1、2010 ASHRAE 87ABSTRACTAirflow change rate utilized in cleanroom facilities ismuch higher than in typical general-purpose buildings, fanenergy saving potential from cleanroom facilities is signifi-cant. Many independent reports have indicated that airflowquantities for cleanrooms are often over desig
2、ned, mainly tomeet the rule of thumb values published in FS-209 and IESTRP-12 for decades. This rule approach solely uses the “roomcleanliness class” to determine air change rate, and ignoresmany other “critical variables” such as room particle gener-ation rate, filter efficiency, particle surface d
3、eposition, particleentry through supply air, particle exit through return andexhaust air. Using existing over-simplified rule could oftencause significant energy waste; however, due to lack of quan-titative methodology, most of design and operating engineersstill choose to obey the existing rule to
4、avoid being challenged.This research team has established a new theoreticalmodel which is more comprehensive and inclusive than previ-ous models during last few decades. The mew model has beenfurther validated through testing in several labs. The compar-ison between the measured and model-predicted
5、results hasshown a good correlation. With this new approach, cleanroomair change rate can be “estimated and provided as needed”instead of “picking an arbitrary rate by rule of thumb”.Detailed analysis including charts, tables and key recommen-dations are provided.INTRODUCTIONCleanrooms utilize about
6、 5 to 50 more times airflow ratesthan for general-purpose buildings. Utilizing high volumeairflow has been mainly to meet the old federal standardFS-209 (versions A through E) and the recommendation byIESTs RP-CC012 (versions 1 and 2) since 1970s. Since thenmany published reports have indicated that
7、 cleanroom filteredair over-supply is a common practice which causes significantenergy waste (Mills et al. 1996, Jaisinghani 2001).The recommended guideline (tables) was based on oldexperience, in which air change rate was arbitrarily determined“only” based on room cleanliness class, disregarding a
8、roomsactual particle generation rate (internal generation and externalintrusion) and other factors. Due to lack of an accurate theo-retical model and related research, this obsolete guideline isstill being used today. For cleanrooms with lower particlegeneration rates, lower-than-recommended air cha
9、nge rates(up to 20% reduction) have been practiced. However most ofdesign and operating engineers still choose to obey the existingguideline to avoid being challenged. Establishment of a moreaccurate model supported with validations is a key to respondthe challenge and to reduce cleanroom fan energy
10、 waste.The fundamental airflow model in cleanrooms is themathematical relationship between the air change rate and theroom airborne particle concentration. In last a few decades,several mathematical models were proposed by Morrison(1973), Brown et al. (1986), Kozicki et al. (1991) and Jaising-hani (
11、2001), however a common shortcoming of these previ-ous models was over simplification due to ignoring manycritical elements and lack of experimental validations, thesemodels could only be used as qualitative indication, but not asa quantitative tool to calculate the required air change rate tomeet a
12、 room air cleanliness class based on the rooms specificairborne particle load, see Table 1. The main project objectivewas to establish a new model which is more descriptive(includes more variables and parameters), and more accuratethan existing models.Development of Cleanroom Required Airflow Rate M
13、odel Based on Establishment of Theoretical Basis and Lab ValidationWei Sun, PE John Mitchell Keith Flyzik Shih-Cheng Hu, PhDMember ASHRAE Member ASHRAE Junjie Liu, PhD R. Vijayakumar, PhD Hiro FukudaMember ASHRAEWei Sun is principal, director of engineering at Engsysco, Inc., Ann Arbor, MI. John Mit
14、chell is vice president at Particle Measuring Systems,Inc. Boulder, CO. Keith Flyzik is the training manager at Micro-Clean, Inc. Bethlehem, PA. Shih-Cheng Hu is a professor at the NationalTaipei University of Technology, Taipei, Taiwan. Junjie Liu is an associate professorat Tianjin University, Tia
15、njin, China. R. Vijayakumaris the president of Aerfil, LLC, Syracuse, NY. Hiro Fukuda is the general manager of Kanomax USA, Andover, NJ.OR-10-011 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions 2010, Vol. 116,
