Saving Energy with Oil-Free Magnetic Bearing Centrifugal Chillers


It’s been more than a decade since oil-free magnetic bearing centrifugal compressors hit the HVAC market. With unheard-of part-load energy efficiency and zero oil-related maintenance, these ultra-quiet machines are totally sustainable because there is no oil to change the heat transfer rates in the heat exchangers. Now there are more than 35,000 of them out there logging over 55 million run-hours, and all of them have an initial cost premium. While the technology was targeted at the HVAC market, which is accustomed to centrifugal compression and larger loads, many lessons have been learned after a decade of seeing the good and the bad applications in other markets. When you apply the technology properly, there are opportunities to save big energy dollars across many sectors — from plastics and pharmaceuticals to foods, metals and manufacturing.

 

Hotel Chillers

Figure 1: Three Chillers Installed by The Arctic Chiller Group Atop a Major Hotel

 

Choose the Right Chiller Loading

During the initial years of market expansion, there were the anxious adopters and the skeptics. Too many applications were designed and negotiated around how the owner could afford magnetic technology with the least premium cost over traditional lubricated, mechanical bearing chillers. So, if a facility had 300 tons of load, a 300-ton chiller was often selected to meet cost versus value assessments, thereby minimizing the upfront premium. Capital cost rebates and financial incentives, such as demand-response, have been created by many public utilities to help bridge the cost gap and reduce overall kW usage across the grid.

Now engineers know that the real advantage of magnetic bearings lies at loads of 85 percent and below the facility load. So, for your 300-ton load, you should choose a 350- to 400-ton magnetic chiller and force it to run in its “sweet spot,” which will help it reach energy levels that lubricated chillers probably cannot achieve. Traditional lubricated chillers must use pressure and velocity to push oil through the entire system and keep it miscible with the refrigerant on its journey. Therefore, its kW/ton gets higher as the load goes down. The opposite occurs when there is no oil — the compressor may operate at minimum pressure ratio to just meet the demand and cool the compressor. Engineers now leverage these facts.

 

The Real Cost of Oil — It’s Not Just About Maintenance

There was and is a lot of excitement about reduced maintenance and the elimination of oil-related mechanical bearing rebuilds and service contracts to maintain warranties. But oil is much more than maintenance cost. Enter the National Institute of Standards and Technology (NIST), the Department of Energy (DOE), the Refrigeration Service Engineers Society (RSES), and Wolverine, each of which recently performed studies of the actual effect of the oils typically used in chillers with R134a refrigerant (the currently favored refrigerant for centrifugal chillers that meets government environmental mandates). The studies verify that oil indeed affects the U-Value inside the heat exchangers. It changes the bubble formation at the tube surface. Additionally, the effect is seen with very small amounts of oil and is linear in the negative effect versus oil concentration.

The newer studies1,2 validated an older American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) study showing that there was about an 8 percent U-Value loss after 4 to 5 years of operation (Figure 2). Important takeaways from these studies include:

  1. With oil typically found in chillers, properties that promote miscibility necessary to lubrication lead to reductions in heat transfer rates in the heat exchangers.
  2. Oil concentrations above 0.5 percent cause reductions in bubble size formation at the tube surface, which reduces heat transfer rates.
  3. The effect was highest between 1.3 percent and 3.5 percent. The negative effect increased in a linear fashion according with the oil percentage above 0.5 percent.

 

Figure 2: Energy Consumption of Various Chiller Types Over Time

Figure 2: Energy Consumption of Various Chiller Types Over Time (from ASHRAE)

 

Oil in Chillers Can Never Be Sustainable

Simple math for calculating the British Thermal Unit per Hour (BTUh) shows the inescapable truth:

BTUh = U-Value x Area of Tube Surface x Logarithmic Mean Temperature Difference (LMTD)

The BTUh is the demand on your chiller that you need to meet. The tube surface area is fixed, so you must change the LMTD between the chiller’s saturated suction temperature and the leaving chilled water. And, once oil has taken out about 8 percent in U-value, it takes more energy to overcome the loss. Worse yet, LMTD is, as its name indicates, logarithmic, so it takes larger changes to make a smaller effect. Any way you cut it, these modern studies show that oil costs you big money — not just in maintenance. The larger loss is heat transfer rate for the life of the machine. This is many times the premium cost of the oil-free chiller, and people are starting to understand that fact and see it in their operating costs.

