research article experimental assessment of a variable

Research ArticleExperimental Assessment of a Variable Orifice Flowmeter forRespiratory Monitoring

Giuseppe Tardi Carlo Massaroni Paola Saccomandi and Emiliano Schena

Universita Campus Bio-Medico di Roma Via Alvaro del Portillo 21 00128 Rome Italy

Correspondence should be addressed to Emiliano Schena eschenaunicampusit

Received 22 March 2015 Accepted 9 July 2015

Academic Editor Kourosh Kalantar-Zadeh

Copyright copy 2015 Giuseppe Tardi et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Accurate measurement of gas exchanges is essential in mechanical ventilation and in respiratory monitoring Among the largenumber of commercial flowmeters only few kinds of sensors are used in these fields Among them variable orifice meters (VOMs)show some valuable characteristics such as linearity good dynamic response and low costThis paper presents the characterizationof a commercial VOM intended for application in respiratory monitoring Firstly two nominally identical VOMs were calibratedwithin plusmn10 Lsdotminminus1 to assess their metrological properties Furthermore experiments were performed by humidifying the air toevaluate the influence of vapor condensation on sensorrsquos performances The condensation influence was investigated during twolong lasting trials (ie 4 hours) by delivering 4 Lsdotminminus1 and 8 Lsdotminminus1 Data show that the two VOMsrsquo responses are linear andtheir response is comparable (sensitivity difference of 14 RMSE of 150 Pa) their discrimination threshold is lt05 Lsdotminminus1 andthe settling time is about 66msThe condensation within the VOM causes a negligible change in sensor sensitivity and a very slightdeterioration of precisionThe good static and dynamic properties and the low influence of condensation on sensorrsquos responsemakethis VOM suitable for applications in respiratory function monitoring

1 Introduction

During artificial ventilation and function respiratory moni-toring the use of flowmeters is required to perform an accu-rate and continuousmonitoring of gas exchangeTheir outputis used to estimateminute volume and tidal volume thereforetheir accuracy is crucial to perform a correct diagnosis and toavoid common side effects related to uncorrectedmechanicalventilation [1] Moreover flowmeters play a crucial role inthe noninvasive assessment of metabolic gas exchange ofmechanically ventilated patients by indirect calorimetry [2]and in noninvasive techniques for cardiac output monitoring[3 4]

In all these applications flowmeters have to accuratelymonitor the patients breathing or the flow pattern deliveredby mechanical ventilators Hence they have to fulfill strictrequirements in terms of both dynamic and static metro-logical properties These sensors are often used to monitorpatients during long lasting ventilation therefore they haveto be able to reject both the influence of gas compositionand the influence of vapor condensation [5] This issue is

crucial because (i) the gas during mechanical ventilationcan experience large changes in its composition (ii) the gasexpired by patients has high relative humidity content thatcan result in vapor condensation and (iii) during invasivemechanical ventilation the inspiratory gases are humidifiedby devices placed within the inspiratory limb of the breathingcircuit between the ventilator and the patients

Among the large number of commercial flowmetersdifferential pressure flowmeters are the most commonly usedin industrial processes (in particular square edged concen-tric orifice meters [6]) and are widely used in mechanicalventilation and in respiratory monitoring Their workingprinciple is based on an orifice plate placed within the pipe inwhich the gas flows The presence of the restriction causes apressure drop across the orifice plate according to Bernoullirsquoslaw Although fixed orifice meters have several advantages(eg they are simple to manufacture robust and accurateand has good dynamic properties) their use in mechanicalventilation and in respiratory monitoring is limited by theirnonlinear response [7] Several patents and studies havebeen focused on the design of orifice flowmeters with novel

Hindawi Publishing CorporationJournal of SensorsVolume 2015 Article ID 752540 7 pageshttpdxdoiorg1011552015752540

2 Journal of Sensors

geometrical features of the orifice (eg fractal shaped [8] andslotted orifice [9]) but the only solution allowing overcomingthe concerns related to their nonlinearity is represented bythe design of variable orifice meters (VOMs) Patents basedon different configurations (eg flap made of flexible sheetmaterials like plastic or stainless steel) have been proposed[10ndash12] In these transducers the flap cut in the middleand placed into the gas stream creates a passage area thatenlarges with the flowrate increase The consequence is thatthe resistance of the orifice is almost constant with theflowrate hence the calibration curve is linear and the rangeof measurement is wider than fixed orifice meters [13]

Other two solutions allow obtaining a linear resistanceflowmeterThese flowmeters called pneumotachographs arebased on a resistance which consists of a number of parallelcapillary tubes or of a finewiremesh proposed by Fleisch andLilly respectively [14 15] Their working principle is under-pinned by Hagen-Poiseuillersquos law therefore the relationshipbetween the pressure drop across the resistance and theflowrate is linear Although these sensors are robust and havegood metrological properties their main drawback is relatedto the risk of vapor condensation within their resistancewhich causes high measurement error [16] In order toovercome this concern a more complex configuration basedon the heating of the capillary or of the wire mesh has beenproposed

VOMs may minimize this concern thanks to the config-uration based on a single restriction

They are used on several commercial mechanical ven-tilators (eg the ventilators Hamilton-C1 Hamilton-C2Hamilton-C3 Hamilton-T1 Hamilton-S1 Hamilton-G5 andGALILEO by Hamilton Medical and the ventilators Bird8400ST and Vela C by CareFusion) Despite the large use ofVOMs in the field of mechanical ventilation and respiratorymonitoring scientific literature does not present studiesfocusing on their metrological characterization and on theinfluence of vapor condensation on their response at least toour knowledge