16、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. 88 ASHRAE TransactionsESTABLISHMENT OF A NEW MATHEMATICAL MODELwhere,V = Space volumeOA = Rate of makeup airflow (volume/ti
17、me)SA = Rate of supply airflow (volume/time)RA = Rate of return airflow (volume/time)EA = Rate of exhaust airflow (volume/time)Q = Rate of leakage airflow (volume/time)Cs= Impurity concentration in space (parts/volume)Co= Impurity concentration in makeup air (parts/volume)Es= Filter efficiency (mass
18、 basis)G = Rate of impurity generation in space, averaged throughout the space (parts/volume/time)D = Rate of impurity deposition from air to surface in space, averaged throughout the space (parts/volume/time)T = TimeIndoor particle balance equation can be established asfollow:Table 1. Comparison of
19、 the New Model with Previous Simplified ModelsComparison of Existing and Proposing Models Existing Representative Mathematical ModelsNew Mathematical Model ModelsMorrison1973Brown 1986Kozicki 1991Jaisinghani2000Sun2009ModelingPurposeShow the particle load contributions from various sources x x x x x
20、Each component of particle load to be quantified asestimation toolxModel inputsIssue Component Included in Model1 Differential equation x x2 Transient state (time as a variable) x x3 Steady state x x x x x4 Supply air particle addition x x x x x5 Room internal particle generations x x x x x6 Return
21、air particle removal x x x x7 Exhaust air particle removal x x x8Particle removal/gain due to air leakage (consider both positive/negative pressures)x9 Particle deposition on surfaces in room x10 Inclusion of particle deposition calculation formula x xModelExpressionKey elements expressed in dimensi
22、onless xAllow quantified and graphical study to exam impact ofeach variable and parameter xFigure 1 Basic cleanroom airflow configuration.VdCSOACORACS+()1 EU()1 EH()dt GVdt+=RACSdt EACEdt QCSdt DVdt 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org
23、). 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 Transactions 89The Particle Surface Deposition Rate D (in unit o
24、f parts/volume/time), in fact, can be calculated indirectly from testresults of particle surface deposition sensing equipment:measure particle counts per unit area for each wall, ceiling andfloor, based on the exposed surface area for each of thesecomponents, calculate the total particle counts in t
25、he space,then divide by the volume of the space. This methodologysuggests that the new model can be expressed in a simplerform by using the current particle surface deposition sensingtechnology.thus,Define m = OA/SA, r = RA/SA, so that, m + r = 1.Based on airflow balance relationship (ignore air den
26、sitydifference among air streams) in AHU unit and in room, so, SA= OA + RA, and SA = EA + Q + RA, then,As Air Change Rate ACR = SA/V, then,Define parameters a and b:m + (EU+ EH EUEH)r = a, andSo,dCS= aCSACRdt + bCoACRdtIf room particle concentration changes from the initialCSOto CSTduring the time i
27、nterval t,Ln(bCo aCST) Ln(bCo aCSO) = aACRt(New General Model in Transient State)where,a = m + (EU+ EH EUEH)r, andIf is defined as the percentage of total particle genera-tion deposited on exposed surfaces, then the surface particledeposition rate can be expressed as a percentage of total roompartic
28、le generation rate G as a logical simplification.Then,If t ,(New Model in Steady State)or,(Simplified New Model in Steady State)where,CST= Impurity concentration at any time in space (parts/volume)CSO= Initial impurity concentration in space (parts/volume)EU= Filters combined efficiency in AHU unitE
29、H= HEPA filter efficiency in room = Percentage of total particle generation deposited on exposed surfacesm = Ratio of Outside Air (OA) and Supply Air (SA), m = OA/SAACR = Air Change Rate (time1)VdCS1 EU()1 EH()RA RA EA Q+()CSdt=1 EU()1 EH()OACOGD()V+ dt+VdCS1 EU()1 EH()rSA SACSdt=1 EU()1 EH()mSACOGD
30、()V+ dt+dCSmEUEHEUEH+()r+ACR CSdt=1 EU()1 EH()mACRCOGD()+ dt+dCSmEUEHEUEH+ r+CSACRdt=1 EU()1 EH()mCOGDACR-+ ACRdt+1 EU()1 EH()mGDCOACR-+ b=dCSbCo aCS- ACRdt=dCsbCoaCS-CSOCSTACR td0t=1a- Ln bCoaCS()CSOCSTACRt=Ln bCoaCS()CSOCSTa ACRt=LnbCoaCST()bCoaCSO- a ACRt=t1aACR-LnbCoaCS0()bCoaCST-=CSTCSOba-CoeaA
31、CRt ba-Co+=b 1 EU()1 EH()mGDCoACR-+=b 1 EU()1 EH()m1 ()GCoACR-+=CSTba-Co1 EU()1 EH()mGDCoACR-+mEUEHEUEH+ 1 m()+-Co=CSTba-Co1 EU()1 EH()mCO1 ()GACR-+mEUEHEUEH+ 1 m()+-= 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transact
