Jump to content

Chinese Timegraphers


Recommended Posts

1 hour ago, spectre6000 said:

Are you saying the Witschi and the Weishi are producing similarly variable results like Vacherin said he was seeing, but the Weishi might be slightly worse?

I'm not sure if the phrase slightly worse is appropriate. Several years ago when I was in a teaching situation of teaching hobbyists watch repair. One of the students loaned me his small collection of pocket watches and his Chinese 1000 machine because he was concerned about the results he was seeing.  this made for a really nice lecture with pictures. Then much later on one of the other students loan me his 1900 machine to do a comparison with.

now of course I acquired a 1000 machine on eBay used for hundred dollars. You can never have enough timing machines.

before we continue I turned both timing machines back on then I was doing something with the witschi you can't do with the Chinese listening to the ticking sound as I was still bothered by the issue of yesterday. Then I wrote down the numbers. Witschi thinks the watch is +3 seconds fast 222° 0.5 ms beat the 1000 machine thinks the watch is +3 seconds fast 226° and 0.5 ms beat.

I'm not sure if it's going to answer your question or not but I have a link to the second page of this discussion it will probably give you some answers and bring up new questions

then when you make comparisons between the differing machines other issues can come. Like what's the consequence of having a small graphical display like the 1000's? Or the Chinese machines typically do not display error messages they will always try to give numbers what's the consequence of that?

The problem with a small graphical display is and we've done experiments on this group somebody had a watch that was perfectly in beat graphically but numerically the number did not agree? This is where on in a timing machine if the numbers do not agree with graphical results you have a problem. so what happened is the watches perfectly and beat graphically but numerically it was a huge number? This is because on the small display when you reach the edge to display it basically rolls over to the top and then you start getting goofy results. See you might have a watch that looks perfect but it's really not.

Then my reference to always giving numbers? I have a saying which is garbage in garbage out relating to timing machines. If the graphical display seems to be random dots and you have nice numbers the numbers are wrong. this is one of the reasons I always ask for pictures of graphical displays because you can see things there that you can't see numerically.

https://www.watchrepairtalk.com/topic/4124-chinese-timegraphers/page/2/

  • Like 1
Link to comment
Share on other sites

On 5/22/2020 at 2:01 PM, quitquitquit said:

 Done this test literally 100 times on 7 different watches and machine behave still the same - I can't replicate results.

I have a 1900 and works very reliably. As mentioned above if your watches have big thick cases you should try adjusting gain or remove the caseback.

Link to comment
Share on other sites

Seeing as how I left the watch on the timing machine and 48 hours later I noticed it was still running reminded me to do another comparison. I'm attaching pictures including one of how two timing machines can time one watch.

Chinese machines do a really good job for the most part. This is also where common sense of the user comes in and that it is important to look at the numbers making sure they agree with the graphical display and visually looking at the watch. Sometimes the numbers might look too good to be true which is why you need to look at the watch.

One of the places the timing machines struggle and not just the Chinese is with amplitude. This is because the machine needs to measure two parts of the waveform correctly. As the amplitude decreases to basically the bottom of where the Chinese machine can measure you will see incorrect numbers. But if your visually paying attention to your watch it's not a problem. So for instance the witschi watch expert two will measure to 70°. The Chinese 1000 and 1900 will measure to 100°. Images attached at 104° the Chinese 1000 is indicating are very happy outstanding 285°. Then dropping below its measuring range 84° now not quite as happy  232°.

Excluding that the amplitude is at the bottom a where the machine supposed to measure still it's doing a really good job for something that cost a fraction of what the witschi did.

 

wc-104a.JPG

wc-104b.JPG

wc-84b.JPG

wc-84a.JPG

wc-mics.JPG

Link to comment
Share on other sites

  • 9 months later...

It depends upon a lot of factors? The Chinese timing machines are very sensitive to noise. So noise will affect things. Witschi machines are better with noise but they can still be influenced. Then of course how much noise are we talking about?

For instance at work typically before they turn the ultrasonic cleaning machine on they will check to see if someone's running the timing machine. What's interesting is sometimes it affects the machine and sometimes it does not it depends upon if we can position the microphone behind something and sometimes it works sometimes it doesn't. It also depends on how to noisy the watches a pocket watch versus a wristwatch seems to be a factor also.

 

Then I have a picture attached you'll notice that the Chinese machine may be displaying numbers but the graphical display looks bad? Then the witschi machine is displaying the correct numbers and there's a battery-powered shaver in the picture. This is because I was reviewing both machines and walked into the room with my shaver and noticed that the Chinese machine was immediately very unhappy from considerable distance away. But the witschi machine is fine even with the shaver inches from the microphone. Then both machines should have shown the same numbers because I was using both microphones clipped to the same watch at the same time while I was making the comparisons.