The aim of this work is threefold (i) to perform thestatic calibration of a commercial available VOM intendedfor breathing monitoring The sensitivity the discriminationthreshold and the characteristic of bidirectionality of thissensor are investigated Moreover the calibration curve oftwo nominally identical sensors is compared to analyzetheir reproducibility also the paper will (ii) experimentallyassess the step response of the VOM and (iii) investigate theinfluence of vapor condensation on the VOMrsquos response

2 Sensors Description Design andTheoretical Background

Orifice meters consist of an orifice plate placed in line apipeline where gas flows The gas which flows through therestriction created by the orifice generates a pressure drop(Δ119875) between the upstream and downstream sides of theplate The sensorrsquos output (ie Δ119875) is usually measured by adifferential pressure sensor connected to two pressure statictaps placed upstream and downstream the plate [17]

The input-output relationship of these sensors is obtainedby considering Bernoullirsquos equation valid

12sdot 120588 sdot V2 +120588 sdot 119892 sdot ℎ + 119901 = cost (1)

where V is the gas flow velocity 119901 the pressure ℎ the heightrespect on a reference line 119892 the gravity acceleration and 120588the fluid density Equation (1) is valid under the simplifyinghypotheses of one-dimensional flows and incompressiblenonviscous and isothermal fluid conditions

The Δ119875 generated between upstream and downstreamsections of the flow obstruction according to Bernoullirsquosequation allows the estimation of the volumetric flow rate by

120601 =1198602

radic1 minus (11986021198601)2sdot radic

2 sdot Δ119875120588

(2)

where 120601 is the volumetric flow rate 1198601

is the inlet area of theflowmeter and 119860

2

is the passage area of flow obstructionThe input-output relationship of the orifice meter can

be easily obtained by (2) and it is a root square function(Δ119875120572120601

2) The nonlinearity of the response is one of the most

important limitations of these kinds of sensors and it isovercome by the development of VOMs

The VOM tested in this work is the ldquoSpiroQuant Prdquo flowsensor produced by EnviteC (Honeywell) Figures 1(a)ndash1(e)

It consists of a flap placed inline a pipeline where gasflows The gas flowing through the restriction constituted bythe flap generates Δ119875 between upstream and downstreamsections of the blades representing the output of the sensorThe symmetrical geometry of the variable orificemeter allowsmeasuring bidirectional flow rate

The flap cut in the middle and placed into the gas streamcreates an opening which widens with the 120601 increase thebigger the flow the bigger the passage area of obstructionin which the fluid flows and vice versa (Figures 1(a)ndash1(d))The increase of 119860

2

with 120601 entails a linearization of the input-output relationship reported in (2)

3 Experimental Setup and Results

The experimental assessment of the above-mentioned VOMhas been carried out with a threefold aim (i) to obtain thecalibration curve of two nominally identical flowmeters andto investigate theirmetrological static properties (ii) to assessthe settling time during dynamic tests and (iii) to investigatethe influence of humidity on the flowmeter output

31 Static Calibration The static calibration of two nominallyidentical variable orifice meters was carried out to evaluatethe sensorsrsquo sensitivity the symmetry of their response andthe sensors reproducibility

The VOM (Figure 2(a)) was connected through a pipeto an airflow controller (F-201C-FBC-22-V by BronkhorstHigh-Tech Figure 2(b)) A differential pressure sensor(163PC01D75 by Honeywell range of measurement plusmn623 Paaccuracy plusmn015 of full scale output Figure 2(e)) was

Journal of Sensors 3

Flow

Variable orifice

Pressure static taps

(a)

(e)

(b)

(c) (d)

Figure 1 (andashd) Restriction area of the variable orifice meter from (a) to (d) the increment of the restriction area with volumetric flowrate isshown (e) Picture of the variable orifice meter and key components

(a)

(b)

(c)

(d)

(e)

Figure 2 Experimental setup used to perform static calibration (a)variable orifice meter (b) air flow controller (c) digital oscilloscope(d) DC Power Supply and (e) differential pressure sensor

connected to the two pressure static taps to measure the Δ119875across the variable orifice The pressure sensor was suppliedby a constant voltage of 500 plusmn 001 V by a DC Power Supply(GPS-3030 GW Instek Figure 2(d)) The voltage output ofthe pressure sensor was displayed on the digital oscilloscope(DL1520 by Yokogawa Figure 2(c))

Two nominally identical VOMs (SpiroQuant P byEnviteC Honeywell) were calibrated using the setup shownin Figure 2

Six sets of experiments were performed on each flowme-ter by delivering constant and dry airflows ranging fromminus10 Lsdotminminus1 to +10 Lsdotminminus1 in step of 05 Lsdotminminus1 Every 5minutes gas temperature at the flowmeter output sectionwas monitored by a type K thermocouple Gas temperatureranged in 260 plusmn 20∘C during the whole set of experiments

The input-output relationship (ie Δ119875 versus 120601) for thetwo VOMs is shown in Figures 3(a) and 3(b) All the resultsare reported as mean plusmn uncertainty the uncertainty wascalculated considering a Student reference distribution with5 degrees of freedom and a level of confidence of 95 (asrecommended in [18]) The relationship between Δ119875 and 120601was investigated by using a linear regression analysis The

sensorsrsquo sensitivity was estimated as the slope of the bestfitting line (ie Δ119875 = 5306 sdot 120601 and Δ119875 = 5232 sdot 120601 for the twotested VOMs ) For each calibration curve the coefficient ofdetermination 1198772 was calculated

