ISO 3170 Petroleum liquids - Manual sampling
11 Package sampling
This clause is based on ISO 2859-1, to which reference should be made for details.
11.1 Statistical aspects of sampling packages
11.1.1 Number of samples to be taken
11.1.1.1 General
The variability of the product both within and between packages, the number of packages sampled, and the precision of the test methods may all contribute to errors in the determination of the properties of the product as shown by the test results. The number of samples to be drawn will depend on the number of units of product, the acceptable quality level (AQL) and the inspection level.

11.1.1.2 Sampling to assess uniformity in a package
Draw spot samples from evenly distributed points within the product. Test each sample using a simple test based on an easily assessable characteristic (e.g. density, colour). Any variations in the test results in excess of the repeatability of the test method indicate that the product in the package is not uniform.

11.1.1.3 Sampling to assess the average quality of a batch
A batch consists of a number of packages of a product of a single composition.
a) Single package: If the product has been shown to be uniform (see 11.1.1.2), take a single spot sample, but, if not, take a sufficient number of spot samples and combine them to give a composite sample.
b) Multiple packages: The accuracy with which an estimate of the average quality of the product within a number of packages can be made is dependent on:
1) the number of packages sampled;
2) the precision of the test method used;
3) the variability of the product between packages.

11.1.1.4 Sampling of whole consignment
If all the packages are sampled, this necessarily involves the greatest amount of sampling, and the error in the determination of the average quality will depend on the testing of the samples.

If each of the samples is tested once, the average of the test results will be a measure of the average quality with the lowest uncertainty. If a composite sample is prepared and is tested (in duplicate), the average result will be a measure of the average quality, but with a much higher uncertainty.

11.1.1.5 Sampling of part of the batch
It is not always possible to sample all of a batch or consignment. Plans are given in this International Standard to enable a decision to be made as to the number of discrete samples which need to be taken from a batch in order that a valid conclusion may be arrived at regarding the quality of the total contents of all the packages.

11.1.2 Acceptable quality level (AQL) (see 3.1)
The AQL shall be 2.5 %.

11.1.3 Inspection level
The inspection level determines the relationship between the lot or batch size and the sample size (number of packages per batch and the required number of samples). The normal inspection level (see ISO 2859-1) shall be used.

11.1.4 Sampling plan
11.1.4.1 General
A sampling plan indicates the number of units of product from each batch which are to be inspected (sample size or series of sample sizes) and the criteria for determining the acceptability of the lot or batch (acceptance and rejection numbers) (see Tables 4, 5 and 6).

11.1.4.2 Single sampling plan
The number of sample units inspected shall be equal to the sample size given by the plan. If the number of defectives found in the sample is equal to or less than the acceptance number (see "Ac" in Table 5), the lot or batch shall be considered acceptable. If the number of defectives is equal to or greater than the rejection number (see "Re" in Table 5), the lot or batch shall be rejected.

11.1.4.3 Double sampling plan
The number of sample units inspected shall be equal to the first sample size given by the plan. If the number of defectives found in the first sample is equal to or less than the first acceptance number (see Table 6), the lot or batch shall be considered acceptable.

If the number of defectives found in the first sample is equal to or greater than the first acceptance and rejection number (see Table 6), the lot or batch shall be rejected.

If the number of defectives found in the first sample is between the first acceptance and rejection numbers (see Table 6), a second sample of the size given by the plan shall be inspected. The number of defectives found in the first and second samples shall be accumulated. If the cumulative number of defectives is equal to or less than the second acceptance number, the lot or batch shall be considered acceptable. If the cumulative number of defectives is equal to or greater than the second rejection number, the lot or batch shall be rejected.

Instructions on the use of the plans to determine the acceptability of the batch should be given with each plan (see ISO 2859-1).

11.2 Procedures for sampling packages
11.2.1 Drums and barrels
Place the drum or barrel on its side with the bung up. If the drum does not have a side bung, stand it upright and sample from the top. If detection of water, rust or other insoluble contaminants is desired, let the barrel or drum remain in this position long enough to permit the contaminants to settle.

Remove the bung and place it beside the bung hole with the oily side up. Close the upper end of the clean, dry sampling tube with the thumb, and lower the tube into the oil to a depth of about 300 mm. Remove the thumb, allowing oil to flow into the tube. Again close the upper end with the thumb and withdraw the tube. Rinse the tube with the oil by holding it steady horizontally and turning it so that the oil comes into contact with that part of the inside surface that will be immersed when the sample is taken. Avoid handling any part of the tube that will be immersed in the oil during the sampling operation.

Return the oil used to rinse the tube back to the drum, and allow the tube to drain.

Insert the tube into the oil again, holding the thumb against the upper end. (If an all-level sample is desired, insert the tube with the upper end open.) When the tube reaches the bottom, remove the thumb and allow the tube to fill. Replace the thumb, withdraw the tube quickly and transfer the contents to the sample container. Do not allow the hands to come into contact with any part of the sample. Close the sample container; replace and tighten the bung in the drum or barrel.

11.2.2 Cans
Obtain samples from cans of 20 L capacity or larger in the same manner as from drums and barrels (see 11.2.1) using a tube of proportionately smaller dimensions. For cans of less than 20 L capacity, use the entire contents as the sample, selecting cans as described in ISO 2859-1.