D. I. Y. Watch Timing Machine.

[Hansi]

On 9/7/2021 at 10:39 PM, matabog said:

I also put a hole in the back plate, in the sixth position :), for the cable - I didn't like it exiting through the side, it was in the way.

Note that only the side pieces were modified, so for the ones that printed the old version, they dont need to print again the entire project, but only those two parts.

ansamblu2.SLDASM 359.18 kB · 1 download

mic 2.SLDPRT 169.19 kB · 1 download

pesa.SLDPRT 226.88 kB · 1 download

sup_1.SLDPRT 302.73 kB · 1 download

sup_2.SLDPRT 254.08 kB · 1 download

 

May I ask why the platform for the piezo is slightly in an angle?

[matabog]

17 minutes ago, Hansi said:

May I ask why the platform for the piezo is slightly in an angle?

Together with the moving piece, which also has an angle, pretty much the same, they will hold the watch in place when turned upside down, for DD position.

[svorkoetter]

I want to let everyone know that I've tested a third-party watch test stand with Watch-O-Scope, and it works great! You can get one from AliExpress, the price is extremely reasonable, and you don't need to build your own microphone and amplifier. I've written up the details on the Watch-O-Scope website: http://www.watchoscope.com/teststand.html

Also, we've fixed a problem in the Long Term test mode in Watch-O-Scope, so it no longer stops after 13.5 hours. All registered users can download the free upgrade.

[guidovelasquez]

Thank you very much for sharing that information. It's very important. Many users stumble over the microphone solution. And we have been able to verify that a good signal allows an easier and more accurate reading.
We are sure that according to your judgment, it will work properly in WOS. But we would appreciate if you could share a 15 second * .wav audio file with us so that we can make comparisons.
And thanks for version 1.4 We have already tested it.

[Tronik]

Hello to all,

i am new here, let me introduce my version of the amplifier ( is the same scheme as Watch o Scope). I go the way all in one sollution, so the piezo is directly on the top of the amp therefore the cable is much shorter and is soldered directy to the amp without any connector, then i take out the potentiometer instead of the trimer inside the box, output 3,5mm jack is for connection to the computer (or headphones if you like to test it) i am curius how much sensiting the amp is.. i am able to hear if somebody is speaking in the room. if i need some aretation for the watch i use the hairrubber acros the screw of piezo, works great and its possible to test it in several position.  Just like to thank you for sharing this, its really awesome idea and work perfectly see the result which i have. (watch is not in 100% condition - i am beginner  ) Just one point which i have is the sensitivity of the amp if i touch whatever.. ( notebook also... ) maybe i will try to desolder the ground from the output jack (maybe is there some ground loop) this problem is great visible on the begining of the scope screen.

[Rob]

Here's a recent video about building a quality studio microphone. Some parts might be of interest when building the amp for a watch timing machine.

 

[JohnR725]

1 hour ago, Tronik said:

i am curius how much sensiting the amp is.. i am able to hear if somebody is speaking in the room.

The problem you're having is you made a nice microphone. The reality is you want to have a vibration sensor to pick up the watch. The ideal pickup for timing machine will only time off the audio vibrations of the watch not the audio in the room. Unfortunately to some degree most timing machine pickups will pick up audio in the room.

1 hour ago, Rob said:

ere's a recent video about building a quality studio microphone. Some parts might be of interest when building the amp for a watch timing machine.

I wasn't really expecting much of the video but it really is a good video. Notice at the very beginning he comments that microphone design is more of an art than a science which unfortunately applies to making a pickup for listing to watches.

Then a minor clarification to something in the video? At about 13:28 he brings up the USB sound device. He talks about how it can have a nice super clean input. But the problem is there's a lot of different designs for the various USB sound inputs some work really well and some are very noisy. But if you get a clean one it works really nice for interfacing to the computer. Then stealing the 5 V power is nice you don't have to have a battery for your amplifier you just plug it in and not worry about the battery aspect.

