D. I. Y. Watch Timing Machine.

[guidovelasquez]

2 hours ago, LittleWatchShop said:

simply bench space.

You are right @LittleWatchShop And it is understood that in some parts the problem of space is very complicated. In my case the problem would be the cost of buying a Weishi. Although it is one of the cheapest. But with the modest investment that goes into building the preamp and an acceptable mic I am very satisfied with the W-O-S software. It's everything a watchmaker could even dream of in the golden years of watchmaking let's call the 60s 70s. There were tools in those years. But compared to what WOS can do, they look like toys. Here I share a photograph of my workspace.

I will apologize for the mess that is appreciated. It is the entropy related to the precision of the clocks. This configuration allows me to be at my workbench and move a few steps to my desk with the computer where I have the microphone stand with a padded insulating support and from there make the readings and sometimes taking my magnifying glass and a tool, do the settings by viewing my computer screen. It is not ideal, but taking precautions is functional. And I explain the black screen, I use an application to be able to invert the colors of the monitor. NegativeScreen.

 

 

[JohnR725]

6 hours ago, LittleWatchShop said:

My issue with the PC based machine is simply bench space.  That is why I like the Weishi (limited as it is)

interestingly have the same complaint. The Chinese timing machine is a self-contained package which occupies relatively little space on a watch bench which is usually short of space anyway.

I wonder how easy it would be to convert the program to run on something like Windows CE? the reason for this comment is that witschi uses that on their machines. So there timing machines are no longer compact hardware-based machines there now compact PCs basically. The reason why comment is you can get 32-bit controller boards that are really small small displays it should then be possible to build a smaller timing machine running on software and have all the nice features without having to have an entire PC or laptop.

[jdm]

50 minutes ago, JohnR725 said:

I wonder how easy it would be to convert the program to run on something like Windows CE?

You can only "convert" (port or recompile) a program of which you have the source code. Not something that you have only in binary form. In that case even the smallest modification is a difficult task done by reverse engineering and patching.

[JohnR725]

16 minutes ago, jdm said:

You can only "convert" (port or recompile) a program of which you have the source code.

that was not what I was intending on doing. I was kinda hoping to plant a seed where the original author might suddenly feel inspired to doing that himself. or maybe a group of people would decide to do an open source timing machine project.

this is because the original complaint I had in somebody else had is that having a laptop on the watch bench is a pain or even next to the watch bench especially mine because I usually get stuff piled up there. So having a much smaller PC or the fact that there's all sorts of inexpensive 32-bit controller boards out there plus touchscreen LCDs it should be relatively easy to make an inexpensive self-contained box that's what I was hoping to plant a seed for.

[jdm]

52 minutes ago, JohnR725 said:

I was kinda hoping to plant a seed where the original author might suddenly feel inspired to doing that himself. or maybe a group of people would decide to do an open source timing machine project.

There is starting point for that, surely it has been posted already.

https://tg.ciovil.li/

[LittleWatchShop]

A Raspberry Pi would be perfect

[LittleWatchShop]

3 hours ago, jdm said:

There is starting point for that, surely it has been posted already.

https://tg.ciovil.li/

yes...this might be the ticket on a Pi!!

[xyzzy]

tg already runs perfectly well on ARM based Linux devices.  I use it on an ARM based Pinebook Pro.  You would not need to do anything to port it to an RPi.

[jdm]

11 hours ago, xyzzy said:

tg already runs perfectly well on ARM based Linux devices.

Yes. But it's very basic, almost to the point of being crude. Unless evolution is done on it, I'm happy with my 1900 and its articulated mic stand.

[praezis]

A win tablet PC is what you want.

Frank

[xyzzy]

On 8/7/2021 at 4:45 AM, jdm said:

Yes. But it's very basic, almost to the point of being crude. Unless evolution is done on it, I'm happy with my 1900 and its articulated mic stand.

What about the 1900 is more advanced?

It just uses simple thresholding to detect beats rather than the much more sophisticated correlation based algorithm of tg.  And it can't display waveforms, much less time and frequency reassigned spectrograms like tg can.

They've both got a paper strip.  Tg has smooth zooming on the paperstrip while the 1900 doesn't.  1900 has different colors for the two beats.  There's a patch for tg to add that.

tg has beat autodetection as well as manual section of common or an arbitrary beat.  The 1000 doesn't have that but I think the 1900 does.

tg can calibrate itself to a quartz watch, or another pps source.  Don't know what the 1900 supports for calibration.

tg will average over a 16 second interval.  I think the 1900 has better averaging interval options.

 

[jdm]

5 hours ago, xyzzy said:

What about the 1900 is more advanced?

