Very confusing Timegrapher results

[JohnR725]

No idea if they have copy protection it would be nice if they didn't. It be nice to the Chinese could clone it or would clone it or even make a clone of the idea of it. Because once you go to a PC-based timing machine all those nice enhancements become relatively easy to do versus a dedicated machine where where limited.

Even witschi understands that like the machine at work when I did the firmware update a new feature appeared which was their time plot capability that wasn't there before. Just because of software it adds a lot of flexibility the things.

The sad problem with witschi and a lot of Swiss companies are as their cost the machines go up the quantity of buyers go down. The Chinese have demonstrated if you make a timing machine for lesson $200 you can sell them all across the planets it be interesting to know how many of those they've sold. It's now really a must have if you're working on watches to have a Chinese timing machine unless of course you just won the lottery than you can a witschi machine warrior in a business. So I guess we can hope that the Chinese will pay attention to what witschi's doing and give us a really nice timing machine at maybe $500 may be less? Add in all is missing features that would be helpful especially in this discussion like to have in our oscilloscope see what's really being picked up.

[VWatchie]

The main reason I bought this used Seiko 5 Sports watch on eBay ($91/£66/€78) was that I wanted to try my hands at the classic Seiko 7Sxxx Japanese workhorse movement, housed in millions and millions of watches. In this case a calibre 7s36b. The watch was beat up, the colours on the hollows on the bezel were all gone, and the movement was definitely in need of a service. Anyway, I thought that if I could successfully service it, it could become a nice daily beater, and I would learn about Japanese movement design in the process. What an exciting project!

Having put my heart and soul into servicing the movement, repainting the bezel, etc., this is how I was" rewarded", or so I thought. Looking at the graph it was “obvious” that something was seriously wrong, likely something related to the escapement. Being passionate about watches and what I do, I immediately blamed myself and I felt more than disappointed. Depressed would be a better word, especially as I had been relentless in my efforts to get everything as perfect as I possibly could. I wore it for a week or so, and to my surprise it did well, losing just a few seconds per day. However, the memory of that graph kept torturing me until I put it away in my “box of failed watches”. At that time, I just didn’t have the energy to try to deal with it.

Same Seiko watch, same conditions, with the only exception of the gain regulator on the TM being pushed to its maximum, and what do you know. Looks like I didn’t do such a bad job after all, and it’s now merrily ticking away on my wrist. Of course, the only reason I came to think of this Seiko was the Omega (the subject of this thread) giving me, yet again, very distressing TM results.

With the Omega I had to push the gain regulator to its minimum to arrive at the video verified truth. With the Seiko I needed to push the gain regulator to its maximum. And it makes sense. The Omega is very loud. Think "Intro-to-the-60-minutes-show-loud". The Seiko on the other hand is very silent.

So again, how can we trust our cheap Chinese TMs? Well, if we have a watch that we, on good grounds, can determine is in a poor condition (like the examples in John’s post) then perhaps we should be suspicious of the results we’re getting if they are looking too good. However, if we know we’ve done a good job servicing and repairing, and we get a consistent result on the TM after adjusting the TM gain regulator, then I believe we can be fairly sure that what we are seeing is near enough to the truth. And for a few hundred dollars that’s good enough for me. At least for now!
 

[jdm]

1 hour ago, VWatchie said:

Same Seiko watch, same conditions, with the only exception of the gain regulator on the TM being pushed to its maximum, and what do you know.

Yes, that happens sometime. The gain control is there for that exactly.

You should be able to get virtually zero beat error and +5 s/d on the instrument. Then check what rate difference you get crown down, it can easily be minus 30 s/d and not easy to correct. Also an amplitude bigger than your is difficult to obtain on used movements. However I would not be surprised if it keeps good timekeeping even like that.

Quote

With the Seiko I needed to push the gain regulator to its maximum. And it makes sense. The Omega is very loud. Think "Intro-to-the-60-minutes-show-loud". The Seiko on the other hand is very silent.

Being silent is a good thing in a mechanical watch. It means that the escapement is precise and efficient, that is, less energy is wasted in producing sounds.

Quote

So again, how can we trust our cheap Chinese TMs? 

I think they do work well and reliably. For the interested reader, below he full topic on the subject.

 

[JohnR725]

3 hours ago, VWatchie said:

Same Seiko watch, same conditions, with the only exception of the gain regulator on the TM being pushed to its maximum, and what do you know. Looks like I didn’t do such a bad job after all, and it’s now merrily ticking away on my wrist. Of course, the only reason I came to think of this Seiko was the Omega (the subject of this thread) giving me, yet again, very distressing TM results.