[Tronik]

7 hours ago, JohnR725 said:

The problem you're having is you made a nice microphone. The reality is you want to have a vibration sensor to pick up the watch. The ideal pickup for timing machine will only time off the audio vibrations of the watch not the audio in the room. Unfortunately to some degree most timing machine pickups will pick up audio in the room.

 

Hello yes its as you wrote. Before i build the amp i fully like it because i understand that it will react (amplify) only the little vibration (sound) of the watch "click" but if i think what is in reality the sound? Its frequency, frequency catch the piezo. In reality it works well when the watch are directly on the piezo beacause it works in both way when the watches is leaving on the piezo ( so it suppress the sound from the room - like if you pres speaker membrane) but it still react on the watch click vibration which is amplifying. So basicaly when the watch is leaving on the piezo it works less like microphone and more like vibration sensor.. Maybe will be interresting to try aply more presure on the watch to see if it will works more like vibration sensor... but because  the amplifed "tick" is in reality during testing much more stronger then the other sound it newer affect the meassurement for me.. 

[guidovelasquez]

On 11/9/2021 at 4:27 AM, Tronik said:

its really awesome idea and work perfectly

Hi @Tronik. Good work and thanks for sharing. I see good results. The waveform looks great. Some noise can be seen, but the signal is clear enough to be distinguished. We would like to hear an audio file of your project. Can you share a recording in wav or mp3 format?

[LittleWatchShop]

I decided to extract the schematic for the Vibrograf B200 amplifier board so that I can simulate it.  Here are the results of my work.  I have included an frequency response simulation.

I did the majority of the extraction myself, but did benefit from earlier work done here.

[guidovelasquez]

Thank you very much for the information and links shared. I only have one question: What does the two curves on the graph mean?

[LittleWatchShop]

4 minutes ago, guidovelasquez said:

Thank you very much for the information and links shared. I only have one question: What does the two curves on the graph mean?

The darker blue is the magnitude in decibels (the scale is on the left side).  The lighter blue is the phase shift in degrees (scale on the right).  You can ignore the phase plot.  The magnitude plot is most interesting.

 

[LittleWatchShop]

This extraction happened because I was puzzled by differences in my Weischi and my Vibrograf.  It is all resolved now and have complete understanding of how the Vibrograf does its magic with gears, double helix (perhaps a misnomer...a helix with two oscillations on the platen) and motor.  Done all the calculations for the beat frequencies and I am now satisfied.  Fascinating.  It is a well-engineered machine.  I like it.  Only wish it were a smaller footprint!!

Edited November 16, 2021 by LittleWatchShop

[guidovelasquez]

 

36 minutes ago, LittleWatchShop said:

fase en grados

Very well thanks. She hadn't seen the scale to the right of phase. I assumed that everything ran in decibels. and that the horizontal was that of Hz.

[LittleWatchShop]

5 minutes ago, guidovelasquez said:

and that the horizontal was that of Hz

correct...horizontal is in units of Hz

[guidovelasquez]

Excellent document. Particularly the answer in herz of the graph.  Confirms a conclusion that we had reached empirically. That the frequencies above 1Khz are the most important in the measurement. And the graph makes it clear there is a strong attenuation at frequencies below 1Khz.

[xyzzy]

I just received, after ordering it two months ago, one of these microphones with rotating stand, https://www.aliexpress.com/item/1005002762635294.html?spm=a2g0s.9042311.0.0.48914c4d5WFIQb

It is powered by a 9V battery and has a 1/8" TRRS plug as output.  Power draw from the battery is about 11.5 mA, so maybe 40 hours from a typical 9V.  The plug is designed for a phone, with four conductors, and the last two, RS, are used as ground and microphone.  The tip and first ring would be line out from the phone.  This means if you plug in a two or three conductor audio cable it will short the microphone to ground.  There is a included a TRRS to TRS cable, but of course, no instructions at all to tell you this.

It is 3d printed and glued together.  It was not hard to separate the lid with a razor blade along the glue joint.  Inside there is a small board glued in place.  I see a 100 µF cap, and 8 pin chip, some kind of trimmer, and a few chip resistors and capacitors.  It's hard to get a good look.  There's a disk piezo in there, but it's encased in expanding foam.

