Jump to content

My Experience With Chronometer Testing and Certification


Recommended Posts

Introduction

Before I jump in to sharing my story I will introduce myself. I’m Matt. I’m the founder, designer, and watchmaker of Typsim Watches. Typsim is a small, independent watch company based out of Seattle, WA, USA. This is my experience with chronometer certification of the Typsim 200M-C.

I chose to have some of the Typsim watches chronometer certified out of my desire to produce the high-quality, accurate watches. For me, chronometers represent watches that use the highest quality materials, are put together with great care, and are rigorously adjusted and tested. But further, they connect a timepiece to the history and pursuit of precision timekeeping

Typsim is a USA-based company, registered in the United States. The watch components come from Hong Kong, China, Canada, USA, and Switzerland. The watches are designed, assembled, and regulated in the United States. Because of this, Typsim watches or movements cannot be tested by COSC, Contrôle Officiel Suisse des Chronométres.

DSC_0562.jpg.44c9c5f1595c3077118c3a410b7be4eb.jpg

The Typsim 200M-C is designed with a Sellita SW300-1 movement. This is Sellita’s comparative movement to an ETA 2892-A2 or Omega 1120:

Diameter: 11.5”’ (25.6mm)

Height: 3.6mm

Frequency: 28,800vph (4hz)

Jewels: 25

Power Reserve: 56hrs

Features: Center Second Hand, Quickset Date

Sellita offers their movements in several grades, Special (Elabore), Premium (Top), and Chronometre. The primary difference between Premium and Chronometre is the precision in which the movement is regulated for the average rate, max positional rate difference, and isochronism tolerance. The Typsim movements were ordered as rhodium-plated, diamond-snailing decoration, blued screws, custom logo engraved oscillating weight, custom date disc, with Premium regulation and Chronometre grade components (balance, pallet fork, escape wheel).

SellitaSpecs2.thumb.jpg.99bf91c95587d2cd9430d56c0f0745a7.jpg

SellitaSpecs1.thumb.jpg.0c5afbc275729cd5b569d82ca2120552.jpg

 

What’s required of chronometer certification?

The chronometer test program is defined by ISO (the International Organization for Standardization). ISO also accredits the testing laboratories that perform the tests. For reference, COSC follows ISO standards and is accredited as a testing agency by Swiss Accreditation Service. There are several other testing agencies around the world that can test and certify chronometers which include Besançon Observatory in France, Glashütte Observatory in Germany, and very recently the Horological Society of New York in the United States. A certification is provided to a wristwatch that successfully meets the requirements of the testing program defined by ISO 3159:2009. Test testing protocol lasts 15 days, tests 5-positions and 3-temperatures.

The minimum requirements are:

ISOProtocol.thumb.jpg.5980913a35b4e344c95a81eaa5c74f6a.jpg

ISOMinReq.thumb.jpg.81572a6c0c70e59dde2d36495a01a7c6.jpg

I selected to have the Typsim watches tested and certified by Observatoire de Besançon in Besançon, France

DSC_0358.thumb.jpg.03b219f9c2ad32738e5c5dd8bde51ecb.jpg

Besançon, the historical watchmaking capital of France, is located in the eastern part of the country near to the Swiss border. It’s history as a watchmaking place began shortly after the French Revolution and culminated in the mid-twentieth century. Unfortunately, the 1970’s rise of quartz-based watches create a sharp downturn in mechanical watchmaking and shuttered much of this city’s watchmaking companies. The city’s expertise in precision work pivoted toward micro technologies. Since the 90’s the watchmaking industry has slowly recovered with a focus on high-end and custom watches. In 2002 the Observatoire de Besançon reopened for chronometer testing. Since then, they test several hundred watches a year, mostly from independent companies like Laurant Ferrier and Voutilainen. The typical success rate hovers around 70% with even the best companies averaging 80% success rate.

DSC_0506.thumb.jpg.18544ef6142cbc1fbe8ce6c015752285.jpg

Testing at the Observatoire de Besançon differs from COSC in that the Observatoire de Besançon only accepts fully assembled, complete watches with all components engaged for testing. COSC on the other hand tests bare movements with the automatic device (if a part of the movement) disengaged. This means testing with the Observatoire de Besançon accounts for more complexities faced by assembling and having an assembled watch.

