Elgin Grade 291: hairspring adjustment

[LittleWatchShop]

Yesterday I was working on an Elgin with poor amplitude.  I looked at under the microscope and could see some vertical wobbling of the hairspring but really could not tell if the hairspring was hitting the balance arm.  Given my limited experience, I was hesitant to mess with the hairspring because, other than this slight wobbling, it looked good.

Nevertheless I removed the balance and observed it outside of the watch and determined that there was a slight bend that was likely causing the hairspring to hit the balance.  I would say it looked like this drawing I show here.

Here is the amplitude I observed.

I made a number of micro adjustments with my hairspring tweezers until I got it pretty level.  This took me about 30 minutes being EVER so careful.

Here is the amplitude after making the adjustment.

Next I have to regulate it.  Once I got the better amplitude, I stopped for the day!!

[LittleWatchShop]

Here are some more pictures...for those addicted to such things.

Noticed some previous horologist put a dial washer under the balance cock to tilt it a bit.  I kept it in when I reassembled the watch.  Guessing that there was a staff replacement that could have been a hair too long.

[mikepilk]

You say "Good amplitude", but it only looks like about 200°.  Is the spring nicely centred and not bunched up on one side?  How is the endshake ?

If you take the pallet out, and give the balance a strong blast from a puffer, how long does it oscillate?

Did you change the mainspring? 

Edited December 29, 2022 by mikepilk

[LittleWatchShop]

20 minutes ago, mikepilk said:

You say "Good amplitude", but it only looks like about 200°.  Is the spring nicely centred and not bunched up on one side?  How is the endshake ?

If you take the pallet out, and give the balance a strong blast from a puffer, how long does it oscillate?

Did you change the mainspring? 

Great questions.

There is Good, Better, and Best.  This is just Good.

Hairspring appears centered to my view.

No, I did not do the free-running test.  Was going to do that before I notice the issue with the hairspring.

Did not change the mainspring.  Frankly, I think it is weak, but that is just amateur observation.

Endshake seemed OK to me.

[watchweasol]

Hi LWS.  In addition to a dial washer I have also found a divot in the plate for the same reason, increasing the endshake. Don’t be tempted to polish them out .

[Neverenoughwatches]

1 minute ago, watchweasol said:

Hi LWS.  In addition to a dial washer I have also found a divot in the plate for the same reason, increasing the endshake. Don’t be tempted to polish them out .

Also seen tin foil used. A good excuse to play with the jewelling tool and push friction jewels

[mikepilk]

About 200° amplitude is not good! It's very bad .  Something is wrong. You should be seeing at least 240° amplitude, and ideally 270° +. With old pocket watches, when I have changed the mainspring, I have often had too high an amplitude (re-banking) due to modern mainsprings being stronger.

I suggest you do the free-running check. My "pass criteria", is that after a strong blow from a puffer, the balance is still rotating strongly (say 180°) after 30 seconds. If you don't get this, you need to examine the jewels/pivots.

If that is OK, then you need to check the gear train for free-running.  

[JohnR725]

Timing machine what happened to the timing machine results?

Then the mere drawing in the description your hairspring is bent The collet. In other words it's no longer flat and when the balance wheel is oscillating you get that peculiar rippling affect out because it's been twice that supposed to be. Putting this into a truing Cala Percy can spin it makes it a lot easier to see when you're trying to get It flat again.

18 hours ago, LittleWatchShop said:

Noticed some previous horologist put a dial washer under the balance cock to tilt it a bit. 

Somebody probably change the balance staff staff numbers 861. Unlike other things in the world were a number refers to an exact parts not so with Elgin staff numbers. The problem is the 861 balance staff comes in a new and old style with a definite link difference so somebody's using the wrong balance staff. I've attached information on your balance staff.

Then it looks like a seven jewel watch which means you could have where on the pivots or the holes someplace. Also somebody probably's played with the banking pins you want to check those and check your pallet fork to see how it is as to whether anyone played with that also.

[LittleWatchShop]

1 hour ago, JohnR725 said:

Timing machine what happened to the timing machine results?

Here you go John...old school from the Vibrograf B200 after 18 hrs.

Initially the beat was way off, but I tweaked it and in the process felt like the HS collet was not super tight...an indicator of some previous work on the staff I believe.

[JohnR725]

17 hours ago, LittleWatchShop said:

Vibrograf B200

I assume if you put a better running watch on the machine the lines are actually straight and not wobbling around? Usually stuff like that could be lack of lubrication on the escapement perhaps? In a case the line should be more straight.

17 hours ago, LittleWatchShop said:

HS collet was not super tight

Yes that's a problem. It's why I tend not to like to even touch the collet on pocket watches because either end up being a loose which you can tighten sometimes or they magically crack sometimes you get it to stay in place and work other times that's a definite problem.

On 12/29/2022 at 11:07 AM, Neverenoughwatches said:

A good excuse to play with the jewelling tool and push friction jewels

Actually a bad idea. In the case of balance jewels like this you really can't change their position at all. The best solution is to restore things back to the factory remove the indents remove the dial washers remove whatever is there. Then new balance staff probably or possibly shortening the pivots depending upon how much needs to be shortened. Or sometimes just leave well enough alone of everything seems to be working.

