all tremendously helpful, and the more tools a watchmaker has (and knows how to use), the better the work will be that's likely to result.

*Again, for the reasons described above, it's not at all unusual for split compensation balance wheels to show slight poise errors as a result of changes in ambient temperature. If a split compensation balance wheel is as perfectly poised as possible at 70 degrees F, it would not be at all unusual for it to show a moderate poise error when the ambient temperature is increased to 75 or 80 degrees, or decreased to 65 degrees.

Hope this helps!

==========================

Steve Maddox
President, NAWCC Chapter #62
North Little Rock, Arkansas

[D N Ravenna]

Ah, now that is a nice treatise Steve, thanks! It certainly puts things in perspective when all the balance screws seem to be equal.

BTW, do you know if the Hamilton 950 used a friction fit staff or one requiring riviting?

Thanks again!

Dan Ravenna

[Steve Maddox]

I thought we could use some better information here about balances, so perhaps the above will suffice.

The balance staffs in Hamilton 950 models are the conventionally riveted type; those in 950 Elinvar (950E) and 950B models are friction fitted.

======================

Steve Maddox
President, NAWCC Chapter #62
North Little Rock, Arkansas

[GandalfPC]

Steve - which method of straightening pivots works best? Does the Seitz Pivot Straightener (Frei and Borel #FB-743) do a good job of it, or is it best done on a lathe or jacot?

Just a side note about timing in positions...

I picked up a small booklet - "Timing Manipulations" by James L. Hamilton - in which he discusses various manipulations of the hairspring, collet, and regulator pins to correct for positional error (amongst other things).

I can't vouch for it's correctness as I haven't had the pleasure of testing his theories yet, but he certainly seems to know what he's talking about...

Any of our more learned members out there know the validity of this book?

Jason Goldman
Apprentice Watchmaker

[Greg Crockett]

One suggestion which is sometimes useful is to push the staff in and check the poise before riveting. Move the staff around until the poise is good. Check again after riveting.

[Steve Maddox]

The Seitz "pivot straightener" is a wonderful little tool, although not necessarily for its stated purpose. I use mine for testing balance pivots, by inserting a staff pivot into a correctly sized hole, then tilting the table a little (about 30 degrees), and using a dust blower to gently spin the wheel. Obviously, the rim shouldn't wobble while spinning, and if it does, it's a clear indication that the staff is bent.

Now, here's where the problem comes in: It's virtually impossible to make a bent staff perfectly straight again (if you don't believe it, just read Ecclesiastes 1:15). If one is really serious about positional adjusting, the only correct solution for a bent staff, is to replace it.

In instances where absolute accuracy might not be essential, and/or where a replacement staff would be difficult or impossible to obtain, the relative straightness of a bent staff can usually be improved, but it's a risky proposition. Most staffs are hardened such that they'll bend a little, but when one tries to straighten them back, they'll usually break. The way to avoid that is to anneal the staff before attempting to straighten it, and then harden it again when the job is finished. With a staff still attached to a balance wheel, that's not easy to do, and most people who anneal a staff to straighten it, usually just leave it that way when they're finished.

A staff (or anything else made of metal) is less likely to break if it can be bent while it's hot. Therein, is one of the shortcomings of the Seitz "pivot straightener." It's impossible to use one of those, and keep the staff hot at the same time. Personally, I have a custom-made "pivot straightener attachment," which fits into an electric soldering iron. When I want to straighten something that's bent, I insert the tool in the soldering iron, and let it heat up for a while. The tip is made of brass, and it has a number of small holes in it, not unlike the Seitz tool, but the soldering iron heats the apparatus so that the item being straightened can be bent while hot. I can't recall a case where this tool has failed me, but it's a lot of trouble to use, and one has to be careful not to burn himself with it!

I also have a wonderful little tool called a "Pivot Master," which is essentially an electric Jacot tool. That thing works great for finishing pivots, but it's a relatively rare machine, and despite the indications on the instructions that come with it, it really won't allow one to significantly straighten a severely bent pivot without breaking it. The same could be said for a standard Jacot tools, balloon chucks for lathes, etc.

As for Mr. Hamilton's book, "Timing Manipulations," it's very good. My copy originally came from the library of the Hamilton Watch Company in Lancaster. All of the information in it is accurate (at least, all I've had occasion to reference), but several of the manipulations are not possible in all watches. For example, I do a lot of Rolex work, and since most Rolex watches are "free-sprung, needless to say, it isn't possible to make regulator pin manipulations to them!

Again, I hope this helps!

======================

Steve Maddox
President, NAWCC Chapter #62
North Little Rock, Arkansas

[GandalfPC]

Could you post some pics of your custom pivot straightener?

Jason Goldman
Apprentice Watchmaker

[Steve Maddox]

Jason,

I'd be more than happy to do that, but I just have a plain flatbed scanner, not a digital camera, and an image of that tool made on a scanner wouldn't be very good. In this case, a brief description should work just as well.

I have a "pencil-type" soldering iron, which uses a tip that fits into a 1/4" hole in the iron, and is held in place with a small set screw. Using an engine lathe, I simply machined (from brass) a little "table" with a short 1/4" shank, and a top that's about the size of a nickel. This top has holes of varying diameters drilled through it, which makes it essentially like the Seitz "pivot straightener," except that it heats up to about 600 degrees F when the soldering iron is plugged in.

The only problem is drilling the tiny holes, but I have some really tiny drills. Using the combination of those and some small pivot broaches, I've made enough holes of varying diameters that I can usually find one that'll work for whatever I'm trying to straighten.

Whenever I come across something I don't have a hole to fit, I just add another hole. You'd be surprised how many tiny holes it's possible to drill through a piece of metal the size of a nickel, and every time I add a new one, the tool just becomes a bit more "versatile" than it was before!

Hope this helps!

---------------------

Steve Maddox
President, NAWCC Chapter #62
North Little Rock, Arkansas

[GandalfPC]

Ok, I get the picture of the table with the holes - what is the method for straightening once you have inserted the bent pivot into the correct hole? Is it just a matter of a gentle bend with tweezers by eye?

Jason Goldman
Apprentice Watchmaker

[Steve Maddox]

Jason,

I use my "pivot straightener" in the same way I use the Seitz tool to test for pivot truth, except that my tool also heats the pivot so it's not as likely to break when straightened.

For "manipulating" the wheel to straighten the pivot, I use a pivot broach (or any similar light instrument). I simply push down on whatever part of the rim is high, thereby straightening the pivot and allowing the wheel to spin without wobbling.

=====================

Steve Maddox
President, NAWCC Chapter #62
North Little Rock, Arkansas

[GandalfPC]

Gotcha :smile: Thanks :smile:

Jason Goldman
Apprentice Watchmaker