Adjusting a Timex

[Shane 13]

I have a ladies seventeen jewel, spring driven Timex with a dial marking of 6201017178.  I am trying to regulate the watch and am finding the auto rate detection function on the software I'm using is having a problem deciding what rate it should be compared against.  I have found NO information online supporting a 171 movement.  It would be great if someone can shed some light on this movement.  I can only get it within about +-20 sec/day before it drastically flips from losing time to gaining substantial time and then I chase it back again.  After this happens once or twice, the software seems to get lost and needs to be restarted but periodically chooses a different frequency to reference.  I would like to lock it into only one frequency and try again.  I am assuming it is a combination of it's very faint ticking and also being out of beat that is confusing things since I have never had this problem with any other "louder" watch.  Please correct me if my grasp on this is incorrect but since it looks like the rate moves (keeping it's relationship with both ends of the hair spring) while adjusting the beat angle, it would be best to adjust the rate before the beat.  I would of course like to upgrade all of my equipment but would, at this time, would find a more professional and dedicated unit for this function difficult to justify.

Thank you for considering my problem.

Shane

[jdm 2,221]

3 hours ago, Shane said:

. I can only get it within about +-20 sec/day before it drastically flips from losing time to gaining substantial time and then I chase it back again.  

Very likely the mov.t is in poor conditions, with insufficient amplitude for decent timekeeping and reading by the application, which has no fault.

In that case a complete cleaning and oiling is needed to try to get  it back to acceptable performances. 

There are various running thread and at least one member expert on Timex working, these should be your main resources. 

[JohnR725 1,357]

5 hours ago, Shane said:

auto rate detection function on the software I'm using is having a problem deciding what rate it should be compared against.

A lot of the software timing apps have issues with basically just working. Then ladies watches tend to be really quiet which is a problem. If you can't pick up a nice clean signal then definitely an issue. What rate or frequency does your app think the watch is?

It would be nice to have a picture of the movement looking straight down in detail.

[Shane 13]

Apparently the watch I am currently working on is actually a DuRoWe INT 1980.  It should run with a beat rate of 21600 but periodically my app incorrectly identifies it as 19800.  Since this app doesn't seem to have the ability to determine a watches amplitude, all I can say is that it is osolating quite enthusiastically.  I put a dot of magic marker approximately 180 degrees from the balance jewel and it appears to osolate 180+ degrees in both directions from that point.  I do know that a subjective observation without being followed by accurate numbers has very little value in trouble shooting but it's all I can say about the amplitude without getting better tools.  Does anyone have a recommendation for an Android app that seems to work for them.  I will do some more research on amplitude and it's influences.  If anyone knows of particular information on the subject that should be read, herd or viewed, I would appreciate a link.

Thanks.

Shane

[Shane 13]

[HectorLooi 229]

Nice movement. Looks Swiss. Do you know where the movement originates?

It has a proper jewelled balance, not the usual Timex pin pivot. How about the escapement? Is a pin pallet?

[jdm 2,221]

On 2/20/2021 at 4:02 AM, Shane said:

Since this app doesn't seem to have the ability to determine a watches amplitude, all I can say is that it is osolating quite enthusiastically.  I put a dot of magic marker approximately 180 degrees from the balance jewel and it appears to osolate 180+ degrees in both directions from that point.  I do know that a subjective observation without being followed by accurate numbers has very little value in trouble shooting but it's all I can say about the amplitude without getting better tools.

Unfortunately if one is serious repairing and even just regulating mov'ts need to spend the money for a cheap timegrapher at least.

What you can do in the meanwhile is take an high frame rate video and post here. The marking will help experienced eyes to make a call. If amplitude is really 180 deg that is too little for good timekeeping, especially when worn and moved around.

[Shane 13]

The manufacturer was to my knowledge a German company and has a Swiss anchor, a full seventeen jewels, no pins.  It's a nice little movement just a bit small for my current skill level.  I have had it completely apart and back together before initially posting.  Since, I have had the balance and it's jewels out at least twice, shoting the Incabloc clips across the bench at least twice per side on each occurrence.

