American PW Puzzle, why does my Waltham run slowly

[MikePilk]

My newly cleaned Waltham 1899 model PSB runs 15 mins a day slowly. I thought the spring looked odd - it had lost it's blue colour.



I had a spare hairspring so changed it - still ran 15 mins a day slow. I'm pretty new to all this, so thought maybe I've got a damaged jewel, bent pivot or something I hadn't noticed.

In the post today I got another PSB from ebay - I thought just for parts, but it all looks OK, so I thought I'd try fitting the cock/balance. And I got the timing spot on.

Sat here with a cold beer pondering what could be the problem ?

 

[richiec]

MikePilk, depends on what kind of beer. Sounds like a power loss in the train, probably needs to be disassembled and cleaned. The mainspring could be gummy, dirty jewels, worn pivots all which will become apparent when you take it apart, after a couple of more beers for courage though.

 

[MikePilk]

All the jewels were cleaned with lighter fluid, then pegged and ultrasonic. I fitted a new mainspring.
Being a newbie, I might have been a bit too generous with the oil.
The puzzle is, why is the timing ok when I put in the cock and balance from the other watch?
Maybe something wrong/bent with the the jewel and original cock - though the jewel looks good to me.

 

[gmorse]

Hi Mike,

That the other balance worked OK points to an issue with the original balance assembly. Did you take the endstones off when you cleaned the balance jewels? They can harbour some really sticky old oil, and lighter fluid by itself can still leave old hardened deposits intact. Looking closely at the pivots and jewels needs some good magnification, at least x20 or preferably more. Another thing to check is the balance amplitude; it should be at least 270 deg.

Regards,

Graham

 

[richiec]

Mike, make sure you take out the balance jewels, usually you have to peg the cap jewel as the oil tends to become hard and not come off in a bath. Is there sufficient endshake in the balance? Too tight and it will run slow, same if it is too loose.

 

[Harvey Mintz]

A couple of observations:

1) the "original" hairspring didn't lose its' blue color: it's not a steel hairspring and was never blue. It appears to be a non-magnetic hairspring; some watches were retro-fitted with non-magnetic hairsprings at the factory (and some were just changed to non-mag hairsprings by watch repairers). If the movement is marked "Non-Magnetic" you'll want that non-blue hairspring back in the movement (be careful - the early non-mag hairsprings are a bit soft, and can deform easily).

2) The balance has 4 mean-time screws, none of which are screwed in all the way. Screwing them in will make the watch run faster, although I can't tell whether there's enough screw available to speed it up 15 minutes worth. The mean time screws are the ones at the balance arms and at 90 degrees to the balance arms; they are there precisely for the purpose of bringing the watch to time without throwing the balance off (tighten them in pairs, the same amount for each screw and the balance should remain poised).

3) By all means, perform all the other checks and cleaning steps being advised before changing the mean time screw positions: if the watch isn't running well, you can't possibly bring it to the proper rate.

As a general rule, you shouldn't move balances between movements as the balances on most movement are actually marked with the serial number of the movement. Changing the balance creates a "Franken-watch", which is generally frowned upon by most collectors. It's OK to do if it's only for a test, though, as long as you change back to the original balance.

 

[Smudgy]

If you have the capability vibrate the balance to see where it vibrates. It is possible that at some time in the past someone altered the balance to compensate for a problem rather than actually address the problem.

 

[MikePilk]

What do you mean by 'vibrate the balance' Smudgy ?

So I took it all apart, re-cleaned, oiled all the jewels, and checked them for damage. Yes, I do take the cap jewels off and give them all a thorough clean, and I examine the jewels under a microscope.

I put the original spring back on, and checked that the impulse pin sat in the middle of the banking pins.
I've no means of measuring amplitude, but it looks strong.

Result - still runs 12 mins slow. I tried the balance/cock from my other PSB, and it was within 10s without any adjustment (and I don't plan to make a Franken-watch !).

I demagnitised it. The balance staff sat in the jewel holes leans about 5 deg - I believe that's about right?

My Conclusions :
1. The gear train, mainspring and balance jewels are OK (as I can get good timing with the other balance)
2. Balance spring is probably OK (I get similar results when I change the spring).

