Back on course
I made the final verge and used brass pallets, but the noise of them hitting was just not acceptable! So I put the aluminium ones back inside the brass ones - I know they will not last a hundred years, but the sound is so much sweeter! I also tried the clock without the crescents and again the noise was terrible!
I will go for a longish run to bed the system down before I try the compensator again. I placed the Crowns at 5.5cms giving a swing of 36°. I added an empty compensator, so that I could experiment without upsetting any wheels . I'm sometimes getting a very quiet noise from the crown arbor, but I do not know what it is!
The humidity started at 53% with a low of 50% and a high of 56% - quite a bit different from the 60s I was getting before.
I don't expect to get another 'good' run like this, so here is how time varied over the 4 days. If the clock always behaved like this I would be very happy! So I shall now try the Compensator again, reducing it's power and seeing what happens. Everything else will stay the same. But first I shall paint the crescent with hardener and see if that has any effect.
The humidity started at 54%, so is roughly equivalent to the previous run, but with a minimum of 47% and only a high of 52%. I've had to offset the humidity by 8 units to fit the graph. The long constant reading was 52% (at the end of the run), where as the previous run ended at 53%
But the little extra weight of the hardener has slower the clock down by about 0.1-0.2 secs.
So lets reset and start again adding the compensator into the equation. With just the weights it followed the humidity pretty accurately.
A small compensator didn't seem to make much difference, but the larger one almost seems to break away a bit. I need a longer run, with more variation in the humidity!
Run started at 50%, with a lows of 45%.
This was a long run, just to confirm that a skeleton compensator did not really help. I had to change camera in the middle of the run and also adjusted the verge distance to speed the clock up a bit.
This run adds a Shilling (5.6gms) to the compensator, but really has little effect at the small diameter of 14.5cms. Although I do not understand why the second half of the run aligns with the humidity, but the first does not.
The creak is back when the crown pin at 48 seconds is vertical - maybe this has slowed the clock down a bit.
I give up - this mad idea just does not improve the accuracy. Next Page
