The Audemars Piguet 150 Heritage Pocket Watch Has A Secret Inside

Audemars Piguet has just announced a whole new series of watches for 2026 and it’s one of the most impressive debuts in the company’s recent history. We have a new chronograph caliber, new openworked perpetual calendars, and perhaps most exciting of all for enthusiasts and clients, a new case design for the first time since the launch of the controversial Code 11.59 collection, in 2019 – the new Neo Frame Jumping Hour, whose shape AP says will be the basis of a new collection by the same name, and which will have shaped cases and window/guichet displays, which is a pretty massive departure from business-as-Royal Oak. However for lovers of complicated watchmaking and of bravura displays of mechanical intricacy, the standout from all the announcements certainly has to be the Heritage 150 pocket watch. AP is clear on the point that exactly how many complications there are depends on what you classify as a complication, saying in the press release:

“A total of 47 functions, including 30* complications. Among the highlights: Grande and Petite Sonnerie, Supersonnerie, minute repeater, flying tourbillon, semi-Gregorian perpetual calendar (day, large date,month, year, astronomical moon, moon phases), flyback chronograph, split-seconds, hours and minutes.**

* These numbers may vary depending on sources and interpretation. Under the FH standard, the total may reach 60.

** This list is not exhaustive and includes the timepiece’s main practical, astronomical and technical complications for the user.”

It’s noteworthy that AP describes the perpetual calendar as “semi” Gregorian; the reason is that while perpetual calendars are often described as being based on the Gregorian calendar, they’re actually not. The Gregorian calendar adds an extra day to the end of February once every four years, to keep the calendar year synchronized with the tropical year; however, this correction, which was also present in the preceding Julian calendar, is not completely accurate, and once every hundred years, the intercalary day is not added when it should be according to the four year rule. This in turn, requires a further correction and so, every four hundred years, the intercalary day is added when it should not be according to the hundred year rule. The perpetual calendar in the 150 Heritage pocket watch incorporates the 100 year rule, but not the four hundred year rule, and is therefore correctly described by AP as “semi” Gregorian. Strictly speaking, any description of a perpetual calendar as “Gregorian” is in fact incorrect, unless of course it incorporates both the 100 and 400 year corrections (most do not).

This perpetual calendar mechanism brings us to a larger discussion of the movement, which is the caliber 1150. This is a movement which was originally released as the caliber 1000, in Audemars Piguet’s Code 11.59 Universelle. The Universelle was announced in 2023 and it was (and is) one of the most complicated wristwatches ever made, by Audemars Piguet or anyone else. (The disclaimer from AP on counting complications which is in the press release for the pocket watch was also used by AP for the Universelle). The caliber 1000 movement introduced a number of innovations.

All of the indications including the perpetual calendar, can be adjusted forwards or backwards without the use of the usual caseband correctors, and the crowns are multi-function “supercrowns” as AP calls them . Start, stop, and reset of the chronograph is traditional, via the pushers at 2:00 and 4:00. The crown at 4:00 can be rotated backwards or forwards to set the month and year, and it will also automatically return to the zero position after it’s rotated in either direction. The central crown can be used to set the time and date, as well as select the strike mode (grand strike, small strike, or silent). Caliber 1000 also has a very high amplitude balance, which rotates through almost 360º, and an innovative moonphase display which shows a more accurate representation of the Moon in the night sky than a conventional moonphase disk.

The 2023 Audemars Piguet ‘Universelle’ Code 11.59

The pocket watch caliber 1500 takes the caliber 1000 as a base, but with some modifications and additions – in the latter case a quite unusual addition. The caliber 1500 is basically the caliber 1000 but with the automatic winding system removed, which offers a less obstructed view of the movement plates and bridges. You might think that the movement would be significantly thinner as well, however that’s not the case – caliber 1000 is 34.3mm x 8.8mm and caliber 1500 is the same diameter, although according to AP it’s actually a tenth of a millimeter thicker than cal. 1000, at 8.9mm (which may be due to either an error in the original press release for the Universelle, or to some minor modifications of the movement in its cal. 1500 incarnation). Removing the rotor doesn’t actually reduce thickness because the original movement, especially given its enormous complexity, is pretty damned thin to begin with, and moreover if you check out the construction of the automatic winding system in the cal. 1000 you’ll see that the rotor is actually suppressed down to the level of the rattrapante seconds wheels at the center of the caliber.

