Jump to content

Recommended Posts

  • 3 years later...
On 2/5/2019 at 12:45 AM, Pip said:

I let the movement run from a full wind straight after reassembly (which was showing at around 240amp and about +1spd) for around 24 hours and measured the following:
https://uploads.tapatalk-cdn.com/20190204/890fc4dea7710769f8b425b3cdafd86a.jpg

The problem with the Weishi 1900 timing machine (and possibly all timing machines, but that's the one I have) is that you can get significantly different results depending on the gain setting and how you place the movement on the microphone stand. The other day I measured a newly serviced 2892-A2 with the movement attached directly to the microphone stand and got around 220 degrees in the dial-up position. Wise from experience, I placed the movement holder (holding the movement, of course) on the mic and got 300 degrees. Video recording the balance in slow-motion the 300 degrees proved to be the correct reading, and I would get the same by lowering the gain on the TM with the movement attached directly to the stand.

TMs are helpful but can be deceiving.

Link to comment
Share on other sites

9 minutes ago, VWatchie said:

The problem with the Weishi 1900 timing machine (and possibly all timing machines, but that's the one I have) is that you can get significantly different results depending on the gain setting and how you place the movement on the microphone stand.

I don't have any experience with Chinese machines with a gain settings so that I don't know.

having looked at schematics for older timing machines they typically have automatic gaining controls built in. Even though there is a volume knob they still have an automatic gain control. Looking at the manual for the witschi machine that I have you tried on your set the knob at the middle position and I don't really remember ever having a problem with that. The manual does state that there's an automatic gain control so basically I don't remember having to adjust because it adjusts itself. Now the machine at work the very expensive witschi only has for you the level the numbers 1 to 4 and at least it has about oscilloscope so for having a timing issue will look at the oscilloscope and we can see why if there's a single problem. Usually with really quiet watches will just increase the number and I'm guessing because the automatic gain control it doesn't seem to make a spectacular difference of was the watches really quiet or the other problem.

The biggest problem at all for timing machine seems to be picking up a good clean signal. So holding a watch in the timing machine pick up holders critical anything that interferes with that is going to disturb the signal while the signal is an audio frequency range it's not picking up audio it's picking up the vibration so anything that interferes with that will mess up with the timing machine sees., That workers with Seiko watches with their plastic movement holding ring case stop the watches will look horrible out of the case they look beautiful. But is not just Seiko watches things like a heavy Rolex watch or any heavy metal case it becomes sometimes problematic to pick up the watch. Sometimes if you put the crown to touching the pickup if you like you get a better signal otherwise the watch has to come out of the case

1 hour ago, VWatchie said:

TMs are helpful but can be deceiving

 

then unfortunately that is a true statement. It's why you need to visually verify that the watch looks like it's doing whatever the timing machine says. The graphical display has to agree with what you're seeing and then the numbers have to agree with that also. So to see random dots on your machine you have a problem the numbers are worthless and you're going to have to troubleshoot the watch. If you go to a super low amplitudes more common on a Chinese machine or the machine at work even if it is more expensive if the amplitude is too low the machine will read the wrong part of the waveform a give you a nice happy amplitude which isn't there. That's where visually observing does it look like it has a nice amplitude versus what the machine tells you. so yes unfortunately you have to be careful on how you interpret what you're seeing and figuring out whether you're actually having a problem with you and your machine or whether it's the watch that's the problem.

  • Like 2
Link to comment
Share on other sites

Thanks for sharing your experience and insights John! Judging by what you say I believe (but now I'm guessing) that the Swiss Witschi machines have more or less solved the problem by adapting automatically to the signal they get. Unfortunately, the Chinese Weishi machines don't seem to be that sophisticated, so we will have to experiment with the gain setting to get near the truth. Based on my experience, it would seem that the gain setting giving the highest amplitude is the correct setting. And, at least to me, that makes some sense if the watch has been serviced well and is also keeping good time.

I'll see if I can take some pictures tomorrow to illustrate the rather dramatic differences between the various gain settings on my Weishi 1900 and the ETA 2892-A2 I mentioned.

Before diving into fault finding it is extremely important to verify that the TM is showing the correct amplitude. Making a mark on the balance that is aligned with the impulse jewel and then recording a slow-motion video of the balance works pretty well, but it can still be difficult to determine the actual amplitude of high-beat movements having small balance wheels.

It would be so very interesting to try out a Witschi.