So it's somethink odd with the balance or staff. The balance looks similar to the other one. Nothing obviously wobbly.

 

[Jim Haney]

mike,
Harvey mentioned it. The Balance screws . Look at the replacement and count how many screws are on each wheel. It they are the same LOOK at the size of them and see if your original has some larger heads than the replacement balance.

 

[MikePilk]

Jim, as you can see from the picture above, apart from the mean-time screws, there are 5 other screws on each arm. My other PSB only has 4 other screws. But then I have a Riverside which has 6 others !
How can that work? Are there different springs depending on the number of screws ?

 

[Jim Haney]

Jim, as you can see from the picture above, apart from the mean-time screws, there are 5 other screws on each arm. My other PSB only has 4 other screws. But then I have a Riverside which has 6 others !
How can that work? Are there different springs depending on the number of screws ?

Mike,
Yes, the higher quality movement used better and heavier Balance wheels & screws and heavier (Gold) screws.

Have you tried moving in the 4 meantime screws as Harvey suggested or Try to solve your problem without messing with switching HS, just lighten up the present wheel by undercutting 2 of the screws, trial and error a little at a time.

 

[richiec]

The hairsprings were "vibrated" to the balance meaning that the strength of the spring was set up to the weight of the balance by getting the length of the spring correct so that the balance swung at the correct rate. I have enclosed a picture of a hairspring vibrator with a balance attached to it. the attached balance was set in motion along with the test balance in the stand and then it was colleted when the balance oscillated the same as the test balance. So it is possible that the balance has an improper hairspring, maybe too weak so it swings too much, hence less beats.

 

[MikePilk]

Richiec, thanks for that explanation. I didn't realise the hairspring and balance were matched like that. I assumed that, being 'machine made' watches with interchangeable parts, all hairsprings were equal.

Jim, I'm fairly new to this, so I don't have the tools to 'undercut' the screws. By undercutting do you just mean removing metal from the head?
I have a spare balance so could try filing some screws down - how close do I need to get the weights?
And is 12 s equivalent to a lot of weight ?

 

[gmorse]

Hi Mike,

I'd like to add a small word of caution here; once you've removed metal from any component, you've changed it permanently, and if you subsequently discover that the cause of your problem is unrelated to what you changed, there's not necessarily a way back. This, and its close relative, "change the part to fit the watch, not the watch to fit the part", is one of the ground rules of good practice; think and analyse before you reach for any tools!

Regards,

Graham

 

[Jim Haney]

Richiec, thanks for that explanation. I didn't realise the hairspring and balance were matched like that. I assumed that, being 'machine made' watches with interchangeable parts, all hairsprings were equal.

Jim, I'm fairly new to this, so I don't have the tools to 'undercut' the screws. By undercutting do you just mean removing metal from the head?
I have a spare balance so could try filing some screws down - how close do I need to get the weights?
And is 12 s equivalent to a lot of weight ?

Mike,
Do you mean is a 12 size balance screw the same threads and weigh as the ones on your 16s balance ? No, is the answer. Did you try screwing in the meantime screws yet?

As Graham has stated, if you are not experienced, forget the undercutting of the balance screws, unless you have extra screws to replace them with and you said you don't have the tools to do it with.

 

[MikePilk]

Sorry Jim, I meant to type '12 minutes'. I have a spare balance from a PSB so have a set of screws to play with. Do you think that screwing all the meantime screws in could five me 12 mins difference ?

 

[psfred]

Examine the balance screws on the "slow" balance for the presence of tiny washers under the screws. These are added to slow the watch down, although with mean time screws I would be surprised to see them. You never know what's been done to an old watch, though.

It is quite possible the hairspring is incorrect, but I'd check the balance very carefully on a poising tool to verify it's in poise and that the arms are not bent -- this is a bimetallic temperature compensated balance, and it's quite easy as a rule to distort them while changing a balance staff. if the cut ends are not in very close alignment with the balance arm side, the watch will likely be impossible to time correctly -- bent outwards means slow, inwards means fast. It looks to me as if one arm is too far in and the other too far out.

Do not adjust the positions of the screws that are not screwed all the way down in the middle of the cut rim sections, these adjust the temperature compensation!

