Sampling, equipment and methods

Contents - Previous - Next

Table 3.6 (see Table 3.6. National and International Standards - Sampling.) shows a number of standard test methods dedicated to sampling. There are at least 30 methods, of which 5 or more are applicable internationally. The remainder are national or regional standards.


The need for sampling

Batches of grain are rarely uniform in quality even when regarded as acceptable. Pests usually occur non-randomly in stored grain. Consequently the only sure way of obtaining complete and accurate information about the grain is to carry out a total examination. This may be possible if the quantity to be examined is small, but is usually neither practical nor economical when a large quantity is involved. The choice is either not to examine the consignment at all or to take samples to obtain some information, acknowledging that anything less than a total examination is bound to affect the accuracy of the results.


Principles of representative sampling

The results of sample analyses can be expressed in precise terms. However, precise analytical results may be of little practical value, and may be misleading if the samples are obtained without taking into account the non-random or aggregated distributions of foreign matter, damaged grains, insects, etc.

Certain principles of representative sampling must be observed:

Many countries adopt the recommendations of ISO 950 "Cereals - Sampling (as grain)". Its recommendation for selecting a proportion of bags is shown in Table 3.7, and for grain in road and rail trucks in Figure 3.1 (see Figure 3.1. Sampling Points in Bulk Grain Carriers.).


Working sample size

In practice, it is necessary to compromise between what is theoretically attainable and the natural desire to obtain results of analyses as quickly as possible. Providing the associated margins of error are recognised and accepted, it is generally suggested that working samples of between 500 and 1000 grains should be used for the determination of common defects such as insect damage, broken grains and discoloured grains. Equivalent minimum working sample weights are:

Maize (small grain) 200g
Maize (large grain) 250g
Sorghum 25g
Black-eyed cowpeas 150g
Wheat 25g
Bulrush millet 10g
Paddy 15g

Samples of these sizes can be analysed in 10 to 20 minutes, depending upon the skill of the inspector and available equipment.

Table 3.7. Selection of Bags for Sampling.

Number of bags in consignment Number of bags to be sampled
Up to 10 Every bag
11 to 100 10, drawn at random
More than 100 Square root (approximately) of the total number of bags drawn at random according to a suitable scheme.

In the USA, minimum working sample weights of 250g and 1000g are required for the determination of ergot and garlic respectively in wheat, while 250g samples are recommended for the determination of smut in both wheat and sorghum. Such 10 to 40 fold increases on the basic working sample weight illustrate what is meant by 'large' and 'small' sample sizes.

There is a need for sampling awareness when dealing with grain contaminated with mycotoxins. As analytical techniques improve, so detectable and tolerance levels are being lowered. To emphasise the link between the standard test method and specification, Jewers et al. (1989) conclude, "When aflatoxin levels are controlled by legislation it is important that sampling procedures and sample sizes are specified."

For the determination of foreign matter and live infestation, samples should be as large as possible. If bagged grain is being tested, the best results are obtained by passing the entire contents of sample bags over a suitable sieve.


Equipment for obtaining primary samples from bagged grain

(i) Simple bag sampling spears

These are the most commonly-used instruments for taking samples from bags, being relatively cheap, simple and quick. The main variations in design are illustrated in Figure 3.2 (see Figure 3.2. Typical Spears for sampling Bagged Grain.). Generally, sampling spears having a maximum external diameter of about 12mm are designed for small grains such as wheat, while 25mm diameter spears are suitable for larger grains. To obtain a good cross-sectional sample the spear should be 40 to 45cm in length.

The tapered type of sampling spear penetrates bags easily. However, it takes unequal portions of grain from along the line of penetration, which could lead to distorted assessments of grain quality. More even sampling is achieved with the cylindrical type of sampling spear.

The main disadvantage of obtaining samples with these instruments is that it does not conform to the basic principles of representative sampling. If foreign matter or defective grain happens to be very unevenly distributed in the bag, the haphazard nature of spear sampling could lead to a distorted quality assessment (Figure 3.3. Inadequacy of Spear Sampling. Black dots represent grain defects).

(ii) Double-tube sampling spears

These spears (Figure 3.2D) comprise two metal tubes, one fitting closely inside the other and each with several common slots. Spears may vary in length from 45cm to 3.5m, and in width from 12mm to 50mm. Turning the inner tube through 180 opens or closes the intake apertures, and so collects grain from a transverse section of the bag.

Double-tube sampling spears are designed primarily for obtaining samples from vertical lines of penetration in bulk grain, although small versions may be used for sampling bagged grain. They are superior in many ways to the simple bag sampling spear, but are still instruments of haphazard rather than representative sampling.

(iii) The Produce-Flow sampler

This sampler (Figure 3.4. Vertical section of Produce-flow Sampler.) was designed at the Tropical Products Institute, now a part of NRI, as a representative sampling device for bagged grain. Grain is tipped into the hopper and falls through onto a cone, which is positioned to ensure that the flow is evenly distributed. Some of the grain is trapped by four vents arranged equidistantly around the base of the cone, and directed via a separate spout into a sample collector. The size of the sample depends upon the dimensions of the vents, which are interchangeable. Sampling of a 100kg bag of grain is complete within 20 seconds of starting the flow.


