An integrated sensor technology for measurements of specific heat ratio of combustion gases in supersonic flow

Wang Tiejin，Chen Jun

（1.China Academy of Aerospace Aerodynamics，Beijing 100074，China；2.State Key Lab.for Turbulence Complex and Systems，Department of Mechanics and Engineering Science，College of Engineering，Peking University，Beijing 100871，China）

An integrated sensor technology for measurements of specific heat ratio of combustion gases in supersonic flow

Wang Tiejin1，Chen Jun2

（1.China Academy of Aerospace Aerodynamics，Beijing 100074，China；2.State Key Lab.for Turbulence Complex and Systems，Department of Mechanics and Engineering Science，College of Engineering，Peking University，Beijing 100871，China）

We present an integrated probe system for measuring the specific heat ratioγof combustion gases by combining Pitot tube and laser beam tracking system.The key benefit of the measurement system is the elimination of the complicated gas chromatography apparatus.The pressure ratio and the angle of refraction of a laser beam across the normal shock wave generated by Pitot tube are simultaneously measured with this system.The angle of refraction of the laser beam is expected to be accurately recorded by the laser tracking system consisting of two CCD cameras and a mobile spatial coordinate measuring system（MSc MS）.Moreover，the density ratio across the shock wave is deduced with Gladstone-Dale relation and Snell’s law.The specific heat ratioγis accordingly quantified by its relation to the pressure ratio and the density ratio across the shock wave.This technique may effectively reduce the time taken to obtainγ.Uncertainty estimation shows that the uncertainty brought by the laser beam tracking system and Pitot tube measurement are limited within 5.5%for any Mach number if the stagnation temperature is lower than 1000K.Feasibility analysis present that the technique is physically realizable but a few technical issues need to be addressed before it can be implemented.

specific heat ratio；combustion gas；supersonic flow；Pitot tube；laser

0 Introduction

High temperature（up to 1700K［1］）gases such as the combustion reaction products of the pure oxygen or air with the fuels，e.g.，hydrogen，methane，ethanol and kerosene can be adopted as the working media in hypersonic wind tunnel experiments.However，the complexity of combustion process makes it difficult to accurately predict the concentration of each component of the combustion gases.In experiments，the characteristic parameters，such as the specific heat and density，are important but cannot be predefined.For example，the specific heat ratio，the ratio of the molar heat capacity at constant pressure to the molar heat capacity at constant volume，is crucial in evaluating the Mach number，as one of the performance indices of a supersonic wind tunnel nozzle.

Commonly there are two methods to obtainγ—one istheempiricalmethodusually adopted in the flow field calculation of engine exhaust system［2］；another isthecompositionanalysismethodused in calibration of the thermal structure of a hypersonic wind tunnel［3］.The empirical method calculates for the components of combustion gases based on an empirical combustion efficiency formula under a similar combustion condition.Thus the resultantγis an approximation under a similar flow condition and its uncertainty is intrinsically unpredictable.The composition analysis method calculates according to the quantities of the components of the combustion gases by analyzing the gas chromatography at different temperatures.Complex gas chromatography is essential for accurate measurement of the quantities of all components of the combustion gases［4］.

In addition to the complicated facilities of chromatography analysis，it is a real challenge for one to probe the temperature of the combustion gases in the heat hypersonic flow［5］.Hence，this method has not been widely employed.

This paper presents a new technique for efficiently obtaining the accurateγof the combustion gas in hypersonic flow field.This technique is proposed to utilize the relations of the fluid densities before and after a normal shock wave induced by the Pitot tube.Feasibility of this method is testified in details.The advantages of this method are the simplicity and economy of the instruments and the high precision of the resultant data.

