The Laserscan Instrument

This chapter covers five topics.

Measurements Provided by the Instrument

Principle of Operation

Precision

Comparison with Airflow

Processing Predictions.

You can access each one individually using the links provided above or in the index, or simply scroll down this page.

Measurements provided by the Instrument

Measuring Mean Fibre Diameter by Projection Microscope is quite laborious. A large number of operators and a large number of laboratories must measure a large number of snippets in order to obtain a precise result. Clearly, for routine measurements this is impracticable, and rapid instrumental techniques are preferred.

Instruments such as Airflow, LASERSCAN and OFDA fulfill this need. They provide more precise results much more rapidly. However, none of these instruments can measure fibre fineness directly. Each instrument produces a measurable output signal which is related to fibre fineness, but which can only be converted to a numerical value by referring the magnitude of these signals to the magnitudes of signals obtained from samples of wool for which the Mean Fibre Diameter is already known. Consequently all these instruments must first be calibrated, using wool tops measured by Projection Microscope.

LASERSCAN measures fibre snippets to obtain:

• Mean Fibre Diameter (microns),
• Coefficient of Variation of Diameter (%), and
• the Diameter Distribution Histogram (% of fibres in 1-micron class intervals).

From the Diameter Distribution Histogram coarse edge statistics such as Comfort Factor can also be calculated and reported. The instrument also provides a measurement of Fibre Curvature (degrees/millimetre), but this measurement cannot be certified at this stage because an IWTO Specification for this parameter is not yet available.

Importantly, in terms of the values of the measurements it provides the LASERSCAN closely emulates the Projection Microscope. By measuring single snippets once it also ensures a length weighted distribution is obtained.

Principle of Operation

The illustration (Figure 8) shows how a LASERSCAN works. Fibre snippets are dispersed into isopropanol-water mixture and the resultant suspension then flows through a measurement cell. As they pass through the cell the fibre snippets intersect a thin beam of light generated by a laser. This beam is directed at a measurement detector. The detector produces an electrical signal that is proportional to the amount of light incident upon it. Therefore, when a fibre snippet passes through the beam this electrical signal is reduced by an amount that is directly proportional to the projected area and therefore thickness or diameter of the fibre. The relationship between the magnitude of this decrease and Mean Fibre Diameter is determined by calibrating the instrument using wool tops where the Mean Fibre Diameter and Diameter Distribution have already been determined by direct measurement using the Projection Microscope.

Figure 8 : Laserscan Schematic Diagram

Figure 9 : Optical Discriminator

The LASERSCAN's optical discriminator is fundamentally designed for selecting snippets for the determination of fibre fineness. It consists of a central detector surrounded by 16 smaller detectors. The instrument will only accept a snippet for measurement if the snippet intersects the central detector and two of the surrounding detectors. However, because the physical dimensions of the detector array is precisely known, the instrument can also differentiate the curvature of the individual fibre snippets, depending upon which pair of the surrounding detectors is intersected by the fibre. The value of a and b for each corresponding pair may be inferred from the physical dimensions of the discriminator.

However, it is important that only snippets that fully intersect the beam are actually measured. It is also important that only one snippet is measured at any instant. Otherwise, the signal from the detector will indicate the snippet is either finer or coarser than it really is.

To ensure that only single snippets are measured and that the snippets fully intersect the beam the instrument uses a fibre optic discriminator. The principle of this device is illustrated in Figure 9. It consists of a ring of fibre optic detectors surrounding a single fibre optic detector. The signal from each of these is continuously monitored. A high-speed computer program identifies when a decrease in signal from the central detector and two of the surrounding detectors occurs simultaneously and matches this event with the signal from the main detector. Events that do not match this selection criterion are rejected. The fibre optic discriminator also provides the instrument with the capability to measure curvature (see Figure 9).

In this regard, the LASERSCAN emulates the Projection Microscope, in that only measurements on individual snippets are used to accumulate the Fibre Diameter Distribution. This is critically important in eliminating bias resulting from selective sampling from the total population of snippets presented to the instrument. The LASERSCAN is the only commercial instrument that has this capacity.

Precision

In an ideal world repeated measurement of the same sample will always provide exactly the same result. In a real world this is not so. There is variation associated with every measurement, arising from the measuring instrument, the sample preparation and the sampling itself. There is also human variation but one of the advantages of well-designed instrumental techniques is that the magnitude of human variation is generally significantly reduced.