DSC_0433.thumb.jpg.49951f3116b7ea37e4c40a85913ce55f.jpg

 

Preparations

Understanding the testing requirements and average failure rates, I felt immense pressure to submit the very best performing watches I could. The first step was identifying the best movements out of the batch of movements provided by Sellita.

Sellita provided movements conforming to their Premium regulation standards:

Adjusted in 5-positions

Average rate of +/- 4 s/d

Maximum difference between positions of 15 s/d

Isochronism of +/- 10 s/d

Per terms with Sellita, up to 5% of the movements are allowed to be delivered outside of Premium regulation specifications.

I created a preliminary, bare movement testing protocol to identify the best movements out of the box. I tested each movement in 6-positions with 20-seconds of stabilization and 40-seconds of measurement per position at both full wind and -24hrs wind. I was able to discover the number of winds to achieve the full wind and -24hr wind thru crown rotations, not waiting 24hrs. I measured the number of winds it took to wind up a watch to full wind after 24hrs. That number was subtracted from the number of winds it took to fully wind a movement to determine how many winds to wind to -24hrs. The instantaneous values from the timegrapher were inputted into a spreadsheet which calculated:

Average Rate

Weight Average Rate

Minimum Rate

Maximum Rate

Delta

DVH

Di

Isochronism

Hypothetical Chronometer Testing Results based on the average between 0hr and 24hr rates

TimingProtocol.thumb.jpg.10eac3c7af8dc6590c9a57dd63fc7c40.jpg

From this information, the best performing movements were selected for assembly. Movements were selected for the amplitude, min/max rate delta, isochronism, and their apparent aptitude for passing the hypothetical chronometer test. These values gave me an idea of how well the movement was fabricated, lubricated, assembled, and regulated from the factory. Serial numbers were engraved on to these movements.

One anomaly that bothered me at first, but I eventually let go, was the rate in the 12H (crown right) position. This is a position not tested under ISO 3159:2009 nor as a part of the Sellita specifications. Frequently, this position presented beat rates 5 – 7 s/d slower than the minimum beat rate in the other 5-positions. The selected movements were selected in part because they did not demonstrate this beat rate delta.

With movements selected, the Typsim 200M-C watches were fully assembled and retested all over again using my pretest protocol. This was done to verify the assembly process was executed without causing issue. The watches were all adjusted based on the pretesting protocol. The beat error was reduced to 0.0 or 0.1 in CH (dial up) position. The rate was adjusted based on the average beat rate between 0hr and 24hr instantaneous rates. The Etachron pins were adjusted to narrow the isochronism down as close to 0 as possible. I would soon find out this was not the best route to make adjustments for this test.

DSC_0124.thumb.jpg.b580906b25cc62853d67a1373d6b7c9b.jpg

DSC_0129.thumb.jpg.c3ac0e5f6e65090d12880d80aecc7763.jpg

DSC_0132.thumb.jpg.cc2558117fec6b8edd73ad1a8fa12e52.jpg

With the watches adjusted I set about testing them in accordance with the ISO 3159:2009 testing protocol. I planned to take measurements every 24hrs and adjust as required. The watches were set against www.time.gov and measured against the same online atomic clock after 24hrs to determine their actual, daily positional rate.

I quickly discovered the movements were not demonstrating a rate that averaged between 0hrs and 24hrs rates but instead represented the instantaneous rate taken at full wind. Said differently, in real world timing condition, the rate was what a timegrapher said it would be at full wind.

If the full wind timegrapher results for CH (dial up) indicated +1 s/d, the 24hrs wind timegrapher result indicated -4 s/d, the watch would demonstrate +1 s/d when measured against the online atomic clock after 24hrs. This was true of all positions. From the point of this discovery onwards, my adjustments were made only considering full wind timegrapher results and actual timing results.

Each movement was tested for actual timing results for two weeks in CH (dial up) and 6H (crown left) positions. These two positions were prioritized for two reasons. First, ISO 3159:2009 emphasizes the rates and differences between these positions in the minimum requirement calculations and number of days tested in these positions compared with the other positions. This meant these positions needed to be as close as possible in their rate. Second, when CH and 6H rates are adjusted to be equal the other 3-positions were reined in to be within +/- 2 s/d, having a 5-position rate delta of around 4 s/d. Confirmation of 5-postiion average rate and rate delta was performed every 4-5 days during this process using the pretest protocol. Achieving a 4 s/d tolerance provided for a daily average rate that was well within ISO 3159:2009 requirements.