Adjust the arms on the balance and get it poised, then see it if still runs very slow.

Peter

 

[Rob P.]

Ya know, I looked at the pic close up and I can't see an upper pivot on that staff. It might be the quality of the pic or it might be that the staff needs replacing. (Yes, it's possible to have a balance still run with a broken pivot. It won't run well and will usually lose a ton of time but it WILL sometimes run with the broken pivot up.)

Swapping in the good balance from the other watch shows that the rest of the movment is working as it should. Thus, your problem is with the balance someplace. Which again leads me to that possibly broken pivot I mentioned.

From there, pay attention to Psfred's warning about the temperature compensation screws. Don't mess with them unless you want a watch that will run fast when it's cold and slow when it's warm or vice versa. If the pivots are good, straight, and undamaged; then you will need to check for excess weight on the balance wheel (timing washers or gold screws swapped in place of brass screws). If nothing out of place, then you need to adjust the mean time screws. Go SLOW and adjust a bit at a time because the screw hole gets enlarged as the screw is tightened and the screw will be loose if you back it out after tightening it.

The big issue is that the meantime screws will only compensate for about 2 minutes/day. 12 minutes is a lot more than you can adjust out with them. That means you need to invest in some balance screw undercutters and balance screw screwdrivers (these hold on to the balance screws while undercutting so you don't drop them) and a poising tool.

To use: Remove the hairspring from the balance. Remove one screw and use the undercutter to remove material from the underside of the screw head and replace that screw on the balance. Now do the same for the opposite screw. Poise the balance. If out of poise, undercut the screw on the heavy side enough to bring to poise. If you remove too much weight from that screw, undercut the opposing screw and continue until in poise.

CAUTION! If you remove too much weight your watch will run fast so remove as little metal as possible from the screws. VERY LITTLE metal will need to be removed to correct for 12 mins/day so once the balance screws are undercut & the balance is poised reinstall the hairspring and check the rate. If still slow undercut some more. How much the timing changed should tell you how much more undercutting you need to do as compared to how much metal you've already removed. (ie: if 1 full turn for each screw on the undercutter gave you 4 mins of rate change, you need 2 more full turns to get 12 mins of total change.)

 

[richiec]

Rob, were you able to enhance that balance photo to get a clear picture of the pivot?

 

[MikePilk]

There are no washers under any of the screws. The arms look to line up pretty well to me.
I've just taken some pics of the pivots - look ok


I also managed to measure the length of the pivot as 5.4 mm, which is correct.

If screwing all 4 meantime screws in won't give me 12 minutes, I'll put this watch on 'hold'.

I need to buy a Poising Tool and some undercutters - and do some research on how to use them.

Thanks for all the advice

 

[karlmansson]

Loss in the power train will give you a watch that runs fast. If the loss is not in the movement of the balance, dried oil etc., that is. This has to do with the amplitude of the balance though so it's easily checked. You don't need a tool to measure amplitude, when you look at the balance you will get a point where the arm becomes visible to you through the motion blur. That will be the turning point and that will tell you how far the arm traveled before changing direction. Half a turn will be 180 degress etc. This is not be confused with older watchmaking literature that will refer do this a 1 turn. 2 turns would in that measurment equal 360 degrees. Anything between 270 and 320 will mean that the aspects controlling the rate of the watch will be the mass of the balance and the qualities of the spring.
Friction will not have a significant impact on time keeping as long as the amplitude is kept between these degrees. Any higher and you risk "knocking" of the impulse jewel.

That being said, if the watch runs well with good amplitude and still loses 12 minutes a day you will have a mismatch between spring and balance. I the spring is not original, that's a pretty strong suspect. I could also have reacted with some sort of chemical while cleaning or maybe it's been annealed at some point. Not sure how invar behaves when heated but if it had been a blued steel spring that could be a suspect too.

One more thing though, if the spring is poorly studded or poorly connected at the balance cock maybe that could account for some slack in the system. Is it on there snugly?

Best

/Karl

 

[gmorse]

Hi Mike,

Well, the pivots don't look too bad, except for the ends, which are pretty flat. This won't give you the slow running, but could result in differences of rate between horizontal and vertical positions.

