All, just a quick note, will follow up when we get some more data.

 Here is Iain's delta T data so far plus asymmetry (I am pretty convinced that asymmetry is a good indicator of spikiness but my note on that is also awaiting a little extra data). The orange line is the asymmetry (left hand axis), the blue one with symbols is the delta T (right hand axis)

 

 There does seem to be some correlation between the two. Here is the correlation, split into the three epochs between c brg adjustments:

 

I would expect to see the data on stright lines, bottoming out at some lower limit of asymmetry. So the simple linear fits I have made are not quite the right thing to do. However, what I see from this is that there is indeed a correlation between spikiness and delta T, which varies according to central bearing state. The red data refer to late 98 - the period at the far right of the time history. The purple points are the period in the middle of the time history where the asymmetry is generally low, the green data are the noisy period we had in last fall. There's a rogue point in the purple data, at about (-1.7, 3). that's the spike in the middle of summer, which does NOT line up with the corresponding delta T spike. More investigation required...

Justin 1/27/98