1、2010 ASHRAE 81ABSTRACTThe general exhaust system of a clean room for hi-techfabrication plants (FABs) contains a great deal of energy formake-up air handling units (MAUs) for pre-cooling and pre-heating usage. The present study employed a concept of thesecond law of thermodynamics analysis to optimi
2、ze the designof a compact heat exchanger installed in a MAU, based on aspecified external dimension and a fixed inlet temperature andmass flow rate for the hot and cold sides of the exchanger.Results indicate that an appropriate compact heat exchangercan be sized on the basis of the second law of th
3、ermodynamicsanalysis within minimum entropy production (or irreversibil-ity), caused by both temperature difference and pressure drop.The optimized compact heat exchanger in this study can save5 and 12% of the energy for pre-cooling and pre-heating in theMAU, respectively.INTRODUCTIONIn subtropical
4、area such as Taiwan, the operating powercost of air conditioning systems for a typical commercialbuilding accounts for more than 50% of the total power bill.Therefore, a significant energy savings can be achieved if heatrecovery technologies such as run-around coils, plate-to-plateheat exchangers, h
5、eat recovery wheels, and heat pipe heatexchangers (HPHXs) are incorporated. There are several typesof heat exchangers available for heat recovery applications.Run-around coils are relatively cheap, but a pump pack and anexpansion tank are needed to run the system. Plate-to-plateheat exchangers are f
6、airly efficient, but are bulky, costly, andextremely difficult to maintain. Also, condensate can betrapped on the plates with a resultant growth of molds. Heatrecovery wheels are difficult to clean and cross-contaminationis always a concern. Apart from these drawbacks, heat recov-ery wheels do not e
7、fficiently drain condensation. For manyyears, heat pipe heat exchangers (HPHXs) with two-phaseclosed thermosyphons have been widely applied asdehumidification enhancement and energy savings devices inHVAC systems in Western countries (Yau 2007), and theapplication of a double heat pipe heat exchange
8、r system in theconventional air handler unit operating in a tropical climate isstrongly recommended as an efficient method for humiditycontrol and energy savings in order to maintain acceptableroom conditions (Yau 2008). However, in a subtropical areasuch as Taiwan, the temperature difference betwee
9、n the evap-oration side and condensation side is not so large. Therefore,the application of a heat pipe as a heat exchanger in the heatrecovery system is not very popular. In semiconductor clean-rooms, therefore, potential energy recovery from the EA isenormous. Furthermore, in the hot and humid sea
10、son, themoisture removal capability of chilled water coil in the heat-ing, ventilating, and air conditioning (HVAC) systems can beenhanced if the supply air is pre-cooled before reaching thechilled water coil.For the optimal design of a gas-to-gas heat exchangerunder the limitations of no difference
11、 of pressure drop of thecold side and hot side, Bajan (1977) conducted an analysis ofsystem entropy generation that taking into account the differ-ence of flow rate but still ignoring the pressure drop, an opti-mal model for the heat exchanger design. However, the modelis only applied in trend analy
12、sis of the model but not in realoperating conditions (Sarangi and Chowdhury 1982). Addi-tionally, a model proposed for minimization of the pressuredrop between the cold /hot sides under a fixed dimension(Ooba 1959).Clean Room Exhaust Energy Recovery Optimization DesignJames J.M. Tsao, PhD Shih-Cheng
13、 Hu, PhDMember ASHRAEWen-Chen Kao Liang-Han Chien, PhDJames Tsao is currently a task leader in Pacific Engineers and Constructors, Ltd., Taiwan. Shih-Cheng Hu is a professor, Liang-Han Chienis an associate professor, and Wen-Cheng Kao was a MS student in the Department of Energy and Refrigerating Ai
14、r-Conditioning Engineer-ing, National Taipei University of Technology, Taipei, Taiwan.OR-10-010 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions 2010, Vol. 116, Part 1. For personal use only. Additional reproduct
15、ion, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission. 82 ASHRAE TransactionsDESCRIPTION AND METHODOLOGYThis study used a process exhaust system in a 200 mm (6in.) DRAM FAB as a design basis. For industrial safety andhygiene cons
