signs of resolution matrix elements?

[js54434]

I'm trying to generate the resolution matrix using both neutronpy and "reselps", but I'm getting some of the matrix elements to be of the opposite sign.

neutronpy code:

from neutronpy.resolution import Sample
from neutronpy.resolution import Instrument

efixed = 14.7
hcol = [50, 80, 50, 120]
vcol = [300, 300, 300, 300]
ana='PG(002)'
mono='PG(002)'

u = [1., 0., 0.]
v = [1., 0., 0.]
sample = Sample(4., 4., 4.,
                             90, 90, 90,
                             mosaic=60.,
                             u=u, v=v)
q = np.array([[2.], [0.], [0.], [4.]])
sample.vmosaic=1.
EXP = Instrument(efixed=efixed, sample=sample, hcol=hcol, vcol=vcol,
                             ana=ana, mono=mono,
                             mono_mosaic=35., ana_mosaic=35.)
EXP.calc_resolution(q)
print(EXP.RMS)

neutronpy result:

[[ 2265.98963477   299.07421529     0.           -43.68486242]
 [  299.07421529  1667.51953788     0.          -145.49237605]
 [    0.             0.            90.07679538     0.        ]
 [  -43.68486242  -145.49237605     0.            14.42250498]]

With similar parameters,

epm = 1
ep = 1
alz, alm, ala, al3 = 50, 80, 50, 120
tm, ta = 1.87325, 1.87325
etam, etaa = 35., 35.
efix = -14.7
etash = 60.
etasv = 1.
source = 1
vbetz, vbet1, vbet2, vbet3 = 300, 300, 300, 300

the reselps code has this result:

[[ 2265.99375019  -299.07567774   -43.6850311      0.        ]
 [ -299.07567774  1667.52554288   145.49293785     0.        ]
 [  -43.6850311    145.49293785    14.42256067     0.        ]
 [    0.             0.             0.            90.07572559]]

(the z and energy indices are swapped). It looks like the y-row and y-column are multiplied by -1. Can you think of a reason for this discrepancy? Is the "y" direction defined as clockwise relative to Q rather than counterclockwise as in the reselps code?

[pseudocubic]

After looking into it, the scattering is calculated for a left-handed coordinate frame (y is defined being clockwise from x). There is an implementation for changing to a right-handed one, but I haven't pushed it due to incomplete testing. In principle, the mondir attribute should control this but it hasn't been setup completely. I will look into this a little more now.

[pseudocubic]

Ok, so I was correct in saying that the default is to use a left-handed coordinate frame, but the way to change it is not through the mondir variable. In fact, dir1 and dir2 are also depreciated in favor of mono.dir, sample.dir, and ana.dir. These are already all implemented in the current release, so to use a right handed coordinate frame:

from neutronpy.resolution import Sample
from neutronpy.resolution import Instrument

efixed = 14.7
hcol = [50, 80, 50, 120]
vcol = [300, 300, 300, 300]
ana='PG(002)'
mono='PG(002)'

u = [1., 0., 0.]
v = [0., 1., 0.]
sample = Sample(4., 4., 4.,
                             90, 90, 90,
                             mosaic=60.,
                             u=u, v=v)
q = np.array([[2.], [0.], [0.], [4.]])
sample.vmosaic=1.
EXP = Instrument(efixed=efixed, sample=sample, hcol=hcol, vcol=vcol,
                             ana=ana, mono=mono,
                             mono_mosaic=35., ana_mosaic=35.)
EXP.mono.dir = 1
EXP.sample.dir = -1
EXP.and.dir = 1
EXP.calc_resolution(q)
print(EXP.RMS)

The above code should output

[[ 2265.98963477  -299.07421529     0.           -43.68486242]
 [ -299.07421529  1667.51953788     0.           145.49237605]
 [    0.             0.            90.07679538     0.        ]
 [  -43.68486242   145.49237605     0.            14.42250498]]

I will update the documentation to reflect the correct usage.