The arrangement of indications on the dial of the 150th Anniversary pocket watch is identical, naturally, to that of the Code Universelle, although the diameter of the pocket watch is quite a bit bigger than that of the watch; the pocket watch is 50mm x 23.4mm while the wristwatch is  42mm x 15.5mm. While 50mm is big if you’re used to wristwatch dimensions, it’s pretty reasonable for a pocket watch, especially a super complication – the Patek Henry Graves, to pick one example, is considerably bigger, at 74mm, and Vacheron’s Berkley Grand Complication, with 63 complications (bearing in mind that how complications are counted can vary) is 98mm in diameter – just shy of a tenth of a meter, so I think credit is due to AP for bringing in such a complicated timepiece in a pocket watch sized to actually fit into a pocket. The extra thickness of the pocket watch is partly due to the necessarily heavier case, but it’s also, and primarily, thanks to the caseback. As is usually the case with pocket watches, the caseback is hinged, and can be opened up to show the movement.

However, the caseback also functions as a mechanical, manually operated computer, which has some very unusual astronomical functions related to lunar and solar cycles.

The caseback is hinged so that it can be opened to a full 180º; generally in pocket watches hinged casebacks only open to slightly more than 90º. This lets the owner lay the watch flat on a table so that all of the indications on both the dial and caseback can be seen at once.

The caseback is organized into concentric rings, some of which are fixed, and some which are moveable. The outermost shows the the equinoxes and the solstices (which mark the astronomical beginnings and endings of the seasons) and inside, there’s a month-and-date ring, with the days marked in five minute increments, and red markers showing the start of each month. These two rings are fixed.

Inside is a ring showing the dates of full and new Moons, and inside that, a ring with indicators for religious observances based on the lunar calendar, solar calendar, or on lunisolar calendars used to fix the dates of the observances.

The solar holidays are Christmas/Sol Invictus, and St. John’s Day/Inti Raymi, which take place respectively on December 25th, and on June 24th, which is the date of the Feast of John the Baptist, and also of the Incan celebration of the Sun God, which is celebrated on the Summer Solstice. The date of the Summer Solstice with respect to the calendar, as with the other solstices and equinoxes, can vary slightly from one year to the next thanks to the discrepancy between the true solar year and the calendar year, but they never shift very far as the discrepancy is corrected by the insertion of a Leap Year once every four years. The holiday  of Sol Invictus was a Roman one, and was a celebration of the “Sun Unconquered” which coincided with the Winter Solstice and was celebrated as the birth of the Sun. These two holidays specifically mark key points in the solar astronomical cycle and were chosen by AP for that reason.

The observances of Diwali, Rosh Hashanah, Pesach, Vesak (the most sacred Buddhist holiday, which commemorates the life, enlightenment, and death of the Buddha), Easter, and Chinese New Year, are all based on calendars which use both lunar and solar cycles to calculate these dates. For example, Chinese New Year takes place on the second New Moon after the Winter Solstice, and so requires both lunar and solar astronomical observations.

The start of Ramadan is shown by a pointer with a tip in the shape of a crescent Moon. Ramadan is based on the Islamic calendar, which is purely lunar, and which consists of 12 lunar months (the time between successive New Moons). Ramadan begins when the crescent Moon is first seen in the night sky on the 29th day of the month of Sha’ban, and so its date is derived purely from lunar observations. Because of the discrepancy between lunar and solar/civil calendars, the date of Ramadan shifts backwards relative to the Gregorian calendar by ten to 11 days per year, completing a full cycle through the seasons every 33 years. Therefore, over the approximately 200 years during which the caseback display will be accurate, the date of Ramadan will make several full revolutions around the dial; hence the need for a moving pointer.