 

New Approach for Magnetic Chillers Unlocks Even More Value

However, that’s only half the picture that has emerged from the oil-free phenomenon. Controls have become more sophisticated and extend beyond just the chiller — all the way to the cooling tower fan, the bypass valves, and the sensors at air handlers and processes. The magic of magnetic bearings includes the very low lift capabilities, the new controls, and the strategies that can unlock unprecedented energy savings. Keep in mind something key about centrifugal systems: no differential pressure, no lift — no lift, no capacity. The idea is to operate at the minimum size “pressure-ratio envelope.” When you are operating at these very low lift conditions, if you quickly try to increase the demand, there is nowhere for the compressor to go. So, a better approach was needed to operate safely where only a magnetic system could go.

 

Direct Chiller Control of the Cooling Tower Equipment

To reach these targets, you no longer approach chiller control based on return water temperature. The preferred approach for magnetic drives is to base chiller control on a floating Saturated Discharge Temperature (SDT), which also reacts faster than water temperature changes, and is now available as a floating set-point. By manipulating the SDT, the operating envelope can be squeezed downward as load changes, which saves energy at the sacrifice of available capacity. In many part-load situations, that is perfectly acceptable.

 

Figure 3: Air-Cooled Chillers Installed by The Arctic Chiller Group at a Hospital Campus

Figure 3: Air-Cooled Chillers Installed by The Arctic Chiller Group at a Hospital Campus

 

By extending this logic outwards and directly correlating this new settable refrigeration cycle SDT to the control of cooling tower set-point and fan speed, an ideal balance can be closely approximated, and you can operate at extremely low lift conditions safely. A controller should have an enthalpy input to provide automatic control of floating cooling tower set-point as it approaches wet-bulb temperature, and optimized fan speed — all controlled by the chiller. This approach enables unprecedented system energy efficiency by this direct tie between the refrigerant compression cycle and tower-side enthalpy performance in real time.

Remember, these are magnetic drives, so this approach gets us into very low and directly controlled lift and system pressure ratios under all load and ambient conditions. Because of centrifugal compression, during cold or inverted start conditions, it is necessary to provide direct control of an automatic condenser water bypass valve so that it will open — and remain open  — once the chillers have reached normal operating pressures.

 

Air-Cooled Chillers Save On More Than Maintenance

Two widely held beliefs have changed. One was that air-cooled chillers could not be as efficient as water-cooled chillers. The other was that due to the higher head pressure and condenser approach, you could never have air-cooled centrifugal chillers. While it is essentially true that water-cooled chillers can deliver better energy performance, proper loading of modern air-cooled chillers using magnetic compressors can result in superior simple energy costs and total cost of ownership. In Figure 4, notice that at 70 to 80 percent of load, the air-cooled chiller saves energy without a cooling tower, fans, condenser pumps or water treatment.

 

Figure 4: Air-Cooled vs. Water-Cooled Chiller Performance

Figure 4: Air-Cooled vs. Water-Cooled Chiller Performance

 

The secret is controlling refrigerant circuits individually with generous condenser surfaces, high-cfm-type electronically commutated (EC) fans, and refrigerant-to-refrigerant economizers to cool the compressor during higher load points and pressures typical for air-cooled chillers. By selecting an oversized chiller, or adding additional condenser/fan banks, a facility can meet its demand and save significantly in simple-energy and maintenance dollars for the life of the equipment.

 

Air-Cooled Chiller Free and Trim Cooling

There are two ways to do free and trim cooling. One is to install chilled water coils before the condensers in the same air stream, which means they share the same fan. When ambient temperatures drop, the condenser needs less and less fan energy, and the free cooling needs more. Therefore, there are two disadvantages to sharing the same fan and cabinet. One is you cannot do trim cooling as effectively while refrigerant circuits are also still running. The other is you always have the pressure drop of both condenser and free cooling coils — even when only running in one mode. The ideal configuration — if you have the space available — is to provide the free-cooling system with its own set of fans. It should also be piped in series with the chilled water return so that it can provide very effective trim cooling whenever the ambient temperature is low enough, eventually taking on 100 percent of the load (if designed for that).

A reasonable compromise is to equip air-cooled chillers with an individual refrigerant circuit for every compressor, with chilled water return coils sharing the same fan. As both the temperature and the load reduce, controls can dedicate each circuit and its respective fan to one duty or the other. Half of the chiller could perform free or trim cooling, and the other half could cool what’s left of the chilled water load with refrigeration.

 

Increase Savings with Chilled Water Temperature Resets

To squeeze energy out of the chilled-water side, facilities should enable multiple levels of chilled water set-point control. That way, as the load is reducing or processes are offline, the chilled water set-point can automatically rise by a degree or two at intervals that make sense for the air handlers or processes affected. Engineers often specify pressure-independent flow control valves to balance and widen the Delta T across loads, and vary the chilled water flow with the real load.

