Copyright This work is copyright. Apart from any use as permitted under the Copyright Act 1986 no part may be reproduced by any process without prior written permission from the Australian Institute of Refrigeration, Air Conditioning and Heating, Level 3, 1 Elizabeth Street, Melbourne, Victoria, 3000. The Australian Institute of Refrigeration Air Conditioning and Heating (Inc) 2017. Disclaimer The information or advice contained in these documents is intended for use only by professionals who have had adequate technical training in the field to which the documents relates. At the time of publication, not all papers had undergone a formal peer review process. These documents have been compiled as an aid only, and the information or advice should be verified before it is put to use. The user should also establish the applicability of the information or advice in relation to any specific circumstances. While the information or advice is believed to be correct, no responsibility is taken by AIRAH for any statements made within. The Australian Institute of Refrigeration, Air Conditioning and Heating (Inc), its officers, employees and agents, disclaim responsibility for any inaccuracies contained within the documents including those due to any negligence in the preparation and publication of the said documents.
WOOLWORTHS INVESTMENT IN THE AUSTRALIAN INDUSTRY TO DELIVER SUSTAINABLE HFC-FREE REFRIGERATION INNOVATION DARIO FERLIN Refrigeration & Sustainable Innovations Engineer, Woolworths dferlin@woolworths.com.au ABOUT THE AUTHOR Honours degree in Mechanical Engineering from Melbourne University. Entered the refrigeration industry in 2001 as a rack design engineer for EPTA (Italy) and collaborated in the first transcritical CO2 projects being piloted in Europe. In 2004 worked for EPTA as an applications engineer in the export department and oversaw turn-key projects for key supermarket accounts in developing countries. In 2009 relocated to Sydney to work for Woolworths; firstly in the capacity of a refrigeration engineer focusing on developing the showcase and refrigeration plant specifications and more recently in the role of innovations engineer. ABSTRACT In the last decade the Australian commercial refrigeration industry has been mired by uncertainty surrounding synthetic greenhouse gas regulation. However, the recent ratification by the Australian Government of the COP21 Paris agreement and the adoption of an 85% HFC phase down strategy by 2036 announced by the federal minister for the environment in June 2016 has provided the Australian industry with some urgently needed direction. The Australian commercial refrigeration industry is somewhat unique in that it is far removed from the natural refrigerants fermenting pot that is Europe and just two retailers command around 80% of the retail grocery pie. Furthermore, the existing technician base has some upskilling challenges which need to be addressed before meeting the demands of a HFC phase down market environment. It is therefore incumbent on the leading grocery retailers to become agents for innovation as well as drivers of a skills step change with local technology partners who will be deliver the refrigeration systems of the future. 1. WOOLWORTHS AND REFRIGERATION INNOVATION Woolworths has been very active in the refrigeration innovation space. In 2006 the retailer took the first uncertain steps away from commonplace R404a systems by opening its first cascade R134a/CO2 store (Bankstown NSW). At last count over 200 cascade R134a/CO2 stores have now been launched. However, the prospect of a HFC-free industry on the horizon has provided the impetus for a new cycle of innovation towards low-hfc and HFC-free refrigeration systems. To that end the initiative has been taken to pilot two promising technologies in 2017. Specifically a low- HFC WaterLoop store in Collins Square Melbourne and a HFC-free transcritical CO2 system at Colebee in Sydney. 1
At the core of the strategy is an investment in the Australian refrigeration industry. An investment which will generate momentum, spark imaginations and develop the local skillset which will be needed to design, build, commission and maintain these new technologies for the next +20 years. Therefore sustain the innovation beyond the store opening date. 2.1 The problem with HFCs 2. RISING TO THE CHALLENGE OF COP21 If we consider that an average Australian car emits 200g/km [1] and travels around 15,500km per year [2] the annual direct emissions footprint is around 3,100kg of CO2. Now consider that 1kg of HFC R404a refrigerant has a global warming potential equivalent to 3,920kg of CO2 [3]. If a medium size supermarket refrigeration system with 1,200kg of R404a loses a modest 10% of its charge to atmosphere per year via system leaks this equates to the same CO2 footprint as 152 average Australian cars (note this footprint does not consider the additional indirect emissions by drawing on fossil fuel sourced energy to run the refrigeration system). It therefore stands to reason that the COP21 agreement included a HFC phase down proposal as part of the response to the threat of climate change. 