Link to comment
Share on other sites

10 hours ago, VWatchie said:

Thanks for sharing your experience and insights John! Judging by what you say I believe (but now I'm guessing) that the Swiss Witschi machines have more or less solved the problem by adapting automatically to the signal they get. Unfortunately, the Chinese Weishi machines don't seem to be that sophisticated, so we will have to experiment with the gain setting to get near the truth. Based on my experience, it would seem that the gain setting giving the highest amplitude is the correct setting. And, at least to me, that makes some sense if the watch has been serviced well and is also keeping good time.

it's interesting is I've done side-by-side comparison with the 1000 1900 and the witschi that I have and I don't recall any problems but then I didn't spend a lot of time with at least 1900. I do have a Chinese 1000 though but mine doesn't have a volume adjustment maybe that's the problem.

On the old paper machines was really important where you set the volume setting even though I do know they had automatic gain control. On the other hand the old paper tape machines were measuring amplitude they just needed to get a clean signal for timing. In the preferred signal is for all the machines the roller jewel hitting the fork which is the quietest sound which is why need the most gain.

When you're playing with your volume settings when it should try another watch and see if that changes anything at the same time. It's when you're doing stuff like this that's why like the oscilloscope feature with the machine at work because I can look at that and see what's going on I get a clue what the timing machine is thinking or why it's not timing things correctly..

out of curiosity and looking at the manual for the 1900 and it claims it will automatically adjust for optimum signal level during testing. now the interesting problem is other then it explained that one arrow makes the whatever go up the other arrow makes it go down there is no hint as to where the volume or gain setting should be a. As I said on my witschi Turner on with the volume knob and turn it to the middle position and you just leave it there and I just don't recall ever having to move it around very much.

 

  • Like 1
Link to comment
Share on other sites

17 hours ago, VWatchie said:

t would be so very interesting to try out a Witschi.

I was looking to see if there's any distributors near where you live and not in your country just neighboring countries. But there's probably watchmakers somewhere nearby maybe could ask nice and see what it looks like on their machine.

https://www.witschi.com/en/contact/standorte/

 

2 hours ago, Kalanag said:

‘Id like to mention that my Weishi 1000 has the same gain adjusting feature (arrows up and down) as the 1900. Other Weishi 1000 owners have been reporting this too.

a classic problem with Chinese products is variations lots of variations. Somewhere in the group we've had this discussion before most of the 1000 machines do not have that adjustment. Mine for instance does not have that adjustments. But some of the machines do have the adjustments. Then unfortunately they do not put revision numbers on the machine to tell us which machine is which. Plus the user manuals tend to not cover any of the stuff either so some new some don't. It tends to be all kinds of variations with Chinese equipment and you never know until you actually get it what you're going to get. And I saying that from experience with other products also

 

  • Like 1
Link to comment
Share on other sites

IMG_9524.thumb.JPG.0444a341ae6ae02fa34fa6bfae77c445.JPG
This is the reading from my newly serviced ETA 2892-A2. As you can see, lowering the gain setting shows an amplitude of 296 degrees, and looking at the balance in slo-mo this seems to be near the truth.

IMG_9531.thumb.JPG.42171555f45045b2122da2d909e37fe5.JPG
Setting the gain to its default value the amplitude drops by almost 25 degrees.

IMG_9530.thumb.JPG.08b37db01425d10be4c6f27062aa5d9c.JPG
Finally, when the gain is set to its maximum the amplitude drops by another almost 40 degrees.

I have also seen the opposite, especially with Seiko movements. That is, to get the correct amplitude I would have to increase the gain by a couple of steps.

So, before starting any fault-finding I always try the various gain settings to see if I can get a decent amplitude. If not it is time to start the fault finding. Also very good and true advice from @HSL " If you measure the amplitude just after servicing you will get a lower reading which is natural. "

4 hours ago, JohnR725 said:

But there's probably watchmakers somewhere nearby maybe could ask nice and see what it looks like on their machine.

Good idea John! To tell you the truth, I have called a few watchmakers on various matters but have always been met with scepticism. Maybe it's because I'm honest and tell it like it is, that I'm just an amateur. They refuse to even consider the problem I describe and say that only if I take the watch in for a complete service will they be able to help me. Very disappointing indeed! They probably think I'm a stingy tinkerer trying to repair my own watch with the tools in my toolbox. Well, there are some very reputable watchmakers here in Stockholm and if I visit them in person I might be able to get better treatment.