The problem is not unique to the Chinese timing machines. I've seen this at work previous watchmaker who was doing Rolex watches was having a bickering match with the boss over a particular watch. Which meant that I got it because I'm the timing machine expert conclusion it was watch case was too big and heavy. I don't remember what we did about the problem sometimes if you make sure the crown is touching where the sensor is it helps but not always.

This comes up for Seiko watches? We get a fair number of Seiko watches in for regulation somebody wants to keep really good time because the factory just adjusts to the specifications they don't adjust to perfect. Put it on the machine looks like total crap or just really bad.. Taken out of the case and usually looks decent.

It's back to the timing machine needs a good clean signal. Heavy watch cases, movement rings that insulate the movement from the case, really quiet movements a variety of things cause the timing machine that have an issue. This is also where it's up to the user to grasp the you're having an issue and see which you can do about it which means take the movement out of the case.

 

 

[VWatchie]

4 hours ago, JohnR725 said:

take the movement out of the case

Solid advice! Thanks John!

[VWatchie]

I've certainly had my doubts about the Weishi machines and I still have. I just stumbled over the below YT video and the guy in it, although not a watchmaker, seems to confirm my distrust of the Weishi machines. I still think the Weishi is useful though to make sure things aren't completely off the scale.

 

[JohnR725]

55 minutes ago, VWatchie said:

I've certainly had my doubts about the Weishi machines and I still have. I just stumbled over the below YT video and the guy in it, although not a watchmaker, seems to confirm my distrust of the Weishi machines. I still think the Weishi is useful though to make sure things aren't completely off the scale.

Only one little problem with this I have done a comparison with my witschi watch expert two and both the witschi 1000 and 1900 and for the most part they do outstanding. It would be really nice in his video if he had explained in a little more detail why the Chinese is crap and the witschi was blessed by 1 million gods Oh wait my bad it was blessed by Rolex. But he didn't tell us which watches they said they were complicated and I don't know exactly what that would mean.

 

 

 

[VWatchie]

1 hour ago, JohnR725 said:

Only one little problem with this I have done a comparison with my witschi watch expert two and both the witschi 1000 and 1900 and for the most part they do outstanding.

I'm well aware of this John and I really don't mean to challenge you on this. I trust you but I do not trust the Weishi machines, or at least, I do not trust my own Weishi 1900. It could be that the electronic circuitry differs between your machines and my machine, who knows? And that might explain our different experiences. Or, maybe I just got one of those Friday afternoon models and perhaps so did the guy in the video. Anyway, it felt reassuring to find at least one other person on this planet who doesn't tune into the choir praising the Weishis. Also interesting to hear that in this guy's experience, the Weishis tend to mislead his customers to believe their watches are up for a service when they're not.

It's really sad that these Weishi machines (or at least my copy of them) are having difficulties consistently showing the truth. It almost killed my interest in learning about servicing and repairing until I realized I was not always the one to be blamed for the poor results.

When it comes to amplitude, a slo-mo video is the only thing I really trust, but even in slo-mo it's difficult to assess the amplitude for hi-beat movements (>= 28.800).

If I ever get myself another TM, it will not be a Weishi. That is unless you're willing to sell me one of yours! 

[nickelsilver]

23 minutes ago, VWatchie said:

If I ever get myself another TM, it will not be a Weishi. That is unless you're willing to sell me one of yours! 

I've mentioned it elsewhere (because Frank doesn't toot his own horn), but Praezis / Frank here has some excellent software that will work with either his own sensor, or with regular timing machine microphones and a Windows computer (or tablet). I'm not associated with Frank, but I am a very satisfied customer, and the price is really reasonable. Just PM him and ask- he doesn't have a site that I know of, but is very responsive.

 

I've compared it to my Witschi Chronoscope S1 and it's bang on, and will also pick up odd escapements like cylinder or detent, or clocks for that matter. It also tells you where the heavy spot is when doing dynamic poise! He also has an optical sensor and another software to do hairspring vibrating, which will do any beat rate. It fits the Luthy vibrating tool, but could be used with a DIY setup and a little fab work. I haven't used it but a friend has and he loves it.

[JohnR725]

5 hours ago, VWatchie said:

I do not trust my own Weishi 1900. It could be that the electronic circuitry differs between your machines and my machine, who knows?

