1. SAMPLING
Sampling:
It is not possible or desirable to test all the raw material or all the final output from a production process because of
time and cost constraints.
Many tests are destructive so that there would not be any material left after it had been tested. Because of this,
representative samples of the material are tested.
Terms used in sampling:
Consignment:
This is the quantity of material delivered at the same time. Each consignment may consist of one or several lots.
Test lot or batch:
This consists of all the containers of a textile material of one defined type and quality, delivered to one customer
according to one dispatch note. The material is presumed to be uniform so that this is the whole of the material
whose properties are to be characterized by one set of tests. It can be considered to be equivalent to the statistical
population.
2. Laboratory sample:
This is the material that will be used as a basis for carrying out the measurement in the laboratory. This is derived by
appropriate random sampling methods from the test lot.
Test specimen:
This is the one that is actually used for the individual measurement and is derived from the laboratory sample. Normally,
measurements are made from several test specimens.
Package:
Elementary units (which can be unwound) within each container in the consignment. They might be bump top, hanks,
skeins, bobbins, cones or other support on to which have been wound tow, top, sliver, roving or yarn.
Container or case:
A shipping unit identified on the dispatch note, usually a carton, box, bale or other container which may or may not
contain packages.
3. Sample:
It is a relatively small fraction which is selected to represent a population.
Reasons for sampling:
To minimize time requirement for testing.
Design nature of many of the tests.
For example :
1) Only 20mg of cotton sample is used from 250kg of cotton:
2) 10 random sample of cones from one container of 15ton of yarns:
4. Aim of sampling:
To produce an unbiased sample in which the population of the different fibre length in the sample are same as
those in the bulk or through sampling systems of each fibre in the bale should have equal chance of being
chosen for the sample.
Sampling methods are governed by:
1. Form of the material (fibre/yarn/fabric).
2. Amount of material available.
3. Nature of the test.
4. Type of testing instruments.
5. Information required.
6. Degree of accuracy required.
5. TYPES OF SAMPLE:
RANDOM SAMPLE:
In this type of sample every individual in the population has an equal chance of being included in it. It is free from bias,
therefore truly representative of the population.
NUMERICAL SAMPLE:
A sample in which the proportion by number of, say, long, medium, and short fibers would be the same in sample as in the
population.
BIASED SAMPLE:
When the selection of an individual is influenced by factors other than chance, a sample ceases to be truly representative of
the bulk and a biased sample results.
6. Causes of bias in sampling:
Bias due to physical characteristics:
Longer fibers always have a greater chance of being selected.
Position relative to the person:
Lab assistant may pick bobbins from top layer of a case of yarn (whether to save himself the task of
digging down into the case or because he has never been told otherwise, we do not know), but the
bobbin chosen will be biased due to their position.
Subconscious bias:
Person selecting cones will pick the best looking ones free from ridges, cub webbed ends, etc., without
thinking about it.
7. FIBRE SAMPLING FROM BULK:
1.ZONING TECHNIQUE:
Handful of samples from at least 40 zones.(x: no. of original handfuls)
Take a quarter from each tuft to make the final sample looking ones free from any damages, etc.
From the bulk, a sample of about 2oz is prepared by selecting about eighty large tufts chosen, so far as
possible, over the bulk.
Divide this sample into four quarters.
Take 16 small tufts at random from each quarter, the size approximately 20mg.
Each tuft shall be halved four times, discarded alternately with right and left hands and turning the tuft
through a right angle between successive halvings. 16 'wisps' are thus produced from each quarter sample.
8.
9. Combine each set of wisps into a tuft.
Mix each tuft in turn by doubling and drawing between the fingers.
Divide each tuft into four parts.
Obtain four new tufts by combining a part of each of former tufts.
Mix each new tuft again by doubling and drawing.
Take a quarter from each tuft to make the final sample.
10. 2.CORE SAMPLING:
It is used for assessing the proportion of grease, vegetable matter in samples taken from unopened bales of raw wool.
It means half way into the bale i.e. samples from center. The tube enters in the direction of compression, so
perpendicular to the layers of fleece.
11. Cutting tip dia is lesser than coring tube.
helps sliding the core upside the tube penetrates.
helps retaining the core as it is withdrawn.
No. of cores are extracted and combined.
Different sizes of tube 14, 15, and 18mm.
After removal cores are kept in air tight container immediately.
Hydraulic coring machine for large number of samples.
12. Fiber Sampling from Combed Slivers/Rovings/Yarns:
Very difficult to obtain “unbiased” samples, because unless special precaution are taken, the longer fibres are more
likely to be taken by the sampling procedures, leading to length-biased sample.
Two ways of dealing with this problem:
Prepare a numerical sample (unbiased).
Prepare a length-biased sample in such way that the bias can be allowed for in any calculation (based on some
assumptions).
Remove all fibres which are started left of A (X zone) .The green fibres will be unaffected.
Again remove fibres, the fibre will be unchanged.
13. If the removal of one sample does not affect the composition of the remaining samples, then it can be considered
as “ numerical sample” and each segment is representative of the whole.
LENGTH-BIASED SAMPLE:
In sample the ratio of proportion of 10mm, 20mm, and 30mm would be 1:2:3.
Removal of length biased sample will change the proportion of fibres in the remaining bulk as longer
fibres will be removed at higher proportion.
In the earlier figure chance of fibre crossing the lines A and B is proportion to its length. If by some way
the fibres crossing this area (between A and B are selected ) then the longer fibres will preferentially
selected.
15. RANDOM DRAW METHOD:
Take out fibre (2mm at each stage) and discard until a distance equal to that of the longest fibre in the sliver has
removed. After that each draw will be of numerical samples.
16. CUT SQUARE METHOD:
Cut all the projected fibres and discarded. The glass plate is then moved back few mm, exposing more
fibres with “natural length” without cut. In each case projected fibre ends must be removed.
17. YARN SAMPLING:
1. Use of random numbers:
Table of random sampling number are normally used a small number of yarn bobbins are to be selected from
comparatively small bulk size.
Total 10 package are to be selected at random from the consignment.
2. a) If consignment contains more than five cases, they are selected at random from it.& then two packages are
selected at random from each case.
b) If no. of cases <5, then 10 packages are selected at random approximately, equal from each package.
2. Count of yarn removed from fabric:
Rectangular strips two for warp and five for weft.
Normal size = 20'' width at least 50 threads.
Different warp or weft in each rectangle.
18. 3. Twist in yarn in package form:
Specimens in equal no. of from 10 packages.
No specimen from within 1yard of the end of package.
Minimum 1yard distance between consecutive specimens.
4. Lea strength of spun yarns:
20 complete leas, one each from 20 packages.
If no. of packages is less than 20, then 20 leas are selected at random approximately equal from each
packages.
FABRIC SAMPLING:
Fabric samples from warp and weft are taken separately.
Warp direction should be marked before it is cut out.
No two specimens should contain same warp or weft threads.
Samples should not be from within 50mm of selvedge.
