Jacot tool and pivot file burnisher question

[JohnC]

On 10/7/2020 at 7:41 AM, nickelsilver said:

When I make a balance staff, which is about once a week, I rough turn the pivot leaving about 0.02mm on its diameter and then count on the Jacot to give me a beautiful cylindrical pivot coming off of the conical portion. The whole point of the tool is to get nice cylindrical pivots- with the exception of the 4th wheel runner, which will purposely produce tapered pivots. If a Jacot tool is making tapered pivots then it is out of alignment*, either from being dropped, or having its runners mixed up with another tool. Older tools aren't adjustable, though some have a "headstock" end that can be raised and lowered, but the more modern Steiner/Horia (Hahn) tools provide means to adjust the height and front-back alignment.

 

* in some cases when doing conical pivots, if the burnisher has too large of a radius it will make horrible unusable "tapered" looking pivots. The Bergeon tool above comes standard with a radius that corresponds well to most wristwatch movements. If buying a secondhand Bergeon carbide burnisher beware that the radius on some older ones is larger and really for pocketwatches. The very old Bergeons have a metal cap for the end, the middle period a purplish bakelite cap, and new ones an anodized purplish aluminum cap.

@nickelsilver almost 2 years later, I have a follow-up question: what is correct alignment between female centre and tailstock runner bed? If they are aligned like I would expect them to be, then I think a slight taper would be almost inevitable. If the spindle centre is aligned with the middle of a groove (halfway between top and bottom), then if the pivot protrudes above the level of the bed, as it must to file or burnish, then it seems to me the necessary result with a horizontal burnisher is a tapered pivot, since at the "far" end the pivot is lifting off the bottom of the bed.

Indeed, this is what I am getting. Although the degree of taper with short pivots is pretty negligible, I find that if I try to burnish a centre wheel, the taper is quite noticeable, and the only way to avoid it is to compensate by rocking the graver.

I know that tools with adjustable head/tailstocks exist, although they are not common. Is the purpose of this adjustment to avoid the problem I have just described? Or should I be able to avoid it with the classic, non-adjustable tool? I am fully prepared to admit that either my tool is defective or I am doing it wrong.

Thanks in advance. 

 

[nickelsilver]

The way the tool is made the headstock runner is aligned to the center of the opposing notch, yes. Horia (Steiner), usually, marks the notches so that a pivot made in that notch will match the nominal same jewel, i.e. have just enough freedom: a 10 notch will produce a pivot a bit less than 10, 0.950-0.975 or so. The idea is you continue to burnish until it won't anymore, so now the pivot is level with the top of the runner. At any rate just a few microns difference even if you don't go that far. So you're right, that anywhere outside of that will be off, and should make a taper, but in practice the taper is so slight it's unmeasurable with any normal equipment. A few microns height difference over say 3mm, measured over a length of perhaps 0.3mm, is invisible. If you are getting visible taper there is an issue with your tool.

 

It is definitely more pronounced on center wheels, true. Especially as the sizes between notches tend to skip by 3 or 4 hundredths. For larger work like that it's really nice to have one of the supplemental tools that works with the jacot tool using a carbide wheel, or a Pivofix or similar. All very expensive of course- but nice.

 

Edit- just sketched it up, with a 0.01mm height difference from center, the total taper over 0.3mm is one micron. But- if using the tool "properly" where the burnisher sort of bottoms-out on the top of the runner, even that gets evened out.

Edited May 10, 2022 by nickelsilver

[JohnC]

@nickelsilverthanks very much, that is very informative. Yes, I think I may just need to be more aggressive and really bottom out the burnisher. But I think it's possible the tailstock runner is not sitting entirely level.

So the tools with adjustable tailstock. This is to correct this issue?

Is it ever appropriate to rock the burnisher inward to compensate? Particularly on centre wheels I sort of feel like I have to do this.

[nickelsilver]

2 hours ago, JohnC said:

@nickelsilverthanks very much, that is very informative. Yes, I think I may just need to be more aggressive and really bottom out the burnisher. But I think it's possible the tailstock runner is not sitting entirely level.

So the tools with adjustable tailstock. This is to correct this issue?

Is it ever appropriate to rock the burnisher inward to compensate? Particularly on centre wheels I sort of feel like I have to do this.

The adjustable ones I have and have seen have had the headstock adjustable; it is to correct the alignment, though it seems some might use it for making tapered pivots on purpose. But that seems iffy in my view; the notch in the runner and runner top are parallel, so it would be a fight. The runner that most tools have for 4th wheel pivots has the notch tapered, as those pivots are tapered on the end. That makes sense.

