Jump to content

Recommended Posts

Posted

I've wondered about regulating a watch myself and came across this. There are other pages about as well.

https://tg.ciovil.li/

Mixed success

The volume control is a linux utility and shows signal strength after the preamp. It was fed into a laptop headset socket via a splitter lead. This is a stable trace off a new Seiko. A 15 year old chronometer usually doesn't give such a stable trace, it's running 15secs a day slow. It still seems to get the timing error correct despite this and playing around with position may give a more stable trace. Any new watch so far does give a stable trace.

General bits bought off amazon UK. The pickup is intended for a violin.

Seiko5 Regulation using a Tg setup

Seems to be 2 problems. Bandwidth of the electronics, too high with an amp like this and probably similar, the mechanical aspects of the pickup. Unfortunately commercial microphones for this use are rather expensive.

If anyone else tries to install on Linux I found that the automated build wouldn't work but the usual ./config etc did and indicated what lib files I needed to install.

John

-

 

  • Like 3
Posted

I have used it a lot, also on Linux. It seems to work well with the builtin microphone of my Lenovo T460, and also with a Nokia phone headset mic. I do need to tweak the levels. more so with quiet watches. It will pickup a Timex pin lever from 20 paces.:P

Compiling from source hasn't caused any issues so far. I was toying with trying to compile it on Android, but I haven't had the courage to drop down that particular rabbit hole yet.

  • Like 1
Posted

I had issues with a guitar pickup and decided to follow a different route:
You can buy the microphone stand that is used with the Weishi timegraphers separately for not too much money. Then you have a stand with the option to measure in different positions and you can also test movements without the case. Additional benefit: the stand has its own amplifier already built in.

Problem: The amplifier doesn‘t work with the 5V provided by the PCs USB-port, so you need a different powersource, therefore an adaptor is needed to convert the microphone-stand to the USB-socket, but this is doable.
Additional note: Taking the power from USB-port might cause some degradation of the signal quality, so I opted for battery power.

This solution works very well for me.

One could argue that for a little more money a Weishi can be bought, but with tg software, it‘s possible to view the frequency display and also listen to the sound of the movement. Both features are not available with the Weishi (at least models 1000 and 1900 which I am aware of...).

Hope, this might be of interest for you.


Gesendet von iPhone mit Tapatalk

Posted
9 minutes ago, Thoast said:

You can buy the microphone stand that is used with the Weishi timegraphers separately for not too much money.

Well, the mic only is about USD 75 Vs 105 for a complete 1000 model, so IMHO the best buy is the latter to get a convenient standalone unit as well in addition to the software based solution.

9 minutes ago, Thoast said:

Problem: The amplifier doesn‘t work with the 5V provided by the PCs USB-port, so you need a different powersource, therefore an adaptor is needed to convert the microphone-stand to the USB-socket, but this is doable.

As explained below. I have got the parts but did't put them to test yet.

 

 

Posted

You are perfectly right.
But if you like to use the tg software with its ability to display detailed graphics and also want to hear anything more than beep-beep, while using a (kind of) proper stand with amplification built in, the described option could make sense.

In fact, it could be that someone already owns a Weishi and wants to add tg to the toolset for the given reasons.
The problem drills down to the making of an adapter cable with power-source, which is about 25 Euro.



Gesendet von iPhone mit Tapatalk

Posted (edited)
42 minutes ago, Thoast said:

But if you like to use the tg software with its ability to display detailed graphics and also want to hear anything more than beep-beep, while using a (kind of) proper stand with amplification built in, the described option could make sense.

Probably everyone know that a good software timegrapher (the one is subject is very basic) is superior to a Chinese box. I just highlighted the convenience of getting the latter at the time you buy an articulated stand microphone.

Quote

The problem drills down to the making of an adapter cable with power-source, which is about 25 Euro.