 

 

timing machine and noise.JPG

Link to comment
Share on other sites

9 hours ago, fjseal said:

When using a timegrapher, does the room have to be completely quiet so as not to affect the microphone, thanks.

Not necessarily. I normally listen to music at moderate volume and the reading is not affected, neither is by talking. But if I drop a driver on the bench where the stand is, it draws some points off. It is sensitive and selective at the same time. 

  • Thanks 1
Link to comment
Share on other sites

  • 3 months later...

This the reading I got from an AS1950/51 on my Timegrapher. The rate is four plus signs. I tried another movement and it was okay. What on earth could this be? Do timegraphers need calibrating at intervals in the same way things like torque wrenches and other measuring instruments do? I don't know if these Weischi's go wrong at all, but I am a bit doubtful about mine sometimes.

Thanks all.

IMG_2877.JPG

Link to comment
Share on other sites

8 minutes ago, fjseal said:

? Do timegraphers need calibrating at intervals in the same way things like torque wrenches and other measuring instruments do?

No they don't, calibration procedure is not even published or the service available.

8 minutes ago, fjseal said:

I don't know if these Weischi's go wrong at all, but I am a bit doubtful about mine sometimes.

Nothing is wrong with the instrument, but something is with the mov.t, as in the mainspring touching or other issue making it run so fast and with insufficient amplitude - if the latter is to be believed at all.

Link to comment
Share on other sites

12 minutes ago, fjseal said:

The rate is four plus signs. I tried another movement and it was okay.

just to be clear one movement we see looks bad and the other one looks fine?

12 minutes ago, fjseal said:

Do timegraphers need calibrating at intervals in the same way things like torque wrenches and other measuring instruments do? 

typically no sort of. There's a quartz crystal in their basically it's so much better than the watch that short-term you'll never notice if it drifted off.

then in situations like this with the timing machine we need to make an adjustment to something which is not the timing machine it's the user.

a simple rule when using the timing machine is that the graphical display should resemble the numeric numbers. So for instance if you see garbage on the graphical screen like you have then the numbers are garbage. The timing machine will attempt to process the signal and unfortunately it has no intelligence. It be so much nicer if the timing machine was a magical crystal ball they could tell you exactly what's going on unfortunately it cannot. this is one of the places where witschi puts a little more intelligence into their machines and if it sees garbage coming in it will tell you that the signals not clean or basically it's having a problem and it will refuse to give you numbers at all. But the Chinese machine will attempt to process and give you numbers so this requires user intervention to say oh dear I'm having a problem.

you need to very carefully examine the balance wheel assembly in the watch. Look at the hairspring make sure it's flat make sure it's not touching the arms. Listen to the watch hold it up to your ear what is it sound like? Are the times when they hairspring is bumping in the stuff to make it interesting clanging noise which timing machines do not like. So you're definitely having an issue with the watch which needs to be fixed before the timing machine can give you? I was a say before can give you usable data but it has given you usable data it's graphically telling you it looks like crap and numerically it's confusing you but is still telling you that you have a serious problem which needs to be fixed.

 

 

Link to comment
Share on other sites

Funny you mention this because I had experienced the same snowstorm after assembling a AS 1920 (Girard-Perregaux 440 movement) without amplitude or rate and trying to figure out what happened to it. I checked everything and tried different positions until the movement stopped itself and the mainspring unwinding itself.

What really happened is that when I installed the pallet bridge, I screwed lightly then checked that the pallet works fine (it snap from end to end) and happy with it, I resumed the assembly with the balance cock before putting in the TG. When it ran erratically, the screw holding the pallet bridge was unscrewing itself due to vibration until the pallet jumped out of the jewels, thus letting the mainspring unwind itself. Really lucky nothing got broken because when this happened, my heart skipped a beat thinking a pallet stone or a pivot broke off. 

I redid the pallet installation by making sure to screw it tight this time and reran it and I still got a bit of snow but less than the first time. I adjusted the hairspring stud until it get in beat enough and the timegrapher stabilized itself into clean lines then i did small adjustments to get the rate I'm happy (at the end around +0s/d with 220 degree of amplitude at 36000 bph - it's a high beat movement).

Hope my experience shed a light about how to find the cause. Mine is a Weishi 1900.

Link to comment
Share on other sites

  • 3 weeks later...
On 3/12/2015 at 5:10 AM, JohnR725 said:

One of the more useful things is the display mode where you see 96 seconds and still another mode which will display 192 seconds.

