First published as an Ozwater'18 Conference Paper.
Logan City Council has combined emerging solar power and battery storage technologies to deliver a reliable, safe solution for water disinfection at a 20ML reservoir in the city’s fast growing south west. The solution, which includes an electrochlorinator powered by 323 solar panels and a 95kwh capacity Tesla Powerpack battery, is maintaining local drinking water quality 24 hours a day. The ‘micro-power grid’ and electro-chlorinator is an Australian first application. The project has also delivered significant capital and operational cost savings to Council, and is safe to operate.
Logan City is located between Brisbane and the Gold Coast, and is home to more than 313,000 people. The current population is growing by an average of almost 2% annually (.idcommunity, 2018). A key growth hotspot is the city’s south west which includes the Queensland Government-designated Priority Development Areas of Yarrabilba and Greater Flagstone. This area receives bulk water with chloramine as the residual disinfectant via a major interconnector main from Mt Crosby Water Treatment Plant. In 2014, the 20ML Round Mountain Reservoir was brought into service to provide drinking water for surrounding residents.
As the area will not be fully developed in the next decade, water age in the reservoir and network is currently longer than is desirable with low residual chloramine concentrations in the network due to nitrification. Council decided to investigate the best way to maintain an effective residual disinfectant and improve water quality in the Round Mountain Reservoir supply zone.
Council’s Logan Water Infrastructure Alliance (LoganWIA) was tasked with the challenge of finding a solution for water disinfection at the reservoir. The alliance is a long-term public and private sector enterprise involving Council, Downer, WSP and Cardno. A key consideration for LoganWIA’s planning and investigations team was that the site is not serviced by a sealed road or connected to mains power.
Round Mountain Reservoir has a short, sealed access road which was constructed at the same time as the reservoir. However, there is a 3.1km section of unsealed road (New Beith Road) between this and the wider sealed road network. Wet weather access along the unsealed section is limited and there is a history of Council vehicles becoming bogged. This has implications for specifying any new infrastructure at the reservoir site which may require regular vehicular access for operation and maintenance activities.
Council’s Logan South Operational Model was used to model and quantify network flows, storage changes, water age and water quality in the Round Mountain Reservoir water supply zone. The model used chloramine decay curves in the water supply zone measured over 12 months for current and future operating conditions. The modelling results confirmed earlier verification testing; showing that over summer there is very little chloramine residual persisting in the supply zone. The outputs from the model were used to quantify the frequency of operation of the chlorinator and estimate the capacity required to meet Council’s minimum residual chlorine targets.
Chlorine dosing at the reservoir complex could either be done in the unmixed reservoir or at the outlet. Dosing into the reservoir was assessed as having a high risk of inconsistent chlorination. Dosing at the outlet flowmeter chamber with the dose rate controlled by flow pacing and feed forward control (with feedback trim if required) was determined as the preferred dosing strategy. Outlet dosing is facilitated by the absence of customer connections within 6km of the dosing point.
There were four options considered for the type of chlorine dosing at Round Mountain Reservoir: sodium hypochlorite, electro-chlorination, gaseous chlorine and calcium hypochlorite. Chlorine gas was ruled out as a potential disinfectant at LoganWIA’s initial stakeholder workshop due to health and safety requirements, and operators’ lack of experience with chlorine gas systems. The remaining options are described below.
Sodium hypochlorite (NaOCl) dosing uses a 10% - 12% available free chlorine solution to disinfect drinking water. Sodium hypochlorite degrades when stored, producing disinfection by-products including chlorates. Degradation rate and by-product formation increases with temperature, making it inadvisable to store hypochlorite longer than 14 days in summer. Sodium hypochlorite dosing is a familiar technology for Council operators as several systems are installed in Logan City.
Electro-chlorination involves electrolysis of a brine solution to produce a low strength hypochlorite solution with 0.6% - 0.8% available chlorine. Electro-chlorination’s low hypochlorite strength reduces disinfection by-product formation but produces hydrogen gas as a waste product. This must be safely released to the atmosphere.
Disinfection using calcium hypochlorite briquettes (tablets) involves spraying the briquettes to form a chlorine solution. Calcium hypochlorite cannot be stored in sealed, unventilated warehouses for prolonged periods. Temperatures more than 32˚C will accelerate the loss of available chlorine from calcium hypochlorite. Maximum storage volumes apply as calcium hypochlorite is classified as an oxidiser and a dangerous good for transport purposes.
Within the three options described, a number of sub-options were identified. These were:
All regimes would be powered via photovoltaic solar panels on the roof of the reservoir, with energy stored in a battery power pack located inside the reservoir control room.
An inspection of the 3.1km unsealed section of New Beith Road was undertaken to determine rehabilitation works required to enable all weather access to Round Mountain Reservoir for materials delivery. The construction works proposed included a 150mm road base, a two-coat bituminous seal, three concrete floodways and two stormwater culverts.