We end up with an interesting problem. One of the ones I'm curious about is like the 1000 machine some of them have again adjustment some of them do not. Does that change the physical or programming or the program that runs the machine characteristics? Then I once gave a lecture on timing machines I track down every single company that ever made a machine at least at the time and I tried to track down the Chinese machine. The number on the box I asked a neighbor who knew how to read Chinese they said it was a phone number. No I didn't call it but to be honest I could never track down where the Chinese machines come from. Even if you find a place you basically find a cool distributor. Even if you find a building well it's a really big building is nothing that indicates they make timing machines. So yes there is bound to be variations and may beer machine is defective somehow perhaps

what would be really nice to have but we can't afford it and no I don't actually know what it costs is witschi makes a test module for testing timing machines basically. It has a whole bunch of programs so you can put it on your machine and see if your machine is working right at least I guess that's what the purpose of it is. Because without a standard test watch that we can all make a comparison whether it's hard to tell what's going on.

So for testing I've actually used to in the 1000 machines too kind souls once loaned me there machines. But thinking about it on the test with those everything was a pocket watch will pocket watches are big in noisy and they tend to work really well

For instance the witschi at work has that public microphone to my microphone except it's mounted on the thing that rotates around and 10,000 positions okayed not 10,000 it only does six. Sometimes I think it's a little noisy. It doesn't like extra low noise in the room in fact none of the timing machines like external noise. Problematic watches some large cased heavy cased watches we've had that with one time a rubber with a Rolex watch one used in the head watchmaker he was being frustrated by the box and he said John what's wrong with this and I look at it I concluded it was probably the case not the watch

Seiko watches I think it's the plastic casing ring. We've had watches come in for regulation and I can't time among the machine be taken out and get to the movement and then they just perfect but not in the case

now from my point of view both the 1000 the 1900 are outstanding machines. Because it puts the timing machine in the hands of everybody because how many people can afford? Firstly I don't even like to think about is set a time when I had money I could spend and not worry about it to get paid $2700 and if I was doing it today I would have the Chinese machine to I couldn't afford the witschi or at have a combination of the software-based machines because some of those are quite interesting. Especially if you get anything that has our oscilloscope So you get a clue of other things going on. I use the oscilloscope a lot of work long with all the other bonus features that art featured in the normal machines unless you go up to the more money

so both the 1000 1900 have been outstanding for everyone because a timing machine is now in your hands. Even when I started I remember going to the material house and wondering if that was a joke the price for the mechanical paper tape machine I believe is several thousand dollars on the other hand the timing machine in question the B200 which is what you found everywhere including Switzerland. I don't remember the new cost but it was thousands of dollars was which is why had to settle for used machines that I had the rebuild which means if everyone only relied on witschi there would be almost no timing machines's ever seen a reference in this group

4 hours ago, nickelsilver said:

Praezis

Which microphone are you using with this? Then his isn't the only software we also have the infinite page discussion of

http://www.watchoscope.com/

I find that's interesting software I like the raw feature for the oscilloscope but the actual oscilloscope and not necessarily a big fan of.

The new yes I do have the free version of Praezis Software. Then there's this software-based machine but at least you get a microphone with it. Or they'll modify an existing microphone found on timing machine so you can rotated around.

https://www.etimer.net/

We still have the same problem though with all the software machines if they don't come with their own microphone and other words you're supposed to make one then it always brings in the question whether there working or not. We really need a good standard test watch. Or we need a witschi's test module that we can't afford it's a shame somebody wouldn't loan it out for a small fee so that you could try it out and see if your machine does exactly what it's supposed to do.

So here's the reality all the software machines will stock if you don't have a decent pick up. We've seen that happen even on this discussion group especially if it's a small watch like a ladies watch attends be hard to pick up anyway.

Oh and one other little thing to remember I sometimes forget the instructor I had at school George in the days of paper tape machines commented about don't these the timing machine. That might actually become more relevant today especially on this group that's obsessed with amplitude. The watch companies tend to be obsessed with does the watch keep time. Mechanical watches even if they are having issues tend to average out some of the issues and will still act to keep decent time. But typically on the group everyone's unhappy because the amplitude isn't I enough which as I said was never anything the watch companies were concerned about they really just wanted the watch to keep time. So maybe people would have more fun if they quit trying to please the timing machine.

 

 

 

 

 

 

 

 

 

[nickelsilver]

27 minutes ago, JohnR725 said:

Which microphone are you using with this? Then his isn't the only software we also have the infinite page discussion of

http://www.watchoscope.com/

I find that's interesting software I like the raw feature for the oscilloscope but the actual oscilloscope and not necessarily a big fan of.

