Measurement of airflow and pressure characteristics of a fan built in a car ventilation system

a Co esponding au ho : poko ny.j@ me. u b .cz Measu emen o ai low and p essu e cha ac e is ics o a an buil in a ca en ila ion sys em Jan Poko ný1,a, Filip Poláek1, Miloš Foj lín1, Jan Fiše 1 and Mi osla Jícha1 1B no Uni e si y o Technology, Facul y o Mechanical Enginee ing, Ene gy Ins i u e, Technická 2896/2, B no, Czech Republic Abs ac . The aim o his s udy was o iden i y a se o ope a ing poin s o a an buil in en ila ion sys em o ou es ca . These ope a ing poin s a e gi en by he an p essu e cha ac e is ics and a e de ined by a p essu e d op o he HVAC sys em (ai duc s and en s) and olume ic low a e o en ila ion ai . To co e a wide ange o p essu e d ops si ua ions, ou cases o en laps se up we e examined: (1) all en s opened, (2) only cen al en s closed (3) only cen al en s opened and (4) all en s closed. To co e a di e en olume ic lows, he each case was measu ed a leas o ou di e en speeds o an de ined by he an ol age. I was obse ed ha he p essu e di e ence o he an is p opo ional o he an ol age and s ongly depends on he h o ling o he ai dis ibu ion sys em by he se ings o he en s laps. In case o ou es ca we iden i ied co ela ions be ween olume ic low a e o en ila ion ai , an p essu e di e ence and an ol age. These co ela ions will acili a e and educe ime cos s o he ollowing expe imen s wi h his es ca . 1 In oduc ion The amoun o en ila ion ai lowing in o a ca cabin is impo an pa ame e , which a ec s cabin indoo en i onmen and HVAC sys em pe o mance. I should be conside ed al eady in ea ly s age o ca design and has o be known be o e he new ca se ies a e p oduced o be able e alua e he e iciency o cabin’s en ila ion, which has close ela ionship o he d i ing com o and sa e y. ”The sa e y o occupan s o educe d i e a igue, ensu e good isibili y and main ain com o is key o he success ul design o such sys ems” [1]. To suppo he ea ly s age o he HVAC sys em design ou esea ch eam has been de eloped a Vi ual Tes ing S and o he Ca Cabin [2], which is a compu a ional model o he calcula ion o he ca cabin hea load in a ious en i onmen s. The i s e i ica ion was done [3], howe e s ill he model needed o imp o e he ad ec ion pa . The amoun o en ila ion ai and i s dis ibu ion o he cabin was impo an ask how o imp o e he ad ec ion scheme o he model. In his pape he measu emen o amoun o sucked ai in o he cabin is p esen ed o wide ange o manual en s se ings. The en ila ion ai dis ibu ion in au oma ic mode du ing a ious seasons is con empo a y published in he pape [4]. Aim o his s udy was o ind co ela ions be ween he se up o an speed, olume ic low a e and p essu e di e ence o he an o a ious en ila ion se ups. This s udy we pe o med on he ou es and he ob ained esul s will be used o subsequen expe imen s wi hin wo k package Human Cabin Cen ed Design o he Czech Republic p ojec Jose Božek Compe ence Cen e o Au omo i e Indus y. 2 Me hods The e a e plen y o me hods how o measu e ai low a es h ough en ila ion sys em, which a e based on a ious physical phenomena, see [5]. Fo example he e a e mechanical low me e s, di e en ial p essu e lowme e s (o i ice pla es, Ven u i ubes, e c.) o eloci y low me e s (Pi o ubes, anemome e s: ane, ho wi e, lase Dopple , e c.). In ou expe imen we u ilized he las men ioned ca ego y eloci y low me e s, which iden i y he low a e by measu ing he ai eloci y and in eg a ing o e he low a ea. To measu e ai eloci y we used a couple o ho bulb anemome e p obes manu ac u ed by Tes o. The me hod is mo e accu a e when he s illing channel is used o ensu e ha ai low h ough a duc ing sys em will be e enly dis ibu ed wi h de eloped ai low egimes. The olume ic low a e ࢂሶ was a e wa ds calcula ed om he knowledge o c oss-sec ion a ea ࡿ o his channel and mean eloci y ࢜ ഥ . Du ing he expe imen s we also kep knowledge o he ba ome ic p essu e and ai empe a u e o be able o calcula e ai densi y ࣋ and hus also mass low a e ࢓ሶ h ough he en ila ion sys em, see Equa ion 1. ݉ሶ ൌܸ ሶήߩൌݒҧήܵήߩൌσሺ௩೔ ೙ ೔సభ ήௌ೔ሻ ேήߩ (kg·s-1) (1) DOI: 10.1051/ C Owned by he au ho s, published by EDP Sciences, 201 / 0 0 (201 ) 201 epjcon EPJ Web o Con e ences , 1 61 6 6 1 100 4 4 2 2 97 97 This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License 4.0, which pe mi s dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.  