Timing Adjustment Problem

[Scouseget]

Hi.

I had previously posted a question about regulator pins, as in how effective are they in timing adjustments, and a lot of folks responded with some great info - as always. Anyway, despite that, I decided I was flogging a dead horse in trying to get this particular Waltham, 1908, size 16S, pocket watch to run with acceptable accuracy so I dismantled it (again), cleaned it (again), replaced the hairspring and balance staff, reassembled and oiled it, inspected all the jewels for problems, etc. It actuallystarted up at the first shake, and then I adjusted it, and therein lay the problem - again.

I adjusted it using my new super, dandy, iPhone App, called Watch Tuner Timeographer, which appeared to be well regarded by reviewers. Using this, I was able to adjust the watch to 28 seconds a day, the beat error showed at around 5ms, the amplitude was 224 degrees. The pictograph wasn't so pretty with random dots with no discernible pattern frankly, a little like a scatter diagram. Also the values I quoted above would vary somewhat, especially the beat error. So then I put the hands on and checked it against a stop watch, and lo and behold, they told a much different story to the Timeographer, which was that the watch was fast by about 1.2 secs/minute, which was were it was before I re-serviced it, and in fact was the value that prompted me to re-service it in the first place.

The hairspring is definitely riding within the regulator pins, however adjusting the regulator makes barely any difference in timing, certainly no more than a fraction of a second. I have screwed out the two timing screws on the balance wheel several turns (exactly the same for both of course) and this adjustment too only changes the timing by a fraction of a second. With all the possible adjustments, i.e., both the regulator and timing screws at maximum adjustments, I still cannot reduce the error to less than 1.1 seconds gain per minute. 

So now my questions:

Has anyone else had experience with such iPhone Apps? Should I just bite the bullet and buy a real Timeograph? Any suggestions about why this watch may be acting the way it is as it's obviously beyond my competence level, and I've run out of ideas, and patience? 

It's taunting me as I type this, daring me to have another go but I'm not biting this time as I'll await all those great suggestions from you all. And if that doesn't work, well there's lots of room to hide it away at the very back of my drawers, er I mean the sliding drawers of my work bench of course. 

This has been too intellectually challenging for me so I'm off to the Rockies tomorrow to pit my intelligence against creatures with brains the size of a pea - I'll let you know who won (the fish, it's always the fish - but I keep trying anyway).

Thanks in advance as always.

[sstakoff]

Hmm. If you plan on working on watches on any type of frequent basis - especially anything over 19,200 bph - I  think a proper timegrapher would be a good investment. I have a Chinese Timegrapher that seems to work very well. I believe I paid around $125 on ebay.

As to your specific issue - have you demagnetized the hairspring? Are you certain that you have the proper mainspring installed? Have you inspected the pallets for proper adjustment (locking, draw, run to the banking, etc)??

 

[digginstony]

Problem is we don't know what has been done to this watch over the decades. Sorry to hear micro adjuster and timing screws have not solved the problem . Does the watch run in all 6 positions reasonably consistent. Your next step to slow the watch is to weight the balance. Used to be able to get small washers to fit under the screws . What I suggest you do is obtain another balance wheel remove a couple of screws (not timing screws) and screw into your balance on equal sides of the balance. If it's then running slow, file those screws down until good timing is achieved. You will need a collet type pin vice to refit screws once filed down.

Sent from my SM-G920F using Tapatalk

[sstakoff]

Excellent point @digginstony - the behavior of the watch in all positions can be a key factor in these types of issues.

[Cad101]

Looking at your picture your regulator pins seem to fall on the inner coils of the hairspring not the outer coil is this the case or is it just the picture. has the regulator arm been changed.

[Cad101]

Oh I see it now its a offset hairspring forget what I said.

[jdm]

Sorry I know zilch about antiques but how's the HS held on the end stud? Can it be repinned a tad longer?

However from my experience if the timegrapher or app shows poor or no pattern, there is no point in trying to regulate.