I couldn't see it being a main spring problem since the amplitude didn't change when gently influencing the center wheel so I stripped and cleaned the whole thing again.  At each location, I chose a lighter oil while reassembling the second time and the amplitude improved significantly.  It still shows a small (almost acceptable) variation in rate but I still can't seem to adjust out the beat error.  I have heard magnetism can show up in odd ways so I have arranged for a borrowed demagnetizer.

I do have so many places to spend money with this hobbie.  I just feel that a dedicated piece of hardware for adjusting away last bit of error is not the best place for me to invest right now (but I will at some point).

Thanks.

Shane

[JohnR725 1,357]

On 2/19/2021 at 7:02 PM, Shane said:

I do know that a subjective observation without being followed by accurate numbers has very little value in trouble shooting but it's all I can say about the amplitude without getting better tools.  Does anyone have a recommendation for an Android app that seems to work for them.

I've quoted too many things and some of them are out of order

here's the problem with your apps. This is based on the people who answer questions on the group we had been led down the path of they don't work very well. If it's all you can afford fine but bad information leads to bad diagnostics. It's basically wasting all of our time so toss the apps and do it the old-fashioned way. Visually you'll have to do it that way. Yes you really can't exist without a app or a timing machine sort of. There are some things a timing machine will show that you cannot see with your eyes at all.

42 minutes ago, Shane said:

It still shows a small (almost acceptable) variation in rate but I still can't seem to adjust out the beat error

The variation in rate a small amount who cares? Is this a Rolex is this a chronometer certified timepiece. Mechanical watches are in perfection devices. What they do really beautifully as they average those problems over time. This is where people would get excited with their timing apps are timing machines when they see stuff but over time they usually get averaged out so we don't care about the rate

beat error even before you have a timing machine you should understand what it's doing what you're trying to do and visually be able to verify that. For one reason even with a timing machine it is so easy to go past zero part and continue in the wrong direction because there's no plus and minus. So it's usually always good to be visually really close. We do have other discussions talking about visually putting it in beat. You can do a search for those otherwise we can find one for you.

43 minutes ago, Shane said:

I just feel that a dedicated piece of hardware for adjusting away last bit of error is not the best place for me to invest right now (but I will at some point).

I covered this above you need to visually put the watch in beat so you're right you don't need a timing machine. But timing machines are interesting devices there now available cheaply versus at one time where hobbyists could not afford them commercial machines used to cost thousands of dollars and still do and there were none of these Chinese machines or somewhat worthless apps out there. So just remember if you're using worthless apps at some point in time we may not help you diagnose a problem if we perceive that it's the apps problem and not the watch problem. But for now you can go without it learn how to do things the old-fashioned way it be much better for you anyway.

On 2/19/2021 at 6:02 AM, Shane said:

software seems to get lost

Notice were more or less I clipped all the same things. Your software sucks okay the software could be fine the classic problem with all of the apps is you need a decent pick up and that's one of the major problems if you can't pick up what you'd need to use for the mid-scene or the software then all of them are totally worthless. This means you can't use your software for diagnostic purposes you have to visually look at the watch like if you wind it up and put hands on it and look at it 24 hours after you did so and after use that the time with something that actually keeps time. A lot of times people use time references that are worthless and that leads to confusion so 24 hours from now what is your watch doing for timekeeping?

On 2/21/2021 at 2:12 AM, jdm said:

Unfortunately if one is serious repairing and even just regulating mov'ts need to spend the money for a cheap timegrapher at least.

This is where the reality of the group is that the timing machine the Chinese 1000 or 1900 is a worthwhile investment. As I said somewhere up above it lets you see things you can't see any other way. It seems to have a very high reliability of diagnostics and yes it does have its faults but it's so much better than the apps. Having done side-by-side comparisons with both these machines and expensive Swiss witschi timing machines that cost thousands of dollars they do really really well. But you also have to have tweezers proper lubrication something to look with something to hold your movements. Just don't expect proper diagnostics until you get a proper timing machine or learn how to do it the old-fashioned way.

Then timing machines are expensive but what about lubrication isn't that expensive what lubricants are you using on this watch?

 

[JohnR725 1,357]

Then I realized that all the quotes I had too much up there anyway. Getting a demagnetizer is good this make sure you learn how to use it otherwise you can magnetize your watch.