1 Principle of measuringγ

When Pitot tube is set in a supersonic flow，the fluid over a small area covering the tip of the tube undergoes deceleration，which makes the flow transform from supersonic to subsonic，as shown in Figure 1.Therefore，a normal shock wave is effectively formed in front of the tube［6］，and the pressure ratio across the shock wave，p2／p1andρ1／ρ2，can be determined by the normal shock wave relations，that are

wherep1andp2are total pressures just upstream and downstream of the normal shock wave，respectively；ρ1is the static density upstream of the normal shock wave，ρ2is the density downstream of the normal shock wave；γis the specific heat ratio of combustion gases，andM1is the Mach number of the hypersonic flow upstream of the normal shock wave.By applying Equations（1）and（2）［7-8］，we established the formulation of the present method.

We first take the pure air as the typical fluid.The principle of calibrating the hypersonic nozzle with pure air can be extrapolated to the multi-component cases.For the pure air flow，γis a predetermined quantity.p1andp2can be directly measured with the Pitot tube［8］.Hence，M1is determined by Equation（1）.However，for the hypersonic nozzle with multi-components，bothM1andγare usually the quantities to be measured.It is necessary to combine Equations（1）and（2）with the measuredp1，p2andr，in order to obtain the specific heat ratio，γ.

Two approximations in high Mach number flows are taken as the conditions to estimate the density ratio，r.They are：

（1）The hypersonic flow across a normal shock is steady；

（2）The combustion gases are perfect gases.

The first approximation is justified based on the fact that the higher the Mach number of the coming flow is，the more steady the normal shock wave generated by the flow is［9］.For example，for the gases ofγ=1.4，the shock wave is always stable ifM1＞1.48.The Mach numbers considered in the present study are far above the instability bound.Hence，the first approximation is rational for most cases of the supersonic flows encountered.

The second approximation——the perfect gas requirement——is the requirement for the applicability of Equations（1）and（2）.We will discuss the errors brought by the imperfect gas condition of the flow in Sec.4.

It is easy to measurep1andp2by using Pitot tube and pressure sensors，but measurement ofris challenging.The following section focuses on the principle and feasibility of the method to measure the density ratiorin detail.

2 Feasibility of density ratio measurement

We developed an optical method to measure the density ratio，r，and the principle is presented in Figure 1.A laser beam is generated from the inner of the Pitot tube，and penetrates the normal shock wave with the angle of incidence，α2.The beam is refracted at an angle ofα1after passing through the normal shock wave.

Fig.1 Schematic of measurement ofρ2／ρ1with a Pitot tube in supersonic flow圖1 在超聲速流動中采用皮托管測量ρ2／ρ1

Snell′s law gives the relationship between angles of incidence and refraction for a light impinging on an interface between two media with different indices of refraction，as follows［10-11］，

wheren1andn2are the refractive indices for the gases upstream and downstream of the normal shock wave，respectively；α1andα2are the angle of refraction for the laser beam and the angle of incidence，respectively.

According to the Gladstone-Dale formula，then1andn2can be determined by the functions of densities，［12］，

whereKis an coefficient varying with the features of the gas.For the same type of gas，the effect of temperature is small，and thusKis usually considered as a constant for the same gas.Combination of Equations（4）and（5）leads to the following equation，

Considering Equation（3）and（6），the reciprocal of density can be expressed as follows：

whereα1andα2are measured by the laser beam tracking system.A container connecting to Pitot tube collects the gas past the shock.A gas density meter（for example，the Sarasota ID900）set up in the container measures the gas density，ρ2.Hence，the refractive index，n2，is obtained using the Gladstone-Dale formula（Equation（5））.It is noted that the Pitot pressure measurement and the gas collection must be operated separately.The density ratiorand the inflow densityρ1are obtained from Equation（7）.

The accuracy ofrmeasurement is based on the fact that the normal shock wave in front of the Pitot tube is so thin that the normal shock wave can be considered as a zero-thickness interface between two fluids with different densities［8］.Since the thicknessof the shock wave is on the order of the magnitude of 10-4mm［13］，this consideration is rational for conventional hypersonic wind tunnels except for lowdensity wind tunnels.