My methodology for equalizing the rates between CH (dial up) and 6H (crown left) was to measure the instantaneous rate at full wind for each position. Depending on the results, the Etachron regulator pins were rotated further opened or closed. If vertical position had a faster rate than the horizontal position, the pins were opened. If the vertical position had a slower rate the pins were closed.

I adjusted the pins with the movement running at full wind, testing for the results after each adjustment. I found, minor rotation of the pins would increase or decrease the beat rate in CH (dial up) position. The smallest rotation I could make would affect the CH (dial up) beat rate by +/- 10 s/d. I discovered that amount of rotation would bring the rates between CH and 6H together by about 1 s/d.

If my CH (dial up) rate was +1 s/d and my 6H (crown left) rate was +4 s/d I would need to open the regulator pins just enough to reduce the vertical rate by 3 s/d. I would rotate the regulator pins until the CH (dial up) rate was approximately +31 s/d. I would then adjust the regulator arm to +1 s/d and measure the resulting rate for 6H (crown left). The resulting rate would typically measure +1 s/d. I would not call this scientific, but the results were consistent enough that I could rely on the action and corresponding results. I have a suspicion the instances that didn’t work out this way were due to my inadvertently moving the regulator arm, hence changing the rate more than the pin rotation alone.

DSC_0565.thumb.jpg.da2767b34ccca435b04baa2e70b143c8.jpg

At the end of the 14 days the watches were performing at +0 s/d to +1 s/d in both positions – both as shown on the instantaneous timegrapher results and with 24hr timing.

 

The Test

Historically, the Observatoire de Besaçon used a Lunette Meridienne instrument to measure elapsed time through precise observation of solar bodies passing through the meridian. This one axis instrument sits on a dedicated, three-story, solid foundation – as do the reference piers outside the building – to maintain instrument stability. Comparing the elapsed time on the watch with the movement of solar bodies allowed the observatory to calculate the rate to the tenth of a second.

DSC_0366.thumb.jpg.827c727e642069286db219acaba7906f.jpg

DSC_0377.thumb.jpg.a10bfb12ac51e2980d0be8d60de3b518.jpg

Today, Observatoire de Besaçon uses three interrelated atomic clocks located in a thermally and seismically isolated room to synchronize the various clocks throughout the laboratory. These clocks are a part of the French national and international atomic clock organizations, setting the most precise time around the world. The watches are provided a QR code, and the hands are digitally synchronized to their timing software with a high resolution, macro camera. Each watch is held in position by holders and placed in a temperature and humidity-controlled vault each day after measurements and winding.

DSC_0351.thumb.jpg.2d5cd57bf93bd14708010d76a8e937bc.jpg

I decided to fly the watches from Seattle to Besançon myself, hand delivering them to the Observatoire de Besançon. Due to the sheer value of the watches being shipped from the United States to France and back again, the shipment needed to be processed through customs with a Carnet. This is a special, temporary export/import procedure allowing for products to be shipped between countries without duty. The caveat to this procedure is the shipment needs to be hand carried or couriered (basically not shipped with FedEx, DHL, etc). I also had concerns that the vibrations of a cargo plane or handling of a courier could displace the precision adjustments in the watches. Finally, I personally wanted to see the historical observatory, meet the staff, and understand every part of the process.

DSC_0383.thumb.jpg.a4c19f860fb91deb455a7d16e1d5408b.jpg

The watches safely made it to France with me in mid-August 2022 and I was notified of their success rate in early September. The 95% of the watches successfully passed the test and were certified chronometers. 70% of the movements had an average daily rate of -0.5 s/d to +2.0 s/d, aligning with what the preparations and pretesting indicated. The remaining successful results were just barely outside that range at -2.0 s/d or +3.0 s/d. The failing results were not far off at around +6.5 s/d, just outside of minimum requirements.

103158760_DSC_05622.thumb.jpg.b69dc2ff2a978e1cfae9ed2906c4d244.jpg

Each movement will now receive a Tete de Vipere, symbol of the viper, the historic marking used by the Observatoire de Besançon that the movements were certified chronometers in Besançon.

VipereGuidelines.thumb.jpg.b75293b1c88991490f772d9c5dcb1ff8.jpg

Vipere1.thumb.jpg.c64ea1373cb84309928e38f5718c96c7.jpg

DSC_0254.jpg

  • Like 4
  • Thanks 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.



×
×
  • Create New...