Figure 3 shows that the sensorsrsquo calibration curve iswell described by a linear model The good linear fitting isconfirmed by the high value of 1198772 (099 for both sensors) Asa consequence the sensitivity should be considered constantin the whole range of calibration (ie plusmn10 Lsdotminminus1) and equalto 5306 PaLsdotminminus1 and 5232 PaLsdotminminus1 for the two VOMs

Moreover the sensorrsquos linearity is kept within a widerrange of measurement In fact in order to investigate thisfeature we carried out trials by delivering airflow rangingfrom minus20 Lsdotminminus1 up to +20 Lsdotminminus1 in step of 2 Lsdotminminus1 thegood linear fitting was confirmed by the high value of 1198772 (ie099) obtained by fitting the data with a linear model

The good reproducibility between the two sensors is wit-nessed by both the low difference between their sensitivities(lt14) and the low value of the rootmean squared error (ie150 Pa) which was calculated as

RMSE = radic119872

sum

119894=1

(Δ1198751 (119894) minus Δ1198752 (119894))2

119872

(3)

whereΔ1198751

(119894) is the pressure drop for 119894th flowrate experiencedby the first sensor under test and Δ119875

2

(119894) is the pressure dropfor 119894th flowrate experienced by the second sensor under test119872 is the number of flow rate used to calibrate the sensor (inour case119872 = 40)

Moreover in order to analyze the performance of thesensor in terms of bidirectionality the gas flow was deliv-ered in the two opposite directions to simulate inspiratoryand expiratory flows The bidirectionality was quantitativelyassessed by the use of the following index of asymmetry (119878)proposed in [19]


minus15 minus10 minus5 0 5 10 15

60

40

20

0

minus20

minus40

minus60

ΔP

(Pa)

R2= 099

ΔP = 5306120601

120601 (Lmiddotminminus1)

(a)

ΔP

(Pa)

R2= 099

ΔP = 5232120601

60

40

20

0

minus20

minus40

minus60



(b)

Figure 3 (a) and (b) Experimental data (dots) and best fitting line (continuous lines) of the two tested VOMsThe equation of the calibrationcurve for each sensor and the 1198772 value are also reported

119878

=

119873

sum

119894=1

10038161003816100381610038161003816(1003816100381610038161003816Δ119875

exp(119894)1003816100381610038161003816minus10038161003816100381610038161003816Δ119875

insp(119894)10038161003816100381610038161003816) (1003816100381610038161003816Δ119875

exp(119894)1003816100381610038161003816+10038161003816100381610038161003816Δ119875

insp(119894)10038161003816100381610038161003816)10038161003816100381610038161003816

119873

(4)

where Δ119875exp(119894) is the pressure drop for the 119894th flowratevalue delivered in the direction of expiration and Δ119875insp(119894)is the pressure drop for the 119894th flowrate value blowing in theopposite direction119873 is the number of flowrate values used tocalibrate the sensor for each direction (in our case 119873 = 20)The asymmetry index 119878 ranges from 0 to 1 by definition thelower its value is the better the symmetry of sensor responseis The 119878 values were 014 and 018 for the two sensors undertest

32 Step Response The dynamic response of the flowmeterwas investigated by a dedicated setup

The VOM was connected to an onoff valve (Series 821-22NCMatrix) with a very short switching time (ie 045msfor opening and 019ms) already employed in previous stud-ies [16 19 20] Flowrate steps were delivered by connectingthe valve to a pressurized source and quickly opening orclosing the valversquos orifice The valve was controlled by afunction generator (AFG313 by Sony Tektronix) Pressuredrop across the orifice was transduced into a voltage bya differential pressure sensor (163PC01D75 by Honeywellrange of measurement plusmn623 Pa accuracy plusmn015 of full scaleoutput) The valve and the pressure sensor were suppliedby a DC Power Supply (IPS2302A by ISO-TECH) A dataacquisition board (DAQ NI USB-6009 by National Instru-ments) connected to a Host PC recorded the voltage outputof the pressure sensor A custom made LabVIEW VI wasimplemented to record signals from DAQ at 1 kHz samplingfrequency All recorded data were postprocessed inMATLABenvironment

The step response of the flowmeter under test is deter-mined by two contributions the contribution related to theVOM which transduced 120601 into Δ119875 and the contributionrelated to the pressure sensor which transduces Δ119875 into avoltage output signal

75 76 77 78Time (s)

Volta

ge (V

) minus32

minus34

minus36

minus38

Vinfin

V0

ST

plusmn5ΔV

Figure 4 Step response of the flowmeter trend of the pressuresensor output when a flowrate step is delivered the estimation ofthe settling time is also shown

Sensor dynamic response showed oscillations around thesteady state (Figure 4) The settling time (ST) defined as thetime required for the process output to reach and remaininside a bandwhose width is equal toplusmn5 of the total changewas evaluated (see Figure 4)

The voltage total change Δ119881 was calculated as

Δ119881 = 119881infin

minus1198810 (5)

where 1198810

and 119881infin

are the voltage output of the pressuresensor at the step instant and at the steady state respectivelyExperiments to assess the step response of the sensor wererepeated seven times (Table 1) The ST value calculated asmean plusmn uncertainty was 66 plusmn 5ms

