[Frost] phi offset

[Michael Dugger]

Hi,

I have something that might be of some interest with respect to the phi 
offset problem. It looks like I can get the type of behavior that Priya 
sees in her plots if I move the measured vertex relative to the true 
vertex.

What I did:
* Created a circle of radius 1 meter centered at zero in the "true" 
coordinate system
* Created a second coordinate system displaced by 0.01 meter in the 
x-direction. I call this coordinate system "measured"
* Plotted the difference in azimuthal angle between "true" and "measured", 
versus the "true" azimuthal angle.

The plot can be found at:
https://userweb.jlab.org/~dugger/frost/phiDiff.gif

Note: Each axis is in units of degree.

As can be seen in the above plot, a displaced coordinate system from that 
of the true coordinate system will give behavior that is very similar to 
what Priya sees in her plots.

This is merely suggestive, and our phi issue may have nothing to do with a 
displaced coordinate system. I just thought it was interesting.

Take care,
Michael