Comparisons Between OFDA4000 and IWTO Standardised Methods for Measurement Made on Wool Tops: Part 2 Length Characteristics
Abstract
A comparison has been conducted of the measurements of OFDA4000 and the IWTO Standard, Almeter on 601 commercial tops from the TEAM-3 project and 301 research tops from processing single sale lots. Results show statistically significant differences between OFDA4000 and Almeter for both average and level-dependent bias for most of the 40 parameters which characterise a top fibre length distribution, supporting previous work by Baxter.
This report uses the estimated the precision of a test for each parameter as a basis for interpretation of the magnitude of the average differences. The Table below sets out average differences (OFDA4000 –Almeter) for Hauteur and 5 other length parameters of interest:
|
|
|
Hauteur |
CV
(Hauteur)
% |
Barbe |
CV
(Barbe)
% |
K25H
(%) |
L5H
(mm) |
|
|
TEAM |
-0.8 |
0.7 |
-0.4 |
0.4 |
1.3 |
0.1 |
|
Sale Lots |
0.3 |
0.3 |
0.7 |
0.1 |
0.5 |
1.6 |
|
Precision |
±1.2 |
±1.9 |
±1.7 |
±1.0 |
±1.3 |
±1.3 |
|
With the exception of L5H for the Sale Lots data, all average differences for these parameters are within the precision limits for a test.
Results also indicate that, on average, the OFDA4000 has estimated the Almeter values of Hauteur and CVH, for the more extreme Sale Lots tops with similar accuracy as the TEAM-3 tops. There are results for a very small number of individual tops which show greater than expected differences between OFDA4000 and Almeter.
For any individual top the OFDA4000 can systematically over-estimate or under-estimate “K” and “L” parameters compared with the Almeter. These biases can range up to the equivalent of the precision of a test.
Citation
"Comparisons Between OFDA4000 and IWTO Standardised Methods for Measurement Made on Wool Tops: Part 2 Length Characteristics", V.E. Fish, A. Balasingam, J.W. Marler, T.J. Mahar, Report SG 03, Technology & Standards Committee, Sliver Group, Biella Meeting, November 2005