The precision of a measurement is an indicator of the magnitude of the measurement variation. It is usually defined as a range between which 95% of all repeated measurements will generally lie. The Test Specifications developed by IWTO for the measurement of wool sliver show that the LASERSCAN provides superior measurement precision when compared to any of the other methods. The precision limits for all these specifications, for wool sliver, are summarised in Table 2.

Table 2 : Precision of Fibre Diameter

* Based on <26 micron and >26 micron respectively

The reproducibility of measurements between laboratories is of major commercial importance to producers, traders and processors. An analysis of the Interwoollabs International Inter-laboratory Round Trials has shown that for wool sliver the LASERSCAN provides superior performance to all other measurements.

As indicated above, the range between laboratories in these trials is best (i.e. lowest) for those laboratories using the LASERSCAN instrument to measure Mean Fibre Diameter.

Comparison with Airflow

The Airflow instrument has been the basis for trading Australian wool for over 25 years. Combing mills have developed relationships between the Mean Fibre Diameter of the raw wool they purchase and the Mean Fibre Diameter of the top they produce from this raw wool. The reliability of these relationships is critical in accurately predicting the quality of the top, yarn and fabrics produced from this raw wool.

As shown in Figure 10, the Mean Fibre Diameter determined by LASERSCAN and the Mean Fibre Diameter determined by Airflow is, on average, virtually the same.

In this illustration, the difference in Mean Fibre Diameter between Airflow and LASERSCAN for approximately 2500 samples of Australian greasy wool is plotted against the Airflow Mean Fibre Diameter. The solid red line represents the average difference. Note that it is virtually horizontal and very close to zero, indicating that on average there is no difference between the two instruments.

For individual samples there are seemingly large differences. However, 95% of the differences are in the range ±0.5 microns. Given the range of diameters involved this is very close to the expected precision limits for greasy wool for both instruments.

Nevertheless, some of these differences for individual samples may also be real. Unlike the Airflow instrument, the LASERSCAN is not affected by medullation and provides a more accurate measurement of the MFD of medullated wool than the Airflow. LASERSCAN is not affected by Coefficient of Variation of Diameter. It is measuring the individual fibre snippets.

Figure 10 : Difference between LASERSCAN and AIRFLOW

The Airflow is affected by Coefficient of Variation of Diameter. It actually responds to the surface area of a mass of fibres (finer fibres have a larger surface area per unit mass). Two samples may have the same Mean Fibre Diameter but very different Coefficients of Variation of Diameter. In this instance the sample with the higher Coefficient of Variation of Diameter will have a lower surface area, and consequently the Airflow will actually provide a slightly coarser result for Mean Fibre Diameter (MFD) than the real value.

It is also known that for very fine wools (less than 16 microns) the LASERSCAN will provide a finer result than the Airflow. However, it has been demonstrated that this is due to errors in extrapolating the calibration of the Airflow and that the LASERSCAN, for these wools, is probably closer to the true result.

 

Extensive Research by CSIRO and AWTA Ltd has shown that LASERSCAN is not significantly biased by: Medullation Curvature Variations in Diameter Distribution Fibre ellipticity Fibre Colour Unlike alternative instruments (e.g. OFDA) the calibration of the LASERSCAN is not affected by the mode of preparation of the calibration tops. Hence only one calibration is required for greasy wool, scoured wool and wool sliver. The LASERSCAN closely emulates the Projection Microscope in that snippets are measured only once thus ensuring a length weighted mean is obtained.

 

Of the two commercial instruments available for determination of MFD and Distribution in Diameter, for Australian greasy wool LASERSCAN is more highly correlated with Airflow, and shows no apparent bias over the range of diameters normally encountered. Commercially this is very important, as it means that the performance of Australian greasy wool, in almost all instances, can continue to be reliably predicted.

Processing Predictions

The availability of CVD information on Australian greasy wool will provide more accurate prediction of the MFD of wool tops; particularly those produced from consignments where the component lots have unusually low or high CVD’s. The CVD information opens up the possibility of extending the range of MFD for the component lots within a consignment, yet still allowing accurate prediction of combing performance.

To understand this, it is necessary to understand how an average MFD for a combing consignment is calculated. For non-mathematicians the calculation is quite complex but greatly simplified in the modern world by the availability of computers. The mathematical equations involved are shown on the left.

When Airflow became the primary method for calculating the MFD of a delivery, it was not possible to use these equations because this instrument did not provide the CVD information. An approximation was made which assumed that the range in MFD and CVD between the component lots was small.