33 Influence of Condensation on Sensor Output Gas flowrate with high content of water vapor (100 of relativehumidity) was delivered to investigate the influence of thehumidity on VOM response This issue is relevant in thefields of artificial ventilation and respiratory monitoring Inmechanical ventilation the airflow delivered to the patient ishumidified by specific devices (eg heated wire humidifiers)At the output of the humidifier airflow has high relativehumidity (in many cases it is saturated) [21 22] also theair expired by the patients has high relative humidity and


Table 1 Settling time of the sensor estimated during the seven trials

Trial

ST[ms]

ST[ms]

mean plusmnuncertainty

1 683

66 plusmn 5

2 6953 6994 7135 5836 5867 686

(a)

(b)

(c)

(d)

(e)(f)

(g)

Figure 5 Setup used in condensation influence assessment (a) airflow controller (b) heated wire humidifier (c) differential pressuresensor (d) DC Power Supply (e) DAQ (f) Host PC and (g) variableorifice meter

in many cases it is saturated Hence it is very importantto know the influence of water vapor condensation on theperformances of the flowmeter in long lasting ventilationThe condensation of water vapor within the restrictionof the VOM may entail a measurement error caused byan unpredictable change in the pneumatic resistance Thisphenomenon causes significant measurement error whenFleisch pneumotachographs are employed [23] Thereforechanges of VOM response and its precision during longlasting trial were experimentally assessed Figure 5 shows theexperimental setup employed to investigate the influence ofrelative humidity on sensor output

An air flow controller (F-201C-FBC-22-V by BronkhorstHigh-Tech Figure 5(a)) is connected to heated wire humid-ifier (MR850 by Fisher amp Paykel Healthcare Figure 5(b))through a limbof a breathing circuit A second limbof breath-ing circuit connects the humidifier to the VOM (Figure 5(g))The VOMrsquos output is monitored by a differential pressuresensor (163PC01D75 by Honeywell range of measurementplusmn623 Pa accuracy plusmn015 of full scale output Figure 5(c))supplied by a DC Power Supply (GPS-3030 GW InstekFigure 5(d)) The output voltage of the pressure sensor isrecorded by a data acquisition board (DAQ NI USB-6009 byNational Instruments Figure 5(e)) connected to a Host PC(Figure 5(f))The humidifier is set to obtain the following gasthermohygrometric conditions gas temperature of 40 plusmn 1∘Cand saturation

Long lasting experiments (ie 4 hours) were carriedout by delivering two constant volumetric flowrates (ie8 Lsdotminminus1 and 4 Lsdotminminus1) During each trial the output of

the pressure sensor was acquired with a sample frequency of10Hz and VOM output was calculated (Figure 6(a)) In orderto investigate if the condensation causes a change in sensorrsquossensitivity the four hours of trial was divided in 24 intervals(each interval lasts 10 minutes) and the mean value of thesensorrsquos output was calculated for each interval (Figures 6(b)and 6(d)) Furthermore in order to analyze the condensationinfluence on the sensorrsquos precision the standard deviation ofthe sensor response was calculated for each interval (Figures6(c) and 6(e))

During long lasting trials performed at the two flowratesvapor condensates within the sensor (Figure 6(f)) but thesensor sensitivity did not experience significant changes themean value of the sensor response is almost constant forthe 24 intervals as shown in Figures 6(b) and 6(d) Thestandard deviation showed a slight increase during the time(Figures 6(c) and 6(e)) for instance the trials performed at4 Lsdotminminus1 show a standard deviation of 066 Pa at the firstinterval (first 10 minutes of trials) and 069 Pa at the lastinterval (last 10 minutes of experiments) This indicates aslight deterioration of precision but it must be noted thatconsidering the calibration curve of the sensor (Figure 3)a standard deviation increase of 003 Pa corresponds to avariation of 6mLsdotminminus1 that can be considered acceptablefor application in mechanical ventilation and in breathingmonitoring

Summing up results demonstrated that the formationof condensation causes a negligible change in sensor sen-sitivity the precision of the sensor experiences a slightdeterioration

4 Discussions and Conclusions

Mechanical ventilation and breathing monitoring are veryimportant practice to mechanically assist patients or to per-form essential diagnoses During both artificial ventilationand function respiratory monitoring the use of flowmeters isrequired to perform an accurate and continuous monitoringof gas exchange Their output is used to estimate minutevolume and tidal volume therefore their accuracy is crucialto perform a correct diagnosis and to avoid common sideeffects related to uncorrected ventilation [24] By flowmetersit is possible to monitor the volumes of gases exchange by thepatients

Since strict criteria must be fulfilled (ie high sensitivitygood accuracy bidirectionality low pneumatic resistancelow volumes added to breathing circuit short response andlarge range of flat frequency response) only few kinds offlowmeters are used in these fields The most employedflowmeters are hot wire anemometers fixed and variableorifice meters Fleisch pneumotachographs and ultrasonicflowmeters [7]

Several studies focus on the performances of hotwire anemometers fixed orifice meters Fleisch pneumota-chographs and ultrasonic flowmeters and on their applica-tion on respiratory monitoring field [16 25ndash28]

Although the first patent on VOM dates back to the 1970sand several mechanical ventilators are equipped with this


1050 15 20 2543

432434436438

44

(b)

(a)

(c) (e) (f)

Interval number

1050 15 20 25Interval number

015

02

025

006

0065

007

0075

008


4Lmiddotminminus1

8Lmiddotminminus1

8Lmiddotminminus1

ΔP

(Pa)