16、ideration, only general exhaust may be taken intoconsideration in this study. From another point of view, it wasfound that a general exhaust system occupies more than a halfof total process exhaust capacity in the FAB, and almost 75%of the heat capacitance. Therefore, the system contains a highpoten
17、tial for energy recovery.The boundary conditions applied in this study wereshown in Table 1, where the conditions of outdoor air are 10C(50F) dry bulb and 30% RH in winter, and 34C (93.2F) drybulb and 80% RH in summer. Under the consideration of elec-tricity-free, easy-to-obtain, and engineering app
18、licability,plate-to-plate heat exchanger is a heat exchanger that meetsthese requirements, as compared with other heat-recoveryexchangers.In most engineering analysis, heat-exchanger-basedenergy recovery systems are all classified by the “energyconversion” (or “quantity”) approach. However, most eng
19、i-neering studies neglect “quality” of energy recovered fromhigh potentials. To avoid this, the second law of thermody-namics is taken into consideration in this study. Evaluating thequantity of “availability of each condition” and “entropyproduction” in the energy recovery system can find the amoun
20、tof “maximum energy” actually recovered by this system.In each location of the heat recovery system, the avail-ability can be considered as follows:i= (hi h0) T0(si s0)(1)Where the subscript “i” denotes any location, and “0”refers to the “dead status” or “ground status.” The efficiency ofthe second
21、law of thermodynamics (II) for the energy recov-ery system may be written as follows:(2)Subscripts “in” and “out” denote the location of the inletand outlet in the energy recovery system, for hot and cold sidesrespectively. If considering the energy recovery system asshown in Figure 1, the entropy c
22、hange between the inlet andoutlet for ideal gas may be written as follows, where Phe= Phi Phand Pce= Pci Pc.(3)The effectiveness of heat transfer () is defined as follows,where and .(4)Equation (3) is used to substitute into the dimensionlessentropy change, .(5a)Rewriting Equation (5) in terns of ,
23、, , and , where theinlet temperature ratio, = Tci/Thi, the heat capacitance ratio, = Cmin/Cmax, the dimensionless pressure drop at cold side, = Pc/Pci, and the dimensionless pressure drop at hot side, = Ph/Phi, gets(5b)Table 1. Inlet Conditions and the Dimensions of a Heat ExchangerItem UnitMakeup A
24、ir (MAU)Exhaust Air (GEX)Volumetric Flowrate CMH 100,000 50,093Inlet temperature, dry bulb C 10 24.1Inlet relative humidity % 30 45Inlet pressure kPa 1.6 1.1Outline dimension m 3 3 1.6IIcold out,cold in,hot in,hot out,-=Figure 1 Layout of a heat exchange system.dSgenmccpcTceTci-ln RPcePci-ln=mhcphTh
25、eThi-ln RPhePhi-ln+mccpcCmax= mhcphCmin=CmaxTceTci()CminThiTci()-CminThiThe()CminThiTci()-=EgendSgen()Cmax=EgendSgenCmax-CminCmax- 1 TciThi- 1ln=1CminCmax- 1ThiTci-+ln+CminCmax-Rcph-1PhPhi-lnRCpc-1PcPci-lnEgen 1 1()+ln 1 1()-ln+=Rcpc-1 ()lnRcph-1 ()ln 2010, American Society of Heating, Refrigerating
26、 and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions 2010, Vol. 116, Part 1. For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission. ASHRAE Transaction
27、s 83If the heat exchanger is very high efficient and has verylow pressure drop when working fluid passes through the heatexchanger at the hot and cold sides, an approximating equationis obtained from Equation (5b), based on the Taylor approxi-mation, and the dimensionless entropy production maybe re
28、written as follows.(6)Three physical phenomena in the right side of Equation (6)refer to “dimensionless entropy production due to flow imbal-ance” (Item 1), “dimensionless entropy production due tostream temperature difference between the hot and cold sides”(Item 2), and “dimensionless entropy produ
29、ction due to streampressure drop” (Item 3). In point of engineering, the dimensionof the heat exchanger is constrained by the size of MAU. Mean-while, since the inlet conditions (volumetric flow rate and inlettemperature) at the make-up air and exhaust air are fixed, thefirst and second terms of Equ
30、ation (6) will remain constant, andthe major parameters which will influence Equation (6) are thedimensionless pressure drop of the heat exchanger at the hotside () and cold side (). Therefore, dimensionless entropyproduction minimization for Equation (6) will reveal that itapproaches the specificat
