NavList:

David,

You may have flipped the off arc columns. The sums of each set should be less than 62' if S.D. is 15.77'. 

I recommend using solar film over the scope objective lense. Focus for a sharp limb and well defined sunspots. Try for matching dim Sun images if using sextant shades. I use both limb to limb and overlapping methods and get identical results. Limb to limb is preferred on a high Sun. Overlap preferred on a low Sun. Spead should be something less than 0.5' on sets of ten.

Greg Rudzinski

From: David Fleming
Date: 2015 Jun 26, 10:17 -0700

To: Bill Morris, BillB, and Fred Hebard

I hope you all see this response in this new daughter thread, the topic was fish out of water in the previous.

Thanks to you all for helping me proceed with evaluating efficacy of IC measurement schemes and the inspiration of your work that you related to me.

I am going to repost my original data with errors of analysis, as pointed out by Bill B corrected and more extensive statistical analysis that is I think not too far from the truth.

To me the bottom line baring mistakes is, as I originally maintained, overlap is supperior in measuring IC compared to edge to edge.

IC measurements 06/05/2015                       on off           # overlap,on   edge to edge 60-off         1 3.2   32.6 28.3 31.7   edge to edge error propagation 2 3.5   31.6 28.6 31.4         3 4.2   32.2 29 31  std IC^2 = 0.015     4 3.3   32 29.3 30.7 std IC= 0.124     5 3.9   32.6 29.1 30.9         6 3.2   32 28.6 31.4         7 4.1   32.4 29.1 30.9         8 3.8   32 29.4 30.6         9 3.1   31.8 30.1 29.9         10 3.5   32.1 29.2 30.8         11 3.4   32.5 28.9 31.1                     IC sum                 (On-60Off)/2 On+60Off     avg 3.56   32.16   30.95 0.61 63.11     std 0.38   0.33   0.48         range 1.1   1   1.8         mean std(11) 0.115   0.099   0.145   Semidiameter Sum/4   mean std(22) 0.079         NA 15.8 15.78               per BillB/Frank 15.77

 

This data compares overlap measurement of IC to edge to edge measurement. All numbers are moa.

The overlap method resulted in an average IC of 3.56 from a .38 stdDev(n-1) population.  The stdDev of the mean( 1/sqrt(11) smaller is .115.  So the IC measured was 3.56 +/_ .115.  Had I taken 22 measurements which was the effort used in the edge to edge case the error would have been +/- .079.

The edge to edge data gives an IC = .61 +/- .124.   .61 = (32.16 - 30.95)/2    +/- .124 follows from the following analysis.

Error of on edge mean .099, as in overlap calculation, and edge off error .145.   IC given by = .5(avg on + avg off).  Error propagation:

SD^2 = ( .5 SDon^2  + .5 SDoff^2 )   so .124=sqrt(sq(.5*.099) +sq(.5*.145)).