 

Depending on what tool you have you don't have to have an adjustable headstock. The Horia/Steiner tools, later design, are adjustable. If you have one of those I'll post how to do it. All this is also a reason these tools are numbered on the runners to match the frame, they were ground and adjusted as a set in the factory.

 

One thing I've frequently seen is the simple runner for the headstock can be quite bent. If you have a lathe that takes collets and can get it in there you can bump it back true.

 

For center wheels, if you find you're getting a noticeable taper (and you know the tool is good), it's no crime to tilt the burnisher. I have an industrial pivot  burnishing machine and a Rollimat (meant for clocks) that I'll turn to personally, and if the (center) pivot is very worn I might touch it up in the lathe with the cross slide to get it clean and straight before finishing.

Edited May 10, 2022 by nickelsilver

[JohnC]

@nickelsilver photos attached. I think I may actually be in luck here?

From what I can see the headstock spindles are straight and true. I hope it's just a question of headstock adjustment.

[nickelsilver]

Ah excellent, that's a nice Steiner there. The little jack screws you see set the height, but different than first impression. By tightening them down, they force the V block in the head upwards, raising the spindle. To lower, you loosen, then tighten down tight with the normal knurled screw on the runner, this forces it down. To make adjustment side to side, you can tilt the whole block, using a tool in the two holes. I have several of these and some don't have the holes; there you need a tool that fits the slot where the jack screw is, but it needs a little recess for the screw. You can see I made a decent tool for the holes, and a quick-and-dirty one for ones without holes- but it works!

 

The Steiners with adjustable height headstocks don't have the jackscrews, as they aren't needed.

 

To see what adjustment is needed, it's nice to have a wheel fairly large in diameter, not too long of an arbor. Set it up in the tool (in the correct sized runner), and check if it's nice and square to the runner, looking down and from the side. That's about as scientific as it gets. If this was a larger machine tool you'd check stuff with dial test indicators and such, but simply no room here. If you get it so that your test wheel (make sure it's running true to itself first, haha) can be taken out an put back and it's always lined up, you're all set.

 

The only thing to really pay attention to is "nod" of the spindle; if you make adjustments to the jackscrews unevenly you can tilt the spindle, so that it hits correctly at one distance but not at others. Easy enough to check, pull the spindle in so that the tail runner is as far forward as it goes, check, do the reverse, check, adjust. In practice I automatically try to use the head spindle in more or less the same position if I can.

Edited May 10, 2022 by nickelsilver

[JohnC]

@nickelsilver that's awesome, thanks so much. I'm gonna try it tonight with a big pocket watch third wheel I think. Looking closely now, I do actually think the adjustment of the jackscrews is uneven, which may be causing my tapering problem. Thanks very much again!

[nickelsilver]

6 minutes ago, JohnC said:

@nickelsilver that's awesome, thanks so much. I'm gonna try it tonight with a big pocket watch third wheel I think. Looking closely now, I do actually think the adjustment of the jackscrews is uneven, which may be causing my tapering problem. Thanks very much again!

Yes a third wheel is good- especially as they often have the wheel quite close to one side of the arbor, so get that side near the tailstock runner for the visual comparison.

[JohnC]

@nickelsilveryeah so with the third wheel, up/down looks good. Left/right needs a small tweak so I need to make one of those tools. Actually I tightened down the brass screws that grip the headstock spindle (left and right of the main knurled knob) and this seemed to help the lateral alignment a tiny bit. But still needs a slight adjustment. Thanks again for your help.

[VWatchie]

On 10/5/2020 at 11:10 PM, JohnR725 said:

Accepting that earlier watches probably were not made to the same tolerances and attempting to modify early watches to fit the perception of a modern standard is not what you're supposed to be doing. Unless you're just doing it to have fun and prove you can do it.

I just serviced a Vostok cal. 2431. It's identical to the cal 2416B but has been modified to make the hour hand rotate 360 degrees only once per day (24h movement). Anyway, in the process, I very carefully adjusted the end-shake to be as small as I possibly could while very carefully making sure all wheels were spinning freely.

After the service, the amplitude is very erratic. In the dial-up position, it varies between 300 degrees and 250 degrees in what I perceive to be a random manner. I can't detect any particular time pattern for the variations. It appears to me like the amount of power that reaches the escapement varies randomly.

