Monitoring the operation of a reverse osmosis (RO) system is an essential requirement for minimizing problems that adversely affect water quality and reduce the useful life of the membranes.
Instrumentation plays an important role in providing the system operator with a means to continuously monitor the water quality and system operating performance. Without it, the operator is unaware of potential problems until they manifest as a deterioration in water quality or a decline in membrane performance.
Equipment manufacturers offer a wide assortment of options for outfitting the RO with the required monitoring equipment. This article discusses some of the essential monitoring instrumentation that should be part and parcel of every RO system. This includes:
Oxidation-reduction potential (ORP)
RO membranes are rated at a standard temperature of 77 F. Cold water is more dense than warm water, which reduces the flux (gallons per square foot per day) through the membrane. At temperatures above 77 F, the flux increases; at temperatures below 77 F, the flux decreases.
Some RO systems are equipped with feedwater pre-heaters to permit the maintenance of a constant temperature. More commonly, the temperature fluctuates within a narrow range. As the temperature varies, the permeate flow will increase or decrease proportionally. This makes it more difficult to determine if a decrease in permeate flow is a result of changes in feedwater temperature or something more severe such as fouling of the membrane surfaces. An RO system rated at 100 gpm when operated at 77 F will produce 80 gpm at 58 F, for example.
Membrane manufacturers publish tables of temperature correction factors that are used to normalize the permeate flow back to the standard 77 F design temperature. Multiplying the actual permeate flow at a given feedwater temperature times the temperature correction factor gives the theoretical permeate flow at standard operating temperature and pressure.
Additionally, if the water treatment system incorporates media filters and/or ion exchange softeners ahead of the RO for the removal of suspended solids and other membrane foulants, temperature instrumentation is recommended to allow the operators to make adjustments to the backwash flow rate to compensate for changes in water density at various temperatures. Most media filters require a 40 to 50% bed expansion to insure removal of filtered solids during backwash. Excessive bed expansion, however, may result in backwashing some of the media down the drain. Filter media and ion exchange suppliers publish charts and graphs that indicate the optimum backwash flow rate expressed as gallons per minute per square foot (gpm/ft2) versus water temperature. Adjustments should be made, as appropriate, to keep the filter or ion exchange media operating on this curve.
Industrial RO systems are generally designed to recover 75% of the feedwater as permeate. The remaining 25% of the feedwater is sent to drain as concentrated brine (also known as reject or waste). Under these operating conditions, the dissolved solids present in the feedwater are concentrated 4 times in the brine stream.
The RO is designed to operate reliably at a concentration ratio of 4:1 without excessive deposition of insoluble salts in the water passages and on the membrane surface. The feedwater quality is specified by the equipment manufacturer for these conditions. Appropriate pretreatment methods such as filtration, softening and/or chemical injection are employed to insure the continuous, reliable operation of the RO at 75% recovery. If the percent recovery is allowed to exceed 75% for an extended period of time, however, sparingly soluble salts and metal oxides may form deposits within the membrane modules. It is also true, that some RO systems may perform reliably at 80 to 85% recovery, if the quality of the feedwater is amenable to this. Since it is more efficient to operate at higher recovery, every effort should be made to maximize the percent recovery without causing problems with membrane fouling and scaling.
At a minimum, flow instrumentation should be installed on the feedwater, permeate and reject streams. The percent recovery is calculated by dividing the permeate flow by the feedwater flow. If the recovery rate exceeds the maximum limit of 75%, for example, the reject flow control valve should be adjusted to increase the flow to drain. Conversely, if the recovery is too low, the reject flow to drain should be decreased.
It is also advisable to record the totalized water consumption for both the feedwater and reject. This provides a record of the gallons of permeate produced by the system, which helps forecast the next membrane cleaning or replacement.
The performance of media filters upstream of the RO system are also affected by flow. Generally, two or more filters are installed, which operate in parallel. Individual flow meters should be installed on each filter to verify that the flow rates are balanced between the filters. If one filter accumulates solids, more flow will be diverted to the other filter(s). Likewise the flow through each filter should be monitored to guard against excessive rinse water flow rates that tend to compact the media and push material through the bed where it can find its way into the membrane pressure vessels and thereby foul the membranes.
Specific conductance measures the electrical conductivity of water, which is expressed as microSiemens per cm or micromhos per cm at 25 C. The higher the specific conductance, the greater the concentration of dissolved solids. This is a very simple test and is routinely used as a general indicator of water quality.
RO membranes are capable of rejecting 99% or more of the dissolved solids in the feedwater. The remaining solids pass through the membrane into the permeate.
Conductivity instrumentation provides a quick and easy method for determining the percent salt rejection and percent salt passage across the RO system. These measurements can be performed by inline instrumentation installed on the feedwater and permeate streams. The readings are displayed on the RO control panel and logged by the system controller. Additionally, hand-held or bench top conductivity meters are useful for testing grab samples from the RO and for comparing the results of inline instrumentation against a standardized lab meter.