To get a better idea of amplitude, try taking a short video of the balance and then playing it back at half speed, on the camera if it has the feature, or else on the PC.

Regards,

Graham

 

[MikePilk]

Karl, thanks for that - I'd wondered if you could tell the amplitude without a tool !
It looks to be about 280 -290 degrees.

The spring does seem to be a snug fit. I tried another spring from a PS Bartlett and got similar results.

It looks like a mis-match between spring and balance.

On the plus side, I spend this afternoon cleaning another PSB I got off ebay for $30. I assumed it would have a broken balance, so I bought it for spares. It turned out the be just dirty. I cleaned it, put in a new mainspring and fired it up - without any adjusting it was within 3 s/day. I don't know how good that is, but it's good enough for me! Just my third watch.

I assume people on here are aware of the mobile phone app WildSprectra? Great free bit of software for those of us who don't have any timing tools.

 

[karlmansson]

Forgot to mention that you should of course use the resting position of the balance arm as a reference. The distance the arm travels after that will be half of the amplitude (the balance will swing back the same distance).

3 seconds a day is very good, especially if it keeps that through all positions. I think the limit for chronometer standard is +-3 seconds over a day under pretty extreme conditions.

The thing about the mobile apps is that they use the phones internal timebase as a reference. I'm not sure how that holds up to real timing machines. There is a free piece of software called eTimer that lets you analyze pre-recorded beats of a watch. As long as it's a .wav file it will run it. It can handle multiple frequencies and also calculates beat error and amplitude. PC only.

 

[Rob P.]

After looking at the pivot pics, I'd have to go with everyone else's opinion that the hairspring is too soft. You will need to either undercut the balance screws to bring that watch to time or find a strong hairspring. Option #2 may still require that you adjust the balance weight. So I'd say to just undercut the screws since you're likely to have to do that anyway if you change the hairspring.

PLEASE don't just file on the screw heads to reduce their weight. I have an Elgin that has had the screws mangled that way by someone on the past. It was an acceptable method at 1 time but that Elgin looks so ugly I'm ashamed of it.

 

[karlmansson]

If you manage to find a stronger spring with a matching collet and terminal curve (or if you have the skill set to re-collet and form the curve yourself) I would recommend using the mean time screws and adding timing washers rather than cutting the screws. Save the old hairspring too, that way everything is reversible if you should find out that you made a mistake of some sort.
Pick a spring that is strong enough to have the watch run a little fast and them poise the balance using washers. I'm sure that there are people here with much more experience with this than me, this is just my own, cautious approach to things. Always give yourself the best abilities to undo your mistakes.

Best

/Karl

 

[Rob P.]

If you manage to find a stronger spring with a matching collet and terminal curve (or if you have the skill set to re-collet and form the curve yourself) I would recommend using the mean time screws and adding timing washers rather than cutting the screws. Save the old hairspring too, that way everything is reversible if you should find out that you made a mistake of some sort.
Pick a spring that is strong enough to have the watch run a little fast and them poise the balance using washers. I'm sure that there are people here with much more experience with this than me, this is just my own, cautious approach to things. Always give yourself the best abilities to undo your mistakes.

Best

/Karl

Karl, the watch is running SLOW. Adding more weight via timing washers will make it run slower.

 

[RJSoftware]

The simple and more elegant solution is to shorten the hairspring by re-pinning the terminal end.

I do not subscribe to doing any undercutting and mean time screw adjusting.

Here is why.

First of all think in terms of likelihood.

Your watch left the factory with a hairspring and balance perfectly vibrated. Vibrated meaning that the hairspring was cut and regulated to as perfect position as possible. This includes poising.

So it may be possible that the hairspring was replaced with a weaker hairspring thus giving the balance more ability to turn a larger arc hence slowing down the escapement action and making your watch run slow.

Typically this would a n.o.s. (new old stock) hairspring of near compatible size.

So let's now knock out the probability that the balance is out of poise.

It does no harm to check by a poising tool, but consider that it is better to add weight by washers and then strengthen the spring by shortening via moving terminal block closer in. Getting the regulation correct is most likely all that is needed. Once you start screwing with the poise things can go downhill quick.

I tell you this because removing material from a balance screws is most likely poised well from the factory can be a mistake.

