alan-jarvis
Hi,
I'm new to the Forum and to Topas.
I have read thru the previous posts, the two manuals for Topas, the information on the Durham Topas site, and
some of the GSAS/EXPGUI manuals/help docs.
I think my questions are new.
I've been given some templates by someone who isn't a Topas expert either, and I'd like to reset the refinement to the "default" values before I run it.
Perhaps "starting" values is a better term.
What I've done is take a template, add my own data and run it.
But using the previous values for the various refined and calculated terms since i don't know what to re-set them to.
Section 3.8 of the Technical Reference talks about this a little bit, but is limited in coverage of functions and parameters.
For example, the background function (Chebyshev) "bkg" I am using has 12 terms: all have been previously refined by the person who gave me the template.
From my limited experience with GSAS I believe you can have anywhere from 2, 4, 6, 8, 10 or 12 terms.
If I delete these previous values (refined from another set of data) and replace all with zeros, it runs and generates DIFFERENT values than it had before.
If I start with all "1"'s then, yes you guessed it, I end up with yet more values.
And if I go with "1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12" ...sigh.
This is not a good thing obviously.
In all cases they converge: I run them several times to check.
In EXPGUI/GSAS the software selects the starting values for you.
So what does TOPAS need for default values for bkg?
Part 2 of my question, does anyone have any ideas for the starting values (or perhaps the potential range) of the following?
I understand they are not really being REFINED, but they need starting values:
r_wp
r_exp
r_p
r_wp_dash
r_exp_dash
weighted_Durbin_Watson
gof
1h
TCHZ_Peak_Type: like if I add a new phase, what should I make the new values for pku etc?
Zero_Error
I've tried deleting the previously refined values that I got from my inherited great-great-grand-father's template and ran with nothing.
Which gave me an error: "expecting a numeric".
So I entered "1" for each one and it seemed to go OK.
And gave me more or less the same values as I had before: but not the same exactly.
It would be really nice to have some sort of macro you could run that would set everything to default values.
Maybe when I get better at Jedit and Topas I shall be able to write one.
Thanks for any help.
Alan Jarvis
Sheffield
alancoelho
Default values for bkg is zero. The number of terms is determined by the number of numbers found after the bkg keyword; ie.
bkg 0 0 0 0 0
will have five Chubychev parameters.
>I understand they are not really being REFINED, but they need starting values:
>r_wp
>r_exp
They don't need starting values; simply enter a zero and the value will be updated after refinement.
Starting value for Zero_Error is typically zero enetered by the user: ie.
Zero_Error(@, 0)
For TCHZ_Peak_Type then try:
TCHZ_Peak_Type(
@, .01,
@, .01,
@, .01,
@, .01,
@, .01,
@, .01
)