32、ions 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. 90 ASHRAE TransactionsFURTHER DEVELOPMENT OF NEW MODEL IN ALTERNATIVE FORMS AND ANALYSISNew Model Expr
33、essed in Configurations Other Than Primary Air Loop AHU SystemThe new model was established based on the simple, mostcommonly HVAC configuration-Primary Loop Air Handlingsystem. For more complex HVAC configurations, such as,Primary loop plus secondary make up unitPrimary loop plus secondary AHU unit
34、 with dual returnsPrimary loop plus secondary AHU unit and tertiarymakeup unit with dual returnsthe Research Team found that in respect to the cleanroomitself, the new model will keep the identical form as for theconventional primary loop system, except for the CombinedEfficiency EU. For those more
35、complex confirmations, thisCombined Efficiency EUwill count for all of the filtersinstalled in series relationship, that is:where, And i is one of the filters installed in AHU systems witha total of n filters in series following the airflow path.New Model Expressed in Room Air Leakage under Pressuri
36、zation or DepressurizationBased on the cleanroom airflow balance, and the AHUunit airflow balance, m = OA/SA, r = RA/SA, SA = OA+RA, SA= EA+Q+RA, so that, OA = EA+Q.If we define: Q/SA = n, as the Leakage Air/Supply AirRatio, and EA/SA = e, as the Exhaust Air/Supply Air Ratio,then m = e + n. If n 0,
37、then m e; if n 0, then m e. Thatis to say, when air leaks from a cleanroom to surrounding area,the cleanroom is under “pressurization”, outside air intake toAHU unit should be more than the total exhaust air from thespaces served by this AHU unit. When air leaks from sur-rounding area to a cleanroom
38、, the cleanroom is under “depres-surization”, outside air intake to AHU unit must be less thanthe total exhaust air from the spaces served by this AHU unit.As m and e are both known during design phase or can bemeasured during operational phase, then n can be calculatedby n = m e. For example, for a
39、 cleanroom room in respect tosupply air SA, if OA percentage is 25%, EA percentage is 20%,then the positive leakage air rate percentage is 5% of the totalsupply air SA to the room. Replacing m with e n in the newmodel, we obtain, where, e = Ratio of Exhaust Air (EA) and Supply Air (SA), EA/SAm = Rat
40、io of Outside Air (OA) and Supply Air (SA), OA/SAn = Ratio of Room Leakage Air (Q) and Supply Air (SA), Q/SA, calculated by formula of n = m eNew Model Expressed in Dimensionless FormAn easy way to express the new model in dimensionlessform is:The ratio of CST/C0can be considered as cleanroom Rela-t
41、ive Cleanliness versus outdoor air. To obtain a lower ratio ofCST/C0, design engineers can focus on the rest of the variablesto identify the cost-effective ways to achieve.Significance of Key Variables on Room Air CleanlinessThe new model developed above is differential-equationbased, it has more va
42、riables and parameters than any existingmodels. The new model includes many important elementsthat were previously ignored or omitted by previous models.Graphical charts (examining the impacts of some parameters)are illustrated in the Figure 2, Chart-1 through Chart-6, asexamples which show each var
43、iable and parameter in newmodel individually varied to test its impact to the room particleconcentration change. Chart-7 illustrates the impact of indoorparticle generation rate on the required air change rate (ACH),comparing with the old IEST RP-CC012.1 recommendedACH without consideration of indoo
44、r particle generationlevel. The significance of each of these variables can beanalyzed individually. If there is a differential change of a vari-able, its significance can be valued by the respective change onthe room article concentration to exam its impact throughpartial differential operations, d
45、iscussion about uncertaintyand sensitivity analysis can be founded in the referenced arti-cle (Sun 2009).CONDUCTING TESTS AND ANALYZING EXPERIMENT RESULTSSUMMARY OF FINDINGSExtensive testing has been conducted in multiple labs inorder to receive more comprehensive and representativefindings. The maj
46、or findings supporting the research objec-tives are summarized as follows:CSTba-1 EU()1 EH()mCoGDACR-+mEUEHEUEH+ 1 m()+-=EU11Ei()i=1n=CST1 EU()1 EH()en()CO1 ()GACR-+en()EUEHEUEH+ 1 e n+()+-=CSTCO-1 EU()1 EH()mGDCOACR-+mEUEHEUEH+ 1 m()+-= 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 reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permis