(d)

22222224226228

23


4Lmiddotminminus1

ΔP

(Pa)

ΔP

(Pa)

ΔP

(Pa)

Time (h)0 05 1 15 2 25 3 35 4

42

43

44

45ΔP

(Pa)

Figure 6 (a) Sensor output during 4 hours of trial at 8 Lsdotminminus1 (b) mean values of sensor output for each interval at 8 Lsdotminminus1 (c) standarddeviation of sensor output for each interval at 8 Lsdotminminus1 (d)mean values of sensor output for each interval at 4 Lsdotminminus1 (e) standard deviationof sensor output for each interval at 4 Lsdotminminus1 (f) picture of the sensor at the end of the 4-hour trial The formation of condensation withinthe sensor is evident

kind of sensor studies regarding its metrological propertiesand the problem of water condensation are lacking

The present work focuses on the static and dynamic char-acterization and also on the investigation of condensationinfluence on the output of a commercial VOM

Experiments show that the response of this sensor is lin-ear hence it has a constant sensitivity and it shows excellentbidirectionality in a wide range of flowrates (plusmn10 Lsdotminminus1)by comparing the performances of two nominally identicalsensors their response shows good reproducibility (thedifference between their sensitivity was 14) Moreoverthe tested VOM shows quiet good dynamic properties thesettling time was 66 plusmn 5ms This value is higher than otherdifferential-producing flowmeters for instance studies showthat Fleisch pneumotachograph and orifice meter have aresponse time slightly higher than 2ms [16 19] Lastly longlasting experiments show that the sensorrsquos sensitivity did notchange and its precision shows a very low deteriorationdespite the formation of vapour condensation

Sensor under test presents many advantages for respira-tory functionmonitoring low dead space (7mL) good sensi-tivity and low discrimination threshold (lt05 Lsdotminminus1) quietgood step response and good bidirectionality and its per-formances are not significantly deteriorated by condensation

Therefore in comparison with other solutions VOMs showsome advantages with respect to the fixed orifice metersthey have a linear response they are more robust andcheaper than hot wire anemometers [29] moreover theyare not prone to sensitivity changes due to condensationdifferently from Fleisch pneumotachographs which requiread hoc solutions to minimize this concern such as the useof an electric resistance These features make VOM reliablefor measurements also in other medical applications Forinstance this kind of sensor may be employed in the fieldof noninvasive methods for cardiac output estimation [3] inapplication related to the monitoring of human intestinalgases [30] and in other industrial processes in particularwhen bidirectional air flow rate must be measured

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Authorsrsquo Contribution

Giuseppe Tardi and Carlo Massaroni equally contributed tothis work


References

[1] J H T Bates M J Turner C J Lanteri B Jonson and P DSly ldquoMeasurement of flow and volumerdquo in Infant RespiratoryFunction Testin J Stocks P D Sly R S Tepper and W JMorgan Eds chapter 5 John Wiley amp Sons New York NYUSA 3rd edition 1996

[2] KMakita J F Nunn and B Royston ldquoEvaluataion ofmetabolicmeasuring instruments for use in critically ill patientsrdquo CriticalCare Medicine vol 18 no 6 pp 638ndash644 1990

[3] S Cecchini E Schena M Notaro M Carassiti and S SilvestrildquoNon-invasive estimation of cardiac output in mechanicallyventilated patients a prolonged expiration methodrdquo Annals ofBiomedical Engineering vol 40 no 8 pp 1777ndash1789 2012

[4] P J Peyton D Thompson and P Junor ldquoNon-invasive auto-mated measurement of cardiac output during stable cardiacsurgery using a fully integrated differential CO

2

Fick methodrdquoJournal of Clinical Monitoring and Computing vol 22 no 7 pp285ndash292 2008

[5] H Jewitt and G Thomas ldquoMeasurement of flow and volume ofgasesrdquo Anaesthesia amp Intensive Care Medicine vol 13 no 3 pp106ndash110 2012

[6] RWMiller ldquoIntroduction to the differential producerrdquo in FlowMeasurement EngineeringHandbook RWMiller Ed chapter7McGraw-Hill New York NY USA 3rd edition 1996

[7] E Schena C Massaroni P Saccomandi and S Cecchini ldquoFlowmeasurement in mechanical ventilation a reviewrdquo MedicalEngineering amp Physics vol 37 no 3 pp 257ndash264 2015

[8] A Abou El-Azm Aly A Chong F Nicolleau and S BeckldquoExperimental study of the pressure drop after fractal-shapedorifices in turbulent pipe flowsrdquo Experimental Thermal andFluid Science vol 34 no 1 pp 104ndash111 2010

[9] G L Morrison D Terracina C Brewer and K R HallldquoResponse of a slotted orifice flow meter to an airwatermixturerdquo Flow Measurement and Instrumentation vol 12 no3 pp 175ndash180 2001

[10] J J Osborn ldquoVariable orifice gas flow sensing headrdquo US Patent4083245 1978

[11] D W Guillaume M G Norton and D F DeVries ldquoVariableorifice flow sensing apparatusrdquo US Patent 4993269 February1991

[12] J Stupecky ldquoVariable area obstruction gas flow meterrdquo USpatent 5083621 1991

[13] M B Jaffe ldquoTechnical perspectives gas flow measurementrdquo inCapnography J S Gravenstein M B Jaffe N Gravenstein andD A Palus Eds pp 397ndash405 Cambridge University PressNewYork NY USA 2011

