The cooling water system for any heat engine test facility has to provide water of suitable quality, temperature, and pressure to allow sufficient volume to pass through the equipment in order for it to have adequate cooling capacity. The water pressure and flow have to be sufficiently constant to enable the devices supplied to maintain control of temperature of engine and transmission fluids. It is essential for purchasers of water-cooled plant to carefully check the inlet water temperature specified for the required performance, since the higher the facility cooling water inlet temperature supplied by the factory, the less work the device will be capable of performing before the maximum allowable exit temperature is reached. In this book the water supplied by the test facility is considered as theprimarycircuit, while the fluid flowing through the unit under test (oil, engine coolant, etc.) are the secondary circuits. There are several different terms used in the English language for the same primary circuit fluid, including mains water, plant water, factory water, and raw water. Most, but not all, automatically controlled heat exchangers in the powertrain test industry have valves controlling the flow of the secondary fluid through a heat exchanger that is fed by a constant flow from the primary circuit, as is shown later in this chapter.
Water
Water is an ideal solvent that is only met in its chemically pure form in laboratories. Unless stated otherwise, this chapter refers to water as the treated, potable liquid generally available in city supplies within a developed industrial country. Water is the ideal liquid cooling medium. Its specific heat is higher than that of any other liquid, roughly twice that of liquid hydrocarbons.
It is of low viscosity, widely available and, provided it does not contain sufficient quantities of dissolved salts that make it aggressive to some metals, relatively noncorrosive. The specific heat of water is usually taken as Cẳ4.1868 kJ/kg$K. (Note: This is the value of the “international steam table calorie” and corresponds to the specific heat at 14C. The specific heat of water is very slightly higher at each end of the liquid phase range: 4.21 kJ/
kg$K at 0C and at 95C, but these variations may be neglected in general industrial use.)
The use of antifreeze (ethylene glycol, H2H6O2) as an additive to water permits its operation as a coolant over a wider range of temperatures. A 50% by volume solution of ethylene glycol in water permits operation down to33C.
Ethylene glycol also raises the boiling point of the coolant such that a 50%
solution will operate at a temperature of 135C with pressurization of 1.5 bar.
The specific heat of ethylene glycol is about 2.28 kJ/kg$K and, since its density is 1.128 kg/l, the specific heat of a 50% by volume solution is:
ð0:54:1868ị ỵ ð0:52:281:128ị ẳ 3:38 kJ=kg$K
or 80% of that of water alone. Thus, the circulation rate must be increased by 25% for the same heat transfer rate and temperature rise. The relation between flow rate,qw(liters per hour), temperature rise,DT, and heat transferred to the water is:
4:1868qwDT ẳ 3600H qwDT ẳ 860H
whereHẳheat transfer rate in kW (to absorb 1 kW with a temperature rise of 10C, the required flow rate is thus 86 l/h).
Required Flow Rates
In the absence of a specific requirement it is good practice, for design purposes, to limit the temperature rise of the cooling medium through the engine water jacket to about 10C. In the case of the dynamometer the flow rate is deter- mined by the maximum permissible cooling water outlet temperature, since it is important to avoid the deposition of scale (temporary hardness) on the internal surfaces of the machine or localized boiling.
Eddy-current dynamometers, in which the heat to be removed is transferred through the loss plates, are more sensitive in this respect than hydraulic machines, in which heat is generated directly within the cooling water.
Maximum discharge (leaving) temperatures are:
Eddy current machines 60C Hydraulic dynamometers 65–70C,1
provided carbonate hardness of water does not exceed 50 mg CaO/liter. For greater hardness values, limit temperatures to 50C.
Approximate cooling loads per kilowatt of engine power output are shown inTable 7.1. Corresponding flow rates and temperature rises are as shown in Table 7.2.
Water Quality
At an early stage in planning a new test facility, it is essential to ensure that a sufficient supply of water of appropriate quality can be made available.
Control of water quality, which includes the suppression of bacteria, algae, and slime, is a complex matter and it is advisable to consult a water treatment expert who is aware of local conditions. If the available water is not of suitable quality then the project must include the provision of a water treatment plant.
