Keys to Proper Water and Steam Sampling
In all industries requiring control of liquid and gaseous streams, the proper design of a sampling system is critical in order to produce a sample...
In all industries requiring control of liquid and gaseous streams, the proper design of a sampling system is critical in order to produce a sample that is representative of the sampled stream. This is specifically the case for utility and industrial steam generation where parts per billion (ppb) concentrations of impurities are controlled. Sample withdraw, transport, collection and handling are often major sources of errors (as high as 1,000%) that can lead to incorrect or unnecessary corrective actions (see Table 1).
A typical sampling system consists of an isokinetic sampling nozzle (see Figure 1), two isolation valves, sample tubing, a primary cooler/condenser, a secondary sample cooler, pressure reduction and total flow regulation valves, a distributor for individual analyzers and grab samples, a back pressure regulator, and a sample drain (see Figure 2). All wetted sampling system components should be made from stainless steel to prevent the components from corroding and contaminating the sample.
What Is Isokinetic Sampling?
Isokinetic sampling ensures that all phases (solid oxides and precipitates, liquid droplets, and vapor) of the sampled fluid enter the sampling nozzle with the same velocity vector (velocity and direction of flow) and the flow velocity into the nozzle is the same as the sampled stream velocity. The main reason isokinetic sampling is necessary is that the sampled stream is almost always a two-phase fluid (gas-liquid, gas-solid, liquid-solid) and the second phase has a very different chemical composition than the steam or water. In addition, the second phase (droplets or particles) has a different density and inertia than the primary phase (gas or liquid) and therefore wouldn’t be proportionally represented in a sample that’s not withdrawn isokinetically.
Locations where isokinetic sampling nozzles should be used include: feedwater, boiler downcomer, saturated steam, and main steam (or reheat steam).
The following must be considered when designing a sampling system (see Figure 2):
Figure 1. Weld-In Style EPRI Single Port Isokinetic Sampling Nozzle Source: Jonas Inc.
- Nozzle Installation Location - the preferred location is in long vertical sections of pipe, away from all flow disturbances (bend, valves, etc.). If a long vertical section isn’t available, the nozzle may be installed in a long horizontal section, provided the nozzle is installed on the top of the pipe between the 10 and 2 o’clock positions to keep the nozzle dry during inactive periods.
- Integrity of the Sampling Nozzle - design must consider the effects of vortex shedding on vibration and strength of the nozzle attachment to the pipe in order to prevent high stresses and potential failures.
- Isolation Valves - should be rated for the application temperature and pressure and provide a minimum change of cross-section between the bore of the sampling nozzle and the valve orifice.
- Sample Tubing between the Nozzle and Primary Sample Cooler - should be as short as possible (not longer than 20 ft) to minimize pressure drop and reduce possibility of impurity deposition in the sample tubing. The inner diameter (ID) of this sample tubing should be close to the nozzle bore size to minimize changes in cross-sectional area. The tubing should form an expansion coil after the isolation valves to allow for any movement or expansion of the pipe. Sharp radius bends should be avoided.
- Primary and Secondary (if Necessary) Sample Coolers - should be of a counterflow design and be sized to ensure adequate cooling capacity, with allowances for reduced heat transfer due to scaling.
- Sample Tubing after the Primary Sample Cooler - should slope downward and have a minimum number of bends. It should be sized so the sample flow velocity is 4 to 6 ft/s in order to minimize deposition in the sample lines and to reduce the time required to achieve equilibrium between impurities in the flowing sample and the tubing.
The length of sample tubing should be as short as possible (<200 ft) to limit both pressure drop and lag time from when the sample enters the sampling nozzle to when it reaches the analyzers. Sharp radius bends in the sample tubing should be avoided.
- Pressure Reduction Valve - used to reduce pressure and therefore control flow of a cooled sample in order to protect on-line instruments and plant personnel. For samples equal to or greater than 500 psig, the pressure reducer should be a rod-in-tube type orifice or capillary. For samples less than 500 psig, the pressure reducer should be a needle valve.
- Thermal Shut-Off Valve - protects personnel and downstream components by automatically interrupting sample flow when the sample temperature reaches a preset limit in the event of an insufficient amount or loss of cooling water or a fouled sample cooler.