Edited by VWatchie
  • Like 2
Link to comment
Share on other sites

5 hours ago, JohnR725 said:

I'm curious as to is this problem occurring on all your watches or just this watch?

I'm afraid the best answer I'd be able to give would be, most watches.

Here's another "funny" thing. Yesterday I cased the ETA 2892-A2 movement. Just before casing it I measured it a final time and got 310 degrees in dial down position. Then, after having cased it, I measured it again and the amplitude then started to fluctuate between 220 and 265 degrees regardless of what gain setting I used. The rate was still consistent.

Measuring cased movements doesn't seem to work well at all. At least not when it comes to amplitude. The rate and beat error seems correct though. I remember well that it was you John that advised us to always measure the movement outside of the case, and I think that is solid advice. My theory is that the housing and/or the movement fixing ring, etc.,  sometimes, but not always, is interfering with the reading.

I've learned to take my WeiShi 1900 with a huge grain of salt. I really don't trust the amplitude reading on it. And I should mention that it almost killed my interest in servicing and repairing as it was giving me poor results after having put down every fibre of my heart and soul into a service. It was only after a tip from @Endeavor to record slo-mo videos of the balance that I realized I couldn't always trust the amplitude shown on my timing machine, and that kept my interest going. I wasn't a such a mechanical idiot after all!

Link to comment
Share on other sites

@VWatchie, when I want to test amplitude on a cased watch, I usually find best results by having the crown of the watch against the metal pick-up of the TG. It doesn’t always work, but most times I get similar amplitude readings as the uncased movement. 
 

OT, but I’m researching this exact movement on an upcoming project so thanks for the info provided regarding expected performance.

  • Like 1
Link to comment
Share on other sites

17 minutes ago, gbyleveldt said:

when I want to test amplitude on a cased watch, I usually find best results by having the crown of the watch against the metal pick-up of the TG.

I share your experience, but I've tried all sorts of positions just to see what effect it might have.

18 minutes ago, gbyleveldt said:

OT, but I’m researching this exact movement on an upcoming project so thanks for the info provided regarding expected performance.

My rule for acceptable amplitude is at least 250 degrees in all four vertical positions and at least 250 degrees (yes, 250) in both horizontal positions.

  • Like 1
Link to comment
Share on other sites

I‘d like to share the behavior of my Weishi 1000 when altering the gain. When testing my JLC in vertical position with max. and min. gain I saw the following displays:

AA67AA07-AC30-4E55-96A0-8EBAE4BAE33B.thumb.jpeg.992cfdb17be4aac0d774004a0b2d6908.jpeg

 

BD46F51A-32AB-420C-B744-87EFF438B8AF.thumb.jpeg.47576c5e8496e84c7280d6a8a84d9ffc.jpeg

As you can see the amplitude is not affected but the beat error is calculated wrong with the low gain!

 

Edited by Kalanag
  • Like 1
Link to comment
Share on other sites

4 hours ago, VWatchie said:

I share your experience, but I've tried all sorts of positions just to see what effect it might have.

My rule for acceptable amplitude is at least 250 degrees in all four vertical positions and at least 250 degrees (yes, 250) in both horizontal positions.

Good luck with Seiko's !  With some of mine, I consider 230° as 'good'.  

For (my) older watches with some wear, I consider 240° acceptable DD/DU and 220° vertical. Then again, you get an old watch, which looks to have had a good working life, and it fires up with a steady 270-300°.

  • Like 1
Link to comment
Share on other sites

29 minutes ago, mikepilk said:

Good luck with Seiko's !  With some of mine, I consider 230° as 'good'.  

Oh yes, I would definitely agree with that. I really should have mentioned that I was thinking about typical Swiss movements such as ETA, Sellita, Unitas, AS, etc.

31 minutes ago, mikepilk said:

For (my) older watches with some wear, I consider 240° acceptable DD/DU and 220° vertical. Then again, you get an old watch, which looks to have had a good working life, and it fires up with a steady 270-300°.

Good and true input! IMO we should be concerned about the amplitude but not too concerned.

  • Like 1
Link to comment
Share on other sites

21 minutes ago, VWatchie said:

Oh yes, I would definitely agree with that. I really should have mentioned that I was thinking about typical Swiss movements such as ETA, Sellita, Unitas, AS, etc.

Good and true input! IMO we should be concerned about the amplitude but not too concerned.

Wise words @VWatchie 😀

And, don't get too concerned with beat error, especially without an adjustable stud holder. 