We take the apparatus in CAAA for example to examine the feasibility of the measurement system.The two CCD cameras of MegaPlusII ES4020（Redlake Company）with the resolution of 2048×2048 pixels（4.19Mpixel，integration time：104μs）are supposed to be applied to record the refracted laser beam and the normal shock wave.The schematic graph for the camera recording system is shown in Figure 2.The field of view for each camera is 15mm ×15mm and the spatial resolution for each pixel is 7.3μm.Considering the width of the laser beam is 50～200μm，the uncertainty for the recorded laser beam is below 15μm.The mobile spatial coordinate measuring system（MSc MS）works to locate the two cameras with the position uncertainty of 2μm［14-15］.Thus the system uncertainty for measuring the position of the laser beam is limited within ±20μm.Taking the FD-3 hypersonic wind tunnel in CAAA for example，the test section for the experiment of the compressible mixing layer is 300mm（width）×40mm（height）×70mm（depth）［16］.

Fig.2 The schematic graph for the laser beam tracking system. The dash line indicates the coordinate system supporting CCD cameras，Camera-1 and Camera-2.Camera-1 is located at（x1，y1）in the coordinate system of MSc MS to capture the image of the refracted laser beam.Camera-2 is placed at（x2，y2）in the coordinate system of MSc MS to record the position where the laser beam is refracted by the normal shock wave圖2 激光光束跟蹤系統(tǒng)示意圖。虛線代表支撐CCD相機(jī)（Camera-1和-2）的坐標(biāo)系統(tǒng)。Camera-1位于MSc MS系統(tǒng)的（x1，y1），記錄折射激光位置；Camera-2位于MSc MS系統(tǒng)的（x2，y2），記錄正激波光線發(fā)生折射位置

Thus the distance between the two cameras is supposed to be 150mm，the uncertainty of the measured angle by Camera-1 is lower than 1.5×10-4in radian number，and the corresponding relative error is under 0.01%.

With primary analysis of the relation between the Mach number and the density ratio as well as the pressure ratio，we found that the measurement is effective at 4＜M1＜8 and 1.1＜γ＜1.4.The specific heat ratio is relatively distinct in these ranges while the angle of refraction is sensitive to the density change.High accuracy of the specific heat ratio measurement can therefore be easily obtained under these test conditions.

3 Uncertainty analysis of measurement of specific heat ratio

Regarding the measurement ofr=1／ra=ρ2／ρ1 as shown in Equation（7），one can see that thoughris determined by bothα1andα2.Whenα2approaches 0°（or 90°），α1is also close to 0°or（90°），and sinα2／sinα1would bring large uncertainty to Equation（7）.The analysis of the angle of refraction under the typical flow conditions（4＜M1＜8）suggests that 2°～10°would give relatively accurateα1.

According to Equation（2），the variance ofrwith different Mach numbers and specific heat ratios can be calculated，as shown in Figure 3 and Figure 4.Figure 3 shows the profiles of the density ratio，r，vs.the Mach number，M1.Note thatrincreases with the increase ofM1.This trend is enforced by decreasingγ.Figure 4 is the profiles ofrvs.γat variousM1.It is shown that，at higher Mach number，ris more sensitive toγ.

The error propagation for density ratio is analyzed here.From Equation（6），we have

Fig.3 Density ratio，r，vs.Mach number，M1圖3 密度比與馬赫數(shù)的關(guān)系

Fig.4 Density ratio，r，vs.γ圖4 密度比與γ的關(guān)系

and then we have the estimation of the error propagation forra，

Considering the relative error of about 0.01% of Camera-1 obtained in the previous section，the relative error of the density ratio is about 0.5% based on the error relation，Equation（9）.The accuracy of the density-ratio-measuring system is satisfactory.