Round Mountain Reservoir is not currently connected to mains power and it is anticipated that no mains power will be available for the ten-year design life adopted for the project. As such, LoganWIA determined that all power requirements had to be supplied using solar panels. These panels had to be of a sufficient size to run all site operations using the available hours of sunlight throughout the year. Sufficient battery backup was also required to ensure power was always available.
Each sub-option was evaluated to determine the most cost-effective, reliable solution for Council. For evaluation, the following assumptions were made:
The initial cost estimate for all proposed options are shown in Table 1. This estimate included allowances for detailed design, project management and delivery. The costings were based on quotes obtained from suppliers and unit rates were verified by LoganWIA’s cost estimator.
Based on cost criteria, Option 1B: sodium hypochlorite had the lowest upfront capital expenditure requirement. However, Options 2B and 2C presented the best overall Net Present Value (NPV) due to lower operational and maintenance costs over the ten-year design life of the system.
A non-cost multi-criteria assessment (MCA) was conducted with consideration given to reliability, water quality, incident risk, vulnerability, operational health and safety, operability and maintainability, constructability, and construction and operational impact on the environment and community. Figure 1 shows the MCA results as agreed by stakeholders at an MCA workshop. Based on non-cost criteria, Options 2B and 2C had the best overall scores.
The Chlorinator Options Study (Logan Water Infrastructure Alliance, 2016) concluded that the most reliable, safe and cost-effective option was an electro-chlorination regime with 28 days of brine storage. Council approved the new chlorination system for design and construction in 2016/17, and commissioning occurred in late 2017.
Key benefits of the solution included:
The preferred option for chlorination at Round Mountain Reservoir required development and implementation of a solar powered electrochlorination system with sufficient capacity to produce a minimum 32kg of chlorine per day to treat up to 12.7ML/d of chloraminated water. The chlorination system delivered consists of a power supply system, electro-chlorinator, water quality analysers and the chlorine dosing system. The capital cost of the solar powered electro-chlorination system was a key constraint on the design. Factors that had to be considered to find a balance between cost and reliability of the system included:
To produce 1kg of chlorine requires approximately 3.3kg of salt and 4.5kWh of electrical power (Black & Veatch Corporation, 2010). The electro-chlorinator must produce a total of 32kg of chlorine per day with the available solar power (minimum 144kWhr/day). In addition, the chlorination facility operates a chiller, recirculation pumps, dosing pumps, fans and lights. The average power demand for the facility is 198kWhr/day. Running the electro-chlorination system at full capacity is not necessary during winter months when cooler temperatures result in better chlorine levels. The critical period is late summer when water temperatures are high, chlorine levels are low and sunlight hours are shortening. A typical solar array output for March is shown in Figure 2. High power outputs are available in the middle of the day but outputs fall off sharply at the beginning and end of the day. It was decided to install two smaller capacity 11kW electro-chlorination units rather than one larger unit in order to maximise the hours of operation. This allowed the first unit to start up as soon as there was sufficient output from the solar array.
The only safe, secure location for the solar panels was on the reservoir roof. The roof support structure was not originally designed to accommodate solar panels and access walkways. However, a condition assessment found the roof support structure would adequately hold the panels. The number of panels was increased slightly to offset performance reductions due to panels being installed flat on the roof, and the panels were offset from the ideal north/south configuration. Increasing the number of panels was cheaper than fitting the panels on stands with the perfect orientation.
The solar panel array consists of 323 panels rated at 270W with 220 panels on the north face and 103 panels on the south face of the reservoir roof. The total installed capacity is 87kW with space to add extra panels if power requirements increase in future. The installation included walkways and edge protection to provide safe access for inspection and maintenance. Figure 3 shows the solar panel arrangement.
The original design concept considered lower capacity electro-chlorinators running 24 hours a day to produce the 32kg of chlorine required. This option was scrapped when the battery capacity needed to operate these high power devices was estimated. A more cost-effective approach was to produce sufficient chlorine during daylight hours using larger capacity electro-chlorinators. The chiller and electro chlorinators would not need to run at night and during cloudy weather. The full power demand of the site with two electro-chlorinators running for eight hours is about 198kWh per day. This reduces to approximately 26kWh per day during low power demand periods when only the water analysis and chlorine dosing equipment is operating. Battery storage was required to operate the facility on low power demand for up to three days with no feed in from the solar array due to overcast weather. A 95kWh Tesla Powerpack was installed to meet this demand.