It's assembled such that the side of the holder the watch is on is on the bottom part of the holder as one rotates it 180° from DU to DD.  This means it passes by the vertical post of the stand and there isn't much clearance.  The band hits.

After opening it, one gains access to the screw attaching the holder to the stand.  I flipped the holder around so now the watch will be on the upper arc of the holder as it rotates and not need to pass by the post.

Photo of insides:

Example beat.  From a Seiko 7S26C.  I've turned off the high pass filter to try to show the entire range.

To compare, here's the same watch with a USB earbud microphone pod.  I had to put in a narrow notch filter at 22.44 kHz to remove a background sound from my monitor that the piezo microphone completely fails to capture.  There is a lot more background noise under 1 kHz.   But also more of the beat above 20 kHz that is now captured.

 

[jdm]

13 hours ago, xyzzy said:

To compare, here's the same watch with a USB earbud microphone pod. 

And the conclusion in practical usage is? Better or worse than earbuds? At EUR43, or 58 if the articulated type) it's just marginally cheaper that the metal articulated mic, in my opinion it cost too much for what is is.

[guidovelasquez]

We would like to hear an audio segment. So you can see the results. Thank you very much for sharing this information. Especially for those of us who have a hard time acquiring one and examining it.

[xyzzy]

13 hours ago, jdm said:

And the conclusion in practical usage is? Better or worse than earbuds? It's just 15USD cheaper that the metal articulated mic, in my opinion it cost too much for what is is.

Haven't used it enough to know.  Having the articulating stand is vastly more convenient than the loose microphone, that's for sure.  I don't see a metal one for only $15 more.  I got this for $47.76 with shipping, the cheapest I see the 1000/1900 microphone alone for is $87.45 with shipping.  But I don't have one to know if it's worth the $40 extra.  I think it needs a preamp too?  Certainly one will need to modify the cable to connect to a computer.

 

7 hours ago, guidovelasquez said:

We would like to hear an audio segment. So you can see the results. Thank you very much for sharing this information. Especially for those of us who have a hard time acquiring one and examining it.

There doesn't seem to be a way to post audio here.  I could do youtube maybe.

I have on my computer's microphone input with the mic boost at +20dB and input gain at 0 dB, which produces about -15dBFS SPPM with the Seiko 7S26.  Dragging my finger on the stand base is about -12 dBFS.  Typing this on my keyboard, about 1' away on the same desk, and it's extremely noisy keyboard the likes of which has not been produced in over twenty years, gets to about -18 dBFS.  Normal volume talking is about -20 dBFS.  The noise floor is about -39.

The Seiko is pretty quiet, so I tried an Elgin grade 173 0s pocketwatch.   For that I need to lower the mic gain to +0 dB to get it to around -17 dBFS.

[JohnR725]

3 hours ago, xyzzy said:

I don't see a metal one for only $15 more.  I got this for $47.76 with shipping, the cheapest I see the 1000/1900 microphone alone for is $87.45 with shipping.  But I don't have one to know if it's worth the $40 extra.  I think it needs a preamp too?  Certainly one will need to modify the cable to connect to a computer.

The preamp is built into the metal microphone that's not the issue. The issue is the last line of what I quoted above. The one from Ally express should just plug-in and be usable. The metal microphone requires modification. The problem with the DIY timing machine is not everyone has the skill set to modify or make a microphone. So a plug-in solution is very nice for a lot of people.

17 hours ago, xyzzy said:

I see a 100 µF cap, and 8 pin chip, some kind of trimmer, and a few chip resistors and capacitors.  It's hard to get a good look. 

Unfortunately that kind of answers my question which I'm going to ask anyway. More than likely that's a standard off the shelf board probably adapted from something else. It doesn't have any hardware filtering as a guess I don't suppose there's anything that resembles anything looking like a number? Just because I'm curious as to what they're actually using.

[xyzzy]

7 hours ago, JohnR725 said:

Unfortunately that kind of answers my question which I'm going to ask anyway. More than likely that's a standard off the shelf board probably adapted from something else. It doesn't have any hardware filtering as a guess I don't suppose there's anything that resembles anything looking like a number? Just because I'm curious as to what they're actually using.

