Before the 2442 G.C.T. became a Kickstarter project, it was simply a personal passion project of mine.
Looking back on this journey, I gained far more from the process than I had originally expected. In particular, my research into 24-hour mechanical movements gave me a much deeper understanding of movement architecture, display logic, and the many compromises behind watch design.
I have always considered myself a pragmatic engineer, and I have always believed in one idea:
“Everything that exists, exists for a reason.”
If a day has 24 hours, why are most watches designed to display time in a 12-hour cycle?
Over the past few years, I have been deeply fascinated by 24-hour watches. On one hand, I find the logic of a 24-hour time display very reasonable. On the other hand, I also wanted to use my own work to express some personal ideas about time, tools, and design.
On paper, this direction sounded reasonable and logical. But once I actually started working on it, I realized that things were not as simple as I had imagined.
Ready-Made Movement Options on the Market
Like most people, my first approach was to look for something ready-made: a three-hand, pure 24-hour mechanical movement that could be used directly.
I searched through Taobao, eBay, and even contacted different movement suppliers to ask whether such a movement was available. Eventually, I found a Swiss movement supplier offering something close to what I wanted. It was based on a modified ETA 6497 movement, with central seconds and modified hour and minute wheels to achieve a true 24-hour gear ratio.
However, the price was simply too high.
Personally, I do believe in the modification capability of that movement supplier. The problem was not whether they could do it, but whether I could afford it at that stage. My own resources were very limited, and I could not justify spending that amount of money on one movement. In the end, I used that budget to purchase more tools and equipment instead, which is another story for another time.
More importantly, once I later decided to turn this personal project into a commercial product, the cost of the movement became directly tied to the total cost of the watch. For that reason, I eventually had to give up this option.
Although I did not choose that route, it gave me an important direction.
When I first started learning watch repair and maintenance, like many others, I began with the classic ETA 6497. It is a mature pocket-watch movement with a relatively simple structure and larger components, making it easier for beginners to understand and work on. The existence of Chinese 6497 clones, such as the ST3600, also lowered the cost of learning and experimentation significantly.
Because of that, I began studying the structure and modification potential of the 6497. I also compared it with the Hamilton 4992B that belonged to my late grandfather. The 4992B is a classic military 24-hour pocket-watch movement, and it further convinced me that a 24-hour display is not simply something made to be different. In certain contexts, it has real practicality and historical reasoning behind it.
In principle, converting a 6497 into a 24-hour movement is not difficult to understand. In a standard 12-hour movement, the hour hand completes one full rotation every 12 hours. If the gear ratio of the hour wheel and minute wheel is adjusted so that the hour hand rotates at half the speed, it can complete one full rotation every 24 hours.
But once I actually started working on it, I realized that the problems were far more complex than expected.
The hour wheel and minute wheel are not just simple gears. They are related to hand height, dial clearance, the keyless works, and the overall assembly of the movement. If even one dimension is wrong, the movement may not become usable, even if the gear ratio itself is correct.
However, this approach still had one limitation that I could not ignore: it was not a central-seconds movement.
The 6497 was originally designed with small seconds, with the seconds hand positioned at 9 o’clock. Although this layout has a strong classical charm, the 24-hour wristwatch I had in mind was something closer to a three-hand central-seconds watch. Therefore, even though the 6497 modification was technically possible, it was still not the final answer I was looking for.
This taught me something important: making a 24-hour movement that can run is one thing; making a movement that is truly suitable for a product is something else entirely.
So I continued looking into other native 24-hour movements.
Russian 24-Hour Movements: Vostok and Raketa
After exploring the 6497 modification route, I began researching Russian movements that were originally designed for 24-hour displays. The names that came up most often were Vostok and Raketa.
The existence of Russian 24-hour watches was not accidental. The 24-hour display is closely connected to use cases such as polar exploration, military use, and navigation. In environments with polar day or polar night, where daylight can no longer clearly indicate whether it is day or night, a watch whose hour hand completes one rotation every 24 hours is not just visually unusual. It becomes a genuinely functional tool.
I also own a Poljot International Polar Bear 24-hour watch. This watch uses a Vostok 2423 manual-winding movement, with 17 jewels, a frequency of 19,800 A/h, and a power reserve of around 38 hours. It is not an expensive or complicated watch, but that is exactly why it gave me a very direct experience of the charm of 24-hour timekeeping.