Large capacity electro-chlorination facilities are uncommon in Australia and this is the first known installation to be used to treat a public water supply system. The initial request for proposals offered to the market asked for a 4kg/hr electro-chlorination unit to produce 32kg of chlorine using eight hours of sunlight per day. Further negotiations with vendors investigated the option of duty/duty arrangements to take advantage of solar power generation at the morning and evening limits of each day. Changing the configuration from one 4kg/hr electrolyser to two 2.2kg/hr electrolysers resulted in a 25% increase in the cost of the required electro-chlorination capacity. This option was preferred as it provided some redundancy when the electro-chlorination system was not being operated at full capacity.
The production of chlorine on site depends on bright sunlight. The chlorination system is therefore susceptible to failure if the volume of stored chlorine is depleted as a result of an extended period of low sunlight. A storage tank mass balance was undertaken to estimate the frequency of failure of various tank storage sizes using historical sunlight data obtained for the last 16 years. The analysis found that increasing the tank storage from three days of storage (12,000L) to five days of storage (20,000L) reduced the frequency of storage depletion from 5.3 events a year to 2.5 events per year.
Hydrogen is produced as a by-product of the electro-chlorination reaction. Being highly flammable, the gas makes the entire facility subject to compliance with Queensland’s hazardous area regulations. Electrical wiring and ventilation ducting of the installed electro-chlorination equipment had to be approved by the Queensland Government Accredited Hazardous Area Auditor.
Commissioning of the electro-chlorination facility started in September 2017. The testing and commissioning of individual components of the facility proceeded without complication as most of the equipment is commonly used in water treatment facilities. The only novel equipment was the solar power system, battery system and the electrolysers. The commissioning of the Tesla Powerpack was not straightforward because this was the first off-grid Tesla Powerpack installation. The Tesla engineers had to implement modifications to synchronise the Powerpack with the solar panel inverters in an environment without mains power.The Tesla Powerpack has operated successfully since commissioning.
The commissioning of the two electrolysers proceeded without significant difficulties. The electrolysers currently produce sodium hypochlorite at the manufacturer’s recommendation of 0.6% concentration. The trade-offs in salt and energy consumption to increase this concentration to 0.8% are currently being considered. The electrolysers use softened water, and the brine waste from the regeneration process is removed from site. The current operation of the water softeners results in a waste stream that needs collection more often than was originally designed.
The combination of solar panels and Tesla Powerpack is exceeding Council’s expectations in terms the reliability of the system. Since commissioning, the charged battery capacity has never dropped below 65kW despite overcast weather and it generally does not fall below 70kWh by the end of the night time draw down. There has been a slight degradation of maximum charge over time. However, this is expected and charging rates will be monitored over time to ensure that the performance remains within specifications. A standby generator for the system is unlikely to be required in the short to medium-term.
The two electrolysers (pictured in Figure 5) have been working to produce sufficient sodium hypochlorite to meet daily chlorine dosing requirements. The chlorination facility’s control system calls for the production of sodium hypochlorite as soon as the storage tank drops below 95% and the Tesla Powerpack has reached 75% of full charge. This normally takes place about 30 minutes after sunrise. The only time the electrolysers have worked at full capacity was during commissioning when they took five days to fill the chemical storage tank (which contains five days of stored sodium hypochlorite).
The success of the solar powered electro-chlorination system can be measured by the reliable availability of chlorine in the chemical storage tank. The mass balance analysis undertaken during design had predicted that there would be days of significant draw down on stored chemicals when overcast weather prevented the operation of the electrolysers. The analysis predicted that the storage would be depleted two or three times a year when running at full capacity. In practice, Council operators have found that the number of solar panels provides sufficient power to charge the battery even in poor weather. This allows at least one of the electrolysers to operate for a few hours each day, maintaining good levels of chlorine in the storage tank. The level in the tank has not fallen below 70% since commissioning (see Figure 4).
The total cost of delivery of this project was $3 million. This was significantly higher than the estimate of costs used for financial options analysis. The additional costs were identified in the design stage and included extra building, switchboard and project management items which applied to all options considered. However, it was found that the cost differential identified in the original analysis was still valid and the preferred option did not change.
Overall, the selected option represents sound value for money for Council as the capital cost of reconstructing 3.1km of New Beith Road was avoided, and annual operating costs were approximately $50,000 lower per year than the more conventional options.
Council has combined emerging solar power and battery storage technologies to deliver a reliable, safe solution for water disinfection at a 20ML reservoir in Logan City’s fast growing south west. The ‘micro-power grid’ and electro-chlorinator system is an Australian first application. The project has delivered capital and operational cost savings to Council and is safe to operate and maintain. The system has maintained chlorine dosing 24 hours a day since commissioning. This has resulted in chlorine residuals being detected for the first time in many parts of the Logan South water supply zone, significantly reducing risk to consumers.
Ozwater is the Australian Water Association's annual international water conference and exhibition which takes place in alternating cities each May. To find out more, visit the Ozwater website.