EPJ Web o Con e ences 2.1 Expe imen al se up The s illing channel added in on o he es ca suc ion inle was a plas ic ube o diame e 150 mm and leng h 2000 mm. Because he suc ion inle is loca ed unde he hood, we had o emo e i o be able p epa e he expe imen al se up, see Figu e 1 on he le ; on he igh side, he e is shown he ing a ound he c oss-sec ion a ea, which was loca ed in he middle o he ube’s leng h o loca e measu ing axis un . The ing we used o be e Figu e 1. Expe imen al se up. On he le : S illing channel du ing measu emen s. On he igh : he ing a ound he c oss- sec ion a ea o ube whe e he p obes we e placed (axis loca ion +x=45 ° +y= 135° om he ho izon al plane). Du ing he es he ca engine was idling o ensu e ha he an be powe ed wi h same powe as du ing eal a ic codi ions (supply ol age abou 13.8 V). Fo he measu emen o he ol age we used classical mul i-me e and o he measu emen o he p essu e di e ence we used Tes o 435 wi h a build p essu e senso . The alue o p essu e di e ence was eco ded manually en imes o he each es case. In he case o ai eloci y we eco ded measu emen au oma ically wi h sampling a e 1 Hz. We used wo ho bulb anemome e s connec ed o da a logge Tes o 350 M/XL. The each poin was measu ed o i e seconds and o he i e seconds we e used o he changing posi ion o he p obes in he ube. The posi ion o he p obes in he ube was se up acco ding o he heo y see [6] using he Equa ion 2. ݎ௜ൌܴήඥሺʹ݅െͳሻ ʹܰ Τ (m·s-1) (2) whe e R = 75 mm is diame e o he ube and N = 7 is numbe o a e sed lines in each hal -axis. Using his he app oach i allows o spli c oss-sec ional a ea o ube (176.7 mm2) in o 28 sec o s o equal a eas. Fo each sec o was measu ed one ai eloci y a i s cen e and hen we calcula ed he mean eloci y ࢜ ഥ simply as he a e age alue o hese local eloci ies. Addi ionally we measu ed he cen al poin [x=0, y=0], so all oge he i coun s 29 poin s o measu e o each es case. Howe e cen al poin no en e o he calcula ion o mean eloci y, because his poin s ep esen s annulus o ze o a ea (see Figu e 2). 2.2 Unce ain ies To ob ain a ep esen a i e iew abou a low egime inside he ube he ai eloci y a he c oss-sec ion‘s a e se lines (x, y axis) was measu ed epea edly. Fo each en ila ion se up i was a e sed i e imes o be able e alua e unce ain y componen s: o ype A. In his case he s anda d unce ain y ࢛࡭ can be associa ed wi h s anda d de ia ion o he mean. We applied his o he ai eloci y and ob ained he Equa ion 3. ݑ஺ൌ݇௨஺ ήටଵ ௡ሺ௡ିଵሻσሺݒ௜െݒҧሻଶ ௡ ௜ୀଵ (m/s) (3) whe e ࢑࢛࡭ ൌ૚Ǥ૝ is sa e y ac o o numbe o epe i ion ࢔ൌ૞. Then he expanded unce ain y o he 95 % con idence in e al is UA = 2·uA, whe e 2 is he ele an co e age ac o , see [7]. The unce ain y o ype B co esponding o he accu acy o measu emen de ices was neglec ed in his s udy. 2.3 Measu emen schedule The measu emen schedule included es cases combining ou di e en an speed (le el 2, 3, 4, 6) and ou di e en con igu a ions o closing en laps (“all en s opened”, “only cen al en s closed”, “only cen al en s opened” and “all en s closed”). We manipula e only wi h he laps o he dashboa ds en s ( wo a cen e and wo a sides) and wi h wo unnel en s o ien ed o he ea sea . These con igu a ion we e chosen wi h espec o he co e a wide ange o en ila ion sys em ope a ional condi ions. We measu ed all men ioned combina ion which ises ma ix 4x4 o examined es cases, see Table 1. 