[dmnc]

For other OS X user, it is better to point out that there is a little confusion in the operations that Roli-poli attempted. To use the homebrew formula prepared by dmnc, as far as I can understand not being a mac user, the correct sequence of operations is

step 1 - install homebrew as indicated in [url]http://brew.sh[/URL]

step 2 - run brew doctor and do everything it asks of you

step 3 - run the magic command: brew install dmnc/horology/tg --HEAD

And this is it. No other step whatsoever is required. Please correct me if I am wrong...

I have made a small addition here as it seems there is some confusion about xcode/dev tools. If homebrew hasn't already warned you, brew doctor will tell you if you are missing anything on your system and give instructions for how to sort it out, but it's usually as simple as xcode-select --install to get dev tools on the newer versions of OS X. You shouldn't have to worry about downloading all of xcode just to get make.

The homebrew formula will deal with the gtk+, ffth etc. dependencies for you so you shouldn't have to worry about any of that.

 

[contrate_wheel]

Thanks dmnc, I will edit my previous post with your correction.

 

[beastomaniac]

How accurate measurements have you got with this software? I've been using it for about 3 months and it seems to be very good, but are those readings close to professional timegraphers measurements?

What kind of microphones have you used, what's the cheapest one to use with this software? I have a DIY measurement microphone with pre amplifier but it's quite big and cumbersome...

 

[Halda]

How accurate measurements have you got with this software? I've been using it for about 3 months and it seems to be very good, but are those readings close to professional timegraphers measurements?

The calculations are probably as accurate as "professional" timers, at least for the rate and beat error values. There might be some disparity with the amplitude calculations at the moment. If so, it will probably be fixed. It's a new feature and still in beta after all.

However, there's no calibration implemented yet. So the accuracy of the numbers you get will depend on how well your sound card performs.

 

[Fred.S]

Hi,
I like the software very good signal detection, Im using it with a microphone from a L&R tickoprint t45, without amplification I just soldered a shielded cable to it and plugged it in my USB sound card. I think it's too good to be true, because when I use Delph escapement analyzer, it doesn't produce a good result, maybe I'll try an amplifier and I need to calibrate the sound card time or frequency, I don't know how, I tried to look it up but all the articles I found are complicated to follow, if someone knows a method, please share it with us, I think it would be very helpful.

One thing I would like from the developer is to add the ability to record audio to check it on other timing as well, taking advantage of the filtering quality.

I posted pictures of my setup.

Thank you WUS.

 

[jappa]

Needed to say thanks for this awesome software! It worked well with my ordinary mic but ordered $5 clip on contact microphone from eBay and it's even better.

Anyways, could somebody post graph from well working 1861 so I could compare? I am a bit worried if I should get my Speedmaster serviced. I bought it few months ago from watchmaker and he said there's no need for service as amplitude is >270 degrees. It was, at the beginning even though bet rate was out-of-spec (something like +15s/d). After few months beat rate is actually better (+10s/d or so), but amplitude looks funny... tg won't even show amplitude anymore, I think difference between tics and tocs are too big? Other value would be 240 degrees and other 300 degrees... Average would be exactly 270 degrees, but is this big difference between tics and tocs ok? I have zero idea...

 

[beastomaniac]

Here's a graph of my 2004 Speedmaster with similar movement (cal. 1863).

 

[Klaus Tickalot]

... but is this big difference between tics and tocs ok? I have zero idea...

Hi Jappa,
did you try to connect the clip to the crown or at least the mic to the caseback if you got a plexi?
Sometimes - depending on the watchcase - I'm experiencing as well some kind of echo which confuses TG as well as biburo or WOS.
In case of a very quiet watch I connect directly to a save place e.g. crown wheel or a stable bridge.

Just clip it to the right place and timegrapher will do its magic b-)b-)b-)b-)b-)

 

[adamant365]

In regards to the accuracy of the measurement: I just did a test and got an interesting result. I found 300 BPM (18000 BPH), 420 BPM (25200 BPH), and 480 BPM (28800 BPH) click tracks on youtube. They were all produced by the same user. Every one of them was showing exactly +33 s/day on TG. I can only assume that the software used to create these tracks was slightly off in their calculations introducing this error. In addition, using TG to measure the rate of my Speedmaster Pro in crown down position gives a result of +1.0 s/day. The observed time difference of the watch from 24 hours ago until now: 0.2 seconds fast. The reason it is +1.0 now is due to the slight increase in rate due to drop in amplitude from the watch being less wound.

I would say it's safe to assume the accuracy of this app is going to be well within 1 s/day, probably even better than that. I would also say the instantaneous measure of rate by TG is going to be way more consistent (precise) than manually measuring the rate after a 24 hour rest in any position.

I can't say this enough...but VERY well done!

 

[jappa]

Hmm I am unable to get good graph using tg... But I am starting to really think that service is now necessary. I listened recording made using piezo mic and theres some extra noise, kling kling kling on some positions... Check this out:

https://www.dropbox.com/s/tavp9t1nz2cbp6t/speedynoise.mp3?dl=0

I am rotating watch to different positions starting from 10 sec and sometims extra kling kling kling is heard. I know nothing about watch movements but I assume this is not normal. What do you say?