2.2 Low-GWP alternatives to HFCs HFCs make for excellent refrigerants. They are generally non-toxic, non-flammable and have operating pressures and physical properties which are very well suited to commercial refrigeration applications. Low-GWP alternatives to HFCs may share some of the physical properties of HFCs but are generally toxic and/or flammable. Retail environments are particularly sensitive to risk. Often the perception of risk is enough to discourage retailers from considering alternatives to common practice solutions. However two low-gwp refrigerants which pose no real or perceived risk to the retail environment include water and CO2. Both water and CO2 present challenges surrounding their physical properties. Water cannot be used in conventional vapour compression refrigeration systems however is a very good medium for transporting thermal energy. Additionally, above around 23 C ambient conditions CO2 stops behaving as a traditional HFC refrigerant and enters the transcritical zone which is essentially defined as a state where one cannot distinguish between gas phase and liquid phase (i.e. a very dense gas) hence condensation (or phase change) which is a characteristic of conventional refrigeration systems is not possible. 2
2.3 WaterLoop refrigeration systems WaterLoop systems employ existing technologies in new ways thereby drastically reducing the refrigerant charge of reticulated refrigeration systems. In a WaterLoop system, each refrigerated showcase and coolroom is serviced by a dedicated HFC condensing unit in a similar fashion to virtually any self-contained showcase or plug-in bottle cooler. The fundamental difference being the method employed for rejecting the heat of condensation. A conventional self-contained system will shed the heat of condensation to the surrounding ambient air via an air-cooled condenser (typically a ventilated finned coil) (refer Fig. 1). Fig. 1: Air-cooled condensing unit courtesy of Tecumseh Fig. 2: Water-cooled condensing unit courtesy of CoolPro On the other hand a water-cooled condensing unit will shed the heat of condensation to a body of water via a refrigerant-to-water heat exchanger (typically a tube-in tube or plate heat exchanger) (refer Fig. 2). The heat shed to the body of water is also ultimately shed to atmosphere typically via a cooling tower (refer Fig. 3), dry fluid cooler (refer Fig. 4) or chiller (refer Fig. 5). All three methods have advantages. Cooling towers and dry fluid coolers are HFC free but the water temperature supplied to the condensing units is dependent on outdoor ambient conditions. So when ambient temperatures rise so too do water temperatures which in turn reduce the efficiency of the condensing units and ultimately require larger condensing unit components. Furthermore cooling towers and dry fluid coolers typically have a large footprint which is not always convenient in a retail environment. Cooling towers also require ongoing maintenance to mitigate the risk of biological contamination (particularly legionella). 3
Fig. 3: Cooling tower courtesy of BAC Fig. 4: Dry fluid cooler courtesy of Alpha Laval Chillers on the other hand have a much smaller footprint than cooling towers and dry fluid coolers. Furthermore chillers provide water to the field at any temperature required by a system designer independently of outdoor ambient conditions. Chillers operate with an albeit limited charge of HFC refrigerant however this could substituted with a small charge of natural or low GWP refrigerant. Fig. 5: Chiller courtesy of Imas Klima 2.3.1 Direct CO2 emissions from WaterLoop systems A typical 3800sqm supermarket employing current best practice R134a/CO2 cascade refrigeration systems nominally features a 900kg charge of R134a (GWP 1,430 [3]) and 400kg charge of CO2 (GWP 1). Hence a system charge equivalent to 1,278 tonnes of CO2. 4
