"Seegrid will be due for a migration to confluence on the 1st of August. Any update on or after the 1st of August will NOT be migrated"
Back to AdxDiscussions


Size distribution

Sizing can change per specimen, its not a batch constant.

I believe the crush size should be a seperate entity to seive size.

-- DavidHester - 03 Sep 2003

Currently this schema supports the information from the Newmont template: the commentary prescribes different quantification for samples prepared by crushing compared with seiving. I encapsulated it in a separate element and gave it the generic name "sizeDistribution" so that we can revisit this later in a clean way.

-- SimonCox - 04 Sep 2003

I know I'm weighing into this a week or so later that the rest of you, but, in my experience it would be unusual as part of the preparation to report a separate sizing per specimen. That would form part of an 'analyte procedure' and the procedure is the sizing, the 'result would be reported at x grams. With Screen Fire Assay's, the sizing 'value' could be treated as an addition to the Assay Specimen Lab Id, with a flag in the related specimen field, for instance, of 'sizing'. Suggestions only as a means of not over complicating process.

SamanthaHussey - 15 Sep 2003

Size distribution is often identified in the preparation procedures, in general the material needs to be prepared to a particular size distribution in order for 'homogenous' sample dissolution etc. It can vary depending on the analytical procedure to follow (eg XRD can be more specific) and may be specified as a pre-requisite for/within the Analyte Procedure. Can be considered constant across specimens (similar to drying temperature in the Assay Specimen Preparation). -- SamanthaHussey - 15 Sep 2003

This we could do with some resolution on: re-reading the discussion above, it is still not clear to me whether a size distribution is
  1. unique to a single prepared specimen
  2. a predictable consequence of a specimen/pulp preparation procedure
  3. a requirement of a particular analyte procedure
  4. several of the above
  5. non of the above

The intention is to record the size distribution in the correct place, and thereby reduce redundancy. In the model proposed, options correspond to the different objects that can be described separately - i.e. the Specimen Preparation Procedure, the Specimen intance, the Analyte Procedure, or the Analysis instance.

Is it possible that the size distribution is normally a predictable consequence of the prep procedure, and thus could be recorded as part of the prep description, but may, occasionally, vary for a particular specimen, so it should be possible to (optionally) record a size distribution locally which overrides the procedure-defined default?

Another question: how do we know what the size distribution of a specimen/pulp is? is every specimen/pulp subject to a seive or difractometer analysis to determine its size dictribution?)

-- SimonCox - 27 Oct 2003

You may need to consider a ScreenFire Analysis.

Report_Analytes +10, -10, +25, -25, +50, -50

Specimen_ID1 value, value, value, value, value, value

Where the +/-10 etc are actually seive size analyte headers (-10/+10 are equivalent chemical element column headers)

-- DavidHester - 03 Sep 2003

I suggest that in terms of the model we are using, these are analyses performed on different specimens, all derived from the same source specimen.

-- SimonCox - 04 Sep 2003

-- SimonCox - 31 Oct 2003
Topic revision: r3 - 15 Oct 2010, UnknownUser
 

Current license: All material on this collaboration platform is licensed under a Creative Commons Attribution 3.0 Australia Licence (CC BY 3.0).