"out.txt" file reports different R_Bragg than reported in the ".INP" file

andreasdrejer

Hi all

When I run a standard Rietveld refinement I also make a text file reporting agreement factors. It works fine for e.g. r_wp, but I recently noticed that the R_Bragg values reported in my text file are different from the ones in the .INP file (I update my INP file after the refinement). For example, in a refinement with two phases, phase one has a R_Bragg value of 12.89 in the INP, but the reported value is 11.51. For the second phase the INP file reports 11.05 while the text file reports a value of 15.73. I suspect the reported values in the text file are from from e.g. the first iteration of the refinement, but I don't know for sure. I tried to set the number of iterations to 1, but the problem is still the same.

From my INP file (how I give r_bragg a parameter name and get it out in a text file):

str
a
b
c
....
....
r_bragg r_bragg1 12.8889705

out "Agreement_factors.txt" append
Out_String("R_Bragg_phase1 = \n")
Out(r_bragg1, "%11.2f\n")

Hope someone can give an advise or solution to my problem. I use TOPAS V6.

Best regards
Andreas

rowlesmr

Instead of giving R_Bragg a parameter name and referencing it in your output bbit, can you try "Out(Get(r_bragg), "%11.2f\n")", and see how that compares?

alancoelho

Hi Andreas

I an not able to reproduce the problem. I just ran test example PbsO4a.INP with the following:

	xdd ...
		r_wp 0
		str...
			r_bragg 0

	out aac.txt 
		Out(Get(r_wp),     " %11.5f\n")
		Out(Get(r_bragg), " %11.5f\n")
		Out(r_bragg_1,     " %11.5f\n")

In the OUT file I get:

xdd ...
r_wp 7.87871059
str...
r_bragg r_bragg_1 2.67249449

and in aac.txt I get

7.87871
2.67249
2.67249

If you send me a simplified version of your INP/data file I will have a look.

cheers
alan

andreasdrejer

Hi again

Matthew: I don't use "Out(Get(r_bragg), "%11.2f\n")" because I want the R_Bragg of each of the two phases present in the data. Even though I just tried to see what result it would give me. And as expected it gives R_Bragg for the second phase (because it is the last in the INP file), but not the value in the INP file. But thanks for the suggestion.

Thank you Alan, I have send you my data/INP files

Cheers
Andreas

alancoelho

The r_bragg reported in the INP file is correct. In the OUT file it is not correct. There is a reporting bug and I am looking into it.

The problem is related to the use XYE files and the use of:

x_calculation_step = Yobs_dx_at(Xo);

There is a work around for your data which is of XYE type data. Your data is of equal x-axis steps with a step size of 0.009228. Instead of using x_calculation_step, use:

rebin_with_dx_of 0.009228

This then reports the correct r_bragg in your TXT file.

I will fix the problem and get back to you.

andreasdrejer

Hi Alan

Thank you very much, the use of "rebin_with_dx_of" works and I now get the same R_Bragg values in the INP and OUT.TXT files

Cheers!
Andreas

alancoelho

I have fixed the problem for Version 7.

For Version 6 the problem occurs when r_bragg is outputted using 'out' and when using XYE files. Typically XYE files are detected as constant step size if the step size is indeed constant. In the present case round-off prevented this. The work around is as given above.

I will put note in the Topics section of the Wiki (when I remember how to logon in that section).

alan

andreasdrejer

I tried to look through some older refinements I did with .XY files and here the problem was the same. Again the use of
"rebin_with_dx_of" worked to get the same R_Bragg values in the out.txt as in the INP file.