[24h]

I'm new to this and I'm looking for a timegrapher relatively cheap, even compared to some of them from China for $120-150.
What microphone and amplifier would you recommend?

Thanks a lot!

 

[24h]

Hi 24h,

I still use and recommend the Conrad kit no. 197688 (for everybody skilled with a soldering rod) and a cheap guitar pick up clip.

br Klaus

Thanks for the info.
The Korg CM200 clip-on microphone looks like a good option for a mic, but unfortunately I cannot purchase the Conrad 197668 here in the US because they won't ship it here.

 

[24h]

Pardon my ignorance - where is the binary for Mac OS? I see that I have to install through the terminal, but don't see the archive to download.

From what I remember, you don't have to manually download the binaries yourself. It's all though Brew, and I think it will just download the source files from the repository name.

Macintosh

A formula for the Homebrew package manager has been prepared by GitHub user dmnc. To use it, you need to install Homebrew first (instructions on http://brew.sh).

Then run the following command to check everything is set up correctly and follow any instructions it gives you:

brew doctor

To install tg, run:

brew install dmnc/horology/tg

You can now launch tg by typing:

tg-timer &

 

[24h]

Are you aware of any downsides (space, security, whatever) to using Brew (not just for this, but for anything)? I remember some other package system that caused some grief back in the 9 or early 10 days.

Not a security expert, but I wouldn't really worry about it too much. Sure, installing this kind of software on your Macbook may open it up more to attacks but I think the risk is minimal.

 

[24h]

The Korg mic didn't work for me until I used a pencil to couple the case back to the mic. I use a Behringer Xenxys 502 mic mixer ($40 on Amazon) into my desktop's mic/line input. The Korg is much noisier than the mics I use for music and voice recording, but the high and low EQ filters on the Behringer allow it to work. I have 13 watches and this rig works on all of them, although sometimes the degree field comes and goes. Sometimes it takes a an adjustment of the mic position and or level controls to get a "green signal," but this setup works and it's pretty cheap.​

You seem to know more about audio than me...
I have some cheap clip on microphones (a few from China and one Korg). When I'm listening to the audio that is picked up, there is a very loud humming noise with some static in the background.
Is this because the wires aren't shielded properly? Is there any way I can cut the current cord out of the microphone and replace it with a better shielded one?

 

[24h]

The Korg and other cheap mics are what's called "unbalanced" (two wire) while professional recording technicians use balanced (three wire) cables and connectors. Balanced setups are much lower in noise. The mics that come with time graphing machines are also balanced, but more expensive and more difficult to connect to your computer. In short, the answer is no you can't rewire a cheap unbalanced mix to balanced, but simple tone controls like on my little Behringer mixer/mic preamp can reduce the noise enough to make the program work. Probably the best solution is to simply spend a little more and buy a balanced mic intended for the purpose for $69. You will still need to buy a preamp to interface this balanced mic with your computer. Now we're up around $100 for the mic and the preamp/mixer, which is close to the cost of a complete machine, but I prefer the software to the box.

What I have is actually a modified pre-amp thanks to the help of Guido from Modifying PYLE PP444 Preamp - Reparación de Relojes
I cut two ceramic capacitors out of the curcuit and replaced one electrolytic capacitor with a 105 ceramic. I also removed the rubber dampener from the clip on mic.
All of these changes increased the overall sound quality and seemed to better isolate the actual sound of the watch.

What I currenty have is definitely usable with all of the available timing software, but simply listening to the audio or trying to add it into a video is an annoyance due to the humming noise.
Perhaps my next project will be to build a proper stand for the watch/mic combo so I can properly time various positions of the watch.

If you're interested, here's are some videos demonstrating all the changes I made to the microphone/pre-amp.
The last half of video 5 is what I'm currently using (after the capacitor replacement) and it sounds decent. Still a lot of unfavorable noise though.
https://www.youtube.com/playlist?list=PLOMif0oGPj4AXfAc7JXbbU6aBayAiXYdW

 

[24h]

That's a phono preamp not a mic preamp. Phono preamps use the RIAA curve to reduce record noise. I know you've modified it, but a proper mic preamp would help a lot, especially in the hum department, as that device is boosting the signal in hum area (60 to 120HZ), and cutting where a watch makes most of its sound, up in the treble region. My setup is a little noisy, but perfectly usable for videos, especially with the EQ (tone controls) applied.

EDIT: OMG I listened to your recordings, that is WAY too much hum!!!! You need to swap that thing out for a proper mic preamp and use it to play records. Get a microphone preamp
that takes an unbalanced input like the Behringer XENYX 502 I'm using.

The last 8 seconds of Test 5 is my best sample with the least amount of hum.
Now that I think about it, I believe that he actually used a shielded RCA cable soldered to the pickup mic and it produced almost zero hum.
I don't have enough knowledge of these cable designs to do it properly.