3 Resul s In he i s subchap e , he e is demons a ed how he olume ic and mass low a e we e e alua ed h ough he measu emen o ai eloci ies, which was he mos ime consuming p ocedu e o he expe imen . The p ocedu e is desc ibed o he es case S6-C1, see Table 1. In he second subchap e a summa y o esul s o all es cases is p esen ed and in he hi d one he esul ing p essu e- olume cha ac e is ics a e plo ed. 3.1 Tes p ocedu e o pa icula es case Du ing he es case S6-C1 ( an speed se up o 6 h le el and only cen al en s opened) he mean ai empe a u e was 25.4 °C and ba ome ic p essu e was 102 375 Pa, which implies ai densi y ߩ = 1.194 kg/m3. Vol age on he an was du ing he es almos cons an U = 11.4 V and p essu e di e ence a an p = 325.2 Pa was qui e s able as well (mos ly i was be ween 320 and 328 Pa). 02097-p.2 EFM 2015 Figu e 2. Tes case S6-C1. Measu ed eloci y p o ile in he ube wi h 28 equal a eas o c oss-sec ion. The measu ed alues o ai eloci ies in speci ic poin s o he c oss-sec ion a ea o s illing channel a e plo ed in Figu e 2. The e a e illus a ed loca ion and measu ed alues o all 28 sec o s, which implies he mean eloci y ݒҧ ൌ͸ǤͲͲേͲǤ͸Ͳ݉Ȁݏ. Hence by using he Equa ion 1 he olume ic low a e was ܸሶ = 106.0 l/s and mass low a e ݉ሶ =126.5 g/s. In Figu e 3, he e a e shown he eloci y p o iles o he x-axis and y-axis, wi h e o ba s showing he alue o expanded s anda d de ia ion ܷ஺ o each poin . I can be seen ha he ai low in he s illing channel had almos ully de eloped u bulen low egime wi h he maximal eloci y in he cen e. Fo example, in he case S6-C1, he highes alue o ܷ஺ was o he poin on y-axis y=+72 mm, whe e = 5.71 ± 1.40 m/s. The indi idual ime samples we e [7.17, 5.91, 4.02, 5.63, 5.81]. On he o he hand he lowes s anda d de ia ion was o he poin on x- axis x=+66 mm, whe e = 5.27 ± 0.18 m/s. The indi idual ime samples we e [5.21, 5.53, 5.22, 5.27, 5.13]. These unce ain ies a he han ou line some luc ua ions in low s uc u e inside a ube, ep esen s he p ecision o he measu emen a gi en poin . Figu e 3. Tes case S6-C1. Measu ed eloci y p o iles and he mean eloci y (poin = 0 mm no en e o i s calcula ion). 3.2 Fan and duc ing sys em cha ac e is ics All iden i ied ope a ing poin s a e no ed in Table 1. Each poin is de ined by an se up, en laps con igu a ion, alue o olume ic low a e and p essu e di e ences. F om his able we e plo ed g aphs in Figu e 4 and Figu e 5, ep esen ing ope a ing poin s o he an and duc ing sys em, espec i ely. The an cha ac e is ics exp esses how he speci ic an o ou es ca wo ks du ing a ious in ensi y o en ila ion, which depends on he an speed and hus also on he ol age o he an. We de i ed o he an a eg ession model using he polynomials o 2nd o de see Figu e 4. I can be seen ha in he case S2 he olume ic low a e is e y sensi i e o he p essu e di e ence. F om his eason we no measu e he case S1, because he ange o p essu e di e ence is in o de o uni s o Pascals which is close o he accu acy o he Tes o de ice. On he o he hand wi h he highe speed o an he ange o p essu e di e ences espec o he ange o olume ic low a e is inc easing and hus he in e p e a ion o esul s should be mo e eliable. Table 1. Ma ix o all measu ed es cases Fan se up C1: all opened C2: only cen al closed C3: only cen al opened C4: all closed S2: U = 4.8 V 42.3 l/s 99.8 Pa 39.8 l/s 110.2 Pa 35.9 l/s 112.4 Pa 23.9 l/s 124.3 Pa S3: U = 6.1 V 53.0 l/s 148.4 Pa 48.2 l/s 160.3 Pa 45.3 l/s 165.3 Pa 31.5 l/s 192.1 Pa S4: U = 7.6 V 75.4 l/s 208.8 Pa 69.2 l/s 214.8 Pa 65.0 l/s 227.8 Pa 47.1 l/s 269.5 Pa S6: U = 11.4 V 106.0 l/s 322.5 Pa 97.5 l/s 368.8 Pa 91.9 l/s 395.1 Pa 76.4 l/s 492.6 Pa Figu e 4. Fan cha ac e is ics 02097-p.3 EPJ Web o Con e ences Figu e 5. Duc ing sys em cha ac e is ics The duc ing sys em cha ac e is ics exp ess he ela ion be ween he di e ence p essu e and he olume ic o he speci ic duc ing sys em. The mos in e es ing cu es a e C1 and C4, which de ines a h esholds o possible low condi ions h ough he gi en duc ing. We de i ed o he he duc ing also a eg ession model based on polynomials o 2nd o de see Figu e 5. The main esul s o his expe imen a e iden i ied eg ession cu es o he an