If these assumptions are correct then the prediction of the MFD of a wool top will generally be quite reliable and this has indeed been the case for many years. It is well known that if the average MFD of the wool top is usually about 0.2 to 0.3 microns coarser than the average for the greasy wool (measured by Airflow) from which it was produced.

Of course there are processing effects. During processing some fibre is removed and generally this wastage is finer than the average for the greasy wool. The extent to which this occurs can be controlled by the mill, but it cannot be eliminated. This is why tops are generally slightly coarser.

However, occasionally, and without apparent reason, the actual differences are much larger than expected. There are many reasons why this may occur, among which is the effect of CVD on the measurement of MFD by Airflow.

Why is this the case? It is known that the Airflow is affected by the CVD of the sample it is measuring. Generally, the magnitude of this effect is small, smaller than the confidence limits of the method. However, the effect increases as the CVD for a given MFD becomes substantially greater than or substantially less than the CVD of the tops used to calibrate the Airflow instrument at the particular MFD. For abnormally high values of CVD, the Airflow estimates of MFD will tend to be higher than the estimates on the same sample obtained by LASERSCAN (and Projection Microscope). For abnormally low values of CVD the Airflow estimates will tend to be lower.

The commercial implications of this are simply stated. The CVD of a sliver produced from a wool blend is determined by two factors:

• the CVD of each of the sub-lots in the consignment used to produce the sliver; and
• the range in MFD between sub-lots in this consignment.

The first factor becomes the more important in special circumstances. Consignments assembled from lots that have a lower than normal CVD, and where the range in MFD between lots is very small, are more likely to have an abnormally low CVD. This situation is most likely to occur in fine wool consignments, particularly where the lots have been classed using objective MFD data for each fleece, or it may occur where the consignment consists of lots assembled from an individual farm. It will result in the MFD of the sliver, measured by Airflow, being finer than measurements made by LASERSCAN (or Projection Microscope).

If it is assumed that the range in Mean Fibre Diameter and Fractional Coefficient of Variation of Diameter between component lots of a consignment is small then the equation for calculating the Mean Fibre Diameter of the consignment is greatly simplified.

Here B is the Wool Base (%) and M is the nett mass (kilograms) of greasy wool for the consignment.

If these assumptions are not correct, then the accuracy of the prediction of combing performance will be reduced. Ensuring that the range in Mean Fibre diameter is small is a simple exercise because the component lot details are known. However, prior to the introduction of LASERSCAN this was not possible for Coefficient of Variation of Diameter. Variation in Coefficient of Variation of Diameter can cause errors in predicting the diameter of the top.

For consignments assembled from visually classed farm lots or dealer lots, the second of these factors is likely to be the more important. The larger the range in MFD between the sub-lots the larger the CVD of the consignment. The MFD of the sliver produced from these consignments when measured by Airflow is likely to be higher than the measurements made by LASERSCAN (and Projection Microscope). Conversely the narrower the range in MFD between the sub-lots the smaller the CVD of the consignment, and the Airflow MFD is likely to be closer to the MFD obtained by LASERCAN (and Projection Microscope).

It must also be noted that situations may arise where these factors can act together to either increase the differences or to reduce the differences between the methods.

The relationship between the calculated diameter, derived from measurements on core samples from the consignment sub-lots, and the diameter measured on the resulting wool top is well understood in most mills.

These comparisons (core/comb) have been derived using Airflow technology. Uncertainty in the relationship may be reduced when the CVD information for the greasy wool is available.

Table 3 shows the magnitude of the errors in MFD measured by Airflow that could be expected for differing Standard Deviation (SD) values. The corresponding CVD is shown in brackets.

In the case of the 20.0 micron wool (measured by LASERSCAN) the error in the Airflow measurement arising from the effect of CVD will be near zero (-0.1 microns) at a SD of 4.0 microns (i.e. CVD 20%). One would expect a bias of -0.4 microns when the SD was 3.0 microns (i.e. CVD 15%) and a bias of +0.4 microns when the SD was 5.0 microns (i.e. CVD 25%). It must be emphasised that these differences only relate to effects from differing SD or CVD.

The magnitudes of the potential errors are small, but they can be commercially significant. The commercial risk, in most instances will be minimised by the availability of data produced by the LASERSCAN instrument, and the more accurate calculation of consignment MFD using the correct formula incorporating the CVD value.