[14] A Fleisch ldquoDer Pneumotachograph ein Apparat zurGeschwindigkeitsregistrierung der Atemluftrdquo Pflugerrsquos Archivfur die gesamte Physiologie des Menschen und der Tiere vol 209no 1 pp 713ndash722 1925

[15] J C Lilly ldquoFlowmeter for recording respiratory flow of humansubjectsrdquoMethods in Medical Research vol 2 pp 113ndash121 1950

[16] E Schena G Lupi S Cecchini and S Silvestri ldquoLinearitydependence on oxygen fraction and gas temperature of a novelFleisch pneumotachograph for neonatal ventilation at low flowratesrdquoMeasurement vol 45 no 8 pp 2064ndash2071 2012

[17] Measurement of fluid flow by means of pressure differentialdevices inserted in circular cross-section conduits running fullPart2 Orifice plates EN ISO 5167-2 2003

[18] Joint Committee for Guides in Metrology ldquoEvaluation ofmeasurement datamdashguide to the expression of uncertainty inmeasurementrdquo Tech Rep JCGM 100 JCGM 2008

[19] E Schena S Cecchini and S Silvestri ldquoAn orifice meter forbidirectional air flow measurements influence of gas thermo-hygrometric content on static response and bidirectionalityrdquoFlow Measurement and Instrumentation vol 34 pp 105ndash1122013

[20] M Giorgino G Morbidoni E Tamilia F Taffoni D Formicaand E Schena ldquoDesign and characterization of a bidirectionallow cost flowmeter for neonatal ventilationrdquo IEEE SensorsJournal vol 14 no 12 pp 4354ndash4360 2014

[21] S N Ryan N Rankin E Meyer and R Williams ldquoEnergybalance in the intubated human airway is an indicator ofoptimal gas conditioningrdquo Critical Care Medicine vol 30 no2 pp 355ndash361 2002

[22] E Schena P Saccomandi C Ramandi and S Silvestri ldquoAnovel control strategy to improve the performances of heatedwire humidifiers in artificial neonatal ventilationrdquo PhysiologicalMeasurement vol 33 no 7 pp 1199ndash1211 2012

[23] M R Miller and T Sigsgaard ldquoPrevention of thermal andcondensation errors in pneumotachographic recordings of themaximal forced expiratory manoeuvrerdquo European RespiratoryJournal vol 7 no 1 pp 198ndash201 1994

[24] D R Hess ldquoApproaches to conventional mechanical ventilationof the patient with acute respiratory distress syndromerdquo Respi-ratory Care vol 56 no 10 pp 1555ndash1572 2011

[25] M Zimova-Herknerova and R Plavka ldquoExpired tidal volumesmeasured by hot-wire anemometer during high-frequencyoscillation in preterm infantsrdquo Pediatric Pulmonology vol 41no 5 pp 428ndash433 2006

[26] P Latzin L Sauteur CThamrin et al ldquoOptimized temperatureand deadspace correction improve analysis of multiple breathwashout measurements by ultrasonic flowmeter in infantsrdquoPediatric Pulmonology vol 42 no 10 pp 888ndash897 2007

[27] Y Snepvangers P De Winter H Burger H A BrouwersJ M Bogaard and K Van Der Ent ldquoCorrection factors foroxygen and flow-rate effects on neonatal Fleisch and Lillypneumotachometersrdquo Pediatric Critical Care Medicine vol 4no 2 pp 227ndash232 2003

[28] M J Heulitt J H Shirley and L Thurman ldquoAccuracy of smalltidal volume measurement comparing two ventilator airwaysensorsrdquo Journal of Pediatric Intensive Care vol 2 no 1 pp 33ndash38 2013

[29] H H Bruun Hot Wire Anemometry in Principles and SignalAnalysis Oxford University Press New York NY USA 1996

[30] J Z Ou C K Yao A Rotbart J G Muir P R Gibson and KKalantar-zadeh ldquoHuman intestinal gas measurement systemsin vitro fermentation and gas capsulesrdquoTrends in Biotechnologyvol 33 no 4 pp 208ndash213 2015

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



geometrical features of the orifice (eg fractal shaped [8] andslotted orifice [9]) but the only solution allowing overcomingthe concerns related to their nonlinearity is represented bythe design of variable orifice meters (VOMs) Patents basedon different configurations (eg flap made of flexible sheetmaterials like plastic or stainless steel) have been proposed[10ndash12] In these transducers the flap cut in the middleand placed into the gas stream creates a passage area thatenlarges with the flowrate increase The consequence is thatthe resistance of the orifice is almost constant with theflowrate hence the calibration curve is linear and the rangeof measurement is wider than fixed orifice meters [13]

Other two solutions allow obtaining a linear resistanceflowmeterThese flowmeters called pneumotachographs arebased on a resistance which consists of a number of parallelcapillary tubes or of a finewiremesh proposed by Fleisch andLilly respectively [14 15] Their working principle is under-pinned by Hagen-Poiseuillersquos law therefore the relationshipbetween the pressure drop across the resistance and theflowrate is linear Although these sensors are robust and havegood metrological properties their main drawback is relatedto the risk of vapor condensation within their resistancewhich causes high measurement error [16] In order toovercome this concern a more complex configuration basedon the heating of the capillary or of the wire mesh has beenproposed

VOMs may minimize this concern thanks to the config-uration based on a single restriction