The choices made for the calendar were very carefully chose and it’s a subtle but wonderful detail – pure lunar, pure solar, and lunisolar observances are all present and moreover, Sol Invictus and Inti Raymi show the importance of solar astronomical observances across centuries of time as well.

Two authors covering this launch have made a couple of interesting points.

Two Authors, Two Perspectives On Complications And Complexity

The first is Mark Kauzlaurich, writing for Hodinkee. As well as covering the watch technically, he discusses some of what he characterizes as philosophical points, relating to the size of the case vs. the size of the movement, and to what counts as a complication. Pursuant to the latter, he specifically discusses what if any reservations we might have about the fact that the caseback lunisolar calculator is not driven by the movement, or connected to it in any way:

“The watch here has an unquestionably creative and impressive technical nature, but I do think there’s a philosophical debate to be had. I wrestle with the question of how integrated a complication must be into a timepiece’s heartbeat to count as a complication. As with any story about a big pocket watch, I’m sure we’ll get the comments “that’s not a watch, it’s a clock.” Eventually, I’ll write the story about how there’s no official definition of either, so the line between the two gets blurred. This watch is far more compact than most. Similarly, I’m curious where a watch ends and a mechanical calculator begins.”

I think he’s got a point – the Universal Calendar is really a kind of orrery rather than  a complication per se – although like Mark I’m not necessarily completely convinced that this is a deal-breaker, even philosophically. While it would have been perhaps more satisfying to have some sort of once per year mechanical coupling with the movement (Mark mentions the Reverso Grande Complication à Triptyque, in which the astronomical indications in the case carrier are updated once a year automatically, by a push pin in the case proper) the lunisolar calculator, or Universal Calendar, as AP calls it, is still interesting and ingenious on its own.

I don’t think that when the rubber really hits the road, the mechanical independence of the Universal Calendar is worth more than a passing note; any drawbacks that it has are as he puts it more “philsoophical” than real and there is probably something to be said for the tactile pleasures of updating the calendar manually – and the calendar, by the way, is accurae back to 1900 and up to 2099, so you can enjoy using it as a sort of astronomical time machine as well. It is in fact, very reminiscent, if you are looking for validating historical antecedents, of the ancient Antikythera Mechanism, which was also a hand operated lunisolar calendar.

The Antikythera Mechanism by the way, is a 2100 or so year-old astronomical computer of great complexity and ingenuity – it may have been designed by Archimedes of Samos –which could be used to display a number of astronomical indications, including several related to lunisolar cycles. It was operated by a hand crank, and its astronomical basis was something called the Metonic cycle. The lunar year and the solar year are, as we’ve said, not the same length and a whole number of lunar months doesn’t fit into a standard civil calendar. However, every nineteen years, a whole number of lunar months – 235 of them, or 235 lunations – fit exactly into 19 years.

We said before that the Universal Calendar showed information derived from lunisolar relationships, and so we would expect to see the 19 year Metonic cycle encoded somewhere in its mechanism. The exact component – and it is a single component; the “secret” mentioned in the title – is a disk, or cam, whose presence and function was noted by Cheryl Chia, in her coverage for Revolution (her technical coverage is uniformly excellent, by the way).

The component in question is the disk in the image below which looks like it has a sort of asymmetrical squiggle engraved on its surface. Below in the first image is the Universal Calendar with indicator disks attached, then the calendar mechanism with the solstice/equinox, month/date, and lunisolar observances ring removed, and then with the year disks removed as well.

The lunar indicator disk for the new and full Moons sits on top of the large outer disk with inward facing teeth, secured by the steady pins on its upper surface. The lunisolar observations disk is inside it, and both disks are carried on small beryllium bronze rollers.

Of that cam, or program disk, Cheryl writes;

“The most striking part of the calendar is the grooved plate at 12 o’clock, which acts as the lunisolar program wheel. Cut with a continuous groove, it converts the selected year within the 19-year Metonic cycle into a precise angular position for the lunisolar festival disc. In other words, it encodes lunisolar offsets. The feeler riding in the groove reads the path and forces the disc into the position in which its printed arrows align with the correct dates for that year.