From Equation（2），we have

For estimating the uncertainty brought from the uncertainty in measuringrandM1，the relative error for the specific heat ratioγis derived as follows，

whereδγ／γandδM1／M1are the relative errors of the density ratiorand that of the Mach numberM1，respectively.Estimation of the relative error ofγis obtained from Equation（12）with Equation（2），（10）and（11）.If the uncertainty ofrandM1are supposed to beδr／r～δM1／M1～0.5%in the range of 4＜M1＜8 and 1.1＜γ＜1.4，δγ／γis always less than 0.4%.Hence，the present method may provide an acceptable accuracy.

Let us analyze the error induced by the perfect gas assumption.Zebbiche（2011）reported that the error by using perfect gas relations is lower than 5% if the stagnation temperatureT0is lower than 1000K for any value of Mach number，and T0may reach up to 3000K when Mach number is lower than 2.0［17］.Hence the present technique using the prefect gas model is effective to estimateγin the hypersonic flow ofT0＜1000K.The gas at high temperatures becomes calorically imperfect gas；bothCpandγvary with temperature.In order to measureγfor high temperature hypersonic flow，one has to introduce a nonlinear implicit equation forTandM1and this takes more time for calculation and for data processing.By using the calorically imperfect gas model，for example theHigh Temperature model［17］，the present measurement system is readily applicable for high temperature flows.

4 Discussion

The core of the technique is the measurement of the density ratio，r，which needs to be validated and verified at first.From Equation（1）and（2），one can see that if the ratio of the specific heat is predefined the two equations become two redundant functions for Mach number.The Mach number can be solved with the measured pressure ratio by Equation（1），and it also can be obtained with the measured density ratio by Equation（2）.This provides a way of calibrating the density ratio measurement system.The calibration procedure may be summarized as follows：（a）obtain the Mach number at the point of interest by probing the pressures before and after the normal shock wave（see Equation（1））；（b）then calculate the Mach number with the density ratio derived from the angle of fraction of the laser beam（see Equation（2）and（7））；（c）finally validate and calibrate the density measurement system by comparing of the two Mach numbers.

In comparison with the gas chromatography technique，the present method may provide real time measuring results.The uniqueness of the pres-ent technique is the ability to obtain the Mach number and the density ratio of a hypersonic flow at the same time with a physically realizable system.With this measurement system the specific heat ratio of gases over the whole flow field can be obtained.In addition，the post processing of these results is more efficient and more extensive than that of gas chromatography，i.e.both spatial and temporal signals can be obtained.This technique is expected to have a wide range of applications in high speed flow measurements，though a few technical issues are yet to be addressed.For example，the structure and the size of the tube，the light path design，the combination of the tube and the laser beam，the thermal protection issues，the effects of the light source and the accurate record of the angle of refraction should be considered.

Several critical techniques in the measurement system are discussed here.First，since the laser beam emerging from the light source in Pitot tube passes through the stagnation and the post-shock regions，the gradient of the density will deflect the beam.The deflection angle can be obtained by measuring the displacement of the laser beam at the presence of the shock wave and simulating the geometrical optics in compressible flows.Secondly，the quality of the laser beam is a crucial factor for obtaining accurate refractive angle of the laser beam.The properties of light such as its reflection，dispersion，and refractive indices are determined by the media，i.e.the combustion gases in the present case.Hence，the frequency and the energy power of the laser beam should be well designed ahead.Thirdly，the physical structure of the Pitot tube should be designed as small as possible to attain the high spatial resolution，but it should still be able to generate a regular normal-shock-wave——an essential flow region making anidealrefraction of the laser beam.Finally，thermal protection of the tube is also a challenge to measurements in high temperature flows.At present this technique is under development in the laboratory.

5 Conclusion

（1）A novel，accurate and simpleγ-measuring technique utilizing laser beam tracking and pressure measurement techniques has been presented.The key benefit of the new technique is the elimination of the complicated gas chromatography system.

（2）The pressures before and after the normal shock wave formed by Pitot tube are measured concurrently with the density ratio by laser beam tracking system.This improves the ease of use and reduces the time to obtain the final data for calibration of a hypersonic wind tunnel nozzle.

