Show from to [ Shift -1 day | Shift +1 day ] Zoom menu

Plot and , as , with .

Comments (see all | edit)

27 Nov 2010 [21:10]
ADJUST: +000 ms/day to 25700 note that we have a new weight configuration
27 Nov 2010 [20:30]
Note that the temperature rose during this exercise - we had a fan heater going because the outside temperature was sub-zero. This should give us a good opportunity to see if the temperature compensation is working as it cools down overnight...
27 Nov 2010 [20:00]
We finally got it all back together after about 5 hours work, only to find that the clock was gaining about 140 seconds per day. We then had to wind down the regulation nut on the pendulum bob by 3.5 turns. We then had to reset the bells, but in so doing we found that the second hand was out-of-synch with the quarters bells. All these unintended consequences of our simple readjustment of the pendulum!
27 Nov 2010 [18:00]
We re-assembled the pendulum, but in so doing we noticed that we had bent the gravity arms of the escapement. In order to prevent further damage we removed the gravity arms. We also re-instated the original suspension (as was removed in October). The repositioning of the gravity arms caused problems with bouncing of the escapement so we decided to adjust the fly on the escapement. This did not work so we had to dismantle the fly.
27 Nov 2010 [16:20]
Today we (Graham Newman, Mark Rainer, Hugh Hunt) had a go at fixing the temperature compensation of the pendulum. We stopped the clock at 16h20, disconnected the barometric compensator, removed the optic sensor, then Mark had made up a neat cradle for the pendulum bob so that we could remove the pendulum shaft while leaving the bob behind. We then dismantled the pendulum. There was a bit of corrosion but not enough to explain a lock-up of the compensator. We found that the zinc tube was bent by about 3mm at its centre - perhaps enough to have locked the mechanism up if it had got twisted?

Download data

Interval between data points: seconds [either 3 seconds (weather data is duplicated) or multiples of 60 seconds (all data is averaged)].

[Raw data: Clock | Weather]

Trinity College, Cambridge 2014