Here I describe a check on the results of the grid test analyses.

These analyses revealed that the direction of positive Phi_Inst_cor is clockwise, in (Q_raw, 
U_raw) space.  (It is counter-clockwise in real space, when looking from SHARP towards M3.)  This 
directional relationship between Phi_Instr_cor and the raw Stokes' parameters is a consequence of 
the manner in which we define these quantities and of the arbitrary choice that was made regarding 
the direction of positive hwp rotation.

However, since we know the direction of positive hwp rotation (I recorded it), we can do a 
calculation to see if the directional relationship I found between Phi_Inst_cor and the raw 
Stokes' parameters is in accord with expectations.

From the SHARP instrument logbook (nights of 1/7/06 and 1/8/06) I read that the direction of 
positive hwp rotation (increasing encoder readings) is "counter-clockwise looking at sky".  So we 
see that "hwp angular momentum vector pointing towards SHARP" = "positive hwp motion".

Keeping in mind that (a) the effect of the hwp is to reflect the E-vector about the fast (or, 
equivalently, the slow) axis, (b) the arbitrary hwp offset has no effect on the directional 
relationship between Phi_Inst_cor and the raw Stokes' parameters, and (c) we can ignore 
reflections between the hwp and M3 since there are an even number, one can see from Figure on HWP rotation 
that the recorded direction of positive hwp rotation is indeed consistent with the directional 
relationship that I found between the Phi_Inst_cor and (Q_raw, U_raw).

So everything is consistent.