More like €5

https://www.aliexpress.com/item/32975005529.htm
https://www.aliexpress.com/item/33039021529.html
https://www.aliexpress.com/item/32892886094.htm

Edited by jdm
Posted

I don't appear to be having any problems down to no amp actually in the pickup. The fact that the preamp I'm using has a 100k input impedance might be helping with that. The main problem with amplification appears to be bandwidth - too much of it for 44khz sampling. Especially on older watches it appears to cause the peaks on the trace to move around / wrong one sensed by the software. I'd guess the bandwidth needs to be severely cut to stop that completely but  a10khz or so limit on that might help a lot.

There seems to be a similar problem on the mic end. Old watch, move things around but watch still face up and the traces become stable. :blink: stick a new one on where it doesn't usually matter and suddenly it isn't stable. This is with the watch and mic resting on 25mm of soft foam.

The as it turned out to be crap built in on board sound on my HP Mini tiny pc thingy could be helping as it's bass response is appalling. It's for a head set where this doesn't matter.  I recently fitted a Soundblaster X-Fi HD usb card and now like the speakers I am using. It should give a better noise floor than the on board but unfortunately the mic input is for dynamic mic's so don't have enough gain.

Problem with buying the china watch mic is price - not that much less than the whole kit. I just don't like that sort of thing. I've found a better than average case holder and am wondering if I can build a spring loaded pickup into it. I'm not well placed for that sort of task at the moment. It's this holder. It could even cope with a pocket watch.

https://www.amazon.co.uk/gp/product/B07TZMH57B/ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&psc=1

Most of the ones amazon sell struggle to hold a 40mm watch. The one I bought and reviewed is all plastic but really solid. At some point I can make some metal pegs and use it as a movement holder as well. It would be a pity if they have switched to metal. Plastic wont transmit sound so well.

Johm

-

  • Similar Content

  • Recently Browsing

    • No registered users viewing this page.
  • Topics

  • Posts

    • Well, my fundamental stance is that I want to go in and out without leaving any trace other than a shining, perfectly running movement. So, no scratchings on the inside of the case back lid, no marred screws, no debris, no fingerprints, and so on. That is, my goal is to make it impossible for the FBI to track me down. As a professional, I suppose you might want to keep track of returning watches, but as @JohnR725 mentioned, we can keep detailed computer records without marking the watch at all. That may not be true for every watch, but luxury and COSC-certified movements do have unique numbers. John also says it’s best to leave no sign you were ever there, and I couldn't agree more. Now, suppose the Sea-Dweller I'm working on is one day scrapped, and you want to sell the case-back separately (perhaps the case was destroyed in a plane crash). Then the scribbles on the inside no longer reflect the current movement inside the case. Also, the engraving will likely halve the market value of the case back. It had been "sleeping" for about a week and a half. Yes, the "debris/old lubricant" theory is my hypothesis as well! It will be interesting to see what I find once I have time to start disassembling the movement.
    • I've repaired a few of these, having some success with stripping and cleaning the mechanism.  They are so cheap though, its hardly worth the effort in many cases.
    • Get well soon Old Hippy, torn muscles.,  not good
    • Id love to see how he has the output shaft mounted to this setup, as I have the same rotary stage and stepper in my build.
    • If you see at least one thread protruding, or even a half, then you may have a chance using this "tool", but don't squeeze hard. Lot's of patience, and if you notice any signs of loctite or other adhesives, then apply several drops of acetone on top of broken stem. Use heat (hair dryer) also. You should be able gradually turn it counterclockwise, just like i did on this broken Timex stem, which was glued with something like nail hardener. Then you can use an extender to save your broken stem. https://m.media-amazon.com/images/I/513DVvkfW5L.__AC_SX300_SY300_QL70_ML2_.jpg   You can also try cutting a small grove / slot on top of the broken stem to use 0.7mm flat screwdriver.   Depending on the Crown material and Stem material, you can try dissolving the stem. 
×
×
  • Create New...