· Rated capacities assume uninterrupted air access to the coil and no hot air recirculation.
· Air entering temperatures are defined as the temperature at the inlet face of the heat exchanger and not the surrounding ambient temperature. Depending upon the installation and environment, the air inlet temperature to the coil may be up to 6K higher than the local ambient temperature.
· This will depend upon the cleanliness of the environment, both in terms of airborne contaminants that may cause surface corrosion and dust, dirt, leaves etc. that may be induced into the incoming cooling air stream. Nevertheless, any fouling or corrosion will decrease the effectiveness of the product and so when such a situation arises, the coil should be cleaned. Light brushing of the finned surface (in the direction of the fins with a soft brush) plus an application of a weak detergent is recommended. It is also possible to use with extreme care, a steam or water pressure system providing that a safe distance is kept to avoid deforming the fins and the spray impinges at 90°to the coil face.
Are there any ‘rules of thumb’ to indicate the noise level reduction when using speed regulation such as an Inverter?
· The reduction is actually product related, but as an indication a 25% speed reduction would represent a noise reduction of approximately -6dBA, whilst running at 50% full speed results in around -15dBA.
So clearly, when a 12 fan unit is considered, the lowest noise reduction attainable is when 11 of the fans are switched off. However this equates to only a -11dBA reduction, which is not as much as when running at 50% full speed and more importantly, provides only 10% capacity as opposed to perhaps 50% capacity!
· Refer to our website for positioning and location diagrams for recommended distances between horizontal and vertically mounted adjacent products, spacing from walls etc.
· Where possible, products should be placed on low solar gain roofs and preferably in the shade to avoid effects that can elevate the air inlet temperature by up to 6K.
· Horizontal airflow products should be orientated to minimise prevailing wind direction implications such as ‘wind milling’ and it is thus recommended that care should be taken when placing product variants fitted with 12 & 16 pole motors. Refer to our website for diagrams suggesting recommended product spacing’s.
· We recommend that motors should be run for a minimum of 5 hours per month to avoid motor bearing stiction problems (Brinelling) and to reduce condensation build-up.
· In severe low temperature environments both motors and electrical panels/switchgear will benefit from the fitting of anti-condensation heaters. Excessive moisture build-up where electrical switchgear is concerned, inside motor housing and bearings etc. can cause component failure or reduced life-cycle.
· For temperatures below -40°C, special materials may need to be considered and in the case of motors, special lubricants/grease used in the bearings.
· At low temperatures fan motor start-up power/current will be affected as a consequence of the higher air density. As an example, the impeller absorbed power (kW) increases by 25% at -40°C compares with +20°C.
· Discharge line dampers for Condensers or flexible couplings for Dry Coolers should always be fitted to avoid over stressing the heat exchanger tubes and/or headers.
· All rotating equipment is both statically and dynamically balanced and thus vibrations are minimal. However, depending upon the location and environment, products can be mounted upon anti-vibration pads/springs to either avoid exceeding published noise levels or preventing transmission of noise or vibrations through the roof.
· If AVMs are used, ensure appropriate flexible pipe work connections are included to allow for product movement.
· It is recommended that flexible connections are always used between the supply pipe work and the connection fittings supplied on the product.
· In the case of industrial dry coolers, where high fluid temperature are often encountered, thermal expansion consideration must be accounted for to ensure that the product is not adversely affected.
· The supply pipe work should always be self supporting to avoid over-stressing the product connections and heat exchanger tubes.
· For condensers, it is advisable to use hot gas discharge line dampers to isolate the product from the high frequency pressure pulses often associated with reciprocating compressors
· Consult the Operating and Maintenance instructions to clarify this issue. Depending upon the product size and weight plus the length of the forks on the forklift, some products can be safely handled providing care is take to avoid damage to the underside of the product and in particular, the coil (heat exchanger). However, products should never be handled or lifted via their headers. Always use the lifting lugs provided in accordance with the O&M instructions.
· Condensers are not usually an issue unless excessive heat is applied when brazing the pipework.
· For dry coolers, care should be taken when tightening screwed (BSP) connections, to avoid over stressing the coil tubes and headers.
· All supply and return pipework should be independently supported and never hung from or in anyway supported by the product, especially if AVMs are fitted to the product.
· Adequate frost protection measures should be taken to avoid coil failure from frost damage. These may include using a suitable fluid that will not freeze, fitting of a ‘winterization’ kit to ensure that fluid temperatures remain above freezing or provision of a system that automatically evacuates and drains the system when temperatures become critical.