On a side note - is there a way, through hardware, to force single channel audio into both the left and right channel? When I record the watch ticking, it only plays on the right side. I know you can put it on both channels through software but what about a hardware solution?

:-!

 

[24h]

That's better, but still a lot of hum. https://www.amazon.com/Mizar-3-5mm-Stereo-Audio-Adapter/dp/B01DZV37VY/ref=sr_1_2_sspa?ie=UTF8&qid=1524155837&sr=8-2-spons&keywords=1%2F8+female+to+rca+male+y+adaptor&psc=1

Wouldn't the female end need to be 6.5mm to accommodate the mic?
The only 3.5mm in my setup is the 3.5mm male that goes into the computer

 

[24h]

You're right, if that's what's on the end of your mic.

Odd...I can't find a 6.5mm female to 2 male RCA.

I guess I'll need to add ANOTHER between the mic and pre-amp
https://www.amazon.com/InstallerParts-Splitter-Compatible-Most-Audio-Systems/dp/B008NCD5WC/

 

[24h]

I'm getting output on both channels from the adapter I just purchased. Now to do something about the buzzing o|

 

[24h]

I'm getting output on both channels from the adapter I just purchased. Now to do something about the buzzing

https://www.amazon.com/Behringer-50...UTF8&qid=1525630292&sr=8-1&keywords=xenyx+502

Managed to eliminate some of the buzzing in my setup with a high pass filter. It's good enough to use with the timing software and OK to listen to but it's still not perfect. I think before I buy another pre-amp I will make a few of my own piezo pickups as I just purchased 20 extras.

Edit: I actually tested with a different higher quality pre-amp and I'm still having a similar sound. I believe the mic is just not properly shielded and/or piezo is picking up other signals from screens/microwave/lights.

 

[24h]

Equalizer APO with Peace plug-in high pass filter seemed to improve the sound tremendously!
Audio: https://clyp.it/ke0zxt03

 

[24h]

Did you do any changes in your mac other than switching to an external microphone as the input? I've tried using both a piezo mic and a Røde M3 mic and I can't get any readings in TG at all... I'm using a Steinberg UR22 connected to my mac and it seems that both mics are connecting and is giving audio (I can record the ticking fine in Logic), but nothing happens in TG timer. I've already calibrated with a quartz watch from the internal mic, but the program just won't pick up anything with external mics.

TIA

I'm nowhere near an expert in this area, but I've been playing around A LOT lately with various timing applications.
Do you have a recording of the audio that your mics are picking up?

 

[24h]

My problem is that I cannot get the software to calibrate.

I am attempting to calibrate by placing the webcam mic on one of my quartz watches. When I attempt to calibrate by checking the "calibrate" box in the upper right menu. the software responds by displaying the red clock and the "cal (wait) 0%" comments. Occasionally (rarely) the clock will turn green for a moment and the dialog will report "cal (acq) 0%" but then immediately goes back to "cal (wait 0%)" and nothing else ever happens. The calibration procedure is never completed.

Unfortunately, the "documentation" in the earlier post doesn't really explain what is supposed to be happening or what these cryptic comments mean, or how to troubleshoot if the system fails. That lack of disclosure makes it impossible to debug the failed calibration procedure.

by any chance is there a debug mode in which additional status information is displayed? or additional documentation somewhere that better explains the calibration procedure?

thanks.

Can you get the software to determine the rate of your quartz watch?
Maybe the audio signal isn't ideal and the software is having issues differentiating the "ticks".

 

[24h]

Sure... though I'm not quite sure of what you're asking for -- I'm familiar with taking snapshots but I'm not aware of a recording mode in the software. If you could explain exactly what you'd like me to do, I'd be happy to do it. My suspicion though, is that we're going to see that none of my quartz watches are being picked up by the mic. I typically see only a screen that shows nothing happening.

My suspicion is that the sofware has no problem picking up the sound of my mechanical watches (see previous images) but that the software just isn't able to hear my quartz watches, even with the preamp gain full-on. It seems that the quartz watches are just too quiet for this type of mic. Of course, this is all guessing on my part, as I'm not getting any clear error message out of the software. That is to say, maybe I just don't know what I should be looking for.

Thanks.

Download Audacity (it's free) and make a recording though that.

 

[24h]

Sorry for the delayed reply.

I am familiar with Audacity, but I have't been able to use it lately due to a library incompatibility problem that's come along with a recent software upgrade. The result is that when I attempt to record, Audacity does not use the sampling rate and bit depth that I specify and the results turn out to be unpredictable and inconsistent. Once I have it working again, what settings do you require as recording defaults?