Most dynamometer manufacturers publish tables, prepared by a water
1. Approaching this absolute maximum outlet temperature range, some machines can experience flash boiling, which can lead to a degrading of control and be heard as a distinctive “crackling”
noise. Running at these extreme temperature conditions will cause cavitation damage to the working chamber of a dynamometer.
Chapter | 7 Test Cell Cooling Water and Exhaust Gas Systems 153
chemist, which specify the water quality required for their machines. The following paragraphs are intended for the guidance of nonspecialists in the subject.
Solids in Water
Circulating water should be as free as possible from solid impurities. If water is to be taken from a river or other natural source it should be strained and filtered before entering the system. Raw surface water usually has significant turbidity caused by minute clay or silt particles that are ionized and may only be removed by specialized treatments (coagulation and flocculation). Other
TABLE 7.1 Estimated Cooling Loads
Heat Source Output (kW/kW)
Automotive gasoline engine, water jacket
0.9
Automotive diesel engine, water jacket
0.7
Medium-speed marine diesel engine
0.4
Automotive engine oil cooler
0.1
Hydraulic or eddy-current dynamometer
0.95
TABLE 7.2 Estimated Cooling Water Flow Rates
Heat Source In (C) Out (C) l(kWh)
Automotive gasoline engine 70 80 75
Automotive diesel engine 70 80 60
Medium-speed marine diesel engine
70 80 35
Automotive oil cooler 70 80 5
Hydraulic dynamometer 20 68 20
Eddy-current dynamometer 20 60 20
sources of impurities include drainage of dirty surface water into the sump, windblown sand entering cooling towers, and casting sand from engine water jackets. Hydraulic dynamometers are sensitive to abrasive particles and accepted figures for the permissible level of suspended solids are in the range 2–5 mg/liter. Seawater or estuarine water is used for testing large marine prime- movers with dynamometers and heat exchangers fitted with internals made from special stainless steels and marine bronzes, but it is not to be recom- mended for standard automotive equipment.
Water Hardness
The hardness of water is a complex property. There is a general subjective understanding of the term related to the ease of which soap lather can be created in a water sample, but the quality is not easy to measure objectively. Hard water, if its temperature exceeds about 70 C, may deposit calcium carbide
“scale”, which can be very destructive to all types of dynamometer and heat exchanger. A scale deposit greatly interferes with heat transfer and commonly breaks off into the water flow, when it can jam control valves and block passages. Soft water may have characteristics that cause corrosion, so very soft water is not ideal either. Essentially, hardness is due to the presence of divalent cations, usually calcium or magnesium, in the water. When a sample of water contains more than 120 mg of these ions per liter, expressed in terms of calcium carbonate, CaCO3, it is generally classified as a hard water. There are several national scales for expressing hardness, but at present no internationally agreed scale:
American and British: 1USẳ1UKẳ1 mg CaCO3per kg waterẳ1 ppm CaCO3
French: 1 Fẳ10 mg CaCO3per liter water German: 1 Gẳ10 mg CaCO3per liter water
1 dHẳ10 mg CaO per liter waterẳ1.25English hardness
(the old British system, 1 Clarke degreeẳ1 grain per Imperial gallonẳ 14.25 ppm CaCO3).
Requirements for dynamometers are usually specified as within the range 2–5 Clarke degrees (30–70 ppm CaCO3).
Water may be either acid or alkaline/basic. Water molecules, HOH or H2O, have the ability to dissociate, or ionize, very slightly. In a perfectly neutral water equal concentrations of Hþ and OH are present. The pH value is a measure of the hydrogen ion concentration: its value is important in almost all phases of water treatment, including biological treatments. Acid water has a pH value of less than 7.07 and most dynamometer manufacturers call for a pH value in the range 7–9; the ideal is within the range 8–8.4.
