POUVILLON RESTORATION PROJECT - September 2011
Calculation and fabrication of equation cam
contours, springs for sunrise/sunset dials
Note that the format of this section of the web
site is designed for the newer wide screen monitors, those using the
conventional 3:2 ratio monitors may need to move the horizontal scrollbar at
the bottom of the screen to see the entire page.
I have asked Buchanan to provide what I call a
'forensic' report. That is to record his observations as he goes along. I
will provide the .MP3 audio file
for each segment but just in case your security settings will not allow you
to open this file I have also transcribed each session. My additional
comments will be inserted into the text from time to time and this will be
in red text. Buchanan refers to each photo by
the number of that photo which can be followed by each photo above the
captioned text. The .mp3 audio file will appear
at the beginning of the photo sequence
in blue text. Click on this
text and you can then follow along with the audio file by scrolling downward
through the photos as they are narrated one by one in the voice of the
Photo 29 009.
Here we have the equation of time differential and compared to the previous
photograph we will notice the rocking lever is in the ‘down’ position
virtually at 9 o’clock it’s rotated the small pinion on the differential
arbor in the clockwise direction and lifted the differential pinion to the
11 o’clock position. And one can then follow the equation of time hand has
accelerated somewhat in comparison with the mean time dial, on the main
dial. Photo 29 010. Is a back
view of the equation of time mystery dial. It gives quite a good impression
of the mystery aspect of this dial.
Photo 29 011,
is the initial marking out of the seventy three steps on the equation of
time cam. The cam blank has been mounted on the seventy three toothed gear
wheel or drive wheel and the teeth are used as indexing points to get our
seventy three equally spaced lines at the correct arc to match the arc of
the leading lever for the equation of time mechanism.
The target diameters for the equation kidney cam
are 73mm outside and 30mm inside. We
have assumed that Pouvillon would have used the latitude for Paris for this
mechanism which is 480 51’N.
Photo 29 012.
Here we can see equation of time blank, again on the marking object, the
dividers have been set at the distance of the pivot point of the reading
lever to the reading point and the point of the divider on the big brass
sheet is at the same distance from the equation cam arbor as on the clock
itself. It’s necessary to mark this line in a curve to match the stroke of
002 is the equation of
time blank with the various points marked out on it
(using center punch marks) which reflect the measurement from the
clock and gives the correct equation of time dial reading for the correct
date. Photo 30 003. We have the
initial or the cutout according to the making out on the one year arbor. The
blue marks on the teeth of the year wheel are to facilitate checking our cam
as we step it around through all 73 positions to compare it with the
equation of time chart.
Photo 31 001.
Here we have my error chart where we have in the first column numbered 1 to
73 we have 73 stations. The next column what the equation of time cam, the
amount of minutes the equation of time cam should lead or lag the mean time
hand. And then in the third column we have the actual error or the
difference between what the equation of time dial should read and what it
actual reads. In other words I now have my errors. On the cam blank you can
see on each station I have marked a dot for each half-minute error. So we
have errors varying from approximately minus 1 ½ minutes to plus 1 ½, or
about a three minute error. One has to, of course, only remove metal so we
then convert these errors to all metal removal amounts and this is what’s
marked out on the cam. Photo 31 002.
Here we have a close up of the cam with the various error marked out I think
you can see our worst error was two minutes at this stage. And we now start
filing more or less away on the portions according to the errors.
Photo 31 003.
I think this is about the third time around where we are still removing
errors. As you can see the errors are becoming more equally spread, and I
always end up with a metal removal. So on the right hand side you can see
one portion where we have zero error and everywhere else metal has to be
Photo 31 004.
Here we have yet a further stage and you will see that our errors are slowly
Photo 31 005.
As you can see our error chart is full and we are getting toward the end of
our whole exercise. Photo 31 006
is the total numbers of tries or adjustments required to produce the cam it
took me at least eleven comparison and filing cycles to get a correct cam.
One can see the laborious process this actually
is. These charts show the methodical process and patience needed to compose
these comparison tables needed to create the correct cam outline.
Photo 31 007.
Here we are approaching the final stages and we’re just removing metal
according the number of dots on each station. It’s a slow process and one
has to be careful not to remove metal where one needs it to remain.
