Moon Unit *

In our last episode your local horologist was offering a surmiseus reversus mea culpa regarding the mystery of the skipping strike on the 18th century Peter Green tall case.

But now we know.

So let us unfold the far less complex (but still wonderful) mystery of the moon dial that "stopped turning."

Moon dials or lunar dials on clocks go back as far as the late 16th century in Germany, becoming more common on English clocks in the late 17th century.

In short the dial, when properly set, will indicate the current phase of the moon (e.g. full moon, crescent moon, "new" moon, etc.). 

Some astronomical trivia:

A single full cycle/phase of the moon is not the same as the time as it takes the moon to make a single cycle around the earth.

It takes the moon 27 days, 7 hours and 43 minutes to make one complete cycle around the earth. This is called the "sidereal" month.

But how long is a moon phase? 

Well over the course of a year it averages almost precisely 29.5 days. That is called the "synodic" month.

Why the difference in time and why "averages"?

The reader may recall that while the moon circles the earth, they both circle around the sun. In any single synodic cycle the moon has to travel just a bit farther to make up the additional distance to complete a lunar cycle around the earth and achieve a similar angle of light hitting the moons surface.

Furthermore the angle of the earth relative to the position of the sun varies seasonally as it orbits around the sun and of course the proximity of the observer to the equator varies that angle as well. 

Lots of details on lunar cycles here.

Horology provides such rich fodder for this amateur scientist.

Now back to clockworks:

The moon dial itself is a complete circular plate (toothed along its edges) with identical images on both halves of its face enabling it to make a full moon phase on 1/2 of a rotation of the plate.

As the moon dial rotates the two half disks of the brass dial plate reveal or start to hide the moon as it appears in the moon dial.

Here's the moon on the dial appearing as a waxing crescent about 5 1/2 days (according to the brass marker/pointer on the dial plate) in the lunar cycle. The viewer might notice the day 20 marker appearing on one side and disappearing on the other. Two full lunar cycles on one dial.

The mechanical underpinnings of a moon dial require that it obey the 29.5 day synodic cycle, thus it must make one complete rotation in 59 days. 

For more on the history and action of moon dials this now 25 year-old article has an excellent overview.

As shown in an earlier post on the Peter Green, its full dial unit is enormous and encompasses the date dial, the calendar ring and the moon dial.

Here is an illustrated view of the back of the entire dial unit.

The attentive reader may have observed that big mechanism at the bottom of the moon dial that engages with its teeth.

The tip of that rod is a wedge that engages with and pushes the teeth and thus rotates the moon dial.

But what moves the rod?

Your photo hound refers the reader back two photos above. That rod is pointing straight at the center wheel hole in the dial plate.

Ah ha.

What else is right around the center wheel?

Why the outer snail of course! 

Once removing the snail unit (comprising both the inner and outer snail) the viewer can see that the outer snail has a large, smooth edge that matches the size of the tip of the rod.

Zo!

It appears that as that snail turns it engages the rod slowing lifting it up as the diameter of the snail expands until the rod finally drops where the snail cliffs off at its end. It's that precipitous edge that YLH is pointing at with the wooden stick in the photo above.

Hang on.

"Drops"

No it doesn't drop. There's too much friction against the plate. It needs to be set back in place.

It became immediately apparent to YLH what wasn't working.

There is a pin on the rod that serves as a hook for a spring that, when hooked over the top of the pin, will press the rod down against the snail.

The observant reader will note that two photos prior highlighting the snail and rod tip, YLH had already engaged the spring correctly.

So if your intrepid horologist were to align the snail to the moon dial, even if its still lying on the workbench, advancing mechanism with the now correctly engaged spring and were he to try turning it...

Voila.

YLH suspects that the last clockmaker who worked on the Peter Green simply forgot to reengage that spring during reassembly.

And in the earliest pictures taken of the clock by this photo nut, when the clock was still fully assembled, the viewer can observe the spring sitting under the rod pin, in effect preventing the rod from pressing back against the snail.

An easy "fix" but almost impossible to really grok while the clock was assembled and your budding horologist having no idea how the mechanism worked.

Oh one last thing.

A bit of horological / engineering math.

We established earlier that 29.5 day synodic lunar cycle x 2 halves of the dial means that the moon dial must make a complete revolution every 59 days.

AND

Because the outer (and inner) snail makes a full revolution every 12 hours (minute and hour hand turning), the moon dial snail mechanism pushes the rod to click one dial tooth twice per day. 

So to achieve a complete rotation of the dial over 59 days...

The moon dial must have 118 teeth around its edge.

And it does.

Neat.

Once that spring was back in place and the movement later reassembled, the moon dial ran perfectly.

* With apologies to Ms. Zappa