The new yes I do have the free version of Praezis Software. Then there's this software-based machine but at least you get a microphone with it. Or they'll modify an existing microphone found on timing machine so you can rotated around.

https://www.etimer.net/

We still have the same problem though with all the software machines if they don't come with their own microphone and other words you're supposed to make one then it always brings in the question whether there working or not. We really need a good standard test watch. Or we need a witschi's test module that we can't afford it's a shame somebody wouldn't loan it out for a small fee so that you could try it out and see if your machine does exactly what it's supposed to do.

So here's the reality all the software machines will stock if you don't have a decent pick up. We've seen that happen even on this discussion group especially if it's a small watch like a ladies watch attends be hard to pick up anyway.

I'm using a Vibrograf B200 mic, it also works fine with a Greiner mic I have. Frank supplies a simple pre-amp (in a cable) that goes between the mic and computer. It picks up great, and seems less sensitive to noise- even ultrasonic- than my Witschi. You can find orphan mics on Ebay or wherever for 50 bucks.

[JohnR725]

1 hour ago, nickelsilver said:

seems less sensitive to noise- even ultrasonic- than my Witschi.

The important thing is the microphone. Then yes the B200 I think is in a metal case it's been a long time since I've looked at one. So that means the sensors and metal and its way less sensitive to extra low influences like witschi I can't even use it at work if they run the ultrasonic.

Oh and thinking about microphones I forgot I had started to play with my Chinese one I recently got but it seems to be extremely sensitive to everything in the universe around it which is not good at all from microphone.

7 hours ago, VWatchie said:

amplitude

One of the problems is or the problem for every single timing machine is to get amplitude you have to pick up two parts of the waveform. You need the initial sound of the roller jewel hitting the fork that's an extremely quiet sound. This is why typically the gain has to be up as high as you can get it. Which means the room has to be quiet the microphone has to be well made and not pick up everything in the universe. So you need to get a really really really quiet signal that's the key to everything. Then you have to pick up the locking signal. If your gain is adjusted incorrectly because the Chinese machines don't have the sophistication of the witschi's sort of you get actually pick up the louder signals and try to do amplitude off the wrong part of the waveform and then that's never going to work

this is where as I keep pointing out at work anytime I get weird amplitude of weird problems which I do I look at the oscilloscope to see what the problem is. You can't do that with you Chinese machine see don't know what it's up to. Then the Chinese machine has a habit of if things are out of the acceptable window basically whatever that may be it will still try to give you numbers you can see garbage on the screen and have numbers which is why always asked for people to give us a picture because if you have garbage on the screen you have garbage for numbers. The witschi at least mine will sit there and say no it won't time at all because it's unhappy that I have to go in and turn off like I went rate only then it might be happy to give you the numbers if it's within its window of operation. I do notice the machine I have at home operates differently than the machine at work because they're basically entirely different designs the machine at work is largely software with fancy audio processing somewhere in the machine at home is its program but it's in a big ship that's basically all hardware-based.

It may just be with the Chinese machines if you have a gain adjustment usable by the user if you don't get that right you're never going get your readings right and if you don't know where it's supposed to be or how to even get it right which would be a problem with the watch was not working right well that maybe it would be better if you just Throw the machine in the garbage bin and the heck with it because it sounds like it would be a lot better for you.

 

 

[nickelsilver]

John- yes, the old Vibrograf and Greiner mics were massive, lots of metal, for a tiny piezo pickup inside. Funny, I have a Greiner mic that turns once every 60 seconds, like 1 turn straight up per minute, and a friend was on a  (multi axis) tourbillon project, so I found a connector we bodged in so this '60s Greiner mic connected to his 90s Witschi Professional and it works great. Even the (semi?) modern timing tools seem to like or rely on simple piezo pickups. I haven't tried it with my Chronoscope (different plug) but I bet a Coke it works.

[praezis]

15 hours ago, JohnR725 said:

We still have the same problem though with all the software machines if they don't come with their own microphone and other words you're supposed to make one then it always brings in the question whether there working or not.

PCTM standard set comes with a microphone Z1 / Z16. It is small, not rotatable but highly sensitive. But as it is not free of cost, some prefer to get the software only and try to reinvent the wheel again (unless you use N.'s well working variant). 

At least for an amateur imo it is no big deal to hold watch and mic together for a few seconds while testing vertical. Even I do it when I am far from the workshop.

[Nucejoe]

On 3/23/2021 at 8:32 PM, Endeavor said:

It may well be that a movement, clamped in a movement holder and subsequently the combination is clamped in the microphone creates additional vibration or resonance.