When the hour hand only makes one full rotation per day, the rhythm of reading time becomes completely different. You are not simply looking at “what hour it is in the morning or afternoon.” Instead, you are reading where you are within the whole day.
Another movement worth studying is the Vostok 2431. It is an automatic 24-hour movement with central seconds and a date function. It is around 24mm in diameter, approximately 6.3mm thick, runs at 19,800 bph, and has a power reserve of around 31 hours. In terms of specifications, it is closer to modern wristwatch expectations than a manual-winding movement, and it has appeared in some Russian 24-hour watches from brands such as Vostok and Sturmanskie.
The Raketa 2623, on the other hand, carries even more historical character. Raketa’s 24-hour watches are closely connected to polar exploration, and its Polar Watch was developed for Soviet Antarctic expedition use. The Raketa 2623.H is a manual-winding, central three-hand movement, around 26mm in size, with a frequency of 18,000 bph and a power reserve of around 45 hours. Its appeal lies in its purity and historical background, but today, it is mostly found through the second-hand or vintage market.
However, although these Russian movements are interesting, they were not necessarily suitable as the final production solution for my project.
First, supply stability was a problem. This was especially true for movements such as the Raketa 2623, which are difficult to source as a long-term production movement. Second, the parts, cases, crowns, and hand specifications are different from the more common Swiss or Japanese movement systems, which limits the availability of compatible components.
For personal collecting or experimentation, these limitations are acceptable. But if the goal is to create a watch that can be produced, serviced, and controlled in cost, these issues quickly become real obstacles.
This part of the research taught me that a movement can be historical, interesting, and even deeply charming, but that does not necessarily mean it is suitable for a commercial product. Vostok and Raketa helped me better understand the tool-watch background of 24-hour watches, but I still needed to find a solution that could be integrated more easily into a modern supply chain.
Thinking Backwards from a GMT Movement: Discovering the HZ6460
After studying Vostok and Raketa, I began to think in another direction.
If it was difficult to find an ideal three-hand pure 24-hour movement, could I approach the problem backwards from a GMT movement?
A GMT movement already has an output that completes one rotation every 24 hours. In normal use, this is displayed as a fourth hand to show a second time zone. But if we look at it differently, the GMT hand is already a ready-made 24-hour hand. The real question is how to turn it into the main display, rather than leaving it as a secondary function.
During this process, I discovered the HZ6460.
It is a Chinese Hangzhou-made automatic GMT movement, commonly understood as being based on the ETA 2836 architecture with an added GMT module. To me, the most interesting thing about it was not simply that it was a GMT movement, but that it offered a very interesting possibility for modification.
Compared with Vostok or Raketa, the HZ6460 had several practical advantages.
First, it is an automatic movement with central seconds, which is much closer to the basic configuration I wanted for a modern wristwatch. Second, its base architecture is similar to the ETA 2836 system, meaning the case, date window, rotor, winding, and setting structures are easier to understand and integrate within a modern watchmaking supply chain. Third, its GMT output already completes one rotation every 24 hours, making it a very worthwhile foundation to study.
However, the HZ6460 is not a pure 24-hour movement that can simply be purchased and used directly. It is still originally a four-hand GMT movement, not a three-hand 24-hour movement. Its GMT module, hand height, dial spacing, stem height, and setting positions all needed to be studied again.
After disassembling it, I found that the HZ6460 is essentially modified from the 2836 architecture. The original day-date structure is removed, and the day corrector is modified into an intermediate wheel that drives a GMT wheel mounted above the hour wheel. In other words, it does not redesign an entirely new 24-hour gear train. Instead, it uses the original day quickset position and structure of the 2836 to add a GMT display mechanism.
The GMT wheel itself is also very interesting. It has a two-layer gear structure. The upper gear is fixed, while the lower gear is originally movable. Between the two layers, there is a specially designed spring structure. This spring is driven through a locking feature attached to the upper gear, allowing the lower gear and upper gear to interact with each other during operation.
In the normal HZ6460 configuration, the modified intermediate wheel drives the lower gear. The lower gear then transmits motion through the spring structure to drive the upper gear. The quickset GMT adjustment function is achieved by acting on the upper gear. This explains why the upper gear and spring structure can be removed in order to reduce the overall height of the GMT wheel.
This structure also explains the biggest casing issue of the HZ6460.
