Estimating # observations or outputting FWHM per HKL

mathiasmorch

Is it possible to calculate the number of good/independent observations directly in TOPAS, accounting for partial peak overlap? Something similar to: Altomare1995, David1999, Sivia2000, Wright2004 or VonDreele2006?
Both for angle-dispersive x-ray and tof neutron.

Alternatively, is there a way to output the calculated FWHM, per HKL after a refinement?
Something like (based on Matthew Rowles pdcif):
for strs { Out_CIF_newfile(ciffile) phase_out ciffile append load out_record out_fmt out_eqn { "\n%4.0f" = H; " %4.0f" = K; " %4.0f" = L; " %V" = phase_id; " %11.6f" = D_spacing; " %12.6f" = Iobs_no_scale_pks_err; " %12.6f" = I_after_scale_pks;
" %12.6f" = fwhm_for_hkl;
} }
Image of data for clarification

johnsoevans

The second question is answered in a couple of earlier posts from Alan - one in 2015 and one in 2016. If you search "output fwhm" they come up. Something like #1 and #4 in the list. You'd have to tweak the scripts for the peak shape function you're using. Notice the caveats about convolution in those answers, which might make it hard in the tof case.

I'm not aware of any built-in macros along the lines of those publications.

mathiasmorch

@johnsoevans
Thank you, i found 2 posts from 2014 and 2015, but no 2016: 2015, 2014.
Although I don't see those convolution caveats you mention?
This should be doable with peak shape i have for synchrotron data.
The TOF data i have where i use wifd_pkshape and tof_sample_peakshape might be more challenging with this approach.

johnsoevans

Those are the ones. The convolution caveat was just the warning that getting the overall fwhm isn't possible if there are multiple convoutions.

mathiasmorch

johnsoevans Ok, just to be clear; it's not possible to extract the total FWHM this way, if I have two (or more) contributions to the peak shape?
For example: Slit_Width(0.015) and LVol_FWHM_CS_G_L(...)

johnsoevans

mathiasmorch That's right. You need to be able to express the fwhm as an equation.