The percent salt passage calculation is performed by dividing the permeate conductivity by the feedwater conductivity. The percent salt rejection is determined by subtracting the salt passage from 100%.
New RO systems typically operate at an overall 1% to 2% salt passage. Interestingly enough, the percent salt passage sometimes decreases as new membranes are “broken in.” This is thought to occur as a result of compaction of the membrane surface resulting in a reduction in pore size.
In any event, it is informative and useful to monitor the percent salt passage over time. As the membranes age, their performance is affected by physical and chemical changes. In some cases, the membranes are adversely affected by inorganic and organic foulants that accumulate on the membrane surface. In other cases, the membrane degrades such as in the case of chemical attack by chlorine and other oxidizing agents. An increase in salt passage is one indicator of the membrane condition. This information is also helpful in estimating the remaining useful life of the membrane modules. Further, documenting changes in the salt passage before and after cleaning is helpful in assessing the effectiveness of the cleaning method in restoring membrane functionality.
The driving force in a reverse osmosis system is the high pressure feedwater pump. The multi-stage RO pump typically operates at over 100 psig. This varies, of course, depending on the feedwater quality and RO design. The general trend is to incorporate variable frequency drive (VFD) into the RO design to improve energy efficiency.
As water flows through the water passages in the RO module, a pressure drop occurs between the high pressure feedwater stream and the brine (reject or wastewater) stream. If the water passages become fouled with inorganic scale, organic materials or suspended solids, the pressure differential increases. It is very common to experience this in RO operation as deposition and fouling of the membrane occurs over the life of the module.
At some point, the membranes must be cleaned or replaced to restore the pressure differential to acceptable levels. Membrane manufacturers publish guidelines for determining when an increase in pressure differential signals the need for cleaning. The general guideline is a 10% to 15% increase. For example, if the normalized pressure differential between the feedwater and reject streams is 40 psig at the time the RO is first placed into service, membrane cleaning should be considered when the pressure differential is in the 44 to 46 psig range.
Industrial RO systems are configured in arrays consisting of one or more stages. A two-stage array may consist, for example, of 4 pressure vessels in stage one followed by 2 pressure vessels in stage two. Pressure gauges should be installed between each stage of the RO array to permit the monitoring of the pressure drop across each stage. It’s possible that the second stage (and third stage, if so designed) will tend to foul more readily since the dissolved solids concentrate as the water flows from stage to stage.
A pressure gauge is also recommended on the permeate stream. The pressure drop is significant as the water permeates through the membrane from the feedwater side to the permeate side. The trans-membrane pressure differential is monitored as the net driving pressure. As the module ages, the resistance to flow through the membrane may increase giving rise to a corresponding increase in net driving pressure.
The pretreatment system ahead of the RO array includes a final 5 micron filter. This may be a cartridge or bag filter canister. In either case, the filter should be equipped with inlet and outlet pressure gauges for monitoring the pressure differential. Filter elements require changing when the differential is in the 15 to 20 psig range. Likewise, media and/or carbon filters should be monitored for pressure differential to determine when backwashing is required.
Depending on the local water quality, many RO systems require an adjustment in feedwater pH to minimize the formation of mineral scale deposits in the water passages. If this is the case, mineral acid (generally hydrochloric, since it forms a more soluble salt) is used. Accurate control over the feedwater pH is important to prevent over- or under- acidification . Inline pH meters and acid feed controllers are commonly used for this purpose. The inline probe senses the pH and sends a signal to the controller to increase or decrease the output of the acid metering pump. With such a control system, it is important to clean and calibrate the pH probe frequently to insure an accurate and reliable reading. Generally, weekly calibration is advised.
As an alternative to acid feed, ion exchange softening of the RO feedwater will prevent the formation of mineral scale deposits. This eliminates the need for pH monitoring and control.
Chlorine will attack thin-film composite membranes resulting in the degradation of the membrane and a corresponding loss of performance as indicated by an increase in salt passage. Membrane damage will occur if it is exposed to any level of chlorine and the effects will accumulate over time. Therefore, manufacturers insist on the complete removal of residual chlorine ahead of the RO system. This is accomplished by either activate carbon filtration or the chemical injection of sodium bisulfite.
Oxidation-reduction potential (ORP) instrumentation is a common method of monitoring and controlling the bisulfite dosage. However, this feed method should be monitored carefully as other factors beside chlorine residual affect the ORP reading. Sodium bisulfite, for example, tends to depress the feedwater pH, which causes the ORP reading to increase. This will signal the chemical feed system to increase the bisulfite pump output resulting in the overfeed of chemical. As with all chemical feed and control systems, diligent monitoring, calibration and control are required to insure reliable and accurate operation.
The installation of essential instrumentation in an RO system provides a measure of the reliability and operating performance of the system. Monitoring and recording the instrument readings provides data that is used to calculate important operating performance criteria such as percent recovery, percent salt passage, pressure differential, normalized permeate flow and net driving pressure. Comparison of this data against the performance metrics provides the operator with the information necessary to extend the intervals between membrane cleanings and/or replacement. This serves to save money and prolong the useful life of the plant equipment.