[Neverenoughwatches]

20 hours ago, JohnR725 said:

The problem is the 861 balance staff comes in a new and old style with a definite link difference so somebody's using the wrong balance staff. I've attached information on your balance staff.

I've noticed a  difference of 0.01mm in length with staffs that are for inca shock versions of the same movements.

2 hours ago, JohnR725 said:

In the case of balance jewels like this you really can't change their position at all.

Thanks john. Thought i had read somewhere that balance jewels could be moved in or out .

Edited December 31, 2022 by Neverenoughwatches

[JohnR725]

10 minutes ago, Neverenoughwatches said:

Thought i had read somewhere that balance jewels could be moved in or out

This is watch repair everything is possible. But on an American pocket watch the balance hole jewel typically is fixed where it goes in. You want to change his location you need some tools and a lathe to trim down the setting to get it closer in. So in this particular case moving them around is not a practical option.

But in some watches you may be able to move the jewels around. But ideally it would help to put the correct balance staff in and adjust its pivots to the situation. I suspect some watches they deliberately made their balance staff pivots long to compensate for these problems and in the watchmaker was supposed to adjust the pivots. But that would assume a watchmaker had a lathe to do that which a lot of them do not.

[LittleWatchShop]

7 hours ago, JohnR725 said:

I assume if you put a better running watch on the machine the lines are actually straight and not wobbling around? Usually stuff like that could be lack of lubrication on the escapement perhaps? In a case the line should be more straight.

Two other watches produced similar linearity.  The Vibrograf is old...even though the design is very robust, there could be some need to clean and lubricate the motor...verify the power supply is clean.

The escapement of the watch has been lubricated.

OK, for the inquiring minds...here are the Weishi results.  Fully wound and 24 hrs (ish) later.

This is suitable performance for the application intended.

Shields up...preparing for incoming.  Parse my words carefully.

[JohnR725]

14 hours ago, LittleWatchShop said:

The escapement of the watch has been lubricated.

OK, for the inquiring minds...here are the Weishi results.  Fully wound and 24 hrs (ish) later.

This is suitable performance for the application intended.

Shields up...preparing for incoming.  Parse my words carefully.

Sorry your shields aren't going to work old technology we have newer technology that will penetrate your shields.

Now what we really need is newer technology than reading a crystal ball. Looking at our crystal ball and trying to predict the future or what's going on sometimes is a bit of well problematic we need modern technology which doesn't exist. Interpreting tea leaves or reading timing machine results is a bit of   basically not the exact science it seems like it should be. I'm attaching a timing machine help but it's not an exact science of help

oh and then there is the other problem. As you're well aware within a digital device you always end up with a plus or minus the lowest amount and conceivably on the pixels we could always see a straight line as being an up down up down because it's not exactly well if you have a different in 10 in your in between you're going to end up with a non-smooth line conceivably so as to how un-smooth that line is could be debatable as to the digital 10 air of the whole thing.

Okay on the attached charts we have numbers 5, 7 and 10. Then further into the document now with five you'll sometimes even see that with lubrication where it's a little bit dry on one side of the escape wheel or one side of the fork doesn't have enough lubrication only one side is lubricated so you will see some of that they're just indicating here it's the pattern repeating and we don't seem to have a repeating pattern on Yours but still I see a rough waveform from not getting enough lubrication everywhere on the escape wheel and the fork both sides

now for number seven is basically it certain degree the same as number five basically up to check everything in the escapement the pallet stones are both  shellacked in place they're not lose. My obsession with getting the lubrication has to be right everything asked be nice and clean no chips flaws scratches old lubrication like shellac on the acting services. A oh and notice the reference to the roller jewel? I've seen this come up and this is really important on American pocket watches are probably all vintage watches check your roller jewel. Try not to break it when you're checking it but grab it gently and see if it's loose. Often times the roller jewel is loose and that wobbling around will give the exact same effect that were seeing.

Then don't even read number 10 because basically it's a everything everything could be at fault and that's why you're seeing this. So all of the previous examples could be the reason in addition to just about everything in the universe. Magnetism a lot of times will produce a really nice pretty sine wave and then it goes away when you  demagnetiz  which you should be doing anyway. Oh and he did notice where there's a reference to letting the paper tape run for a while this is where if you have expensive machines or software-based machines you can run a time plot and then you can look for wheels that are out of round. So basically if it looks this bad you're really screwed and. But yours doesn't look this bad euros looks like a mild case of this.

So if you haven't invested in a crystal ball this year you might think of getting one or learning how to read tea leaves because same skill set is required to read a timing machine graphical display and come up with definitive answers other than it points you in the direction of you have a problem.

 

Timing-Machine-Charts.PDF

[LittleWatchShop]

I opened it up and ran an open loop oscillator test.  The balance oscillated for a minute and a half before I stopped the video.  It was still going.

When I do the pallet fork test, it just feels sluggish to me.  I checked the endshake and clearance...pulled it out...gave it another one dip.  Looked at the pivots...all appeared good.  Made there was no drag from the second hand.

I think a new mainspring is in order.