It's like cutting a leg on a ladder that you think is longer, then to keep cutting leg after leg to get the ladder to once again stand as perfectly as it did from the factory.

The mean time screws are not only for speeding/slowing up but are also calculated along with temperature compensation. So it's not as simple as screwing in/out in equal amounts on both sides. It is a much longer process and compromise. At least that is the way that I understand it.

See you can shorten the balance hairspring in small increments by re-pinning and you have the job done with no damaging permanent alteration. Also understand you do not need to cut the hairspring you simply have a longer end sticking out from the terminal.

The only other thing you have to do once re-pinned is to re-establish the beat. The shortening throws the position of the roller jewel position (the beat gets uneven -favors one side).

Here is simple steps.

1. With balance installed, balance cock on, pull the terminal block pin out with tweezers and pull end of hairspring end out about 1/8th inch.

2. Push terminal block pin back lightly and adjust hairspring true-ness of the flat (hairspring coils level with each other). The fully insert.

Setting beat: Notice the jewel/bushing hole for the lever fork. That is exactly where you want the roller jewel to land when the balance is relaxed.

So with the whole balance assembly off and laying on it's back you can set the pivot of the balance in the top balance cock jewel and tickle the balance to see where the roller lands in relation to that spot.

What I do sometimes is push the regulator prior to taking off the balance assembly and have some part of the regulator pointing to the location of the fork lever jewel bushing.

That way when I flip the balance over on it's back the portion of the regulator that I designated to point at the hole can be seen. Either that or I use a magic marker or small scratch only visible by loupe.

Now here's the trick that makes setting beat much easier.

With the balance laying on it's back I flip the balance wheel over and then prize the balance off by the hairspring collet.

Then let the hairspring stand floating above the balance cock laying on it's back. I also take time then to center up the hairspring collet over the balance cock bushing/jewel by small adjusting bends.

The idea is to have the hairspring in a perfectly relaxed state with the collet dead center over the balance cock bushing hole. So that when you have the roller in the correct spot there is no tension or unnecessary bending to have the collet dead center and the roller jewel landing exactly where it needs to be.

So the trick is in the install below.

3. Now to install, I line up the roller with the designated spot on the balance cock (the regulator spot) and then do a small press of the balance cock into the hairspring collet.

Just a small press by inserting the pivot into the jewel.

This is not installing the hairspring, but instead it is just getting the hairspring collet to grip without losing the position. Just enough so that the collet grip won't too easily come off. You don't press hard because you could break the jewel.

You could set something like a small piece of paper so to add more grip to the hairspring collet to grab the arbor, but it's not hard to do without.

4. Then unscrew/unpin the terminal block from the balance cock.

5. Carefully turn balance over with hairspring still connected lightly and then push collet down with staking tool or other to more firmly grip. Try not to turn the collet any when pressing it down.

6. Re-assemble and test watch speed with regulator at dead center.

Repeat steps 1-6 as needed.

The ideally perfectly regulated watch should have regulator at dead center. You can get there by adjusting the termination point and re-establishing the beat.

Doing the above steps is much better than turning the collet by inserting something into the slot. It is too easy on some hairsprings to break the collet or distort it by turning the collet by slot to establish the beat. Some collets are old and tender for lack of a better word. Once they start to crack is sucks.

Prizing up a collet I like to do with an Exacto razor knife. The sharper the edge the better it is at getting under the collet. This is particularly important when watches get smaller.

This is much better than turning the collet by the slot to remove also. Breaking a collet makes life a pain.

RJ
RJ

 

[karlmansson]

Karl, the watch is running SLOW. Adding more weight via timing washers will make it run slower.

Which is why I suggested he select a spring that is strong enough to make the watch run fast with the balance in its current state. Then add washers for poise and timing.
One thing worth mentioning when it comes to mean time screws, balance screws and washers is to keep regulating symmetric. That is, always add the same amount of weight to opposing sides of the balance to keep the poise. Your balance is a solid one, it would be worse if it was a split arm balance with temperature compensation. Yours might still have that but different principles apply, the one of your type will probably deform to compensate while a split arm balance would move the tips of the arms inwards and outwards.
Anyhow, that means that you need to keep the poising and adding/removing weight as close to the balance arms as possible to avoid messing with the temperature compensation. Same goes for mean time screws, only turn those directly above the balance arms.

