Apr-2019

Designing a sampling system

So you thought that delivering process samples was a simple task? The useful performance of an analyser system relies on how the sampling system is designed.

PARTHA S MONDAL and ARCHAN SINHA
Fluor Daniel India Pvt. Ltd

Article Summary

The main objective of a sampling system is to provide a representative sample to the analyser for which it is designed. Other key objectives of a sampling system include:
1. Extract a representative sample from the process line with respect to the measuring parameters
2. Transport the sample to the analyser location in a timely manner
3. Condition the process sample per the analyser need
4. Incorporate stream switching as per the requirement
5. Provision for analyser online calibration
6. Proper disposal of sample after analysis.

Components of the sample system should be selected in such a way that it is reliable and requires least maintenance. The total cost of the system should be economical, taking into account any mandatory components. And the system must be totally safe from the standpoint of operation and maintenance.

The representative sample
A representative sample is a small quantity of something that accurately reflects the larger entity. If the sample is not representative of the process, there is nothing an analyser can do to correct the situation, and the analytical data cannot be used for control purposes. The sample reaching the analytical sensor should be identical in composition to the process stream, or differ in composition in a known and predictable way. Known and acceptable differences may be due to the removal of certain components from the process stream by filtration, condensation and so on, or because of dilution, dosing or other planned sample treatment.

Sample compatibility
For ideal representative sampling, the process sample should be presented to the analyser in exactly the same condition as it exists in the process sample, to be modified in some way to ensure that it is compatible with the measurement technology employed. A typical example would be cooling a furnace effluent sample from over 700ºC to a little above ambient before passing it into the optical cell of an infrared analyser. Another example would be reducing the pressure from above 500 psig to about 50 psig to ensure that the measurement cell will not rupture. Any change in the sample to ensure compatibility must be carefully engineered to ensure that additional incompatibilities are not created.

Sample system response time
One of the major factors for a sample transportation system is to minimise the sample time delay to improve the response time. Response time can be divided into two main stages: one is the time delay in the sample transportation from tapping point to analyser shelter/cabinet; the second is the processing time for the analyser output to respond to any change in measurement. So to reduce the time delay we first need to see how close we can site the shelter/cabinet with respect to the tapping point. Next, the sample velocity can be improved by considering the correct tube size and system considerations like closed loop or open loop.

Also, conditioning components such as filters, coalescers and knockout pots add to the time delay and it is worthwhile eliminating or reducing the size of larger vessels and unpurged volumes where mixing can occur. The time delay that takes place inside the analyser depends on the technology that is being used. Each type of analyser has its own defined cycle time to process the sample. Sometimes we need a faster response from the analyser depending on the dynamics of the process. It is therefore essential that the application requirements are discussed with process in detail before any design work is started.

Reliability and cost
To maintain high reliability, the sample system should be designed in such a way as to minimise maintenance and to preserve or enhance the analytical information contained in the sample. But the fact is that over a period the reputation of a sample system is very poor in terms of reliability. The majority of problems in an analyser system (85-95%) are because of the sample system. One of the major reasons is that a lack of attention is given to the initial design. Much care is exercised in the selection of analysers, but the sample system often is an afterthought.

Safety
The safety of analyser systems is a subject of high importance and a number of aspects require consideration. All components of the system should be accessible for maintenance, preferably at ground level. Personnel should be protected from coming into contact with toxic or corrosive substances and very hot or very cold surfaces. Equipment must be rated for all possible pressures and temperatures and must meet the hazardous classification of the area.

Sampling system design
The following items need attention when designing any sample system:
•    Selection of the sample tap/probe
•    Transport system design
•    Design of sample conditioning system
•    Sample recovery system.

All of these play a significant role in achieving an efficient and reliable system as a whole. One needs to consider not only normal operating conditions, but upset conditions too. Protection against process upsets may involve specifying additional equipment in the sample system to overcome the situation.

Selection of the sample tap/probe
While selecting the sample tap and sample return point, several points need to be considered. Firstly, be sure that the process composition has not changed between the sample point and the point where the sample is going to be measured inside the analyser. Secondly, the process time lag between the unit operation being monitored and the sample point chosen must be acceptable. Sometimes a complex sample system may be necessary to achieve the desired response time. Next, the sample extracted must be representative. Finally, when considering the sample point, the distance to the analyser must be considered.

When a number of analysers are present in a plant it is economical to group all the analysers inside a shelter but this makes fixing the sample point location and run lengths even more difficult and a compromise must usually be sought.
Different types of sample taps and probes are described below.

Simple sample tap
This type of tap can be used when the process fluid is clean. The location of the sample tap is very important. Top placement can be used if the process fluid contains vapour and suspended solids. For liquids tapping from the side is recommended since this will provide the best protection from suspended solids, and from gas bubbles in the process liquid. This type of simple sample tap should never be installed at the bottom of the process line. This is inviting disaster since all types of particulate matter are likely to enter the sample system. For installing this kind of sample tap we have to keep in mind that the internal diameter of the line should be the same as the internal diameter of the nozzle, with the internal edge radiused to remove burrs. If the nozzle is to penetrate the pipe wall, it should be set so that its end is in line with the inside pipe wall. These precautions will help to reduce the accumulation of deposits which tend to build up at sharp edges in the sample path, and which might eventually block the tap. The isolation valve for the tap should be in accordance with the piping specification. A correct choice would be a gate or full-bore ball valve. Then, if a sample probe is subsequently found to be necessary, it can be inserted through the valve without difficulty. Following the valve, there should be a pipe tee with the connection opposite the probe capped off. This will allow the sample tap to be rodded out when required. A typical sample tap arrangement is shown in Figure 2.