Now it is generally accepted that average leak rates in commercial refrigeration systems are in the vicinity of the following values: centralized supermarkets (multiplex): 15% per annum packaged chillers: 4% per annum integral cabinets: 2% per annum The above number point to a direct emissions impact for a typical 3800sqm Woolworths supermarket of 193 tonnes of CO2 per year. If we consider the same supermarket employing water-cooled R410a (GWP 2,090 [3]) condensing units to service the showcases and coolrooms and a chilled water reticulation (i.e. WaterLoop system) serviced by an R134a (GWP 1,430 [3]) chiller the system charge is now equivalent to around 200 tonnes of CO2. The direct emissions impact is reduced to around 4 tonnes of CO2 per year. Hence, notwithstanding the use of HFC refrigerants, the potential for direct carbon emissions abatement is enormous with a WaterLoop system even when compared to a current best practice system. 2.3.2 Leveraging the energy potential of water The thermal transport properties of water, it s abundance in nature (hence low cost), non-toxic and non-flammable characteristics and abundance of commercially available control devices offer the design engineer many opportunities to add value to a refrigeration system. All supermarkets require a degree of climate control for both the comfort of occupants and reliable operation of showcases and coolrooms. Hence space heating (delivered by the store mechanical services) for temperature and/or humidity control is a characteristic of virtually every Australian supermarket. This heat is generally provided by a gas-boiler or an air-sourced heat pump (older stores still feature electric duct heaters). All these systems could be considered energy intensive. The return leg of the WaterLoop reticulation which is transporting the heat of condensation of the water-cooled condensing units provides for a ready source of thermal energy which could be recovered for space heating. A water-to-water heat pump could, for example, be used to transfer thermal energy from the return leg of the WaterLoop reticulation (typically at around 23 C) and use this otherwise wasted energy to heat a space heating hydronic system to required temperatures (typically at around 45 C) for use in an air-handling unit. The energy required to drive such a system is significantly less than conventional space heating systems. Several chiller manufacturers now offer free-cooling options on standard model units. This option is essentially an on-board dry-fluid cooler whose purpose is to temper the return leg of a water reticulation before it is mechanically cooled by the chiller s vapour compression refrigeration system. Recalling that the return leg of the WaterLoop reticulation is warm the opportunity is available for tempering the flow via a free-cooling coil whenever the outdoor ambient temperature is below the return water temperature. If we consider that in Sydney the number of hours in a year that the outdoor ambient is below 20 C is in the order of 5,560hrs [4] (out of a total of 8,760hrs in the year hence 63% of the time) the potential for free-cooling for a WaterLoop system is significant. 5
Fig. 6: Concept drawing of a WaterLoop supermarket with free-cooling chiller courtesy of Carel 2.4 HFC-free transcritical CO 2 systems Transcritical CO2 systems are not new. The first commercial refrigeration systems were installed in Europe close to 20 years ago and by last count there are now over 8,700 installations [5]. The challenge with transcritical systems is implicit in the very name; when a system is running in transcritical mode it is no longer able to condense the compressor discharge gas into useful liquid. To obtain useful liquid, the discharge fluid from the gascooler (i.e the heat exchanger shedding the heat of rejection) the refrigerant flow must first be throttled via a high pressure valve. The throttling process creates useful liquid refrigerant and a proportion unuseful vapour (i.e. flash gas ). The diagram in Fig. 7 illustrates this process for a transcritical cycle operating in a 35 C ambient environment. A typical 3K approach temperature would yield a gas-cooler outlet temperature of 38 C. Furthermore, several published studies [6] point to an ideal gas-cooler pressure of 94bar at these temperatures. For a system operating with a -10 C evaporator pressure the proportion of unuseful vapour is around 50% of the system mass flow. 6