The preparation of a full specification of the chemical and biological properties of a given water supply is a complex matter. Many Chapter | 7 Test Cell Cooling Water and Exhaust Gas Systems 155
compoundsdphosphates, sulfates, sodium chloride, and carbonic anhydridedall contribute to the nature of the water, the anhydrides in partic- ular being a source of dissolved oxygen that may make it aggressively corro- sive. This can lead to such problems as the severe roughening of the loss plate passages in eddy-current dynamometers, which can cause failures due to local water starvation leading to plate distortion. Note that the narrow passages in eddy-current dynamometer loss plates are particularly liable to blockage arising from the use of inappropriate chemicals used in some water treatment regimes. Water treatment specifications should include the fact that, if used with water brakes, the treated water will be subjected to highly centrifugal regimes and local heating that may cause some degrading of the solution.
Control of water quality also includes the suppression of bacterial infections, algae, and slime. British Standard 4959 [1] describes the additives used to prevent corrosion and scale formation, with chemical tests for the control of their concentration, and gives guidance on the maintenance and cleaning of cooling water systems. A recirculating system should include a small bleed-off to drain, to prevent deterioration of the water by concentration of undesirable compounds. A bleed rate of about 1% of system capacity per day should be adequate. If no bleed-off is included the entire system should be periodically drained, cleaned out, and refilled with fresh water. Finally, consideration should be given at the design stage of a cooling water system to the consequences of a power failure. Consider, for example, the possible effects of a sudden failure in the water supply to a hydraulic dynamometer absorbing 10 MW from a marine propulsion diesel engine operating at full speed. Even when the shutdown system operates immediately the fault is detected the engine system will take some time to come to rest, during which the brake will be operating on a mixture of air and water vapor, with the strong possibility of serious over- heating. Therefore, in the case of any large engine test facility, some provision for a gravity feed of water in the event of a sudden power failure is advisable.
The supply pressure to hydraulic dynamometers should be stable or the control of the machine will be affected. This implies that the supply pump must be of adequate capacity, and having as flat as possible pressure–volume characteristic in the normal operating range.
Types of Test Cell Cooling Water Circuits
Test cell cooling water circuits may be classified as follows, with increasing levels of complexity:
1. Direct mains water supplied systems containing a portable dynamometer and cooling column that allow heated water to run to waste.
2. Sump or tank-stored water systems that are “open”, meaning at some point in the circuit water runs back, under gravity and at atmospheric pres- sure, into the sump by way of an open pipe. These systems normally incor- porate self-regulating water/fluid cooling modules for closed engine
cooling systems filled with special coolant/water mix and, if required, for oil temperature control. They commonly have secondary pumps to circu- late water from the sump through evaporative cooling towers when required.
3. Closed pumped circuits with an expansion, pressurization, and make-up units in the circuit. Such systems have become the most common as most modern temperature control devices and eddy-current or electrical dyna- mometers, unlike water brakes, do not require gravitational discharge.
Closed water cooling systems are less prone to environmental problems such as the risk of Legionnaires’ disease.
4. Chilled water systems (those supplying water below ambient) are almost always closed, although some may contain an unpressurized, cold, buffer tank.
Direct Mains Water to Waste Cooling
These are systems where the plant water is directly involved in the heat extraction from the process rather than from an intermediate heat exchanger. In most cases the water runs to waste on exiting the process and is most commonly used in occasional engine testing benches when portable dynamometers and cooling columns (see below) are used. While must of these systems run the water to waste it is possible to circulate the water through a gravity feed sump.
One drawback of these systems is that the coolant is mains supply water and therefore cannot be dosed with any additives.
“Open” Plant Water-Cooling Circuits
The essential features of these systems are that they store water in a sump lying below floor level from which it is pumped through the various heat exchangers and a cooling tower circulation system. The sump is normally divided into hot and cold areas by a partition weir wall (seeFigure 7.1).
Water is circulated from the cool side and drains back into the hot side.
When the system temperature reaches the control maximum, it is pumped from the hot sump and through the cooling tower before draining back into the cool side. A rough rule for deciding sump capacity is that the water should not be turned over more than once per minute; within the restraints of cost the biggest available volume gives the best results.
