Microfiltration and ultrafiltration membrane (UF/MF) fibers and modules have improved substantially over the last few years, and several new brands have become available. Each manufacturer touts unique membrane characteristics, hydraulic properties, fiber durability, warranty support, resistance to fouling, and operational flexibility.
The abundance of competitive brands on the market has driven costs down, even as quality improves. These factors have caused several water plant owners to consider retrofitting their plant when it comes time to replace part or all of their old membrane system modules.
Membrane retrofits fall into three main categories:
- Direct module replacement
- Integrating new modules into existing framework, with minimal plumbing and controls changes
- Major equipment replacement, including membrane module racks and supporting equipment, as needed
Direct module replacements are typically the fastest way to retrofit a system, but aren’t always in the owner’s best interest. Options 2 and 3 above require a preliminary design assessment performed by an equipment manufacturer and an engineer to determine the feasibility and economics of the retrofit. Assuming the economics work, a comprehensive design process is followed.
Considering the Economics: Retrofit vs. Maintaining the Existing System
The costs to perform the retrofit (equipment, construction labor, engineering, system downtime) need to be closely scrutinized and compared to the costs of maintaining the existing system. Maintaining an existing system could mean more frequent module replacements with outdated membranes, high operating and maintenance costs, and more frequent downtime for maintenance or replacement.
The following table includes some very simple figures that an owner used to assist with the decision to retrofit their plant. At this particular plant, they were replacing their original modules every 3-4 years, and expect to only have to replace the new brand every 10 years.
|Year||Original Brand||Cost||Retrofit Option||Cost|
|0||Replace half of existing modules||$172,560||Change plant over to new brand||$347,991|
|4||Replace full set of existing modules||$345,120||n/a||n/a|
|8||Replace half of existing modules||$172,560||n/a||n/a|
|10||n/a||n/a||Replace full set of new modules||$110,000|
This table shows the capital costs of the modules and the equipment retrofit, but doesn’t include the pinning labor, cleaning chemical costs, or plant downtime associated with the original plant, nor does it include any contractor or engineering fees associated with the retrofit. All of these factors must be taken into consideration, along with capital costs.
These economic analyses are performed during the preliminary design assessment, and several different brands may be evaluated during this step.
The early stages of design typically consist of the following activities:
- Listing module options, and each brand’s impact on the ability to reuse existing components
- Determining the available footprint and height for the module skid, transition skid (if required), and any other process equipment that may change
- Assessing the costs of the proposed options, and comparing it to the costs of retaining the existing brand
During the preliminary design, the first step is typically creating a short-list of possible module candidates. The list may be based on the owner, engineer, or equipment manufacturer’s preferences, based on experience and relationships with module manufacturers. The list could also be based on successful installations or pilot studies in the area that have demonstrated the effectiveness of that particular brand.
When evaluating the different low-pressure hollow-fiber membrane module brands typically seen in municipal drinking water and industrial water treatment plants, keep in mind that even though most of the brands are converging on a common hydraulic flow path and shape, there are still several physical operational differences that make each module. The following table shows parameters that can be compared at this stage in the assessment process.
|Fiber properties||Surface area||m2 / ft2|
|Fiber outer diameter||mm|
|Fiber inner diameter||mm|
|Fiber wall thickness||mm|
|Operations||Typical flux range||gfd|
|Max chlorine tolerance||ppm|
|Air scour rate||scfm|
|Max inlet module pressure||bar / psi|
|Max operating transmembrane pressure||bar / psi|
|Physical properties & connections||Module dry / wet weight||kg|
Feed / filtrate / backwash
The module selection process includes collecting design projections, warranty terms, and costs from each brand of interest. The module manufacturer’s application engineers will assist with the preliminary design, as far as determining suggested design flux and the resulting number of modules installed.
Once the number of modules is determined, the equipment manufacturer typically takes the lead on sketching out the full process, determining how to retrofit the module skids, and determining which process equipment can be reused and which must be replaced.
The hydraulic process for the new modules may be substantially different than the previous modules, which would impact the design of the retrofit skids.
In some situations, the feed, filtrate, and backwash waste headers are in such drastically different locations (e.g. Figure 1), that a transition skid is required to redirect the feed, filtrate, or backwash supply piping from top to bottom or to split any of that piping into multiple branches. This transition skid needs to be as compact as possible – since it adds to the overall length of the skid – and the additional fabrication and assembly costs must be integrated into the retrofit pricing.