· In theory, yes. However, the tube length, product orientation, circuitry pattern and surface tension considerations of the fluid may prevent the coil(s) from fully self-draining. In this event, ensure that the fluid is evacuated under pressure. If sufficient fluid remains in a particular circuit, serious frost damage can result from low ambient temperatures.
· When applied to air cooled condensers and dry coolers, is a generic term usually referring to alternative methods of adjustment of the air volume to match the desired capacity of the product, which may vary throughout the day.
In the case of condensers the controlling criterion is usually the ‘Head Pressure’ or Condensing pressure monitored at the inlet to the condenser and for dry coolers, the criterion is usually the fluid leaving temperature – also known as the ‘process entering’ temperature.
The sensor monitoring the head pressure or fluid leaving temperature provides a signal to one of the following air flow control systems which endeavours to match the ‘set point’ condition…
o Fan step controller - also referred to as a fan cycling controller - starts and stops individual or pairs of fans in sequence or randomly, resulting in a variation in the cooling load. This ‘wide band’ control methodology imposes different cooling behaviour over the different sections of heat exchanger surface as fans are activated and deactivated. The method is simple and usually cheap but does not necessarily provide ‘close control’.
o Fan speed regulation relates to varying the speed of all of the fans simultaneously, thus adjusting the cooling load equally over the entire surface of the heat exchanger. Clearly this is a preferable control methodology but requires a more sophisticated and thus more expensive control system.
Currently the favoured alternatives are electronic variable voltage or frequency inverter solutions. Both result in the ability to adjust the speed of the motors/fans and hence reduce the air volume / cooling load to match the set-point condition.
· A fan step/cycling controller in its simplest form can comprise a number of thermostats (dry cooler) or pressure stats (condenser), each with its own sensor and which is adjusted to a slightly different incremental set-point condition. As each set-point condition is reached the thermo/pressure stat activates one or more motor contactors, thus providing additional (incrementally stepped) air cooling to meet an increasing thermal load.
Although for ‘course’ capacity control this system operates adequately, the inherent hysteresis and required ‘dead zone’ between each step, does not make it suitable for ‘close control’ scenarios.
However, today it is more common to provide electronic step controllers utilising only one sensor and typically offering 2 through to 8 steps all integrated into one electronically programmable device. Thus much closer control is achievable.
Furthermore, the programming functionality allows for sequential, rotational or randomised triggering of the motor relays which ensures more even usage of the motors fitted to the product and helps equalise the motor lives.
· Unlike a step controller, speed regulation provides capacity control by infinitely varying the air flow across the heat exchanger; by simultaneously modulating the fan speed of all the fans fitted to the product; to match the required thermal load.
The speed regulation can be achieved by either varying the voltage to the motors or more commonly by modulating the AC supply frequency via an inverter. Frequency modulation provides a very efficient method for adjusting the fan speed without the inherent electrical inefficiencies associated with voltage modulation.
A sensor (temperature or pressure) monitors the operating conditions of the dry cooler or condenser and electronically compares the signal with a set-point. The inverter microprocessor translates the differential into the necessary frequency/motor speed increase or reduction. This in turn modulated the cooling air flow to maintain stable operating conditions.
Generally the electronic and programming functionality provided by an inverter allows for a high degree of system customisation and control behaviour besides the ability to interface with Building Management Systems (BMS), which are becoming more and more commonplace.
· Simple starter control panels for dry coolers and condensers often use current overloads fitted to the motor contactors to provide protection in the event of a motor related problem. Often this is more than sufficient if the motor runs at full speed when operational.
However, if the motor is speed regulated and thus can operate anywhere between zero and full speed, then when running at reduced speeds it is possible that the protecting current overload may not sense a problem and thus not ‘trip’ even though the motor windings become excessively hot. Overheated motor winding can result in either a breakdown in the winding insulation, reduced motor life or in the worst case, motor burn-out and failure.
To avoid such a scenario, ‘thermal contacts’ can be embedded into the windings which can be connected to the motor control circuitry to isolate the motor in the event of excessive winding temperatures, either when the motor is running at low speed or if a fault occurs.
Thermal contacts are usually small bi-metallic relays which react to temperature. When the temperature exceeds the set-point, the contacts open and provide a break in electrical integrity. On its own, a thermal contact will not do anything, but when connected to a control relay or in serial with the supply voltage to a motor contactors solenoid coil, will provide a break in the electrical continuity when it overheats. Thus the motor will be isolated and if provisioned for, an appropriate alarm signals generated.