By selecting the right chiller loading, and resetting set-points strategically (by leveraging the floating SDT with a floating cooling tower set-point and fan speed as the temperature approaches wet bulb), you can achieve significant energy savings. These techniques are a culmination of best practices learned since the magnetic technology emerged.

 

Combined Cooling, Heating and Power

Co-generation is coming back with a strong natural gas supply at sustainable prices for the foreseeable future. DOE suggests that industrial applications for waste heat recovery and combined duty equipment will save billions of dollars over the coming years. Modern co-generation includes natural gas turbines and diesel generators with absorption chillers to capture the waste heat and run at very steady-state loads, while using magnetic chillers that can handle part loads and trim cooling very reliably. The magnetic chillers also provide redundancy to the absorption system. Figure 5 shows a complete system.

 

Figure 5: Co-Generation System with Natural Gas Turbines, Diesel Generators, Absorption Chillers and Magnetic Chillers

Figure 5: Co-Generation System with Natural Gas Turbines, Diesel Generators, Absorption Chillers and Magnetic Chillers

 

Emerging technologies, such as binary-cycle and organic Rankine cycle equipment, will significantly reduce waste heat by operating at lower temperatures and producing electricity directly. These systems are described below:

Binary Cycle Co-Generation Systems: These systems use lower temperature 300°F geothermal and industrial waste heat across heat exchangers and heat transfer fluids to heat a rapidly expanding gas to drive turbines. Thermal fluids or waste gas never contact the generator impellers (Figure 6).

Organic Rankin Cycle (ORC) Systems: This type of system can be described as a magnetic compressor operating in reverse. ORC uses low-grade heat or gas and generates economical DC electrical energy in a variable-speed, oil-free environment. ORC can use renewable sources like geothermal and solar.

 

Impellers

Figure 6: Sample Impellers from a Co-Generation System

Bridging the Financial Gap

While rebates help when available, they are not uniform across the markets. By doing the due diligence of correlating your facility load profile with the available ambient air temperatures, energy usage and rates with any cash rebates, owners can determine the amount of premium required to meet their payback objectives.

There are many energy service companies in the market that analyze the overall energy-related opportunity and provide up to 100 percent financing for solid energy-saving projects. Some specialized companies provide complete turnkey engineering, construction and life cycle support for energy-related systems and the assets.

Taking this one step further, there are businesses now emerging that sell the thermal effect of the equipment as a utility with the premise that the thermal utility can be acquired at less cost than purchasing, owning, maintaining and operating infrastructure equipment that is not considered core competency. Under this model, there is no capital purchase or lease — even the service is outsourced.

The technology and funding channels are here and available now. Modern approaches to magnetic compression and the correlated control of the chiller ecosystem have unlocked unmatched sustainable value for facility owners. After all, that’s what it is all about — sustainable owner value.

 

About the Arctic Chiller Group

With factories in the U.S. and Canada, The Arctic Chiller Group is a world leader in chillers and chilled water systems. Arctic manufactures ultra-high efficiency chillers using magnetic bearing oil-free compressors. The product range includes water-cooled chillers up to 1500 tons and air-cooled chillers up to 400 tons with trim and free-cooling options. These products provide facility owners with unmatched savings in energy, noise and total lifetime cost of ownership. The ArctiChill division is the world leader in modular, medical and process chiller systems with many features, options and owner benefits. Scroll, screw and oil-free centrifugal models are available with pumping, free-cooling and controls.

About the Author

Jackson Ball is Group Vice President and co-owner of Arctic Chiller Group. With more than 25 years experience in process and HVAC heat transfer, he is a well known evangelist of oil-free, low-energy cooling designs and leading edge control technologies. He is also a regular speaker at energy functions around North America.

 

For more information contact Jackson Ball, Group Vice President, The Arctic Chiller Group, tel: (678) 234-2821 or visit www.arcticchillergroup.com.

 

To read more about Chiller Technology, please visit www.coolingbestpractices.com/technology

 

References
1. Kedzierski, M.a. "The Effect of Lubricant Concentration, Miscibility, and Viscosity on R134a Pool Boiling." International Journal of Refrigeration 24.4 (2001): 348-66. Web.
2. Thome, John R., Dr. "Chapter 16: Effects of Oil on Thermal Performance of Heat Exchangers." Engineering Data Book III. N.p.: Wolverine Tube, 2004-2010. N.