They are used on several commercial mechanical ven-tilators (eg the ventilators Hamilton-C1 Hamilton-C2Hamilton-C3 Hamilton-T1 Hamilton-S1 Hamilton-G5 andGALILEO by Hamilton Medical and the ventilators Bird8400ST and Vela C by CareFusion) Despite the large use ofVOMs in the field of mechanical ventilation and respiratorymonitoring scientific literature does not present studiesfocusing on their metrological characterization and on theinfluence of vapor condensation on their response at least toour knowledge

The aim of this work is threefold (i) to perform thestatic calibration of a commercial available VOM intendedfor breathing monitoring The sensitivity the discriminationthreshold and the characteristic of bidirectionality of thissensor are investigated Moreover the calibration curve oftwo nominally identical sensors is compared to analyzetheir reproducibility also the paper will (ii) experimentallyassess the step response of the VOM and (iii) investigate theinfluence of vapor condensation on the VOMrsquos response

2 Sensors Description Design andTheoretical Background

Orifice meters consist of an orifice plate placed in line apipeline where gas flows The gas which flows through therestriction created by the orifice generates a pressure drop(Δ119875) between the upstream and downstream sides of theplate The sensorrsquos output (ie Δ119875) is usually measured by adifferential pressure sensor connected to two pressure statictaps placed upstream and downstream the plate [17]

The input-output relationship of these sensors is obtainedby considering Bernoullirsquos equation valid

12sdot 120588 sdot V2 +120588 sdot 119892 sdot ℎ + 119901 = cost (1)

where V is the gas flow velocity 119901 the pressure ℎ the heightrespect on a reference line 119892 the gravity acceleration and 120588the fluid density Equation (1) is valid under the simplifyinghypotheses of one-dimensional flows and incompressiblenonviscous and isothermal fluid conditions

The Δ119875 generated between upstream and downstreamsections of the flow obstruction according to Bernoullirsquosequation allows the estimation of the volumetric flow rate by

120601 =1198602

radic1 minus (11986021198601)2sdot radic

2 sdot Δ119875120588

(2)

where 120601 is the volumetric flow rate 1198601

is the inlet area of theflowmeter and 119860

2

is the passage area of flow obstructionThe input-output relationship of the orifice meter can

be easily obtained by (2) and it is a root square function(Δ119875120572120601

2) The nonlinearity of the response is one of the most

important limitations of these kinds of sensors and it isovercome by the development of VOMs

The VOM tested in this work is the ldquoSpiroQuant Prdquo flowsensor produced by EnviteC (Honeywell) Figures 1(a)ndash1(e)

It consists of a flap placed inline a pipeline where gasflows The gas flowing through the restriction constituted bythe flap generates Δ119875 between upstream and downstreamsections of the blades representing the output of the sensorThe symmetrical geometry of the variable orificemeter allowsmeasuring bidirectional flow rate

The flap cut in the middle and placed into the gas streamcreates an opening which widens with the 120601 increase thebigger the flow the bigger the passage area of obstructionin which the fluid flows and vice versa (Figures 1(a)ndash1(d))The increase of 119860

2

with 120601 entails a linearization of the input-output relationship reported in (2)

3 Experimental Setup and Results

The experimental assessment of the above-mentioned VOMhas been carried out with a threefold aim (i) to obtain thecalibration curve of two nominally identical flowmeters andto investigate theirmetrological static properties (ii) to assessthe settling time during dynamic tests and (iii) to investigatethe influence of humidity on the flowmeter output

31 Static Calibration The static calibration of two nominallyidentical variable orifice meters was carried out to evaluatethe sensorsrsquo sensitivity the symmetry of their response andthe sensors reproducibility

The VOM (Figure 2(a)) was connected through a pipeto an airflow controller (F-201C-FBC-22-V by BronkhorstHigh-Tech Figure 2(b)) A differential pressure sensor(163PC01D75 by Honeywell range of measurement plusmn623 Paaccuracy plusmn015 of full scale output Figure 2(e)) was


Flow

Variable orifice


(a)

(e)

(b)

(c) (d)


(a)

(b)

(c)

(d)

(e)










RMSE = radic119872

sum

119894=1

(Δ1198751 (119894) minus Δ1198752 (119894))2

119872

(3)

whereΔ1198751


2





60

40

20

0

minus20

minus40

minus60

ΔP

(Pa)

R2= 099

ΔP = 5306120601


(a)

ΔP

(Pa)

R2= 099

ΔP = 5232120601

60

40

20

0

minus20

minus40

minus60



(b)


119878

=

119873

sum

119894=1

10038161003816100381610038161003816(1003816100381610038161003816Δ119875

exp(119894)1003816100381610038161003816minus10038161003816100381610038161003816Δ119875

insp(119894)10038161003816100381610038161003816) (1003816100381610038161003816Δ119875

exp(119894)1003816100381610038161003816+10038161003816100381610038161003816Δ119875

insp(119894)10038161003816100381610038161003816)10038161003816100381610038161003816

119873

(4)





75 76 77 78Time (s)

Volta

ge (V

) minus32

minus34

minus36

minus38

Vinfin

V0

ST

plusmn5ΔV




Δ119881 = 119881infin

minus1198810 (5)

where 1198810

and 119881infin





Trial

ST[ms]

ST[ms]


1 683

66 plusmn 5

2 6953 6994 7135 5836 5867 686

(a)

(b)

(c)

(d)

(e)(f)

(g)














1050 15 20 2543

432434436438

44

(b)