· A klixon is a term often incorrectly used to infer a thermal switch.
In fact Klixon is both a company name, product range and a trade name for ‘snap-acting bi-metal discs’ used as electrical switch gear, often built into hermetic compressors. However, the common misuse of the phrase by this industry has resulted in klixon, bi-metallic contact and thermal contact referring to the same functional device.
· A thermistor is a small semiconductor device which is a type of a resistor used to measure temperature. Depending upon its makeup, it can either provide an increase (PTC) or decrease (NTC) in the resistance as the temperature rises.
Unlike the thermal contact, it can not be directly connected in a circuit to provide a physical break in electrical integrity, but has to be powered and wired to some other device that can sense the positive or negative change in resistance before an action can take place. Therefore, depending upon the sophistication of the control circuitry required, a thermistor based system tends to be a more expensive option or provides less sophistication when considering individual motor protection.
· Generally motors for dry coolers and condensers are either rated as IP54 or 55 (incl. drain hole), whilst liquid unit coolers and evaporators are rated at IP44 or 54.
· Lockable safety switches are nominally rated at IP65, but as a matter of necessity often include a drain hole to prevent condensation. Thus such electrical switch gear is effectively rated at IP54.
· Electrical control panels housing equipment such as speed regulation, step control, Dri-Batic control etc. are initially IP65 or indeed IP67 prior to inclusion of the electrical equipment and on occasions, cooling fans. Thereafter, the enclosure’s IP rating drops to typically IP55 or even 54 or whatever the lowest IP rating of any such component that is fitted to the control panel which may affect the overall IP rating.
· FlexSpray is the terminology we use when the heat exchanger element of a dry cooler or air cooled condenser is sprayed with water to depress the normal dry bulb air inlet temperature (pseudo dry bulb), thus widening the effective operating TD (temperature difference) resulting in a boost in the capacity or alternatively, requiring less surface to dissipate the thermal load. Often this technology is used to meet abnormal ‘peak load’ performance from a product sized for normal ambient conditions negating the need for oversized or redundant dry coolers.
This particular ‘sprayed coil’ technology should not be confused with adiabatic cooling technology where dry bulb temperature can be depressed close to saturation (wet bulb) conditions e.g. cooling towers.
Distribution of the water is achieved by a matrix of ‘hollow cone’ spray nozzles served by a series of galvanised steel sparge-pipes which spray the water in the direction of air flow and directly onto the coil surface. The nozzles are sized to provide the correct amount of water in relation to the available mains water pressure. Mains water pressures from 0.5 to 10 barg can be accommodated.
The FlexSpray process is intentionally designed as a ‘total loss’ system providing typical water droplet sizes of 100-240 microns which is significantly greater than the 5 micron generally considered as the limit for human inhalation. Both these factors are important in relation to eradicating Legionella bacteria related issues.
Spraying the coil does indeed depress the air inlet temperature but not as efficiently as with a truly adiabatic solution. Thus we do not claim that we can depress the air dry bulb temperature to equal the wet bulb temperature i.e. reach saturation. However, we are able to reach saturation efficiencies of between 60 and 80% depending upon the air inlet temperature, fans speed/air flow. Nevertheless, significant performance improvements can be achieved.
· A FlexSpray variant of a dry cooler or air cooled condenser is a solution to accommodate abnormal peak load ambient variations - which seem to be occurring more frequently these days. Thus if adopted, it is recommended that epoxy coated fins are used to resist lime scale build-up (deposition of minerals suspended in the sprayed water) onto the finned surface which over time, can affect the product performance. Consequently it is recommended that such a system should be operational for less than 200 hours per year during which time, regular inspections and cleaning of the heat exchanger surface should be scheduled.
· Spraying water onto a coil is the last stage of capacity control to match the cooling demand when either all of the fans are activated via a step controller on a multiple fan product, or alternatively all the fans are running at full speed in the case of a variable speed regulation system. Thus the solenoid valves that activate the water spray system are interlocked such that they will not operate until the full air volume is achieved,
When the product is running at full design air volume and the cooling demand can still not be met, the spray system is triggered. This imposes a ‘thermal shock’ to the system often resulting in over-cooling and thus the control system will sense this behaviour and shut the main water solenoid valve. Also interlocked with the mains solenoid valve are air vent and drain valves that allow the water in the distribution system to naturally drain from the pipework - eliminating any Legionella ramifications.