I didn't say it's more advanced and did not even tried to compare features.
Actually, I should have not referenced a standalone, commercial instrument in the context of discussing a PC software, as they are very different things and one's preference can be based on other factors.
And in fact the same PC software running on a self-contained platform costing easily more than €100 before a professional microphone constitute yet another solution which can, or can not, be appealing to different people.

[praezis]

On 8/8/2021 at 2:56 PM, praezis said:

A win tablet PC is what you want.

Some pictures to illustrate my short statement. Doesn't take more space than a TG, imho.
Any Windows program / TM software will run on them.

Old 10" tablet with TIMOMAT microphone

7" tablet with PCZW microphone

Frank

[xyzzy]

On 8/9/2021 at 12:56 AM, jdm said:

I didn't say it's more advanced

You said tg was very basic, almost to the point of being crude, and implied that the 1900 was not as basic.  I was using advanced as the opposite of basic.

Certainly I see how tg is more basic than a considerably more expensive Witschi machine or some of the other software that is available.  But the 1900 seems very basic to me too.  What about tg is more basic than a 1900?

[jdm]

5 minutes ago, xyzzy said:

What about tg is more basic than a 1900?

If you are asking me, I have already answered above to the best of my communication abilities 

[LittleWatchShop]

21 hours ago, praezis said:

Some pictures to illustrate my short statement. Doesn't take more space than a TG, imho.
Any Windows program / TM software will run on them.

Old 10" tablet with TIMOMAT microphone

7" tablet with PCZW microphone

Frank

Very nice idea.  Guessing that the software supports stylus or touch input so that a keyboard is not required...

[praezis]

3 hours ago, LittleWatchShop said:

Very nice idea.  Guessing that the software supports stylus or touch input so that a keyboard is not required...

Not needed, any tablet / OS does. You need no special software, as mentioned. 

Frank

[pawsan]

On 3/7/2018 at 11:14 AM, matabog said:

So I played a bit in Solidworks, and this is what I got. A friend of mine helped me with the 3D printing.

 

sup_2.STLUnavailable

mic 2.STLUnavailable

pesa.STLUnavailable

sup_1.STLUnavailable

 

the editables:

 

mic 2.SLDPRTUnavailable

pesa.SLDPRTUnavailable

sup_1.SLDPRTUnavailable

sup_2.SLDPRTUnavailable

ansamblu2.SLDASMUnavailable

 

Hi Maya big nice work. Can you please share STL files.

[jdm]

52 minutes ago, pawsan said:

Hi Maya big nice work. Can you please share STL files.

Just FYI, we have a dedicated section where it's considered polite for new members to introduce themselves first. 

[LittleWatchShop]

1 hour ago, pawsan said:

Hi Maya big nice work. Can you please share STL files.

Ignore the word "unavailable" and click.  But introduce yourself first.

[watchweasol]

I have already got one of these printed and working using Watch O Scope and it’s very good

[matabog]

I made a new version... The problem with this 3D print design is that it gathers too much noise from the environment. Ideally you need you sensor stuck to your watch/crown, then some soft/insulator/dampen material, the a heavy support.

 

So I modified it a bit for the sensor not be glued directly with superglue by the plate, but with a layer of foam-double_side-tape. Also I decreased the plates sizes a bit because not all the bracelets fit.

[matabog]

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

mic 2.SLDPRT

pesa.SLDPRT

sup_1.SLDPRT

sup_2.SLDPRT

 

Edited September 7, 2021 by matabog

[watchweasol]

Hi I also used double sided tape for the sensor so it can be replaced in case of failure. The re routing of their cable makes sense.

[matabog]

For when I measure only the movement, I bought an old timegrapher microphone. I cleaned it and I changed the piezo sensor with a bimorph one (like here https://www.watchrepairtalk.com/topic/3002-d-i-y-watch-timing-machine/page/6/

 

 

But for the best environment separation remains the first diy - a plastic chinese holder, some nuts and bolts and a heavy Ikea lamp base (more details here: https://www.watchrepairtalk.com/topic/3002-d-i-y-watch-timing-machine/page/10/)

 

 

 

Also, these are my schematics, because I made 2 amps - one cut at400Hz and one at 8kHz - the 8kHz is more noise  friendy https://www.watchrepairtalk.com/topic/3002-d-i-y-watch-timing-machine/page/9/

 

Sorry, but I needed to recap, because there is so much noise on this thread (pun intended :)) ) and I am very happy with my results.

Also, I'd like to take this ocasion, again, to thank Stefan for Watch-o-scope and filter/boards designs and microphone designs and basically for all his work in this direction!

 

Edited September 7, 2021 by matabog