 

[adamant365]

Hmm I am unable to get good graph using tg... But I am starting to really think that service is now necessary. I listened recording made using piezo mic and theres some extra noise, kling kling kling on some positions... Check this out:

https://www.dropbox.com/s/tavp9t1nz2cbp6t/speedynoise.mp3?dl=0

I am rotating watch to different positions starting from 10 sec and sometims extra kling kling kling is heard. I know nothing about watch movements but I assume this is not normal. What do you say?

I'm no expert, but that does not sound normal. It's almost like the balance is hitting something...i.e. doesn't have enough clearance. Is there a dramatic decrease in amplitude when the pinging begins??

 

[Klaus Tickalot]

kling kling kling

... is the sound of the hairspring which seems to be dirty (minimal amount of oil) and probably sticking short time at the pins of the regulator. Adhesion ends when the balance travelled a few degrees after zero crossing, the hairspring abruptly unsticks and this will cause an extra (bad) vibration. You can even see this under the loupe.
You should visit you watchmaker to clean the hairspring as a gesture of goodwill.

 

[saltddirk]

I just stumbled upon this thread and of course I had to download and play with it!
Although I have no, but absolutely no idea of what I am looking at, this is a fantastic piece of software, Well i managed to get the data from my only mechanical watch that I have with me.
I have learned an awful lot in the hour or so that it took me to read through the thread and the links provided.
For a complete NOOB as me, while I get the visual representation of the Tick and tock on the right, what is the graph on the left showing?

As we are all more or less serial hoarders it would be great to have an user input field on the right in the title so that you can personalize the graph with let say the watch name before taking a screen shot and save it for future reference

I tested the SW on a Windows 7 with my Logitech skype headset and on the laptop which runs Win 8.1 with the Bose quiet -something- phone headset. Both worked great, although I got the impression that the signal pick up was more stable on the laptop.

Thanks for all the hard work!

D

 

[jappa]

... is the sound of the hairspring which seems to be dirty (minimal amount of oil) and probably sticking short time at the pins of the regulator. Adhesion ends when the balance travelled a few degrees after zero crossing, the hairspring abruptly unsticks and this will cause an extra (bad) vibration. You can even see this under the loupe.
You should visit you watchmaker to clean the hairspring as a gesture of goodwill.

I went to watchmaker and he said that movement is in excellent shape and full service would not do any good, could do the opposite. Only thing was that balance wheel or something was not centered or something like that, he suspected that there has been slight impact or something... He tinkered around with it a bit and no more extra sounds, and, well take a look at yourself...

I would say that this is pretty good for 15 year old movement designed 50 years ago, huh? Though in real life there is positional differences but anyways, much better than before. He did not charge anything even though I tried to pay...

 

[eeki]

Thanks for the effort developing the software, just used homebrew to compile the OS X -version and it works well enough even with a simple headphone mic:

One small suggestion however - it seems that the software does not allow for decimals in the lift angle, would it be possible to change this? Many older Seiko calibers have lift angles of 54.5 (as the 6117 above) and one even has a lift angle of 52.36 (cal 6159), so if it would be possible to use two decimals it would be nice.

Also, it might be helpful to include the Mac OS instructions already in the first post - this topic is getting a bit lengthy and the search functions on the forum are not exactly ideal for searching within topics.

Sent from my iPhone using Tapatalk

 

[contrate_wheel]

Hi my friends!

Thanks for the suggestions, I just wanted to tell you that I have not disappeared, I just had a lot of work in the past months. There are new versions of tg to come: I have parallelized the algorithm, which makes it much snappier on multi core machines, and a fellow named wahlstedt on GitHub has made improvements to the user interface.

A question: does anyone prefer the tg-lt version to the regular one?

 

[adamant365]

Hi my friends!

Thanks for the suggestions, I just wanted to tell you that I have not disappeared, I just had a lot of work in the past months. There are new versions of tg to come: I have parallelized the algorithm, which makes it much snappier on multi core machines, and a fellow named wahlstedt on GitHub has made improvements to the user interface.

A question: does anyone prefer the tg-lt version to the regular one?

I only ever use the TG version. I don't use tg-lt at all. Is the new version available or is it "coming soon?"

 

[Deafboy]

Hi,
I'm new to this forum. I used this very nice software on several of my watches with good success. It really works well. I'm impressed how it digs the watch beat signal from ambient noise (does it use some kind of synchronous filtering?). Anyway, I have a request. I also would like measure the beat of a pendulum clock to analyze escapement characteristics. A major difference is that beat time is 1 beat per second, which equates to 3600 beats an hour. The current software seems limited to, I think, about 11000 bph. Can you modify the software to have this extended range?

Michel

 

[ChrisTheEngineer]

Hi my friends!

A question: does anyone prefer the tg-lt version to the regular one?

I use the tg-lt version on an old win xp box. I expect it is extra bother to make two versions, so I would understand if it went away.

 

[Deafboy]

A question: does anyone prefer the tg-lt version to the regular one?

Regular for me.