Once the spray system is deactivated, the residual cooling thermal inertia may cause the step controller to turn off one or more fans or in the case of a speed regulation controller, may cause all the fans to reduce in speed. However over time, the insufficient cooling provided by the fans will cause the FlexSpray spray system to be activated again and the whole cycle to repeat itself.
Smaller system hysteresis can be achieved by using a staged spray system where increasing portions of the coil surface are sprayed to more closely match the cooling demand and minimise over-shoot scenarios.
The deactivated FlexSpray system is designed to be free from any residual water, however the incoming water mains supply pipework should either be adequately insulated, provided with ‘trace’ heating or an automated evacuation/’blow down’ system to drain the water should be allowed for to eliminate frost damage. These latter considerations are not included in the scope of supply of the FlexSpray equipment and thus are the responsibility of the End User.
· A FlexSpray control package ordered in addition to the ‘supplied loose’ sparge pipe system and nozzles, comprises a filter, solenoid valves, regulation valves and pressure gauge and thermostat, but we expect that the incoming water feed is clean, free from waterborne particles and preferably ‘soft’ rather then ‘hard’.
Water quality levels should comply with …
o pH levels between 5 and 7.5
o Hardness 6.5-8°fH (3.5-4.5°dH)
o Ca / Mg carbonate levels < 120 ppm
Should it be determined that the water is ‘too hard’, it is recommended to fit a water softener to the mains water supply line.
· Wind-milling refers to the condition where axial fans that are ‘off line’ and should be stationary are in fact rotating due to localised air circulation or prevailing winds. This condition is usually associated with upright units (horizontal air flow) but can affect Vee type products.
If the undesirable driving air flow is through the coil and thus emulating the normal mode of operation, then wind-milling is not usually a problem. However, if the air flow direction is against the normal operating mode, then the fan(s) will wind-mill backwards and on occasions, faster than their normal operating speed. In such cases, when the fan(s) are required to start they have to overcome the ‘backward’ inertia before they can run-up to full speed in the correct direction. This situation can become critical if 12 & 16 pole motors are in use and the resultant start-up current drawn exceeds the electrical overload settings. The tripping of the current overloads result in some or indeed all the fans remaining idle and thus no forced draught cooling is available. Consequently the set-point conditions can not be fulfilled and in the case of a condenser, the system may trip-out on high pressure or for a dry cooler, the fluid leaving temperature exceeds the allowable operating temperature resulting is system shut-down.
· When a process produces a hot fluid or fluids at more than one temperature that should preferably be handled in the same cooling device, then this product/dry cooler is often referred to as a sandwich cooler.
Usually the resultant product is a variation of a standard dry cooler where perhaps the 6 rows of coil and associated surface area are divided into 2 rows for the high temperature process and 4 rows allocated to the low temperature process.
Usually the cooling (ambient) air stream passes through the low temperature section first and then through the high temperature section.
Clearly, in this mode of operation it is not possible to accurately control each process fluid leaving temperature by air volume modulation via speed regulation, alone. If the primary process temperatures are control via the air flow capacity control system and the secondary process controlled via 3-way valve fluid mixing, then some degree of dual-mode control is viable.
If however, the processes are not critical, then capacity regulation controlled via the primary process temperature may be good enough.
· The simple answer is ‘with difficulty’!
Firstly, if this less than ideal scenario is the only viable possibility, one of the operating sections has to be defined as the ‘master’ section and its fluid outlet temperature determines the behaviour of the capacity control system. Therefore the ‘slave’ section performance will follow accordingly.
· If both processes call for ‘close control’ then a ‘sandwich’ cooler concept is not a viable option. The alternatives are...
o Each process should be handled by its own dry cooler with its own capacity control system
o Alternatively, use a product with 2 lines of fans and 2 discrete heat exchanger sections where each line of fans is controlled independently by the heat exchanger lying beneath.
This latter solution has its own shortcomings relating to heat transfer by conduction between the adjacent high and low temperature sections. Thus often, an oversized product has to be provided to match the combined capacity requirements.
· Standard product ranges utilising copper tubes are not suitable as Ammonia is highly aggressive towards copper. However, in such a case alternative tube materials such as stainless steel or aluminium are able to provide a solution.
· EN13487 is the preferred sound level standard adopted by Eurovent to present sound pressure levels at a distance in a free field environment. This methodology provides for the correction to the certified/approved overall A-weighted sound power level (LwA) with respect to parallel-piped enveloped constructed around the product, typically at a 10 meter distance, resulting in the overall A-weighted sound pressure level (LpA).