[digginstony]

See pictures below of a Waltham mod 1908 hair spring. The study is friction fitted between 2 legs. Often you will observe a slight kink before and after the study to prevent any possibility of the students moving. One assumes this to be done by the person regulating the movement

Sent from my SM-G920F using Tapatalk

[digginstony]

Should say stud not Study !!

Sent from my SM-G920F using Tapatalk

[digginstony]

And not students should be stud. Think this phone needs to go in the bin !

Sent from my SM-G920F using Tapatalk

[Cad101]

Has the watch been de-magnetized at all ?

[digginstony]

Working on the assumption it has. If not and a de-magnetizer is not available. Take a very small screwdriver and without damaging the coil of the hair spring, try pushing the coils together and see if any sticking occurs. If so, magnetism is the prime suspect

Sent from my SM-G920F using Tapatalk

[matabog]

Check the regulator legs-hairspring in-between space. The hairapspring has to be in the middle and oscilate from one leg to another. With a Breguet hairspring I think this issue is critical! The HS also has to stay in the middle throuhout the regulators' arc

Edited July 8, 2016 by matabog

[StuartBaker104]

If you have a random pattern on the timegrapher then there is no point trying to regulate.

Mark has some good videos about troubleshooting escapement issues. Checking operation in positions might lead you to a quick diagnosis, but if not you need to go through the whole process step by step. If you prefer a written explanation then these might help you...

http://www.tp178.com/jd/watch-school/5/article[1a].html

http://www.mybulova.com/sites/default/files/file/Joseph Bulova School of Watch Making - Unit 10a.pdf

You should also check that the pallet and balance pivots are not bent. As matabog says, the hairsprng must be free between the regulator pins, but only just, however it sounds like your issue is further upstream.

S

[clockboy]

If when you adjust the regular but it is not making any difference then it can not be touching the hairspring. I would check this first you will need really good magnification. I did come across this problem a few years ago.

If I remember correctly I found that one of the regulating pins was bent.

[oldhippy]

Have you oiled the pallets if so where and what oil are you using.

[Scouseget]

Thanks for all your responses. I have spent a lot of time on this as it's my "trial by fire" way of learning things so it's time well spent even if it doesn't actually fix a problem, which is the case with the Waltham. To recap what I've done so far.

I installed a new balance staff.

I disassembled, cleaned, and lubricated the train and escapement - again. Lubrication was with D5 for the center and third wheel, and 9010 for everything else, including the pallet stones faces. I did not lubricate the pallet fork pivot jewels.

I installed a new hairspring and ensured that it was centered between the regulator pins.

I adjusted the hairspring collet to center the roller jewel correctly.

I exchanged the balance pivot jewels with those of the spare Waltham.

I actually think that the problem might be the balance pivot jewels as there appears to be a fair amount of slop, even though I tried the spare Waltham's jewels. Is there a measurement method I can use to ascertain if this is actually the case, perhaps some type of gauge for the jewel holes? I do have both a micrometer and a vernier caliper for the pivots themselves.

I have purchased yet a third Waltham from my very best friend EBay, and will use this to cannibalize parts from when it arrives, in the meanwhile, I'll set it aside and cogitate over it.

Thanks again for all the great advice, and please keep it coming if you can think of anything else I should try.

 

[jdm]

This is a lot of work that you did. In my opinion when chasing problems like the most important objective is the knowledge and manual ability that you develop, more than achieving the goal of having something work perfectly, thing that for many reason beyond one's control, may simply be impossible.

[digginstony]

Going right back to the beginning of the thread. Have you demagnetised ??? A complete scattered pattern on a timegrapher and gaining time is a classic symptom. If you have 1 reasonable line and one scattered or wavy, is indicating problems with the pallet stones to escapement wheel ???

Sent from my SM-G920F using Tapatalk

[Scouseget]

Sorry, I forgot to mention that I did indeed demagnetize the movement, twice actually.

Also, I did get it to work for a while quite accurately dial up, but the moment I moved it away from the DU position, it would noticeably slow down or stop; I verified that the hairspring was not catching on anything when this happened. Not sure what that means exactly but I did check the balance jewel holes and noticed the slop that I mentioned above, hence the reason I bought a third Waltham to allow me to change out the jewels. 