31、ion value. That is, if taking an extremevalue of Equation (6), an optimized heat exchanger will beobtained under the restricted installation dimensions whenvolumetric flow rate passes through the heat exchanger with aminimum pressure drop.A cross-flow membrane-type compact heat exchangerwith dimensi
32、ons H Lh Lc(neglecting the thickness ofmembrane) was shown in Figure 2. If the dimension fractionat the cold and hot sides is c and (1 c) in the H direction,respectively, the dimension of the heat exchanger at the coldand hot sides will become cH and (1c)H, respectively. Thus,the face area of the he
33、at exchanger at the cold and hot sides isas follows, respectively:Ac,c= (Lc)(cH 7)Ac,h= (Lh)(1 c)H (8)The face velocity of the heat exchanger at the cold and hotsides is as follows, respectively:(9)(10)Defining the volumetric flow ratio r as(11)If calculating the following conductance (UA) betweenth
34、e hot and cold sides,(12)by the Number of Transfer Unit (NTU, UA/C), Equation (12)may be rewritten as(13)Introducing the Colburn J-factor into the NTU expres-sion,(14)The pressure may be written in the following expression,(15)If the ideal gas equation, P = RT, is used to substitute intoEquation (13
35、), this expression may be rewritten as(16)Considering that the property of air flow at the hot andcold sides is very similar, the following physical properties canbe assumed to be the same.EgenEgenEgen f,Egen T,Egen P,+ln 1 1-Item 1ln+=+ 1()21 () + - Item 2Rcpc-Rcph-+ Item 3VhmccAcc,-=VhmhhAch,-=Fig
36、ure 2 Heat exchanger for optimization.rmccmhh-=1UA-1UA()h-1UA()c-+=1NTU-1NTU()h-vNTU()c-+=NTU jPr23 LDh-=P 4 f w22g-LDh-=LDh- gRT2fV2-PP-=j 2f()hj 2f()cch Prc23Prh23, 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transacti
37、ons 2010, Vol. 116, Part 1. For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission. 84 ASHRAE TransactionsAccording to the assumption and definition describedabove, Equation (13) can be s
38、implified as(17)If assuming the specific total NTU value constant for theheat exchanger, the NTU value for both sides exists in thefollowing expression.(18)The value of dimensionless pressure of the heatexchanger at the cold and hot sides can be obtained in terms ofF and c.(19)(20)whereAccording to
39、the assumptions in this study, the entropyproduction for the flow imbalance and stream temperaturedifference between the hot and cold sides remains constantdue to the fixed inlet conditions; thus, the dimension fractionof the cold side (c) and the NTU ratio (F) between the cold andhot sides may infl
40、uence the entropy production for the heatexchanger. The following extreme values c and F may beobtained.(21)(22)Therefore, the relationship between c and F is as follows:(23)or(24)The heat transfer relations were obtained from the resultsof literature (Wang 2007). For the heat transfer mechanism inN
41、usselt numbers, the correlation is given byNu = 5 + 0.012Re0.83(Pr + 0.29) 0.6 Pr 0.9 (25)The friction factor for a rectangular type heat exchangeris given as follows (Rohsenow et al. 1998).f = 0.1268Re0.35000 Re 1.2 106(26)Since there is no divided membrane in the flow channel,the configuration of
42、the effective calculation may consider thefollowing type (Bejan 1993):(27)The relation of pressure drop is simplified from the resultsof literature (Kays and Landon 1984).(28)Where v denotes a specific volume, and the subscript mdenotes the mean value for the physical properties of the inletand outl
43、et.Results and DiscussionFigure 3 shows an external dimension of the heatexchanger. According to results of optimizing the height c atthe cold side, and introducing the inlet conditions in this study,it was observed that the entropy production became minimumwhen c = 0.509. The specification of the h
44、eat exchanger, basedon the previous design parameters, was calculated in Table 2.Though the height fraction of the heat exchanger at the coldand hot sides remained fixed, increasing the number of flowchannels might increase the number of the transfer rate and theheat transfer rate; Figure 4 shows th
45、is trend.According to the heat exchanger efficiency of the second-ary law of thermodynamics defined in Equation (2), Figure 5shows the calculated availability difference at the make-up air(MAU) and exhaust air (GEX) sides under different NTUs.Results indicated that the availability difference at the exhaust1NTU- Vc2Prc23j 2f()ccP P()cgRTci-=Vh2Prh23j 2f()hhP P()hgRThi-+Pr23j 2f()gR- Vc2P P()cTci-Vh2P P()hThi-+=FNTUc()NTUh()-P P()cP P()h-TciThi-r2Lc2Lh2-c21 c()2-=PP-c1B- F+()TcLc2c2H2-=PP-h1B-r2ThLh21 c()2H2- F+F-=B1NTU-j 2f()gRmcc()2Pr23-=