- Pressure and Temperature Gauges and Flow Indicator - provides the operator with verification the system is working properly.
- Back Pressure Regulator - used to maintain a slight pressure (~ 20 psia) in the sample tubing before the grab sample location. This will ensure adequate flow to the on-line instruments.
- In-line Sample Filters - during commissioning, or any other time when high concentrations of corrosion products (iron, copper) are present in the sample, in-line sample filters should be installed to protect on-line instruments. These filters should be installed after the grab sampling line (see Figure 2) or must be removed when obtaining grab samples for iron and copper analysis.
- On-line Analyzers - the sample flow rate, temperature, and pressure must all be within the instrument manufacturer’s specifications.
- Booster Pumps - may be required for long sample lines (high pressure drop) or low-pressure samples (condensate).
- Pressure Drop - must be low enough to ensure that there is enough pressure to provide adequate flow velocity. A high pressure drop through the system could result in insufficient sample flow at the sample panel, or the deposition rate in the sample lines could be high, which could result in plugging of the sample line or a sample that is not representative of the conditions in the pipe.
- Grab Samples - should be obtained in accordance with ASTM D3370 and ASTM D4453. Samples to be used for the analysis of iron (ASTM D1068) or copper (ASTM D1688) should be preserved with nitric acid (HNO3) to a pH of 2 or less (~2 ml/L) immediately at the time of collection. Samples to be used for the analysis of total organic carbon (ASTM D2579) should be collected in a glass bottle with TFE fluorocarbon-lined or aluminum-lined cap and acidified to pH = 2 or less. If stored more than 24 hours, all samples should be kept refrigerated and analyzed within a week of sampling.
A meticulously performed analysis is of little value if a bad sample is used. The proper design of a sampling system is critical in order to produce a sample that is representative of the sampled stream in order to avoid incorrect or unnecessary corrective actions in relation to the cycle chemistry. IWW
For additional information on this subject, see:
- “Standard Practice for Sampling Steam,” ASTM D1066, 1997.
- “Standard Practices for Sampling Water from Closed Conduits,” ASTM D3370, 1995.
- “Standard Specification for Equipment for Sampling Water and Steam in Closed Conduits,” ASTM D1192, 1995.
- O. Jonas, “Development of a Steam Sampling System,” EPRI (TR-100196), Palo Alto, CA: December 1991.
- Guidelines Manual on Instrumentation and Control for Fossil Plant Chemistry, EPRI (CS-5164), Palo Alto, CA: April 1987.
- O. Jonas and J. Mancini, “Sampling Savvy,” Power Engineering, May 2005.
About the Authors: A registered professional engineer, Dr. Otakar Jonas, founder of Jonas Inc., of Wilmington, DE, has over 40 years of corrosion, steam generation, water chemistry, materials engineering, instrumentation, teaching, research and development, and troubleshooting experience. He’s active with ASTM, NACE and the International Association for Properties of Water and Steam, having played a significant role in related standards development for many years.
Joyce M. Mancini has been an engineer at Jonas Inc. for nine years and is experienced in nozzle and sampling system design. The company provides products, services and training for the water and steam chemistry issues affecting utility and industrial systems.
Table 1. Causes of Sampling Errors (in Order of Priority/Impact)
- Sample withdrawal...
- - not representing sampled stream (wall effects, stratification, not isokinetic, mixing, etc.)
- - not representing all phases (solid, liquid, gas)
- Deposition in sample line (could also result in plugging)
- Deposit release (leading to spikes)
- High pressure drop resulting in insufficient sample flow
- Changes in sample flow rate
- High sample temperature (causing pH and conductivity errors)
- Chemical reactions in sample lines or coolers (oxygen reduction, etc.)
- Deposition in the pressure reducing device
- Dissociation of water in the pressure reducing device - forming O2 + H2 (high pressure and temperature water)
- Leakage of cooling water into the sample - leaking sample cooler
- Corrosion of the sampling system - generation of corrosion products
- Filters in the sampling system interfere with sampling suspended solids
- Sorption on sample tubing and suspended oxides removes a portion of monitored chemical species
- Leakage of air into grab sample container increases O2 and conductivity, reduces pH, possible introduction of bacteria)
- Dissolution or precipitation and plating-out of metal species in the sample tubing and containers
Source: Jonas Inc.