There is, above the solder points for GND and OUT.  118362G.P567 or something like that, it's small and poorly printed.  The first 8 could be a 0.  The "G." might be a "6." instead.  The .P5 is partially damaged by solder, it could be ".2" or ".8", and the second digit might be S, or maybe anything with a horizontal bar on the top of the character.

There's another number under that, 191123.

I tried looking at the signal with a scope but at this point it's getting a bit too analog for me.  I see an output of a 12.6 MHz 1.5 Vpp signal at 2.1 V.  Tapping on the device doesn't seem to change the frequency or amplitude, but does move the mean up to 2.5 V.  Some aspect of the amplification I guess?

7 hours ago, JohnR725 said:

The metal microphone requires modification.

I didn't think it had a power source and it looks like a two pin cable.  Does it use phantom power?  Or is there a third pin?  If it has a preamp, then it seems like one must modify it to supply power, in addition to merely changing the connector to an an 1/8" phono.

Edited December 1, 2021 by xyzzy

[JohnR725]

2 hours ago, xyzzy said:

I didn't think it had a power source and it looks like a two pin cable.  Does it use phantom power?  Or is there a third pin?  If it has a preamp, then it seems like one must modify it to supply power, in addition to merely changing the connector to an an 1/8" phono

You're right it doesn't have a power supply it needs one of those. Your incorrect in that the connectors three pins. You can purchase the connector on eBay it's exactly the same as what comes with the machine.

I think there's a separate discussion at least one probably more on modifying the microphone. 

The power supply depends upon the op amp they put in your microphone. the one that was in mine is being powered by 6 V on the timing machine. But As somebody pointed out the specifications for that op amp was a minimum of 7 V. But I think somewhere seen that they may use a different op amps now which should work at a lower voltage it have to take the microphone apart book and see or just use a higher voltage.

 

[praezis]

Thanks for your interesting review, xyzzy 

Stimulated by the hint to that mic, I ordered one, too. I wondered if the Chinese had some tricks that I dont't know yet (they do not!).

It is the type without movable stand (don't need that), and has a LiPo battery already. 2/3rd of the circuitry is needed to charge the battery.

It came without any connecting cable and - of course - without any manual. The seller was not at all interested in my complaint about the missing cable.
I suppose, anyone without electronic skills will not be able to get this set working.

The amplifier is made from a LM386, an older headphone amp IC. Circuit is exactly like its data sheet application. No filter, just the piezo properties add a bit HP filtering.   

I made some measurements, using my software TM and my tic-noise generator and compared the chinese mic to a simple Z1 mic (piezo disc, 1-transistor amp, no battery needed)

Both mics produce about the same output voltage from a given ticking strengh. But the chinese mic needs a much louder watch because of its worse signal/noise ratio.
For testing mics, the watch is substituted by a piezo disc driven by a tic noise voltage.

Test results
Needed for full recording level (volume = max, +0 dB):
  CN mic: 0.75 Vp / Z1 mic: 0.3 Vp @ stimulating piezo disc

Sensitivity for "amplitude measurement possible":
  CN mic: 45 mVp / Z1 mic: 9 mVp @ stimulating piezo disc

The chinese mic seems to react a bit sluggish to the stimulating tics.
Some scope views (stimulated by 0.15 Vs @ piezo, equals a medium loud wristwatch):

Frank

Edited December 2, 2021 by praezis

[katolik]

21 hours ago, xyzzy said:

There is, above the solder points for GND and OUT.  118362G.P567 or something like that, it's small and poorly printed.  The first 8 could be a 0.  The "G." might be a "6." instead.  The .P5 is partially damaged by solder, it could be ".2" or ".8", and the second digit might be S, or maybe anything with a horizontal bar on the top of the character.

 

I identified this board, this is a ready-made amplifier based on LM 386, the green connector has been removed from it, on Aliexpress they are sold for US $ 0.32

Edited December 2, 2021 by katolik