Reverse Osmosis System Failure Analysis

Overview

Many factors can cause reverse osmosis (RO) equipment to malfunction. Like a doctor diagnosing a patient, you must gather as many symptoms as possible before proceeding with further analysis.

Introduction

This article focuses on troubleshooting RO equipment. We assume the equipment is designed with appropriate instruments and sampling points to meet troubleshooting and cleaning needs, a common design in industrial systems. However, this is not necessary for household or light industrial equipment. While the market price of instruments and sampling valves in small RO systems can be quite high, the cost of frequently replacing RO membrane elements is much lower. However, for larger RO systems (15 gallons/minute or more, i.e., 35 m³/hour or more), frequent replacement of RO membrane elements becomes significantly more expensive than adding instruments, sampling valves, and cleaning equipment to the initial investment cost.

The best way to avoid equipment malfunctions is through prevention from the outset. Here are some relevant RO design recommendations:

A complete water quality analysis should be included when designing an RO system. If there are seasonal variations in water quality (common in surface water) or changes in water source (common in municipal water supplies), obtain all available analytical data and ensure they are from the most recent materials.

Conduct a 15-minute SDI (Soil Density Index) test on-site to determine the likelihood of colloidal fouling.

If you want a good night's sleep, ensure that adequate pretreatment is included in the design of your RO system.

Allow margins in RO system design (especially in areas with potential fouling). A more aggressive design can be used for RO systems using clean well water as a feedstock than for surface water systems.

Conservative RO system designs should use lower flux rates, as reducing the amount of water produced per unit membrane area reduces contaminant buildup on the membrane. For systems using well water as a feedstock, the design flux should be between 8 and 14 gfd (plus/square foot/day). For well water systems, the design flux should be between 14 and 18 gfd. The recovery rate should be conservatively set to minimize contaminant concentration.

A conservative design should maximize both the feed crossflow velocity and the concentrate velocity. Higher crossflow velocities accelerate the diffusion of salts and contaminants from the membrane surface into the bulk solution, thus reducing their concentration.

Select the appropriate membrane element type for different applications. Sometimes, when treating difficult-to-treat surface water and industrial wastewater, electrically neutral CAB (cellulose acetate) membrane elements are superior to negatively charged CPA (polyamide composite) membrane elements.

Investigate if any reverse osmosis systems in your vicinity use the same feed water source.

Troubleshooting:

We recommend running "standardization" on your recorded operating data to identify patterns of system fouling, allowing you to schedule flushing and confirm system malfunctions. Membrane element suppliers have developed "standardization" software that can calculate standardized permeate flow rate, salt permeate flow rate, and feed-concentrate pressure drop. These standardized parameters are obtained by comparing key daily operating data, such as temperature, feed water TDS, recovery rate, and pressure, with the operating data from the first day, and adjusting accordingly based on changes. For example, if the standardized permeate flow rate on day 100 is 80 gpm (gallons per minute), while the first day's flow rate is 100 gpm (gallons per minute), it indicates that the membrane elements are fouled and have lost 20% of their permeate flow rate, thus recommending cleaning. If RO system parameters fluctuate, the wisest and most reasonable approach is to determine the true condition of the reverse osmosis system.

Was the RO system shutdown correct? When the system is running, the concentrate should be flushed out; otherwise, contaminants will deposit on the reverse osmosis membrane surface. It is best to use RO product water as the flushing source.

Was the RO system shut down for too long? If water remains stagnant for extended periods (especially in warm climates), it can cause serious microbial contamination problems.

If acid is added to lower the pH to control calcium carbonate (lime) scaling, have you adjusted it to the required pH level? Confirm that the pressure drop increase between feedwater and concentrate does not exceed 15%. Otherwise, it indicates feedwater contamination, limited membrane surface flow, and the system requires cleaning. Monitoring inter-section pressure drop helps determine which section is fouled, thus identifying potential contaminants.

Confirm that the pressure drop between feedwater and product water does not exceed 15%; otherwise, it indicates the reverse osmosis membrane surface is fouled and requires immediate cleaning.

Confirm that the increase in product water conductivity does not exceed 15%; otherwise, it indicates the reverse osmosis membrane surface is fouled and requires immediate cleaning. Confirm that all instruments are calibrated.

If possible, measure the product water quality of each section and each pressure vessel. Some contaminants will contaminate the first half of the system; others will contaminate the second half. The RO system fault analysis table (attached) can help identify the type of contaminant.

Sample from the product water line and measure the product water conductivity to check for O-ring damage. When sampling, insert a 1/4-inch plastic tube into the sampling tube and measure the insertion depth.

Check the pre-RO security filter for contaminants, as this is relatively easy to do.

Check the RO membrane elements for contaminants or damage.

Sample and analyze the RO feed water, concentrate, and product water, and compare the results with the design values provided by the membrane element manufacturer.

When the RO system malfunctions, if external damage can be ruled out, it's necessary to deduce the type of contaminant and perform one or more cleaning operations accordingly.

Collect the cleaning solution and analyze the removed contaminants, color changes, or pH changes. The cleaning effect can be verified when the RO system is restarted.

If you don't know what the contaminants are and don't want to conduct on-site experiments to select suitable cleaning solutions and methods, companies that supply specialized cleaning agents and provide off-site RO membrane element assessment services can offer this service. Such services are especially valuable during the initial RO cleaning.

If all the above methods for checking for contaminants in the RO membrane elements are ineffective, then membrane element dissection and analysis are necessary. In this case, the membrane element is disassembled, and the membrane surface and contaminants are analyzed and tested to determine the problem.