If you're amplitude is below at least 200° preferably higher then you don't necessarily worry about anything else. The reason for this is as the amplitude decreases problems look much much worse. When you have low amplitude beat errors can increase dramatically. Then typically do see everyone recommending 250° the minimum but unless you get that amplitude up beat error is not going to be concern except?

On 2/19/2021 at 6:02 AM, Shane said:

Please correct me if my grasp on this is incorrect but since it looks like the rate moves (keeping it's relationship with both ends of the hair spring) while adjusting the beat angle, it would be best to adjust the rate before the beat. 

One of the problems with movable stud holders is that as you've observed moving it moves everything. The fantasy is if they're designed correctly that you can move them independent of each other. But that isn't always the case.

I would start by visually putting this watch in beat.

That is providing we can get the amplitude up. But if you're grossly out of beat which I don't think you are then that takes priority because if it's adjusted to the maximum of one end or the other which visually watch does not look like it is then that's going to affect the amplitude. To a certain degree everything affects everything.

So the easiest thing to do now is visually put it in beat and don't worry about the timekeeping. There were also going to worry about amplitude but because your visually putting out of beat and not using a timing machine then amplitude isn't going to upset what you're seeing. If you're using a timing machine I mention this above horrible amplitude will make everything look worse and that would have to worry about the amplitude.

[HectorLooi 229]

A Chinese timegrapher is really affordable. If you were planning on only regulating this one watch in your lifetime,  then getting a timegrapher is probably an expensive toy. But I saw that you have a whole box full of watches in your original introductory post, so you will be needing a timegrapher sooner or later. 

And as JohnR725 said, oils are expensive. Just getting a basic set of oils will cost more than the timegrapher. 

But a word of caution, if you have other Timexes in your box, the type with pin pivots and pin pallets, don't expect to see a nice single line trace on your timegrapher. That's not going to happen. 

[Shane 13]

I would like to say that I appreciate all the input.  It shows significant interest and willingness to help.  Thank you.

Yes, I know using an app yelds the lowest quality information from the audible signal possible and are justifiably looked down upon by those with other options.  Duley noted.  I have always tried to put all the watches I work on visually as close to in beat as possible during their assembly and attempt to confirm it dynamically once running, adjusting if necessary.  As previous mentioned, with this ladies Timex, I found it small enough to make this and other efforts difficult and I was looking to improve on something that I could not see.  After re-reassembling, the current error is now well within what I personally would and have excepted in the past (less than 60sec over the last 24 hours).  It is really is amazing the finer your ability to measure your problems the bigger your problems look in comparison.  I have gone back and "retested" (for lack of a better term) other movements and have not found this inability to resolve the rate and/or beat among them.  My curiosity is mainly within the anomalous behavior of this movement and I was looking for input.  If we all had the same knowledge, had the same skills, used the same tools and solved problems the same way, we would have nothing to aspire to and no reason to talk to others about it.

PS.

On the first assembly, I used Moebius 9501 on all the keyless work.  Keystone light mainspring oil on the mainspring.  Moebius 9010 on the train.  Moebius 941 on balance jewles and pallet stones.

I was concerned about the Keystone oil being to thick while assembling but influencing the center wheel didn't change anything.

On the second assembly, I stayed with Moebius 9501 on all the keyless work and Moebius 941 on balance jewles and pallet stones.  I changed to a lighter clock oil on the mainspring, the label fell off some time ago (I would need to look it up) and Moebius 941 on the train.

[JohnR725 1,357]

On 2/19/2021 at 7:02 PM, Shane said:

DuRoWe INT 1980.

It looks very close to the link below except the one below has a Incabloc Protection system and yours is not. Also found the link below timing specifications which look pretty good. Then based on the link below to have to start over with my message because timing looks pretty good for Timex doesn't it? So obviously you can do better we just need to figure out why yours isn't doing with this one is?

https://17jewels.info/movements/d/durowe/durowe-1980-int/

A lubrication looks fine we could probably get nitpicky over something but it should work so that's probably not the issue

7 hours ago, Shane said:

not found this inability to resolve the rate and/or beat among them. 

Think I lost track of where we are in this discussion?

