ISO 3171 Petroleum liquids - Automatic pipeline sampling
5 Selection of sampling point (including stream conditioning)
5.1 General
As has been emphasized in clause 4, it is essential that the sampling probe is presented with a portion of the liquid flowing in the Pipeline which is representative of the whole. This depends on the extent of mixing within the pipeline, which in turn depends on a number of factors including flow rate and pipe configuration upstream of the sampling location.

The guidance given in this clause for the selection of the sampling point assumes that the pipeline remains full at all times.

5.2 Initial selection of sampling probe location
5.2.1 The forte of gravity tends to promote stratification in horizontal pipelines whereas distribution tends to be more uniform in vertical lines. For this reason, given the choice between vertical and horizontal, and assuming identical operating conditions, the preferred location is in a length of vertical pipe, provided that the pumping rate is significantly higher than the water and sediment settling rate (see annex A, A.2.2).

Reference should be made to annex C for a useful guide to the initial screening of potential sampling locations but annex A is preferred as it provides a more detailed treatment.

5.2.2 The turbulence.which exists naturally in a,long horizontal pipeline tan be, but frequently is not, sufficient to provide adequate mixing. The minimum natural turbulente for adequate mixing depends on flow rate, pipe diameter, viscosity, density and interfacial tension. If the natural turbulente is not sufficient for representative sampling, then the additional mixing offered by specific pipeline components should be introduced (see 5.3 and annex A).

5.2.3 The minimum acceptable flow rate and/or the minimum energy to be provided by a mixing device should be determined. Suitable procedures using formulae or nomographs described in annex A may be used.

5.3 Mixing devices
5.3.1 General
Mixing devices produce homogeneity by the addition of kinetic energy from outside or the conversion of pressure energy from within the pipeline.

In situations where there is an insufficient rate of energy dissipation for pipeline fittings or static mixers to be effective, consideration should be given to the use of powered mixers.

Reference should be made to annex C for a useful guide to the initial screening of potential sampling locations.

5.3.2 Pipeline fitthgs
Pipeline fittings may be used as in-line mixing devices. The fittings which may be suitable in this respect are valves, orifices, pipe reducers and expanders, manifolds, T-junctions and metering stations. These, however, vary in their effectiveness as mixers and may create concentration zones, leading to zigzag concentration profiles over a downstream distance equivalent to about 20 pipe diameters.

5.3.3 Reduced pipe diameter
If the velocity is too low for sufficient mixing, a length of pipe of reduced diameter tan often be installed to advantage but not to the extent that the obstruction caused by the sampling probe will distort the water distribution (see 7.2).

5.3.4 Vertical loops
The improved distribution afforded by a vertical length of pipeline (see 5.2.1) tan be obtained by inserting a vertical loop into a horizontal line. If sufficient line pressure is available the flow rate may be increased by reducing the diameter in the vertical loop to below that of the main line. In a vertical loop, the sampling probe should be preferably installed in the downstream leg, at a minimum distance of three, but preferably greater than five, pipe diameters from the upstream bend and a minimum distance of 0.5 pipe diameters from the downstrealn bend.

lt should be noted that water tan accumulate at the bottom of vertical piping until these accumulations resch such a size that they are carried fotward in the form of slugs. As a result of this phenomenon, water transients occur and this should be taken into account.

5.3.5 Static mixers
Static mixers are commercially available devices specially designed in one or more stages to provide adequate distribution and dispersion of the pipeline constituents prior to sampling.

Reference should be made to the manufacturer for advice on the minimum flow rate at which the mixer is still effective. The maximum flow rate will be limited by the acceptable pressure drop across the mixer, but a wide flow range tan be accommodated using a variable geometry mixer.

5.3.6 Powered mixers
Powered mixers tan be dynamic versions of the static mixer in which the mixer elements are driven from an external power source to introduce a high shearing action which effects the maximum dispersion of the contacting media. Alternatively, mixing may be achieved when a portion of the flow is reintroduced into the pipeline via a pump and high-velocity injector(s), upstream of the sampling probe.

5.4 Position of the sampling probe
5.4.1 If the contents of the pipeline have been adequately mixed the position of the sampling probe relative to the pipe wall is relatively unimportant. However, to avoid wall effects, which could influence the flow rate as well as the water concentration, the sample should be drawn from the centre of the pipe. In horizontal or vertical pipes the sampling probe should be positioned to withdraw from the shaded areas shown in figure 3.

5.4.2 Under some conditions, centrifugal forces occurring at bends may act to separate the dispersed phase from the bulk liquid. Therefore the exclusive use of Single bends to provide mixing upstream of the sampling location is not recommended.

5.4.3 The distance between the mixing device and the sampling probe should be sufficient to avoid the effects of swirl and asymmetry generated by the device but not so great as to lose the mixing effect provided by the device. Recommended distances are between one-half of a pipe diameter and 8 pipe diameters, depending on the type of mixing device being used.

5.5 Checking the location of the sampling probe
Whatever location is chosen for the sampling probe, or whatever additional form of mixing is employed, it is recommended that the concentration profile at the sampling location be checked. The test procedure for this check is described in clause 6. Other methods of checking may be used if they provide equal assurance that the sample taken will be representative.