“A consistent theme across the module is the use of pinions with paired teeth, separated by a gap. They are in turn driven by a finger with two flanking grooves which allows the pair to rotate in both directions. This bidirectional design underpins the entire system, enabling the calendar to be adjusted freely forwards or backwards.”

You would expect a Metonic cycle program wheel to have 19 segments, and while some parts of the program wheel are concealed under other components, it seems pretty clear that there are 19 high and low points, and so the groove is basically a circular graph of the Metonic cycle.

This is an interesting point philosophically as well. When I wrote about the Berkley Grand Comp, I thought, not for the first time and obviously not for the last, about how the universe is ultimately chaotic, and I mean that in a specific technical sense. The cycles encoded in clocks, watches, and calendars are all temporary over long enough spans of time. One of the great achievements of the Berkley Grand Comp, was its ability to show the correct date of the Chinese/lunar New Year. However, there’s a problem inherent in trying to encode a fundamentally irregular sequence of events in a calendrical or mechanical approximation. On the Berkley, I wrote:

“You have at this point probably begun to realize that creating a mechanical Chinese perpetual calendar is something that requires a certain willingness to tackle extremely frustrating problems. The cycles of the Chinese calendar are highly irregular thanks to the need to keep lunar months synchronized with solar years, and to the fact that neither cycle consists of a whole number of days (and in fact, it would be a rather miraculous coincidence if either of them did, if you think about it). We can see a similar problem in the calculation of the date of Easter, which also relies on reconciling lunar and solar cycles and which does have a cycle of dates – these however, repeat only once every 5,700,000 years.

“This cycle moreover is based on a simplified version of the definition of a full Moon, based on a calendar, rather than on direct observation, or an observational algorithm, and 5.7 million years is still the best it can do in terms of regular cycles. The Chinese New Year has no regular cycle of dates that I have been able to find, and so a Chinese perpetual calendar, like Patek’s date of Easter complication in the Caliber 89, would have to rely on program cams.”

The Universal Calendar, then, is another heroic but ultimately doomed attempt to make the irregular regular. It’s no less interesting for that, though. The 200 year window that it covers, is about as far as you can push the Metonic cycle before the same problems that extending the perpetual calendar past 400 years presents. Like the rules for the Gregorian calendar, the Metonic cycle is an approximation – a very good approximation but an approximation nonetheless. The Metonic cycle defined as 235 lunar months, is 2 hours, 4 minutes and 58 seconds longer than 19 years. This means that after 11.36 Metonic cycles, about a full days’ error will accumulate – that works out to roughly 220 years, and on top of everything else, the agreed on length for a lunar month is itself only an approximation, since the actual time it takes for the Moon to return to a given point in its orbit varies from month to month thanks to a complex number of perturbations which I don’t recommend looking up unless you either already have a PhD in orbital mechanics or are looking to get one.

But as the superhero android Vision says in one of the Marvel movies when told that mankind is doomed, “A thing isn’t beautiful because it lasts.” This whole conundrum is really only a specific instance of the larger, unbeatable general problem of perfect precision in horology. If we’re looking for an apropos quote from horology specifically, maybe this is a good time for me to trot out this quote from George Daniels again; I know, twice in one week, but if the shoe fits, right? Of sympathique clocks, Daniels wrote:

“There is more mystique surrounding the sympathiques than any other of Breguet’s products. Relatively few people have seen one and only perhaps half a dozen know exactly what they do or how they work. [I am not sure that number is all that much higher today.] In Watches, written jointly by this author and Cecil Clutton, they are described as ‘Breguet’s most advanced flight of misapplied ingenuity’ and in the context of the history of the development of horology, this description may stand.”

May the future continue to bring us misplaced flights of ingenuity. Watchmakers love records; speaking of the Moon, we already have a moonphase complication courtesy IWC which is accurate to a one day error in 45 million years. Maybe someone already has a Metonic cycle-based complicated lunisolar watch that adds a one day correction every 220 years in the works.