What is the usual reason that a movement changes its behaviour in different positions? And should we always regulate the movement in these different positions, and if they're out considerably, what should be done about it? I realize that this is a loaded question, and perhaps something that requires a lengthy answer, but I'll ask it anyway.

Thanks again for the help and support.

[digginstony]

There are guy's on the forum far better to answer your question than me. But basically a watch to keep reasonable time keeping in all positions every moving parts in the watch is interrelated with the other and adjustment to be perfect. The weakest or most vulnerable part of the watch is the balance assembly it's also the part that does the most work within the movement. Also the balance has to cope with gravitational pull. The movement of the balance staff within the jewel hole should not exceed a third of the diameter of the pivot for correct function. A brand new movement can show slight variations at different positions but your Waltham movement is 100+ years old ! I think you've done a marvellous job but I think it's time to consider what's acceptable for you . If you can get the watch to run in positions that you can use and enjoy the watch, then to me that's job done. If in those positions it's running fast, remember it's not a solid wheel balance, it has screws that act to weight the balance and you can increase the weight to slow the watch as previously described.

Sent from my SM-G920F using Tapatalk

[digginstony]

But do check the pallet stones are in the correct place when the wheel balance comes to rest. There are two inspection holes on the bottom plate on the model 1908.

Sent from my SM-G920F using Tapatalk

[digginstony]

  On 7/12/2016 at 8:21 PM, Scouseget said:

Sorry, I forgot to mention that I did indeed demagnetize the movement, twice actually.

Also, I did get it to work for a while quite accurately dial up, but the moment I moved it away from the DU position, it would noticeably slow down or stop; I verified that the hairspring was not catching on anything when this happened. Not sure what that means exactly but I did check the balance jewel holes and noticed the slop that I mentioned above, hence the reason I bought a third Waltham to allow me to change out the jewels. 

What is the usual reason that a movement changes its behaviour in different positions? And should we always regulate the movement in these different positions, and if they're out considerably, what should be done about it? I realize that this is a loaded question, and perhaps something that requires a lengthy answer, but I'll ask it anyway.

Thanks again for the help and support.

Hi,

Further to my post I've done some timegraph readings for you. See pics.

The first one is a cheap watch i bought from Amazon. Manual hand wind, cost £10.99. This has issues from the picture and the timing of the pallet stones seems to be going far to deep into the escape wheel and generally out of time.

Then I have my swatch. It's about 10 years old. It's fitted with a quality ETA movement. Never been out of its case. Been to the middle east with me several times. Don't know the water proof rating for the case, if any but has been under a water fall in the Dominican with me. No issues with this

Don't know the lift angle of these watches so used the standard 52° setting. But I suspect the Waltham may be quite different and that might explain the difference between your graph software and stop watch. Hope this helps for future reference.

Tony

Sent from my SM-G920F using Tapatalk

[Cad101]

Very old watches can be a pain having no incabloc system which helps hold the staff in a more stable position and help keep the balance in position when gravity takes hold.

I would firstly check.

1  you have the correct staff does it move excessively up and down and side to side

2 Is the balance wheel original to the watch or a replacement it may not be poised correctly

3 your jewel holes would have worn over 100+ years check your jewels for excessive side movement ( If you do have the correct staff )

4 Has your balance cock been packed  or is there any flea bites under it.

5 Is your mainspring new or the one that came with the watch it may not be transferring enough power.

6 Does your hairspring lay centrally between the regulator pins when in the normal position before any regulating takes place and before the hairspring starts vibrating making hit the inside pin when coiling and the outside pin when uncoiling ( is it just a over coiled hairspring or true Breguet hairspring with a flat portion before the coil starts)

just a couple of things I've had to deal with in the past

[jdm]

  On 7/14/2016 at 9:23 AM, digginstony said:

Don't know the lift angle of these watches so used the standard 52° setting. But I suspect the Waltham may be quite different and that might explain the difference between your graph software and stop watch. Hope this helps for future reference.

Expand  

If the lift angle is not set correctly on the timegrapher the only difference is a very minor inaccuracy in calculating amplitude. Nothing else is affected.