Hornik.cc

Performance of the OFDA 100 compared to other instruments
1. Measurements available
The OFDA instrument is used to certify mean fibre diameter (MFD) and diameter distribution (SDor CvD) under IWTO-47. It can also be used to measure medullation and the percentage ofobjectionable fibres under IWTO-57.
Additional measurements available from the instrument include mean fibre curvature, andparameters that relate to the amount of residual material on the surface of the fibres.
2. Precision when measuring greasy wool
Greasy wool may be certified for diameter under IWTO rules using Airflow, OFDA or Laserscan.
Each method is documented in the respective IWTO test methods IWTO-28, IWTO-47 andIWTO-12. The relative precision values of the complete measurement process (including someelements of sampling and preparation), as documented in these methods are as follows: 95 % Confidence Limits - MFD on greasy wool, from IWTO-
28, IWTO-47 & IWTO-12
95% CL (µm)
MFD (µm)
3. Precision when measuring tops
When used for certifying sliver or tops, the documented precision values in the IWTO testmethods are as follows for IWTO-6, IWTO-47 and IWTO-12: ** It should be noted that the precision data in IWTO-6 has not been updated since 1980 95 % Confidence Limits - MFD on sliver / tops, from IWTO-
6, IWTO-47 & IWTO-12
95% CL (µm)
MFD (µm)
With the exception of the data from IWTO-6, all the above data was obtained in carefully-controlled round trials under the auspices of IWTO1,2.
Another source of data for top testing is the 6 monthly Interwoollabs trials. Approximately 100airflow laboratories, 30 OFDA laboratories and 20 Laserscan laboratories participate in thesetrials. In comparison with the IWTO round trials, these groups of laboratories encompass a muchwider range of skill levels and throughputs, and therefore these trials would not be expected toproduce exactly the same levels of precision as the IWTO trials. The summary data produced in 1 Harig, H., report of the 1995 IWTO Round Trial, Part I (Raw wool) and Part II (Wool Tops), IWTO report15, Harrogate, June 1995 2 Marler, J.W. and Baxter, B.P., Addendum to Report 15: Confidence limits for Measurements made on rawwool and wool tops, IWTO Harrogate, June 1995.
these trials minimises within-laboratory variance and focuses on between-laboratory variance.
They cannot, therefore, be thought of as simply reflecting instrument performance.
Last 4 Interwoollabs round trials: between-laboratory
performances
Est. 95% CL µm
In all cases it can be seen that laboratories using OFDA and Laserscan perform morereproducibly than those using Airflow.
4. Comparison between the two new technologies
The 1995 IWTO round trials allowed comparison to be made between the grand mean results ofOFDA and Laserscan over a range of laboratories (4 core test labs in the case of the greasy woolsamples, and 16 core test and mill laboratories in the case of top samples). In terms of meanfibre diameter, these trials showed substantial equivalence between OFDA and Laserscantechnologies: 1995 IWTO trial - OFDA-Laserscan differences - Greasy
wool cores
OFDA-Laserscan (µm)
Grand mean MFD (µm)
1995 IWTO trial - OFDA-Laserscan differences - Tops
OFDA-Laserscan (µm) -1.0
Grand mean MFD (µm)
Data plotted in a similar manner for the last 4 sets of Interwoollabs trials (97/2, 98/1, 98/2 and99/1) indicates a similar level of agreement: Last 4 Interwoollabs round trials: differences between
OFDA and Laserscan grand means for MFD
Difference O-L (µm) -1.0
Grand mean MFD (µm)
In each of the above plots, the grand means over all laboratories have been used. Theimprovement in precision of comparison can easily be seen – the first plot is for grand meansfrom 4 laboratories, the second is for 16 laboratories and the third is for an average of 30 OFDAlaboratories and 20 Laserscan laboratories. It would seem clear that whilst some individual woolsmay give slightly different results on each of the systems, the overall trend is towardsequivalence.
5. Comparison against airflow
In view of the fact that there are 100 airflow laboratories participating in the Interwoollabs trials,perhaps the most precise comparison between the two newer technologies and airflow can beobtained from this data set. The plots below compare for each sample grand means from 100airflow laboratories with grand means from either 30 OFDA laboratories or 20 Laserscanlaboratories: Last 4 IH round trials: between instrument comparisons
for MFD: OFDA & Airflow
OFDA minus airflow (µm) -1.5
Grand mean MFD (µm)
Last 4 IH round trials: between instrument comparisons
for MFD: Laserscan and Airflow
Lsn minus airflow (µm)
Grand mean MFD (µm)
It is obvious that the level of agreement with airflow is similar for both instruments, but thecomparisons with airflow are somewhat less precise than the level of agreement between OFDAand Laserscan. In numerical terms, the average differences of the grand means were as followsfor these 32 samples tested over a period of 2 years: This clearly suggests that the overall differences are of little practical significance. There is noevidence for major fundamental difference between the average results from these instrumentson tops.
6. Other data
Results have been published elsewhere that suggest that there may be significant differencesbetween the instruments on greasy wool. Such suggestions are incorrect – differences that havebeen published refer to possible differences in results from different test methods, and it mustalso be noted that different laboratories have obtained different trends.
It must be observed that with the exception of the IWTO round trial quoted above, all the recentresults published to date comparing OFDA and Laserscan on greasy wool were obtained inindividual laboratories, and may therefore be influenced by the sample preparation systems andinstrument calibration uncertainties in these respective laboratories. It is unrealistic to rely onsuch data to predict what may happen outside those laboratories, and therefore the abovecomparisons so far provide the most reliable data available.
There have also been a number of papers published on the apparent differences between theinstruments (test methods) at the ultrafine end of the spectrum. Despite a significant change tothe Laserscan calibration algorithm in 1997 (equivalent to a change of approximately 0.5 µm at14 µm), specifically to address this issue, there does still remain the possibility of some relativelyminor differences3. Additionally, another detailed study4 has highlighted the errors inherent in theairflow method at this end the spectrum, and it is therefore invalid to draw comparisons betweeneither of the new instruments and airflow on ultrafine wools.
In conclusion, it might be appropriate to quote Sommerville, the author of the last 2 studies: IWTO now has a number of Test Specifications for specifying fibre fineness, all of which are fit for
purpose, provided the basis for their use in commercial trading of wool is adequately specified.
Each of these instruments uses different geometric definitions of fibre fineness. This is a fundamental
difference, which must contribute to the differences being observed, particularly where the characteristics
of the wool being measured differ in some way from the calibration wools and/or the fibre fineness is
outside the range of the calibration wools. This does not limit the usefulness of any one of the
instruments, provided the same instrument is used in all instances where comparisons must be made.
3 Sommerville, P.J., Measurement of the Fineness of Superfine Wool: Effect of the revised Laserscancalibration function on comparisons between airflow, Laserscan and OFDA, IWTO report CTF 04, Nice,Dec 1998 4 Sommerville, P.J., Fundamental principles of Fibre Fineness Measurement: the Airflow instrument, IWTOreport CTF 03, Nice, Dec 1998

Source: http://www.hornik.cc/download/Performance%20OFDA.pdf