Sufficient excess sump capacity, above the normal working level, should be provided to accommodate drain-back from pipework, engines, and dynamom- eters upon system shutdown. There is a continuous loss of water due to evap- oration plus the small drainage to waste, mentioned above; therefore, make-up from the mains water supply needs to be controlled by a float valve. It is important to minimize air entrainment in the pump suction; therefore, the minimum level of the sump when the pressure and return lines are full should be sufficient to discourage the formation of an air-entraining vortex. The return flow should be by way of a submerged pipe fitted with an air vent. The Chapter | 7 Test Cell Cooling Water and Exhaust Gas Systems 157
arrangement shown diagrammatically in Figure 7.1is a classic arrangement of which thousands of similar systems are installed worldwide, but care has to be taken to keep debris such as leaves or flood water “wash-off” from entering the system. All ground-level sumps should have the top surface raised at least 100 mm above ground level to provide a lip that prevents flooding from groundwater.
A sensible design feature at sites where freezing conditions are experienced is to use pumps submerged in the sump so it can be ensured that, when not being used, the majority of pipework will be empty.
Closed Plant Water-Cooling Circuits
The design and installation of a closed water supply for a large test installation is a specialist task not to be underestimated. It may require the inclusion of a large number of test and flow balancing valves, together with air bleed points, stand-by pumps, and filters with changeover arrangements. By definition these closed systems have no sump or gravity draining from any module within the circuit. Such a system does not suffer from the evaporative losses of an open system and is less prone to contamination. Typically it uses one or more pumps to force water through the circuit, where it picks up heat that is then dispersed via closed-circuit cooling towers, then the water is returned directly to the pump inlet.
Engine Cooling
tower
Cooling tower pump
Cooling water
circulation pump Engine coolant temperature control unit Make up supply
Hot sump
Cool sump
FIGURE 7.1 Simple open cooling water system incorporating a partitioned sump.
It is vital that air is taken out and kept out of the system. The whole pipe system must be provided with the means of bleeding air out at high points or any trap points in the circuit. To achieve proper circulation and cope with thermally induced changes of system volume, also to make up for any leakage, a closed system has to be fitted with an expansion tank plus some means of
“make-up” and pressurization. These requirements can be met by using a form of compressed air/water accumulator connected to a pressurized make-up supply of treated water. “Balancing” of water systems is the procedure by which the required flow, through discrete parts of the circuit, is fixed by use of pressure-independent “flow-setter” valves having test points fitted for commissioning purposes. Valves are required for each subcircuit because they have their own particular thermal load and resistance, and therefore require a specific primary system flow rate. The balancing of closed cooling systems can be problematic, particularly if a facility is being brought into commission in several phases, meaning that the complete system will have to be reba- lanced at each significant system addition. None of the devices fitted within a closed plant water system should have “economizer” valves that themselves regulate the flow of the primary (plant) water, since that variation may continually unbalance the primary system. To avoid such unbalancing, devices in the circuit, the temperature control valves, should work by regulation of secondary fluid and have constant primary (plant water) flow. Freezing protection must be considered by the facility designer, even if the whole system is within a building. Closed, pressurized water systems can be filled with an ethylene glycol and water mix to prevent freezing but, as mentioned at the beginning of this chapter, the cooling efficiency of the mixture is inversely proportional to the concentration of glycol. It should also be remembered that some materials in seals and pipes, such as natural rubber derivatives, may deteriorate and fail if exposed for long periods to glycols at elevated temperatures. The pipework of open water systems can be trace heated.
Such systems consist of special heating tape being wound around the pipes in a long pitched spiral under insulating material. The control is usually fully automated such that the heating current is regulated according to ambient temperature.
Engine Coolant Temperature Control: Cooling Columns
If special engine coolants are not required, a cooling column is a simple and economical solution commonly used in the USA (seeFigure 7.2). It can be portable and located close to the engine under test. The column allows the engine outlet temperature to run up to its designed level; at this operating point a thermostatic valve opens, allowing cold water to enter the bottom of the column and hot water to run to waste or the sump from the top. The top of the column is fitted with a standard automotive radiator cap for correct engine pressurization and use when filling the engine circuit.
Chapter | 7 Test Cell Cooling Water and Exhaust Gas Systems 159