Would in such a case,  graphs be nice and smooth in both cases as VW's first and second graphs?  

 

16 minutes ago, Nucejoe said:

Would in such a case,  graphs be nice and smooth in both cases as VW's first and second graphs?  

 

By first graphs I meant FU ones and second graphes in verican position. 

In other word does the stand quit producing resonances in vertical position?  Can we blame the software filtering resonances out. 

[JohnR725]

8 hours ago, Nucejoe said:

In other word does the stand quit producing resonances in vertical position?  Can we blame the software filtering resonances out.

This is where we really need better ways to look at what's coming into the machine. Like oscilloscope what does the signal look like. I'm always suspicious when I see people clamping watches in Movement holder placing them on the timing machine because a lot of times it doesn't look like they're getting a really good pickup or the movement holder is plastic which doesn't transmit vibrations as nice as metal. Or there's a variety of other things I've seen where they put plastic pins are worried about scratching the case or things that are basically going to interfere with getting a good clean signal.

20 hours ago, nickelsilver said:

a tiny piezo pickup inside

One of the difference with the vintage microphones are the little square piece of piezo Is specifically designed to pick up vibrations. Not that it won't pick up audio that manages to make it in their but it's not his huge disk that has a lot of surface area really good for picking up anything that happens to come through the case.

20 hours ago, nickelsilver said:

Even the (semi?) modern timing tools seem to like or rely on simple piezo pickups.

So the witschi seems to have what I'm guessing is a biomorphic sensor as it looks like the right shape and size. We also put a metal shield around the end with the sensor is in the sensor is close to the part that sticking up that the watch movement goes in. The Chinese at least the 1000 1900 have something similar and a sensor that looks similar. They just don't have the metal on the end to shield its.

But if you look at like the new USB one it's a disk. It's a disk in a super lightweight box that during my brief time a playing with it everything in the room it would pick up I had actually move it off the table with the laptop because I think it was picking up the fan. Of course the other problem with that is there is no filtering on the USB design. The Chinese 1000 1900 does have a little bit of control filtering. Then who knows what which he has inside their microphone because I refuse to take mine apartment. Then yes I know the machines is supposed to do in trouble filtering a processing but I'm guessing if you do a little filtering at the pickup point it would minimize the noise and crap that goes down the cable.

 

 

[MCBwatchrepair]

On 3/23/2021 at 9:02 AM, Endeavor said:

As JohnR725 always says with timegraphers; rubbish in is rubbish out. The microphone receives a signal, passes it on to the processor and it's up to the software to make any sense out of it. It may well be that a movement, clamped in a movement holder and subsequently the combination is clamped in the microphone creates additional vibration or resonance (?) If that would be the case, the software may or may not correctly interpret the received data.

Which one is correct? My two cents; as per design, the movement directly in the microphone.

Do I know for sure: no I don't

 

Sorry for hashing an old post but with there being excess vibrations from other items with the timegrapher, would you still have the movement in or out the case (with out the bracelet) when regulating? 

[Endeavor]

12 hours ago, MCBwatchrepair said:

would you still have the movement in or out the case (with out the bracelet) when regulating? 

I do the fine-tuning with the movement in the watch-housing (with or without bracelet), but one has to be aware of the fact that a housing can cause extra vibrations and those extra vibrations could cause interference / "corrupt" the timegrapher signal.

When the movement runs fine without the housing, and the build-in went flawless, but the timegrapher results are slightly "off", I take in consideration that it may be the housing which causes the interference.

Most of the times, at least in my experiences, housings don't cause a problem, but if the signal on the timegrapher suddenly changes after the build in, the housing may be the culprit.

There could be other reasons too, like one of the hands touching the inside of the crystal.

Changing the position of the housing in the timegrapher microphone, or having the crown against the pick-up may help too.

Edited March 24 by Endeavor

[VWatchie]

On 3/24/2024 at 7:00 AM, Endeavor said:

but if the signal on the timegrapher suddenly changes after the build in, the housing may be the culprit.

If it's a thick case (diver), the signal can become too weak. The result is then often that the dots on the graph become more scattered (Weishi). Increasing the gain (Weishi 1900) a bit can then help. I've also experienced that the readings can become unreliable if the signal is too strong (noisy movement). This usually only happens when the movement hasn't been cased.

TMs are very useful but can't always be trusted. In my opinion, a TM can't replace testing the watch on the wrist or on a Cyclotest watch winder.

[Endeavor]

43 minutes ago, VWatchie said:

TM can't replace testing the watch on the wrist

Wise words