Photo 31 008 is the final cam
replaced in the blank and shows the final amount of adjustment required,
possibly due to the lack of my skill but we have the correct adjustments in.
The third photo shows stock watch balance wheel packets.
Next the spring is attached to the indicator hand's arbor used in the length
of day/length of night dials.
The next photos show the
retaining taper pin used just like it would be in a watch to secure the
other end of the spring. We are careful to use the posts and holes that were
already present to reverse engineer what we believe were present when the
clock was originally finished. These springs are used to eliminate lash in
the gear meshing so the hands will read off the dial with greater accuracy.
Photo 33 001.
At the bottom of the picture, out of focus, we have the hand we’re hoping to
replace. The coil on the right hand side is the spring we have selected to
use as material of suitable thickness and on the left hand side the actual
piece we have cut off. We’ve drilled a hole in the foreground we’ve started
to round off to make the boss of the hand.
So here we have gone from the
delicacy of a watch balance spring, to using the material from a full
fledged clock spring material. In fact this material is very well suited for
the making of clock hands.
Photo 33 002 is the embryo hand
and the hole drilled in the end fitted onto a pair of filing buttons used to
control the width of the rim around the boss.
Photo 33 003.
Another view of the filing buttons. The tubular section pushes up the spigot
and prevents one from filing any further than is required.
Photo 33 005
is the vise we use for filing and in the foreground the binocular magnifier
used to see filing right up to the line in other words to ensure suitable
accuracy. Using diamond files here as first of all the hand is small and
secondly I have not softened this material as I want to preserve its
helps prevent bending during working
working (filing). Just next to the file on the bench itself you can see the actual
copy hand or the hand we’re trying to copy and the brush on the right hand
side used to get rid of the filings so we can see the scribe line clearly.
Photo 33 006. A closer view of
the embryo hand held in the vise you can see the filing buttons just to the
right hand side of the jaws and you can see we haven’t started to do too
much of shaping as yet.
Photo 33 007,
A close up of the blank for the hand. We’ve filed the boss around until
we’re approaching the root of the hand.
Photo 33 010.
Here we have the two hands,
(for the length of day/length of night dials).
The hand on the left hand side is the new hand and the one on the right hand
side is the original. You can see there are subtle differences it’s also
difficult to match the differences perfectly.
Photo 33 011. Another view of
the two hands. The hand in the foreground
is the new hand and the one in background is the original to be
supercritical you can see I haven’t thinned down the pointer section or the
front section of the hand quite as thin as the original. This should be
attended to finally. Photo 33 012,
shown). Another view of the two
hands side by side. These
are about 1.5cm in length or just over ½ inch.
Photo 34 012. We have the lever
operating mechanism for the sunrise/sunset dials
(shutters) and also the length of day/length of night dials. The
arbor projecting rightwards from the drop down pillar is the one year arbor
that will finally revolve the five operating cams. Working from the right
hand side the brass lever with the two little pins just above the arbor the
equation of time mechanism. Then the two straight black levers operating the
sunrise shutter and the sunset shutter and then we have the round wire with
the counter weight which operates both the length of day and length of night
mechanism. Photo 34 013. We have
another view of the sunrise/sunset-length of day/length of night dial and
its operating mechanism. In the background we can see the one year arbor.
Photo 34 014
is another view of the operating levers for the sunrise/sunset-length of
day/length of night dial. What we are showing in these
three photos are the pair of 'C' shaped springs Buchanan
used to give bias to the sunrise/sunset shutter levers needed to keep these
levers seated on their respective cam perimeters. As was done in the case of
the cam pack assembly as well as the small hair springs installed earlier we
were careful to design our parts to comply with what was already present in
the clock. If you look closely where the spring ends are anchored they fit
neatly onto the grooved area on the drop down post as well as their
counterparts on each lever. Without actual photographic evidence we had no
idea what Pouvillon had originally done. It is also possible that he simply
used a set of conventional coiled springs; these too would equally fit
within the original clock's configuration. We think this design is more
elegant than coils; notice the beautiful blued finish.
Photo 34 016. We have a front view of the length of day/length of night
dial. Showing some fairly extensive tarnish or corrosion as well as the
shutters lifted far too high.