Equipment for obtaining primary samples from bulk grain

Bulk grain is sampled either when it is static, i.e. when it is contained in a truck, barge or storage bin, or when it is on the move, i.e. when it is being discharged through a spout or on a conveyor belt. A wide range of sampling equipment has been developed to meet the special requirements of these various situations, some for small-scale operations and other items for situations where grain is handled in very large quantities.

(i) Equipment for sampling static bulk grain

Double-tube sampling spears (see also above)

Spears 1.8m long and 3.5cm outer diameter are commonly used, but longer 3.7m double-tube spears are available for sampling grain in exceptionally deep trucks and barges.

The sampling spear should always be inserted into bulk grain at a slight angle from the vertical, with the slots facing upward. The slots must be opened only when the spear has reached the sampling position, and must be closed before it is removed.

Manually-operated deep bin probes

The simplest probe of this type consists of a hollow spear head, which serves as a sample cup, with a spring-loaded cap attached to a metal or wooden rod about 1 metre long. Extension rods are attached to increase the depth of penetration. When the sampling point has been reached a slight upward pull on the rod lifts the cap of the spear head, allowing grain to fill the cup. The probe is then withdrawn completely and the sample removed. A single probe yields up to 300g of sample material.

The deep bin fin-probe consists of a double-tube sampler with a set of extension rods. When the sampling position is reached a twist of the extension rod opens the sample intakes. This action is facilitated by the fin which prevents the outer tube from turning. A reverse twist closes the sample intakes before the probe is withdrawn from the grain. Up to 600g of sample representing a 1.5m long vertical 'cut' may be obtained.

A considerable amount of physical effort is required to push any of these probes into grain. None can be expected to penetrate more than approximately 5 metres.

Pneumatic grain samplers

Pneumatic grain samplers overcome the main disadvantages of manual operation by using powered-suction to penetrate the static bulk of grain, and by taking a continuous sample. They are quicker to operate than manual samplers, and can be used easily to obtain samples from the sides and floors of bulk grain containers. An example is shown in Figure 3.5 (see Figure 3.5: Pneumatic Sampler.).

Auger-type sampler

The sampler consists of a tube approximately 1.4m long and 5cm wide, open at the bottom end and housing a motor-powered auger. Grain lifted by the screw is collected in a bag at the outlet spout. It is necessary to insert the device into the grain at an angle in order to obtain sample material. There are no extension pieces which would permit sampling deeper than the half metre or so the sampler penetrates. The sampler is therefore of limited usefulness.

(ii) Equipment for sampling moving grain.

The Pelican sampler

The Pelican sampler (see Figure 3.6: Manual sampling of moving grain. A Pelican sampler. B Ellis Cup sampler.) consists of a cowhide pouch attached to a metal frame at the end of a hardwood or tubular metal handle. It is used to obtain samples from freefalling grain, e.g. from a spout discharge to the hold of a ship.

If the spout is sloping, the components of the grain stream are likely to be stratified. It is important, therefore, to cut the sampler through the stream from one side to the other in a single motion to obtain a good sample.

The force behind a stream of grain may be very great. It is essential to observe appropriate safety measures when sampling in this manner.

The Ellis Cup sampler

This is a hand-held scoop (Figure 3.6B), designed for obtaining small samples from bulk grain on moving conveyor belts. When properly used, the cup will obtain a vertical section of the flowing grain at the point where it is inserted into the stream. Samples taken in this way are used for making spot checks on the condition of grain and are not intended as substitutes for representative samples obtained elsewhere in the system.

Sampling with the Ellis cup is hazardous. Extra safety precautions are necessary, as with the Pelican sampler.

Limpet-type sampler

This type of sampler is clamped or bolted to the outside of the delivery spout. A tube is inserted through a hole drilled into the spout wall. The tube usually is open at both ends and has an inlet slot in the upper side projecting into the grain stream. Sampled material is removed either by means of a motorised worm screw, or a plunger operated by compressed air. Worm screw extractors can be made to operate continuously or at intervals. Plunger sample extractors remove samples of fixed size at intervals.

The limpet sampler is capable only of extracting material from part of a grain stream. Figure 3.7 (see Figure 3.7. Limpet sampler.) shows the auger principle in operation. If there is any appreciable stratification of material in the stream, samples cannot be regarded as representative.

The diverter-type sampler

The diverter-type sampler is probably the best device yet invented for obtaining representative samples from bulk grain. The sampler (Figure 3.8. Principle of operation of the Diverter sampler.) is designed to take a complete cross-section of a stream of grain, by means of a powered diverter head which takes a cut through the stream, on a preset schedule. During periods of inactivity the aperture of the diverter head is sealed to prevent it collecting dust.

Grain extracted from the main stream by the sampler may be fed directly into a secondary sampler, which reduces the sample to a manageable size before it is delivered via spouting to the grain inspection laboratory. Figure 3.8 shows the principle of operation.

Contents - Previous - Next