RJ,
There are two more aspect to shortening the spring. First off, just pulling the spring through will only work if the regulator sweep is concentric with the balance staff. If not, there will be regulator pins interfering, even pushing the spring at the wrong places. As this is an overcoil hairspring that will mean even more trouble, with the overcoil being displaced above the body of the spring.
The second aspect is that the colleting point of the hairspring will have moved in relation to the terminal point. If they are not arranged correctly the watch will run more inconsistently over different positions. I can't remember the theory of this now but it has to do with whether the balance starts on an "uphill" stroke or "downhill" stroke in relation to gravity and the balance stud.

Best

/Karl

 

[gmorse]

Hi Karl,

... I can't remember the theory of this now but it has to do with whether the balance starts on an "uphill" stroke or "downhill" stroke in relation to gravity and the balance stud...

It's explained in chapter 14 of "Practical Watch Repairing" by de Carle. It was discovered by Jules Grossman. With the watch in its "normal" position, (pendant up for a pocket watch), the spring should exit the hole in the centre collet upwards, with the exit point on a horizontal level with the centre of the staff.

Regards,

Graham

 

[karlmansson]

Hi Karl,



It's explained in chapter 14 of "Practical Watch Repairing" by de Carle. It was discovered by Jules Grossman. With the watch in its "normal" position, (pendant up for a pocket watch), the spring should exit the hole in the centre collet upwards, with the exit point on a horizontal level with the centre of the staff.

Regards,

Graham

Yes, that's it! page 146 for those interested. DeCarle also deals with measuring amplitude and measuring amplitude in "turns" in that chapter.

Fo those interested in reading more about it I found these two articles: http://people.timezone.com/library/horologium/horologium631675494030170118
http://raulhorology.com/2012/11/the...rt-2-vibrating-the-spring-using-a-luthy-tool/

I know I found a source that explained the theory behind it but I can't remember it now, will post if I find it.

/Karl

 

[karlmansson]

Not sure if this is it either but it does discuss hairspringing more. From the post:

"As you might remember from the Watch School article that dealt with poising the balance (Part 5), it is critical for the balance to be in poise (the weight evenly distributed about the axis of rotation) in order for it to keep good time in the vertical positions. Likewise, the hairspring can also be poised by ensuring that its vibrating point lines up with the point of attachment at the collet. This means that the active length of the hairspring consists of whole coils, thus being more or less evenly distributed with regards to mass.There are a variety of different theories on points of attachment and the pursuit of poise of the hairspring is only one of them. Another approach is to attempt to align the vibrating point approximately 90 degrees (86.5 degrees I believe) from the point of attachment at the collet and in this way ensure a flatter isochronism curve (the graph of the rate against amplitude). The tradeoffs of positional performance versus isochronism can probably be debated ad infinitum, but most modern watches are pinned at whole coils in my experience."

http://www.tp178.com/jd/watch-school/6/article.html

 

[RJSoftware]

Hey Karl.

I do understand that pulling/shortening the hairspring by re-pinning at the terminal can effect the position of the regulation portion of the hairspring. A simple adjustment handles this.

But I can not see how repining could possibly effect the hairspring exit position at the collet. And I think Fried is saying that the angle is to be near 90 degrees perpendicular to the staff arbor. Or maybe I am reading that wrong.

Even so if there where some desire to have the exiting at the collet tilted slightly upward (as maybe to support hairspring body?) repining should not effect that either.

That and the repining probably will be slight to gain great effect. I think much better solution than adjusting meantime screws (which had been calibrated to include weather conditions) and/or undercutting.

Plus, there is nothing to lose. If incorrect simply restore old pinning location.

RJ

 

[karlmansson]

As you rotate the spring, which will in effect be what you do when shortening it at the stud, you will also rotate the collet in relation to the balance cock. The physical point of attachment on the collet will of course not change but the point of attachment in relation to the resting balance (when in beat) will change and that is the change that counts.

Take a look at the links I posted above, they explain it better than I do.

