Post by ausimax on Nov 24, 2015 20:11:58 GMT -8
Has anybody had any experience with watch timing software?
I came across an old post on SCWF by Feca67 about using Bilburo software with a home made mike/amplifier and decided to have a go,
bought a cheap guitar mini amp & some peizoelectric guitar pickups and constructed the hardware.
I was surprised at the sensitivity of the pickup picks up the TV in the next room as background noise.
I downloaded the Bilburo software and proceeded to run some test on a Seiko 7019 movement (uncased without dial or hands) so I had no idea of its timing. The main problem I found was that even though I was taking readings with the watch on the pickup back down the slightest movement of the watch, even just rotation changed the amplitude of the recording and that effects the result of the test because you then have to alter the software settings.
Tried altering the watch beat error and time error, however you would get it fairly good then a couple of readings later its all over the place.
I also tried using Delph software, only problem is the freeware version can't be used in real time, you have to make and save a WAV file recording in Audacity and then run those files in the software, this seems to give a more stable reading all I need to be able to do is interpret the data.
All this is because I don't want to spend $250 on a Timegrapher, a lot of loot for the use I have for it.
Anyway I will show some of the results below.
Bilburo
Ist run.the auto beat rate didn't!
Shows up as running slow.
Adjusted the timing.
Only change watch rotated 180, timing out again.
Another rotation and back on time again.
another rotation and its going fast.
All these changes are just from rotation of the watch centred on the middle of the pickup with the watch in the back down position.
This shows the sensitivity of the pickup, wife switches on TV in next room, I place a thermofoam box over the watch to cut background noise.
Generally though I can't seem to get reliable enough results to realistically adjust a watch.
Delph Software
The first 2 readings were taken from 2 16sec recordings made back down with no changes to watch position between.
The next recording is made face down.
These readings seem to be more useful than the Bilburo ones, still need to do some more work on it.
Interested if anyone else has tried these methods with any success.
Max
I came across an old post on SCWF by Feca67 about using Bilburo software with a home made mike/amplifier and decided to have a go,
bought a cheap guitar mini amp & some peizoelectric guitar pickups and constructed the hardware.
I was surprised at the sensitivity of the pickup picks up the TV in the next room as background noise.
I downloaded the Bilburo software and proceeded to run some test on a Seiko 7019 movement (uncased without dial or hands) so I had no idea of its timing. The main problem I found was that even though I was taking readings with the watch on the pickup back down the slightest movement of the watch, even just rotation changed the amplitude of the recording and that effects the result of the test because you then have to alter the software settings.
Tried altering the watch beat error and time error, however you would get it fairly good then a couple of readings later its all over the place.
I also tried using Delph software, only problem is the freeware version can't be used in real time, you have to make and save a WAV file recording in Audacity and then run those files in the software, this seems to give a more stable reading all I need to be able to do is interpret the data.
All this is because I don't want to spend $250 on a Timegrapher, a lot of loot for the use I have for it.
Anyway I will show some of the results below.
Bilburo
Ist run.the auto beat rate didn't!
Shows up as running slow.
Adjusted the timing.
Only change watch rotated 180, timing out again.
Another rotation and back on time again.
another rotation and its going fast.
All these changes are just from rotation of the watch centred on the middle of the pickup with the watch in the back down position.
This shows the sensitivity of the pickup, wife switches on TV in next room, I place a thermofoam box over the watch to cut background noise.
Generally though I can't seem to get reliable enough results to realistically adjust a watch.
Delph Software
The first 2 readings were taken from 2 16sec recordings made back down with no changes to watch position between.
The next recording is made face down.
These readings seem to be more useful than the Bilburo ones, still need to do some more work on it.
Interested if anyone else has tried these methods with any success.
Max