(a)

(c) (e) (f)

Interval number


015

02

025

006

0065

007

0075

008


4Lmiddotminminus1

8Lmiddotminminus1

8Lmiddotminminus1

ΔP

(Pa)

(d)

22222224226228

23


4Lmiddotminminus1

ΔP

(Pa)

ΔP

(Pa)

ΔP

(Pa)

Time (h)0 05 1 15 2 25 3 35 4

42

43

44

45ΔP

(Pa)












References





2






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Flow

Variable orifice


(a)

(e)

(b)

(c) (d)


(a)

(b)

(c)

(d)

(e)










RMSE = radic119872

sum

119894=1

(Δ1198751 (119894) minus Δ1198752 (119894))2

119872

(3)

whereΔ1198751


2





60

40

20

0

minus20

minus40

minus60

ΔP

(Pa)

R2= 099

ΔP = 5306120601


(a)

ΔP

(Pa)

R2= 099

ΔP = 5232120601

60

40

20

0

minus20

minus40

minus60



(b)


119878

=

119873

sum

119894=1

10038161003816100381610038161003816(1003816100381610038161003816Δ119875

exp(119894)1003816100381610038161003816minus10038161003816100381610038161003816Δ119875

insp(119894)10038161003816100381610038161003816) (1003816100381610038161003816Δ119875

exp(119894)1003816100381610038161003816+10038161003816100381610038161003816Δ119875

insp(119894)10038161003816100381610038161003816)10038161003816100381610038161003816

119873

(4)





75 76 77 78Time (s)

Volta

ge (V

) minus32

minus34

minus36

minus38

Vinfin

V0

ST

plusmn5ΔV




Δ119881 = 119881infin

minus1198810 (5)

where 1198810

and 119881infin





Trial

ST[ms]

ST[ms]


1 683

66 plusmn 5

2 6953 6994 7135 5836 5867 686

(a)

(b)

(c)

(d)

(e)(f)

(g)














1050 15 20 2543

432434436438

44

(b)

(a)

(c) (e) (f)

Interval number


015

02

025

006

0065

007

0075

008


4Lmiddotminminus1

8Lmiddotminminus1

8Lmiddotminminus1

ΔP

(Pa)

(d)

22222224226228

23


4Lmiddotminminus1

ΔP

(Pa)

ΔP

(Pa)

ΔP

(Pa)

Time (h)0 05 1 15 2 25 3 35 4

42

43

44

45ΔP

(Pa)












References





2






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











60

40

20

0

minus20

minus40

minus60

ΔP

(Pa)

R2= 099

ΔP = 5306120601


(a)

ΔP

(Pa)

R2= 099

ΔP = 5232120601

60

40

20

0

minus20

minus40

minus60



(b)


119878

=

119873

sum

119894=1

10038161003816100381610038161003816(1003816100381610038161003816Δ119875

exp(119894)1003816100381610038161003816minus10038161003816100381610038161003816Δ119875

insp(119894)10038161003816100381610038161003816) (1003816100381610038161003816Δ119875

exp(119894)1003816100381610038161003816+10038161003816100381610038161003816Δ119875

insp(119894)10038161003816100381610038161003816)10038161003816100381610038161003816

119873

(4)





75 76 77 78Time (s)

Volta

ge (V

) minus32

minus34

minus36

minus38

Vinfin

V0

ST

plusmn5ΔV




Δ119881 = 119881infin

minus1198810 (5)

where 1198810

and 119881infin





Trial

ST[ms]

ST[ms]


1 683

66 plusmn 5

2 6953 6994 7135 5836 5867 686

(a)

(b)

(c)

(d)

(e)(f)

(g)














1050 15 20 2543

432434436438

44

(b)

(a)

(c) (e) (f)

Interval number


015

02

025

006

0065

007

0075

008


4Lmiddotminminus1

8Lmiddotminminus1

8Lmiddotminminus1

ΔP

(Pa)

(d)

22222224226228

23


4Lmiddotminminus1

ΔP

(Pa)

ΔP

(Pa)

ΔP

(Pa)

Time (h)0 05 1 15 2 25 3 35 4

42

43

44

45ΔP

(Pa)












References





2






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











Trial

ST[ms]

ST[ms]


1 683

66 plusmn 5

2 6953 6994 7135 5836 5867 686

(a)

(b)

(c)

(d)

(e)(f)

(g)














1050 15 20 2543

432434436438

44

(b)

(a)

(c) (e) (f)

Interval number


015

02

025

006

0065

007

0075

008


4Lmiddotminminus1

8Lmiddotminminus1

8Lmiddotminminus1

ΔP

(Pa)

(d)

22222224226228

23


4Lmiddotminminus1

ΔP

(Pa)

ΔP

(Pa)

ΔP

(Pa)

Time (h)0 05 1 15 2 25 3 35 4

42

43

44

45ΔP

(Pa)












References





2






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










1050 15 20 2543

432434436438

44

(b)

(a)

(c) (e) (f)

Interval number


015

02

025

006

0065

007

0075

008


4Lmiddotminminus1

8Lmiddotminminus1

8Lmiddotminminus1

ΔP

(Pa)

(d)

22222224226228

23


4Lmiddotminminus1

ΔP

(Pa)

ΔP

(Pa)

ΔP

(Pa)

Time (h)0 05 1 15 2 25 3 35 4

42

43

44

45ΔP

(Pa)












References





2






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










References





2






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









research article experimental assessment of a variable

Documents