Expert: Kenneth Saunders - 7/25/2008

Question
QUESTION: This is my second question posed here about this clock, which I originally
inherited in a state of disrepair. I have successfully re-attached the pendulum
which disconnected from the suspension spring.

Next problem:

The main spring (assembly), apparently, was over wound, and appears to be
locked up. It will not accept any further winding. I didn't force it further. When
the pendulum is manually put into rotation, the clock actually runs, the hands
advance, until the momentum depletes.

Is there any way to release the tension in the main spring. I see a large pawl
that engages a gear tooth on the housing of the main spring, but I'm not
completely sure what is actually turning. Is the housing fixed and the internal
spring unwinding inside, driving the main-shaft that passes through it or,  
what I suspect, the spring turns the housing and the gear teeth I see on the
outside, drive the clock mechanism.

Back to that pawl, it seems if I could disengage it from the gear train on the
main spring housing the mainspring could release. I'm afraid to do it because
it might unwind to violently if at all.

I suppose the main spring could be just fully wound, and something else is
causing the clock mechanism to not perpetuate its' motion. I still suspect
mainspring lockup. What does an expert think?

ANSWER: something else is
causing the clock mechanism to not perpetuate its' motion.

this is the more likely answer since the clock runs, power is getting from the mainspring to the hands.

likely first to be out of balance then the upper fork may be located in the wrong position or be crimping the escapment lever that is engaged in the fork.

If you wind the clock with the key you may be able to get enough slack in the pawl to move it out of the way of the the ratchet gear on the winding shaft- and gently but with an iron grip on the key, allow a few teeth to un-wind until re locking the pawl and moving your hand 180 degrees for another 1/2 turn. But I repeat- this is not your problem and is a tricky move.

read up here:Initial Beat Setting

"A simple check can be made by first allowing the pendulum to come to rest.  Then, by hand, carefully rotate the pendulum until you observe an escape wheel tooth drop onto a pallet.  Release the pendulum.  If it rotates until the opposite pallet contacts the escape wheel tooth (escapes) and then stops, the clock is in beat.  If it does not escape, turn the suspension saddle slightly in the direction of the side that did not escape and repeat the procedure.  If the opposite pallet does escape but the pendulum continues moving (“overswing”) then turn the suspension saddle slightly in the opposite direction you initially allowed the pendulum to turn (swing) and repeat the procedure.

Checking the Overswing and Fine Tuning the Beat

When the clock is perfectly in beat the pendulum should turn exactly the same amount to either side after each pallet locks (this is called overswing).  Overswing can be checked visually using any of the following three methods (a, b, & c), first turn the pendulum by one half turn in either direction to start the clock:

a.      Use a beat amplifier to hear the "tick" as each pallet locks.  Place a beat adjusting quadrant under the pendulum (see below and this is found on page 52 of the 10th edition), and following one point on the pendulum.  Mark its position on the quadrant at the point of each tick; also mark the end of swing in that direction.  The difference between the point of tick and the end of swing on each turn is the amount of overswing.  Looking from the top of the clock from the back, turn the suspension saddle clockwise if overswing is too little to the left, and counterclockwise if it is too little to the right.  Saddle adjustments can make a large change in overswing because any increase of overswing in one direction is decreasing the same amount in the other direction.

b.      Watch the escape wheel action through the observation holes, or by looking at its reflection in the front plate.  You will be able to see the escape wheel "jump" as each pallet unlocks / locks, mark the pendulum positions accordingly.  Adjust the overswing as necessary.

c.   With the motion works and hands in place, set the minute hand at 20 minutes to the
hour, and watch the action of the minute hand.  It will "jump" as each escape wheel tooth is released, which corresponds to the lock of the pallets.  Adjust the overswing as necessary.

When you are finished here, there should be a total of at least 1" overswing (1/2" in each direction at the pendulum outer periphery).  If this is not met, there is some reason for power loss or maladjustment and the clock is unlikely to run very long.


above is from :ftp://atmos-man.com/atmos/400-day.doc

---------- FOLLOW-UP ----------

QUESTION: I will observe the engagement mechanism for evenness and adjust the saddle
as needed, to restore. Obviously balance is every thing, especially in a clock.

Although, I completely accept your logic of suspecting the clocks equilibrium,
I don't understand how it would be determined, that the manually rotated
pendulum is deriving energy from the main spring. Couldn't the inertia
imparted (to the pendulum) drive the clock (all by itself) albeit, not for very
long. That's all that I see happening. The clock runs until the periodic inertia
is gone.

If I was forced to answer my own question, I would say, that there is no
practical way (for me) to make a lock up determination with the assembly
(still) in the clock and unreleased. By first confirming proper equilibrium of
the clock mechanism, and eliminating that as a possibility, only then could
things like a locked up main spring even begin to be considered.


Answer
you said "When
the pendulum is manually put into rotation, the clock actually runs, the hands
advance"

the hands can advance only when the clock is ticking and the escape wheel is releasing teeth as the pendulum rotates from side to side-- ergo, there IS power driving the escapement and the power comes only from the spring being wound and uncoiling releasing power to the gear train.


I have NEVER in 36 years in this biz found or heard of a mainspring being fully wound and "locked".

  The clock may or may not run due to some other cause but not because of a locked fully wound spring. In your case as described, your clock DOES run -just doesn't stay running.A dirty, gummy mainspring may not impart full power in theory but not so much an issue in the real world.