Fig. 7: Vapour fraction in a warm climate transcritical system The proportion of unuseful vapour increases as the outlet temperature of the gas-cooler increases. In other words the higher the outside ambient temperature the more unuseful vapour is produced. This vapour cannot be simply disposed of; vapour compression systems are closed circuit systems. Hence the vapour must be re-compressed and passed through the gas-cooler again (at the expense of energy) but to no refrigeration effect. In essence, transcritical systems become energy intensive and deliver a reduced refrigeration effect at high ambient conditions. Hence the low take up of these systems in warm climates. However, in low ambient environments, these systems are particularly efficient and very effective. Hence the high take up in colder climates. 2.4.1 The transcritical CO2 equator The refrigeration industry has coined the term transcritical CO2 equator to define a geographic line in both hemispheres of the world which define the threshold beyond which climates are considered too warm for transcritical CO2 systems to be viable. Innovation is however gradually closing the gap between these threshold lines and unlocking opportunities for what were once considered unlikely markets such as Australia. These innovations primarily centre around efficient management of unuseful vapour. Fig. 8 shows a modified cycle where the throttling process is undertaken to an intermediate pressure of around 35bar. Useful liquid refrigerant is supplied to the showcases and coolrooms at this pressure and the vapour is re-compressed by a dedicated compression group. This compression group is commonly referred to as parallel compression and works on a pressure differential from 35bar to 94bar. In the original cycle the unuseful vapour was re-compressed by the compressors operating at -10 C evaporator pressure (for CO2 this equates to 25bar) hence working with a pressure differential from 25bar to 94bar. The benefit of parallel compression with reduced compression ratio is immediately apparent and results in reduced energy consumption. 7
Fig. 8: Parallel compression Furthermore, the parallel compressors become redundant and cycle off at lower ambient temperatures when the system is operating in a conventional sub-critical cycle. A more recent innovation which is gaining industry traction in the management of unuseful vapour is the ejector (refer Fig. 9). Ejectors use the Venturi effect of a converging-diverging nozzle to create a low pressure zone that draws in and entrains a suction fluid. After passing through the throat of the ejector, the mixed fluid expands and the velocity is reduced which results in recompressing the mixed fluids. Fig. 9: Ejector In a transcritical CO2 system there is an ample supply of high pressure fluid which can be used as the motive fluid in an ejector. In fact ejectors have been successfully used not only in lieu of parallel compressors (at no energy cost) but also in lieu of liquid refrigerant pumps to enable safe operation of evaporator liquid overfeed and therefore hence driving system efficiency. 8
3. CONCLUSIONS The prospect of an Australian refrigeration industry whose supply of HFC refrigerants will begin dwindling as of 2018 is very real and will quickly drive up the price point of installing and maintaining HFC systems. There are therefore real business drivers as well as the often stated environmental drivers for retailers to start seriously considering low-hfc and HFCfree systems. WaterLoop systems afford retailers with a very low-hfc solution which is resilient to warm environments. Transcritical CO2 systems provide retailers with a HFC-free solution. Both systems have scope for integration with mechanical services. These non-conventional systems have already been successfully implemented in the Australian market. However their non-conventional nature presents challenges to the local industry which has had little or no exposure to these developing technologies. The onus is therefore on the local industry to invest in these non-conventional systems and above all invest in developing the human resources to meet the technological challenges of COP21. 4. NOMENCLATURE COP21-21st Conference of the Parties of the UNFCCC (United Nations Framework Convention on Climate Change) in Paris HFC - Hydrofluorocarbons are organic compounds commonly used as refrigerants in air conditioning and refrigeration systems GWP - Global Warming Potential 5. REFERENCES [1] www.drive.com.au/motor-news/australian-cars-co2-villains-report-20130313-2g1oq [2] www.roymorgan.com/findings/australian-moterists-drive-average-15530km- 201305090702 [3] IPCC (Intergovernmental Panel on Climate Change) Fourth Assessment Report (AR4) [4] Australian Bureau of Meteorology (BOM) data for Sydney 2009 [5] Shecco publication - Accelerate Australia 2016 Spring edition [6] Vollmer "Paper for Refrigeration Engineering Research Council" 1996, Liao & Jakobsen "Optimal heat rejection pressure in transcritical CO2 AC & heat pump systems" 1998 and Ge & Tassou (20011b) - extract from Brunel Uni thesis by I Dewa Made Cipta Santosa 9