Understanding the Effects of Changes over Time to the OFDA 4000 Software
Abstract
During the development of the OFDA 4000 there have been a number of iterations of the OFDA software. This is a logical process in being able to satisfy IWTO requirements for the OFDA 4000 to demonstrate ‘equivalence’ for Hauteur, Barbe, their CV values and the respective distribution statistics.
Baxter (2005) has shown that OFDA 4000 values generated for the average values Hauteur, CVH, Barbe and CVB are in statistical agreement with the Almeter. There is new evidence emerging concerning the range within these parameters (Balasingham et.al. this meeting). Couchman’s (2005) findings generally concur with this but there remains a continued debate concerning what appear to be systematic differences in the distribution parameters (Baxter 2005, Couchman 2005 & Balasingham et.al. 2005) and the methodology used to convert optical length and scan step diameter readings to Hauteur and Barbe.
The IWTO questionnaire that was sent to IWTO national committees after the Hobart meeting, seeking their feedback re the commercial importance of a range of MFL parameters, did not receive many replies. This low level and conflicting response makes it difficult for technicians to advise IWTO members of the importance of the technical differences being observed between the Almeter and the OFDA4000. Significant differences are occurring in the K (short fibre) and the L (long fibre) values so commercial members need to decide if the differences will cause trading problems.
This paper reports on observations of the effects of the software iterations over a 2 year period and on the current situation regarding attempts to match the cumulative distribution curves which generate all the ‘fibre’ length attributes reported by the Almeter.
The current OFDA 4000 software (Ver."5.18" "Calb date = 30 Nov 04") was used by Baxter (2005) in reporting the results of an international round trial and thus results for earlier versions are by and large, redundant except in the context of understanding the changes that have been made to the software and why they are important in trying to match the Almeter’s cumulative capacitance also referred to as the cumulative distribution or frequency curves.
Citation
"Understanding the Effects of Changes over Time to the OFDA 4000 Software", R.C.Couchman, Report SG 01, Technology & Standards Committee, Sliver Group, Biella Meeting, November 2005