Once the module options are identified, the equipment changes are listed out, and the costs are gathered, the owner can review the total capital costs of the retrofit and compare against maintaining their existing plant. In addition to capital costs, the owner must consider any engineer fees that may be incurred, any costs associated with securing approval from the state, construction costs, and piloting costs – if required.
Even with all of these factors considered, the economics and benefits of retrofitting the plant frequently work out in the owner’s favor.
This is also frequently the point where a pilot study would be performed, especially if the performance of different module options needs to be evaluated. The goals of the pilot study would be to confirm the flux rate and other operational set-points for the different module brands, confirm projections on cleaning cycle frequency and chemical consumption, and secure approval from the state if required.
If the water is challenging to process, the results of the pilot study can be very important in selecting the proper membrane module.
Detailed Design Process
Once the preliminary design phase has been concluded and the owner has secured funding to retrofit the plant, the detailed design phase begins. This phase involves a coordinated effort between the engineer and the equipment manufacturer, as every detail of the process is evaluated and re-designed as necessary.
The following pieces of equipment must be scrutinized closely, to determine if they are sized appropriately to support the new modules:
- Module skids (typically replaced, but may just be modified)
- Feed pump
- Feed strainer
- Backwash pump
- Air compressor or blower
- Chemical pumps
- Clean-in-place equipment (heater, CIP pump, supply and return connections, instrumentation)
- Skid and plant instrumentation
- PLC and HMI panels, skid control panels, and other SCADA equipment
Each component is carefully evaluated and modified/replaced as necessary. Process differences between the different brands may not only mean upsizing the equipment, but also upsizing the interconnecting piping and possibly the power feed to the equipment. Also, the retrofit is an opportune time for the owner to replace outdated or poorly-performing equipment, and make other system or process upgrades in their plant.
Most retrofits will also involve an overhaul of the PLC and HMI programs to account for the different process steps for the new brand of modules. Even though the module brands are converging on a similar shape and similar fiber properties, the operational sequences that each brand uses can be fairly different and requires a competent programmer to modify or replace the program to accommodate the new processes.
During this phase, the engineer, owner, and equipment manufacturer must communicate very well to ensure a successful retrofit. Minimizing system downtime is critical on these projects, so all details of the retrofit must be carefully thought out and detailed, to minimize errors and surprises.
When all factors have been accounted for properly and all equipment and materials are available for the retrofit, the actual retrofit can transpire rapidly. Experience has shown that the retrofit system can be back online within 24-48 hours of the shutdown.
Life Cycle Cost Benefits Often Outweigh Retrofit Costs
Recent trends and advancements in UF/MF membrane materials, configuration, and process design have improved the operational range and performance flexibility of low pressure membrane systems. Additionally, there are more high quality membrane modules available in the market than ever before.
With more options for high quality membrane modules available, a systematic process is required in selecting which membrane is the best fit for a specific retrofit application. Membrane system retrofits are carried out to address expansion capacity in the existing system framework, technical obsolescence of original equipment, high pricing of proprietary module replacements, or unsatisfactory membrane performance.
Although retrofit costs can be significant, ranging from 20-50% of original equipment capital cost, there are often life cycle cost benefits associated with achieving higher production rates and/or decreasing operations costs through reduced fiber breakage and fouling.
Dye, J. Nay, E. Linton, (2017) The Art of Retrofitting UF/MF Systems: A Comparison of Strategies, Costs, and Results, Proceedings of the AWWA/AMTA Membrane Technology Conference & Expo, Long Beach, CA
Nay, E. Linton, A. Richard, J. Berryhill, D. Dye (2017) Retrofit and Expansion of a 10 MGD UF System in Granbury, Texas, Proceedings of the AWWA/AMTA Membrane Technology Conference & Expo, Long Beach, CA
Nay J, Linton L, and Housley L. (2015) Side-by-Side UF Membrane Comparison and Retrofit of a 3 MGD System, AWWA AMTA Membrane Technology Conference, Orlando, FL
Vickers J, Zylstra D, and Owens E. (2016) West Basin’s Universal Membrane System – Pressurized PVDF Performance Pilot Program Particulars, AWWA AMTA Membrane Technology Conference, San Antonio, TX