Adjustments First lets see how moving the verge weights, in and out, effects the timing.

Here a plots for different distances of my standard 2 weights hanging from the bottom of the test verge. It works out that the time changes linearly with distance - I would have thought it was dependent on inertia which is proportional to the distance squared.  Wrong again! This works out that change of 1 second per minute is equivalent to moving the weights by 3.3mm.  Assuming we want an accuracy of 1 minute per day, then we need to position the weight to within 0.14mm of the correct position - quite a tall order!

Now lets see how the distance of the verge from the crown changes the timing with the verge weights at 9.5cm and roughly set the verge distance to gain around a second. The thread pitch is about 0.7mm per revolution, so moving in increments of 90° moves the verge by 0.175mm - not a lot!  With the wee dial it should be easy to position it to say 2.5° or 0.005mm! Moved the verge dial +90° after 5 hours and +180° after 15 hours.  But because of the humidity change this does not give me very accurate answers.  The averages were -0.77, -0.62 and -0.22 for 0, +90° & +180° changes - sort of indicates that it is not linear like the previous run. But I must repeat this to be sure and get some idea of how to calibrate this wee dial.   However I noticed that the pulley shaft was bending under the weight of 3kg as the big pulley was 10cm out from the frame.  So I added a block under the mounting plate to reduce this.  I also place the verge weights at 9.4cm. I also brought a digital humidity gauge which is more sensitive (and accurate?), although I am only taking spot reading each hour and I notice it can vary quite a bit within minutes!     I guess the big humidity change on the last run (yellow) made it artificially high and the average should have been more like -1.6.  However, there is a sweet area on this non-linear curve where a 90° change results in 0.1secs/min = 0.1min/hour =2.4mins/day.  It is going to be difficult to calibrate this dial as it obviously depends on the initial distance of the verge from the crown as well as the humidity - a lot of patience will be required! Thus a 2.5° change would alter the clock by 4 seconds a day - if only humidity did not enter into the equation! Time for a long run as the weather is meant to be getting unsettled - fingers crossed we can see what humidity does to the timings.  This will also test how round the big pulley is if there is a pattern every 48 hours.. The clock time is recorded to the nearest 4 seconds - a pin's worth - and instant humidity reading. I have scaled the humidity in an attempt to match the time, and I think 5% is around the right amount.  The humidity started off at 58%, went as high as 66%  and as low as  54%. I reckon this shows a pretty good relationship between humidity and clock time! There are a number of points I must remember and assuming my arithmetic is correct then, 0.1 seconds/minute difference in time is equivalent to a) 2.4 minutes per day b) 0.33mm change in position of the verge weights c) 15° - 9° (depending on the initial position of the verge) on the verge dial, or d) 0.03mm - 0.02mm change in the distance of the verge from the crown wheel e) approximately 3% change in humidity! Next Page Content Menu