Hi @JohnR725 . Reading through this thread again, and early on you posted about the facility to replay recorded data on the 1900. I'm sure I've done this myself in the past, but now I cannot remember how, and the feature isn't even mentioned in the instructions. Any ideas?

Also, what information do you get from graph type B that type A doesn't have?

Link to comment
Share on other sites

3 hours ago, oldhippy said:

Going off subject. Have you watched Kalle’s video on Ultrasonic cleaners? Many on here need to change how they clean watch movements.

That is why in many of my posts I've been warning people not to clean delicate parts in the ultrasonic cleaner for long periods. Never more than 4 mins. And never put anything with shellac in it, like pallet forks and balance wheels. I cringe when I read about people cleaning things in the ultrasonic for 30 minutes. 

In my most recent post, I was talking about building a revolving bottle to put into the ultrasonic to minimize the damage caused by cavitation at the antinodes of standing waves. I haven't tested it out yet. Hopefully if I put a piece of aluminium foil in the bottle and rotate it, it will come out with fewer holes than without any rotation.

  • Thanks 1
Link to comment
Share on other sites

On 7/24/2021 at 7:35 PM, Klassiker said:

facility to replay recorded data on the 1900

I worked it out for myself, by pressing random buttons (my standard method of problem solving with electronic devices). Here's how to display (not replay) the data from the most recent measurement:

1. Stop the measurement by pressing the red button

2. Immediately press either the up-arrow, or the down-arrow.

3. Toggle between graph-type A (last 80s of the measurement) and graph-type B (last 160s) using the arrows.

Unfortunately, the measured rate, amplitude and beat-error are no longer displayed.

Link to comment
Share on other sites

On 7/24/2021 at 7:35 PM, Klassiker said:

what information do you get from graph type B that type A doesn't have?

Here's my attempt to answer this question myself, by comparing curve type A (what the manual calls "normal"), and type B ("complex") of the Weishi 1900 maschine.

When a watch is running nicely, with clean traces, the type B graph doesn't show anything which cannot be seen better with graph type A. Type B only shows one trace, representing the rate. Type A displays two traces so we get a graphical representation of the beat error as well as the rate. I didn't see much point in showing this comparison, as type A is obviously superior. I looked for a watch in my collection which isn't running so healthily, and put that on the machine. The first pair of photos (1A and 1B) show the watch straight out of the drawer (fast rate, intermittent bad interaction between a pallet jewel and an escape-wheel tooth). The second pair are after magnetising the watch.

1A:

161122733_TypeAmag0.JPG.26084ad692f5e41714c31533dc5ec7b4.JPG

 

1B:

582790202_TypeBmag0.JPG.0946db5d3defc3b86c86d7f6a77c2eab.JPG

 

2A:

602391556_TypeAmag3.JPG.3bf218de6a58be9c97035e7c7d937696.JPG

 

2B:

387122102_TypeBmag3.JPG.b5a3172c541b140af782363e7bda0e0d.JPG

I don't know why the manufacturers refer to the type B curve as complex, when in fact it is simpler and doesn't display as much information. Looking at the comparisons above, this may be to its advantage, when a watch is running so badly that the type A curve is such a mess it becomes difficult to interpret, In this case it may be wothwhile looking at the recording of curve type B.

Link to comment
Share on other sites

  • 4 weeks later...

Hey,guys…Recently bought a Weishi timegrapher and would like to know what the smaller numbers mean that are under the main readout numbers..have searched the net but no luck…thanks in advance..Lou

66E1D6F3-A55E-43C9-94DE-8FD325093A61.jpeg

Edited by Kanie
Add searched
Link to comment
Share on other sites

6 minutes ago, Kanie said:

Hey,guys…Recently bought a Weishi timegrapher and would like to know what the smaller numbers mean that are under the main readout numbers..have searched the net but no luck…thanks in advance..Lou

66E1D6F3-A55E-43C9-94DE-8FD325093A61.jpeg

Although your main readout says zero seconds a day, probably because in the main menu you have it set to 999 sec/day instead of a more accurate 99 sec/day, those small numbers are telling you that the movement is losing one second per day. In the main menu you can change how often those smaller numbers are displayed, such as every 2 seconds, 4 sec, 10, 20, and so on. I like to set mine to have a readout every 2 seconds, but that is just my preference.

Once you set the main menu to 99 sec/day, the main display that currently say zero sec/day will then become more accurate and reflect those numbers that say -1 sec/day, because at present it is not giving a reading other than a whole number and you want to see 0.1 sec/day (as an example) rather than rounding it to the nearest whole number which would be zero.

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...