Plus I was going to suggest a quality problem in other words this is inexpensive watch normally its quality may not be the same as a more expensive watch and that quality difference can affect timekeeping. But the above link indicates that it should be of reasonable quality except? Yours isn't exactly the same as that we might still have a quality issue.

The beat error is going to be a problem if it's visually set where it's supposed to be then it's fine if whatever app you're using says it's not fine minutes the apps problem.

Timekeeping could still be a quality difference. For instance if the wheels are not perfectly round you'll get uneven power transmission to diagnose some of that requires more expensive timing machines. You can sometimes tell if you watch the rate and as you're watching you'll see it go up and that on that's usually a power problem through the gear train.

Before the software update for the timing machine at work I would run a full cycle. This is with a basically a $10,000 timing machine the full automatic microphone. I run a full cycle look at it push the button run it again in every single time the numbers were all different.

Then they did a update apparently the timing machine is running Windows CE the update brought in the time plot capability. To run a time plot it shows things differently and that's where you can tell if you have wheels that are out around you can see the amplitude fluctuate the time fluctuates and over 24 hours with the hands on probably will never see it ill average out. It just does interesting things on the timing machine if you're not paying attention to what's going on you get all concerned about things that can't really be changed.

 

[watchweasol 1,457]

You are getting some sterling advice from John and his comment regarding apps etc are right on . If you want to operate a timing machine and don't want to run to a timegrapher at this point, consider the PCTM opensource software  or the Watch-O-Scope program, With both of these you need an amplifier and a pick up. I have built the Watch-O-Scope pickup and amplifier and it works very well even the free light version, the licensed version allows printing etc It gives you all the features of the genuine machine although PC based.  It is another option. As John said the timing machine gives a graphic representation of performance of the  watch and picks up things you cant see, Like as mentioned out of round wheels etc

[Shane 13]

That is the page I found several days ago based on the markings on the movement.  I don't know enough about all the different shock systems to have noticed the differences.  I hadn't considered gear to pivot concentricity in something manufactued within a "modern" factory or that it could be off enough to be noticable.  I guess tollorences can stack up to produce bad assemblies at times.  The changes I saw where synasodel in nature.  I was thinking bent piviots but it didn't repeat consistently enough to suggest that any one wheel was a problem.  I dismissed the idea of a shock hard enough to bend multiple piviots without distroying the balance but now not sure.  If a concentricity was off in multiple gears, especially if they all have a hunting pattern, the variation could be quite complex and unaddressable.  Are timing machines capable of showing something problems like that?

[Shane 13]

Thanks watchweasol.  I will look into them.  My frugal nature feels that leveraging something that I already own (my PC) for this task would at least save space if not money.

Shane

[JohnR725 1,357]

Somewhere I've considerably more but this will give an idea. Is a collection of four images from the Hamilton 992B. Fully wound up and 24 hours later.  You can see that there are amplitude variations but timekeeping stays really really tight.

Then there's a image from a 17 size Elgin pocket watch well over 100 years old who knows when it was last  serviced. The dramatic sawtooth the fact about every five minutes that's the mainspring barrel meshing with the center wheel pinion.

Then would you can see with the last four poor quality watch such as the laco  versus Rolex.

On a normal timing machine you'll see this as the rates will slowly change sometimes. Though is if you look at the rate slowly go up and down other times the timing machine may just average at all out you won't see it at all. I find it really fascinating whatever I have a problem is to run a time plot and it's like all that's why I'm getting so much fluctuation in other words if you see a dramatic issue at least on a time plot you'll see it on the timing machine but over time the watch will just average it out usually. Just depends on how The problem is.

 

[Shane 13]

Please do not think I'm being rude or dismissive for not responding.  You have all given me some information that must be digested before I can respond with anything remotely resembling an informed opinion and I have not had much free time.  I have an inquiry in with watch-o-skope and am waiting to hear back from them before going further.  This sounds, for better or worse, as the path that I will try next.

Thanks you for your time.

Shane

[watchweasol 1,457]

No problem Shane   I think Watch-O-Scope would be a good investment as it utilises the PC you already have and a purpose built Amp is cheap enough.       cheers

[jdm 2,221]

So far the good advice about timegrapher, a very important instrument.
But I think that in the end the app the OP was using has done its job already: the watch runs very poorly. There lies the fault.