The mean time screws halfway along the rim will affect temperature compensation but not the ones directly over the balance arms. Unless the arms themselves are used for expansion and contraction in compensating for temp. changes. That would be news to me though.

/Karl

 

[Rob P.]

I take the description to mean that if you draw a line between the end stud and the center of the collet, the point where the HS is pinned & exits the collet will be on that line. Or it will be at 90* to that line.

I believe that the overcoil is an attempt to move "the working coils" in relationship to the above "line" for isochronism purposes. How that works exactly I am not sure.

 

[karlmansson]

After re-reading your last post RJ, I think we mean different things by "rotating" and "point of attachment". The point of attachment in this case is the point where the hairspring meets the collet at its circumference (in relation to the balance (when in beat) and with the spring fixed on the balance cock.

The schematics in the TimeZone article shows it well.

 

[MikePilk]

I'm enjoying the discussion, but I've just had a look at the end of the hairspring, and re-pinning that is someway beyond my current ability !

 

[RJSoftware]

I re-read the De Carles Practical watch repair Chapter 14 on Grossman's discovery.

Well, he kind of waffles on it's validity a bit as he states not all watches this effects the same. But then later states that the phenomina should not be dismissed. (similar words anyway).

I try to picture the why and the how of how it works. Most I could come up with is perhaps the balance wheels are not really perfectly poised. So as a heavy side of the balance wheel tends to go downward, maybe having the hairspring exiting horizontal and then continuing upward counter-acts the wheel heavy side downward motion.

I see it sort of like a small diving board, the extra weight bounces better when the diving board is horizontal than vertical. Besides I would hate to jump up and down on a vertical diving board as my feet might slip and the diving board edge crush my privates... well never mind...



RJ

After re-reading your last post RJ, I think we mean different things by "rotating" and "point of attachment". The point of attachment in this case is the point where the hairspring meets the collet at its circumference (in relation to the balance (when in beat) and with the spring fixed on the balance cock.

The schematics in the TimeZone article shows it well.

 

[penjunky]

Did you try screwing in the meantime screws yet?

HI Jim

This is the first I've heard that these screws are called meantime screws in a watch, my understanding they are weights. Do these meantime screws have any relation with Greenwich meantime (GMT) Seems they both are related to a certain point at a certain time. Just curious.

Roger

 

[penjunky]

My newly cleaned Waltham 1899 model PSB runs 15 mins a day slowly.

Hi Mike

Did you get this watch fixed, if so what was the problem. I'm asking because my 16s waltham is currently running about one minute slow per hour--over 24 minutes a day even though it has great amplitude. Letting it run down to see if there's any time change before I get back into it. The first thing I'm going to check is the balance wheel as I had to change it because the wheel that was in it was loose on the hub and the hub was cracked.

Roger

 

[Jim Haney]

HI Jim

This is the first I've heard that these screws are called meantime screws in a watch, my understanding they are weights. Do these meantime screws have any relation with Greenwich meantime (GMT) Seems they both are related to a certain point at a certain time. Just curious.

Roger

Yes that was a 5 year old post to help him out but you can read this thread for an explanation..


What are meantime screws?.

 

[penjunky]

Yes that was a 5 year old post to help him out but you can read this thread for an explanation..


What are meantime screws?.

Thanks Jim, I'll keep it for future reading.

I'm just now working my way to the balance and trying to take in all I can about it. All this information I have found on this forum is like throwing an alka-seltzer in a glass of water, it's making my brain fizz. I never dreamed so much knowledge went with the few parts of a pocket watch.

Roger

 

[praezis]

. ..is currently running about one minute slow per hour--over 24 minutes a day

... the balance wheel as I had to change it because the wheel that was in it was loose on the hub and the hub was cracked.

Roger, these are problem and reason in the same post.
Balance wheel and its hairspring must be kept as a unit.

Frank

 

[penjunky]

Roger, these are problem and reason in the same post.
Balance wheel and its hairspring must be kept as a unit.

Frank

Thanks Frank

Since the wheel and spring both was damaged beyond (my) repair I replaced them with a wheel and spring from another watch. After checking more I found the fourth wheel is stopping often and just today checking why.

I do now understand that the wheel and spring were (made for each other) as a way of looking at it.

Roger