and duc ing sys em inside ou es ca . I we plo he an cu es and duc ing sys em cu es o e hemsel es, we ob ain om hei in e sec ions he ope a ing poin s, which we al eady measu ed. Howe e he ad an age is ha i is possible o change he ype o an on he cu en duc ing, o also i is possible o change duc ing sys em o he cu en an o ob ain new ope a ing poin s. 4 Discussion The expe imen s shown ha he ol age o he an is linea ly p opo ional o he use se ings o an con olle , as well as he olume ic low a e see Figu e 6. I is e iden ha he speed o an no de ine olume ic low a e unambiguously, because he e is s ong in luence o how he en ila ion duc ing sys em is h o led by en laps se ings. This is he main bene i o ou expe imen , o conside ed p essu e - olume ic low a e cha ac e is ics o en ila o (Figu e 4) and duc ing sys em (Figu e 5) o be able co e all possible se ings o en ila ion con ol (de os , au oma ic mode, closing o manual se up o en laps, e c.) Figu e 6. Speed o an e sus olume ic low a e. The posi ioning o he p obes we e done manually, so he e o o as ening and posi ing could occu and sligh ly in luence pa icula measu ed poin . S ill he s anda d de ia ion (con idence in e al 95 %) du ing he measu emen we e accep able gene ally and he anemome e based me hod shows a good epea abili y. The highes s anda d de ia ions o eloci y we e nea he wall. We assume ha is due o he highe eloci y g adien a he wall, which caused ha he measu ed alues we e mo e sensi i e o p ecise p obe loca ion. The highe accu acy o he measu emen could be achie ed by inc easing o epe i ion o measu emen om 5 o example o 10. Howe e we chose n = 5 due o he ac ha he measu emen would be oo ime consuming in case o 10 and he ݇௨஺ ac o would no change signi ican ly. The s illing channel imp o ed he s abili y o low a lo and educe he luc ua ions which occu di ec ly be o e he suc ion inle o he en ila ion sys em. The peak in he eloci y p o ile (Figu e 3) a he cen e o channel c oss-sec ion can be explained by he ac ha all poin s we e measu ed by wo p obes simul aneously, howe e in case o cen al poin i was used only one p obe. Ne e heless he alue o cen al poin was no conside ed in he calcula ion o he mean eloci y and olume ic low a e. 5 Conclusion We iden i ied an and en ila ion duc ing sys em cha ac e is ics desc ibing co ela ions be ween olume ic low a e, se up o en ila ion sys em and p essu e di e ence a he an. Ob ained co ela ions a e alid only o an and duc ing sys em o ou es ca , bu e en so hey will allow e alua e olume ic low a e o ai di ec ly om he an p essu e di e ence. This app oach will acili a e and educe ime cos s o he consequen expe imen s wi h ou es ca subs an ially. Acqui ed knowledge is possible o use no only o clima ic chambe es s, bu e en o eal a ic condi ions measu emen s. Acknowledgemen The esea ch was suppo ed by he p ojec Reg. No. FSI- S-14-2355 o he B no Uni e si y o Technology; he p ojec LO1202 Ne me Cen e Plus wi h he inancial suppo om he Minis y o Educa ion, You h and Spo s o he Czech Republic unde he "Na ional Sus ainabili y P og amme I" and he p ojec o he Jose Božek Compe ence Cen e o Au omo i e Indus y TE01020020. Re e ences 1. S. Daly. Au omo i e ai -condi ioning and clima e con ol sys ems. 1s ed. Ox o d: Bu e wo h Heinemann (2006). 2. J. Poko ny, J. Fise , M. Jicha, Ad ances in enginee ing so wa e 76 (2014) 02097-p.4 EFM 2015 3. J. Poko ny, Fise , M. Jicha, EPJ Web o Con e ences 92, 02072 (2015) 4. M. Foj lin, M. Planka, J. Fiše , J. Poko ný, M. Jícha. Ai low Measu emen o he Ca HVAC Uni Using Ho -wi e Anemome y. P oceedings o Expe imen al Fluid Mechanics 2015, P ague, Czech Rep. ( o be published). 5. H. D. Good ellow, E. Täh i. Indus ial Ven ila ion Design Guidebook. Pa 12. Expe imen al echniques, (2001). 6. M. Pa elek. Expe imen ální me ody echnice p os edí. 3. yd. B no : Akademické naklada els í CERM (2007). 7. P.Nmeek. Nejis o y mení. 1. yd. P aha : eská spolenos p o jakos (2008). 02097-p.5