Now, these movements aren't exactly known for their precise tolerances. Could it be that these movements don't handle super fine end-shake as well as Swiss ETA movements? Perhaps I should try to increase the end-shake to say 0.06 mm or so? Of course, and as always, the problem could be something entirely different, but as I did adjust the end-shake so precisely that's what I suspect could possibly be the problem.

I guess my question is, can end-shake be too small even when all wheels are perfectly free?

EDIT: I've just started a new thread to put some more focus on this question.

 

Edited August 14, 2022 by VWatchie

[JohnR725]

18 hours ago, VWatchie said:

After the service, the amplitude is very erratic. In the dial-up position, it varies between 300 degrees and 250 degrees in what I perceive to be a random manner. I can't detect any particular time pattern for the variations. It appears to me like the amount of power that reaches the escapement varies randomly.

Now, these movements aren't exactly known for their precise tolerances. Could it be that these movements don't handle super fine end-shake as well as Swiss ETA movements? Perhaps I should try to increase the end-shake to say 0.06 mm or so? Of course, and as always, the problem could be something entirely different, but as I did adjust the end-shake so precisely that's what I suspect could possibly be the problem.

one of the problems are dealing with a mechanical watch is the gear train.  You're always going to have a power fluctuation with the gear train..  Typically will average out typically won't see an issue.. but anything that can cause a wheel to be out of round. in other words as is revolving around it will have a little bit of a binding issue causing a power loss that will show up as numbers changing. So basically bent pivots wheels that were not perfectly round or non-jeweled holes that are no longer exactly where there supposed to be in other words there out of round. Then you start to see a pattern of power fluctuations.

Typically on a timing machine you'll see that the numbers will change. If you look at the amplitude you may see it going up and down not settling down to something. At work because we have really expensive timing machine made by witschi I go to the time plot feature. Then it plots the amplitude versus time in a different fashion and I can look for patterns. So I'll see a pattern if there's a wheel out of round for instance in the case of vintage watches it just tells me why I'm having a problem but I don't worry about it. Because typically even if you find a pattern the watch will average out the problem hopefully.

 

[xyzzy]

On 8/14/2022 at 4:15 PM, JohnR725 said:

Then it plots the amplitude versus time in a different fashion and I can look for patterns. So I'll see a pattern if there's a wheel out of round for instance in the case of vintage watches it just tells me why I'm having a problem but I don't worry about it. 

What does this plot look like?

I've been trying to get tg to show this kind of data.  I've got a Seiko 2539A that seems to have a lot of out of round wheels.

Here's about four minutes' worth.  The line is amplitude.  The short length oscillation in the time plot is 30 ticks long.  So likely caused by the escape wheel.  But there's also a cycle in the amplitude that looks like it's one minute in length.  That's probably the fourth wheel.  It looks like there's also a cycle in the time plot that's one minute in length as well.

I can plot the amplitude over a long period of time.  There's clearly some kind drop in amplitude that happens every 2 hours.  There's also something else, where the amplitude slowly drops over a period of a bit less than two hours, then jump backs up before slowly dropping again.  I wonder what could cause that?

Zooming in over a period of a few minutes, the cycle with period of one minute is clearly visible.

[JohnR725]

10 hours ago, xyzzy said:

I've got a Seiko 2539A that seems to have a lot of out of round wheels.

What's interesting with a mechanical watch is it will average out all those problems up to a certain point. So over time everything tends to average out. They can be a problem depending upon  the rate of the cycle if it's a short term on the timing machine. So in other words  you time a watch you are watching it you'll see things going up and down in.. Or if is like it work where Rodda six position timing and every time it's a little bit different because each time it's in a different power cycle.. But you put the hands on and they usually keep time.

10 hours ago, xyzzy said:

I can plot the amplitude over a long period of time.  There's clearly some kind drop in amplitude that happens every 2 hours.  There's also something else, where the amplitude slowly drops over a period of a bit less than two hours, then jump backs up before slowly dropping again.  I wonder what could cause that?

The problem with things that occur over long periods of time is it's not the gear train typically it could be the mainspring barrel  yet figure out the gear ratio and how many revolutions that makes per whatever. Then you'd have to go to the dial side of its running a calendar

by the other thing that becomes important is does the amplitude variation cause you a timing problem that you're concerned with? The words you see that there is your amplitude variation but if you do a time plot at the same time as amplitude plot are you seeing a dramatic timing problem because of the timing stays relatively consistent then it's not really a problem..