Because the added GMT wheel is taller than the original day-date structure, the center position of the setting stem, relative to the dial side, is about 0.33mm higher than that of a standard 2836. This may sound like a very small difference, but for a watch case, crown, and stem alignment, it is already enough to affect whether the movement can be installed directly into a standard 2836 case.
To further verify this, I designed a dedicated tool to disassemble the GMT wheel. After taking it apart, I found that the upper fixed gear and spring structure account for a significant part of the added height, with the upper gear itself measuring about 0.28mm thick. Since the general tolerance for setting stem height in a 2836-style case is around ±0.2mm, removing the upper gear and spring structure brings the HZ6460’s setting stem center to only about 0.05mm higher than a standard 2836. This falls within an acceptable tolerance range.
However, removing the upper gear also means that the original interaction between the upper and lower layers is lost. In this experimental setup, I fixed the originally movable lower gear by soldering it with solder paste. By doing so, the lower gear is no longer free-moving, allowing the modified GMT wheel to function as a fixed 24-hour driving component while also reducing the overall height of the module.
Another detail worth noting is the dial spacer. The dial spacer supplied with the HZ6460 is usually taller than the one used on a standard 2836, in order to accommodate the original higher GMT module. If the goal is to make it compatible with a standard 2836 case, it is necessary to use a standard 2836-style dial spacer instead.
Through these modifications, the HZ6460 can largely return to the casing logic of a standard 2836 and may become compatible with many cases originally designed for the 2836. For me, this was a very important discovery. It meant that the HZ6460 was not just another interesting GMT movement, but a movement that could potentially be transformed into a relatively reasonable 24-hour movement solution in terms of cost, supply, and case compatibility.
Finally, the HZ6460 has one unexpected advantage: its GMT hand hole size is 2.00mm. This is different from common GMT movements such as the NH34 or ETA 2893, but it happens to be compatible with the hour hand hole size of the ETA 6497. Although the hand height is slightly different and may still require adjustment by sanding or modification, it at least provides a practical direction for sourcing suitable hands.
If the 6497 represented an approach from a traditional pocket-watch movement, and Vostok and Raketa represented the search for native 24-hour movements, then the HZ6460 represented a third path: using the existing 24-hour output of a GMT movement, then disassembling, understanding, and modifying it into something closer to a pure 24-hour display.
This later became one of my most important experiments.
It taught me that sometimes the answer does not necessarily exist inside a movement with the “correct name.” Sometimes, it is hidden inside a movement originally designed for a different purpose, but whose structural logic can be understood in a new way.
Why Not the HZ6460 in the End?
The HZ6460 experiment was very important to me. It proved that thinking backwards from a GMT movement could indeed lead to a solution that was closer to a pure 24-hour display.
But as the 2442 G.C.T. gradually moved from a personal passion project into a Kickstarter project that would actually be delivered to supporters, I could no longer think only in terms of whether something was technically possible.
I also had to consider supply stability, serviceability, consistency in production quality, and whether users could wear the watch with confidence over the long term.
The HZ6460 was a very interesting movement for experimentation and research. But if every movement had to be disassembled, modified, and adjusted again before use, the risk would still be too high for a small independent brand.
That is why I ultimately chose the Miyota 9075 as the base movement for the 2442 G.C.T.
Strictly speaking, the Miyota 9075 is not a pure three-hand 24-hour movement. It is a modern automatic GMT movement. However, it offers stable supply, mature quality control, reasonable cost, and a modern movement architecture that is easier to service and understand. For me, these factors became extremely important at the product stage.
The 2442 G.C.T. does not achieve its 24-hour concept by physically modifying the movement itself. Instead, it redefines the way a GMT movement is used through the overall design of the watch.
It takes advantage of the GMT hand’s 24-hour rotation and transforms what is normally a secondary time-zone function into the main language of the entire watch.
Looking back, every step of this journey mattered: modifying the 6497, studying Russian 24-hour movements, and disassembling the HZ6460. Even though none of these became the final production solution for the 2442 G.C.T., each step helped me understand what direction was truly suitable for this work.
To me, the 2442 G.C.T. is not simply the result of choosing one movement over another. It is the result of a long process of exploration and compromise.
As a pragmatic engineer, I still believe that everything that exists, exists for a reason.
And this journey helped me understand one thing even more clearly: good design is not only about pursuing the most ideal mechanical solution. It is also about finding the balance between an idea that can be imagined, a mechanism that can be built, and a product that can truly be manufactured, delivered, serviced, and worn for years to come.