Back to the subject of calibration -- I'm having trouble understanding why we have a quartz-referenced calibration procedure in the first place, for two reasons:

First, Quartz isn't all that accurate. That is to say, Quartz is better than most mechanical movements, but it's nowhere near as accurate as a computer clock that's running at megahertz frequencies. Quartz watch accuracy is on the order of +/- 500 milliseconds/day. Although that's better than many mechanical movements, 1/2 spd isn't a good accuracy reference if you're trying to measure a watch that has an accuracy on the order of 2 spd. The Nyquist theorem tells us that sampling at 2x the desired frequency is the absolute minimum acceptable standard. With a quartz reference we barely have enough accuracy to measure the performance of a mechanical chronometer with a resolution of +/- 2 spd.

Second, any modern computer is equipped with a Network Time Protocol based system clock that operates at megahertz frequency. My plain-Jane Fedora system uses Chronyd, which performs regular time-syncs, polling it's upstream time server every 2^6 seconds, and between polls it automatically compensates for any skew in the system clock. The following query shows that the NTP-calibrated system clock on my PC is 1.78 nanoseconds slow of NST.gov's atomic NTP time as I type this:

[FONT=monospace][COLOR=#5454FF][B]#[/B][/COLOR][COLOR=#000000] chronyc tracking                                                                                                  [/COLOR]
Reference ID    : 0A0A0A01 (*********)
Stratum         : 2
Ref time (UTC)  : Mon Aug 13 22:58:52 2018
[B]System time     : 0.000001780 seconds slow of NTP time[/B]
Last offset     : -0.000003389 seconds
RMS offset      : 0.000005357 seconds
Frequency       : 37.157 ppm fast
Residual freq   : -0.002 ppm
Skew            : 0.077 ppm
Root delay      : 0.036606569 seconds
Root dispersion : 0.047560256 seconds
Update interval : 64.6 seconds
Leap status     : Normal
[/FONT]

Nanosecond resolution is millions of times higher resolution than millisecond resolution. Using a quartz based reference that has accuracy on the order of 500 msec doesn't seem worth the effort when a call to the system clock can obtain time at 1.78 nsec resolution. That's actually a difference of what, 280 million times higher frequency?

Given that any modern PC uses NTP time sycing (Microsoft, Apple or Linux, take your pick), nanosecond level time accuracy is available on any modern PC. I don't understand why the timing software is querying the soundcard clock as a time reference, as the soundcard clock is uncalibrated. It also doesn't make sense to me why the software expects a user to use a quartz-based calibration procedure using an external time source (watch) when polling the system clock would provide a reference source for calibration that's accurate to the nanosecond level and is presumably hundreds of millions of times more accurate. Polling the soundcard for timing seems like it would provide a lesser result than polling the system clock on any NTP-based PC system.

I have to assume that I'm misunderstanding something. Thanks for your time.

While I don't fully understand how a computer's clock functions related to the processor, I believe that in the case of watch timing software we are solely relying on the sound card clock.
I find the best way to calibrate TG and Watch-O-Scope is to start a timing test for a quartz watch with an app like Toolwatch.io for an ENTIRE month while storing your watch somewhere with a consistent temperature.
The readout in this app is your KNOWN RATE.

For this example, let's say that the known rate is +0.4 s/d.
Now, start a calibration in the software. We'll use a CALIBRATION readout of -0.6 s/d.
Subtract KNOWN RATE from CALIBRATION.
-0.6 - 0.4 = -1.0

Since the software is effectively 1 second slow we need to correct it with a positive number.
Enter the calibration value of +1.0 in the software.
If your number after performing the subtraction is positive, use a negative number for the calibration value.

 

[24h]

Anyone having problems measuring a rolex because of the chiming echoe sound of the movement ?
I pass the sound via Ableton to EQ it a bit but its a bit jumpy. My Omega is way easier and clean (sound).

I use a guitar piezo for tuning etc. Via a mixer with dedicated Mic input.

If you could share an audio file to browse we may be able to offer help.

Some of my watches suffer from this problem also. I think the culprit is mostly the echoing from your case/movement ring/caseback.
Also it's possible that if you are using a standard piezo disk, it it deformed from the force of your watch crown and the audio signal is distorted.

 

[24h]

Hi guys,
Anyone know what happend with download links on project's page? I used this program some time ago to regulate my 7s26 and it's doing great gaining +3 seconds per day on avarage. Now I wanted to download and use it for my next project and all links disappeared.

Works for me: https://github.com/vacaboja/tg/archive/master.zip

 

[24h]

deleted

 

[24h]

Direct link to the audio file.

https://relojesdetorre.weebly.com/uploads/1/0/1/3/10134874/skeleton_black_15sx.wav

Your site is still not loading. Maybe an issue on my side, but everything else loads 🤔
Edit: It's working now