Thus LpA = LwA – 10 log10 S, where S is the surface area (m2) of the constructed parallel-pipe envelope. Depending upon the size of the product in question, the surface area reduction component for a distance of 10 m is typically between -31 and -33 dBA
· The Hemispherical approach is to apply exactly the same correlation as above, but in this case the surface area (S) of the envelope under consideration is a hemisphere with a radius (R) of a defined distance, typically 10 m.
Note that this methodology assumes that the noise source is a ‘point source’ … clearly not the case with dry coolers or condensers with more than one fan ! Indeed this product type is neither a line source, but a ‘line of point sources’ and thus subject to more rigorous analysis.
However, this is where ‘artistic licence’ is often applied because the normal defined radius of 10 m can either be assumed to be from the centre point of the product or more often and thus commercially more acceptable, a radius equating to the distance from an observer’s position 10 m away from the extremities of the product, to the centre point of the product. Clearly, this latter scenario is a greater distance and thus a greater surface area and hence lower noise level !
Thus considering point source theory applied to the centre point,
LpA = LwA – 10 log10 (2pR2) which further reduces to LpA = LwA – 20 log10 R – 8
Now when R = 10, LpA = LwA – 28 …. apparently +3 to +5 dBA noisier than the EN13487 prediction !
Alternatively for a product size 4 m x 2 m x 1.5 m and a distance 10 m from the end of the product, the effective radius would be R = 12 m and then the above relationship results in LpA = LwA – 30 … mysteriously becoming 2 dBA quieter !
Both solutions are correct, but it depends upon the definition of R that dictates the theoretical noise level.
· The third method often used to present noise levels is referred to as just ‘free field’ and is usually an empirical variant of the theoretical prediction and takes into consideration the directivity and constructional characteristics of the product.
If one was to plot a 3D sound pressure level map around a real product, then it would be clear that the sound level changes circumferentially around the product and indeed at various elevations.
The measured values at a given distance from the header/return bend end (short side) of the product will be slightly lower than measurements from the long side at the same distance. Furthermore, the measured sound level above the unit increases still further. These variations are generally not identified when presenting overall average sound pressure data because theses are derived from overall average sound power levels measured in accordance with ISO 3744 or ISO 9614-2, which by definition are included within the overall average figures.
Nevertheless, sound levels presented as free field at 10 m from the header end (short side) of the product can be up to -3 dBA lower than figure derived from overall average sound power levels.
Finally, certain product variants are subject to far greater directivity implications; for example Vee type products that have vertically inclined coils and thus radiate sound horizontally rather than reflection from the ground. Furthermore, the sound levels at the header end of the product can be up to -4 dBA lower than from the open coil side of the product. This characteristic results from the improved attenuation from the sheet metal end sections where far more noise is attenuated.
· The best case scenario is ‘not a great deal’, only -3dB
If a product was able to be suspended in free space then the measured sound level would be -3 dB less than the ‘real’ scenario, where there is reflection from the ground.
Noise emitting equipment such as a fan will generally radiate sound energy and thus ‘pressure waves’ is all directions. Clearly the sound energy is often channelled vertically through the fan orifice and thus a proportion of the noise exits in this fashion. But approximately 50% of the noise energy travels backwards through the coil and is reflected off the surface below.
Clearly the ‘absorption coefficient’ of the surface upon which the product is mounted will govern the strength of the reflected sound pressure waves. Thus an observer at a distance from a product receives one sound pressure wave directly from the discharge of the fan/motor and a second - partially attenuated - pressure wave reflected from the underside surface of the product. Hence if the discharge (fan side) noise level is fully attenuated, then the observer still receives the reflected noise level … approximately -3 dB lower than without attenuation.
· To provide levels of attenuation greater than -3 dB, the inlet side of the product has to be attenuated in some way. Marginal improvements can be achieved by acoustically insulating the inside of the fan deck assembly but if the break-out through the coil is not addressed, there will be no significant attenuation.
Clearly, placing acoustic louvers etc. on the air inlet tract to the product invariably creates pressure drop and thus reduces the cooling air volume and thus capacity and thus is often not a solution. Furthermore, changing the fan operating condition may decrease its efficiency and thus ’increase’ the generated noise level, partially offsetting the whole purpose of fitting attenuators !
Erecting barriers (even acoustically lined) around the product is a simple method of providing significant attenuation. However, if positioned too close to the product, there may be an impact upon the product’s performance.