[pdDMG] A definition for F_meas_squared for powder data

[kaduk at polycrystallography.com]

This seems to have converged on the right definition.

Jim Kaduk

On 2017-03-22 08:11, Robert Von Dreele wrote:
> Good
> On 3/21/2017 5:43 PM, James Hester wrote:
> 
>> A further refinement to the definition based on feedback:
>> 
>> ;
>> The structure factor amplitude for the reflection derived by
>> partitioning
>> the background-subtracted observed intensity
>> _pd_proc.intensity_net between
>> reflections in the same proportion as those reflections
>> contribute to
>> the corresponding background-free calculated point in
>> _pd_calc.intensity_net .
>> 
>> ;
>> 
>> On 21 March 2017 at 16:37, James Hester <jamesrhester at gmail.com>
>> wrote:
>> 
>> Examination of the GSAS-II GetFobsSq routine suggests that a more
>> precise definition would note that the background should be
>> subtracted first.   I propose the following revised definition:
>> 
>> ;
>> The structure factor amplitude for the reflection derived from
>> a powder
>> histogram by partitioning observed intensity between peaks at
>> each point
>> in the same proportion as the model peaks contribute to that
>> point.
>> Model background is subtracted from both observed and
>> calculated values
>> before partitioning.
>> 
>> ;
>> 
>> On 13 March 2017 at 15:43, James Hester <jamesrhester at gmail.com>
>> wrote:
>> 
>> Dear Powder Dictionary Management Group,
>> 
>> It is now recognised formally that CIF dictionaries may need to
>> redefine datanames, and DDLm has been equipped with some facilities
>> to handle this.  In the case of the DDLm version of the powder
>> dictionary, as far as I can tell the only two datanames from the
>> core that are implicitly different are refln.F_complex and
>> refln.F_meas_squared.  _refln.F_complex is different purely because
>> the phase id has to be considered when performing the calculation
>> (unlike in the core case). _refln.F_meas_squared is different both
>> because of the need to consider phase_id, but also because the
>> procedure for calculating f_meas_squared from the powder histogram
>> is completely different to the single crystal case.  Could you all
>> please consider what I have written below for F_meas_squared, and
>> suggest any improvements. Ideally, the definition text should be
>> clear enough for a programmer to implement the calculation.
>> 
>> While I would appreciate feedback from at least some of you, if I
>> don't hear from you within a week or so, I will assume that the
>> definition below is adequate.
>> 
>> all the best,
>> James.
>> 
>> (New cif_pow definitions follow)
>> 
>> save_refln.F_complex
>> 
>> _definition.id [1]                          '_refln.F_complex'
>> loop_
>> _alias.definition_id
>> '_refln.F_complex'
>> '_refln_F_complex'
>> _definition.update                      2017-03-07
>> _description.text
>> ;
>> The structure factor vector for the reflection calculated from
>> the atom site data for the phase given by phase_id.
>> ;
>> _name.category_id                       refln
>> _name.object_id                         F_complex
>> _type.purpose                           Measurand
>> _type.source                            Derived
>> _type.container                         Single
>> _type.contents                          Complex
>> _enumeration.default                    0.
>> loop_
>> _method.purpose
>> _method.expression
>> Definition
>> ;
>> If (_diffrn_radiation.probe == "neutron")  _units.code =
>> "femtometres"
>> Else If (_diffrn_radiation.probe == "electron") _units.code =
>> "volts"
>> Else                                            _units.code =
>> "electrons"
>> ;
>> 
>> save_
>> 
>> save_refln.F_squared_meas
>> 
>> _definition.id [1]                          '_refln.F_squared_meas'
>> loop_
>> _alias.definition_id
>> '_refln.F_meas'
>> '_refln_F_meas'
>> _definition.update                      2017-03-07
>> _description.text
>> ;
>> The structure factor amplitude for the reflection derived from
>> a powder
>> histogram by partitioning observed intensity between peaks at
>> each point
>> in the same proportion as the model peaks contribute to that
>> point.
>> ;
>> _name.category_id                       refln
>> _name.object_id                         F_squared_meas
>> _type.purpose                           Measurand
>> _type.source                            Derived
>> _type.container                         Single
>> _type.contents                          Real
>> _enumeration.default                    0.
>> loop_
>> _method.purpose
>> _method.expression
>> Definition
>> ;
>> If (_diffrn_radiation.probe == "neutron")  _units.code =
>> "femtometres"
>> Else If (_diffrn_radiation.probe == "electron") _units.code =
>> "volts"
>> Else                                            _units.code =
>> "electrons"
>> ;
>> 
>> save_
>> 
>> --
>> T +61 (02) 9717 9907 [2]
>> F +61 (02) 9717 3145 [3]
>> M +61 (04) 0249 4148
>> 
>> --
>> T +61 (02) 9717 9907 [2]
>> F +61 (02) 9717 3145 [3]
>> M +61 (04) 0249 4148
> 
> --
> T +61 (02) 9717 9907
> F +61 (02) 9717 3145
> M +61 (04) 0249 4148
> 
> _______________________________________________
> pdDMG mailing list
> pdDMG at iucr.org
> http://mailman.iucr.org/cgi-bin/mailman/listinfo/pddmg
> 
>  		 [4]
>  		Virus-free. www.avast.com [5]
> 
> 
> 
> Links:
> ------
> [1] http://definition.id
> [2] tel:+61%202%209717%209907
> [3] tel:+61%202%209717%203145
> [4]
> https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient&utm_term=icon
> [5]
> https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient&utm_term=link
> _______________________________________________
> pdDMG mailing list
> pdDMG at iucr.org
> http://mailman.iucr.org/cgi-bin/mailman/listinfo/pddmg