I'm intrigued by @Kalanag's problem with gain.  According to Witschi to first beat is used for rate AND beat error, so it's hard to see how one can vary and not the other.

Maybe with low gain, it's missing the first beat and sometimes detecting the second or third (which is used for calculating amplitude). But then you would expect variation in rate AND beat error 😧

image.png.5f1158854fe767492840199551d7c554.png

  • Like 1
Link to comment
Share on other sites

8 hours ago, VWatchie said:

My rule for acceptable amplitude is at least 250 degrees in all four vertical positions and at least 250 degrees (yes, 250) in both horizontal positions.

you would have been disappointed with the lecture I attended yesterday. The lecture was understanding what goes into servicing a Rolex in a modern Rolex service agent shop. so what the purpose this lecture is is to explain to people why it costs so much to get a Rolex serviced everything that goes into it parts that are changed etc. Then for timekeeping they don't actually follow Rolex specifications. Rolex will accept usually negative number like -1 and of course except zero seconds but this place will not it has to be plus something. In a previous version of this lecture several years ago the watch they showed timing specifications for was keeping zero seconds per day and he said that would fail quality control and have to go back and be redone for timekeeping. then amplitude 310 maximum fully wound up and half wind which is 24 hours ground down minimum of 200°.

so typically when you see the spec sheets that list timing specifications they are not obsessed with a specific amplitude other than the minimum maximum and of course they are more obsessed with timekeeping.

3 hours ago, mikepilk said:

I'm intrigued by @Kalanag's problem with gain.  According to Witschi to first beat is used for rate AND beat error, so it's hard to see how one can vary and not the other.

Maybe with low gain, it's missing the first beat and sometimes detecting the second or third (which is used for calculating amplitude). But then you would expect variation in rate AND beat error

technically that isn't quite correct. Everything for timekeeping is depended upon the first sound which is the roller jewel hitting the fork. Unfortunately that's also the most quiet sounds.  then the sounds are more complicated with the typically show although they do have a breakdown of the sounds that I've snipped out images for you. So you can see there's five sounds two of them are supposed to overlap. But what if they don't overlap like what if locking occurs and is a time delay before it crashes into the banking pins. Or what if something else is going on or any other external noises or noises in the watch what is that due to your signal.

This is where it be nice if the Chinese machines had audio where you could listen and see if the signal sounds nice and clean. It's amazing how noisy the hairspring can be if it bumps into a rubs on something.

then the problem with the example of how they measure amplitude of beats said right is officially an absolute perfect idealized waveform. What if the waveform isn't perfect then what happens. As you mentioned if your gain is not high enough and you don't pick up the first signal that everything's going to be really screwed up. Or if you're amplitude is superlow the waveform get stretched out the timing machine will look for locking at the wrong place pick it up in the middle which is quite common on the Chinese machines and their very expensive witschi at work with bikes to do that quite often and then you end up with an amplitude that's much higher than what it should be.

timing signal page 3.JPG

timing signal page 2.JPG

timing signal page 1.JPG

Witschi Training Course.pdf

  • Like 3
  • Thanks 2
Link to comment
Share on other sites

  • 4 weeks later...
On 10/23/2022 at 12:00 PM, VWatchie said:

Yesterday I cased the ETA 2892-A2 movement. Just before casing it I measured it a final time and got 310 degrees in dial down position. Then, after having cased it, I measured it again and the amplitude then started to fluctuate between 220 and 265 degrees regardless of what gain setting I used. The rate was still consistent.

Gentlemen and ladies, I'm embarrassed to say that the reason for the sudden drop in amplitude was not the fault of my WeiShi 1900, but my fault. The tip of the seconds hand was ever so slightly touching the inside of the crystal at certain points, making the amplitude fluctuate. Once rectified and the gain setting set to its default value, the TM gave the same results with the movement cased.

  • Like 2
Link to comment
Share on other sites

11 minutes ago, VWatchie said:

Gentlemen and ladies, I'm embarrassed to say that the reason for the sudden drop in amplitude was not the fault of my WeiShi 1900, but my fault. The tip of the seconds hand was ever so slightly touching the inside of the crystal at certain points, making the amplitude fluctuate. Once rectified and the gain setting set to its default value, the TM gave the same results with the movement cased.

Never be embarrassed to say you’ve made a mistake and things are your fault. We're all human watchie and it happens to the best of us. Admitting fault never hurts honor only enforces to show that you have plenty of it. Pleased to hear you've sorted it out and happy to have you back matey 👍

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.



×
×
  • Create New...