（3）Uncertainty estimation is illustrated with the examples of the available apparatus.The uncertainties of the technique are acceptable——the upper bound of the uncertainty is assured below 5.5% for any Mach number withT0＜1000K.

（4）Feasibility analysis shows that the technique is physically realizable with several technical issues to be addressed.The present technique can be applied for other purposes，such as the realtime measurement of hypersonic flows.

［1］Felderman E J，Akers D T，Liu C T，et al.Hypersonic wind tunnel nozzle survivability for T＆E final report：T＆E／S＆T project［R］.AEDC-TR-06-17，2006

［2］Liu G.Principle of liquid rocket engines（in Chinese）［M］.Space Publishing Company，1993.

［3］Huebner L D，Rock K E，Voland R T，et al.Calibration of the Langley 8-foot high temperature tunnel for hypersonic airbreathing propulsion testing［R］.AIAA 96-2197，1996.

［4］Amirav A，Gordin A，Tzanani N.Supersonic gas chromatography／mass spectrometry［J］.Rapid Commun.Mass Spectrom，2001，15：811-820.

［5］Roseberry C M.Arc-heated gas flow experiments for hypersonic propulsion［D］.University of Texas，2005.

［6］Balachandran P.Fundamentals of compressible fluid dynamics［M］.Prentice-Hall of India Pvt.Limited，2006.

［7］Ames research staff.Equations，tables，and charts for compressible flow，Tech rep［R］.NACA-TR-1135，1953.

［8］Tropea C，Yarin A L，F(xiàn)oss J F.Springer handbook of experimental fluid mechanics［M］.Springer-Verlag，2007：216-228.

［9］Matsuo K，Yaga M，Mochizuki H.Stability of normal shock waves in diffusers［J］.AIAA Journal，1987，25：1515-1517.

［10］Klein M V.Optics［M］.New York：Wiley，1970.

［11］Driscoll W G.Handbook of optics［M］.New York：McGraw-Hill，1978.

［12］Tropea C，Yarin A L，F(xiàn)oss J F.Springer handbook of experimental fluid mechanics［M］.Springer-Verlag，2007：473-474.

［13］Anderson J D J.Fundamental of aerodynamics［M］.New York：McGraw-Hill Company，1984.

［14］Franceschini F，Galetto M，Maisano D，et al.Mobile spatial co-ordinate measuring system（MSc MS）——introduction to the system［J］.Int J Prod Res，2009，47：3867-3889.

［15］Franceschini F，Maisano D，Mastrogiacomo L.Mobile spatial coordinate measuring system（MSc MS）and CMMs：A structured comparison［J］.Int J Adv Manuf Technol，2009，42：1089-1102.

［16］Wang T J，Shi X T，Chen J，et al.Multiscale structures in compressible turbulent mixing layers［J］.Mod Phys Lett B，2010，24：1429-1432.

［17］Zebbiche T.Thermodynamics-interaction studies-solids，liquids and gases［C］.November，2011，Ch.Effect of Stagnation Temperature on Supersonic Flow Parameters with Application for Air in Nozzles Zebbicc，2011，16：421-444.

王鐵進(jìn)（1970-），男，河北故城人，研究員。研究方向：空氣動力學(xué)實驗與測試技術(shù)、實驗設(shè)備、空氣動力學(xué)應(yīng)用技術(shù)。通訊地址：北京7201信箱57分箱（100074）。E-mail：tiej701＠163.com

陳 軍（1968-），男，吉林省吉林市人，北京大學(xué)湍流與復(fù)雜系統(tǒng)國家重點實驗室高級工程師。研究方向：湍流物理機(jī)制及流動結(jié)構(gòu)、氣動光學(xué)效應(yīng)、高速流動測量技術(shù)、流動顯示技術(shù)、熱對流物理機(jī)制、風(fēng)工程。通訊地址：北京大學(xué)工學(xué)院力學(xué)與工程科學(xué)系（100871）。E-mail：jun＠pku.edu.cn

（編輯：張巧蕓）

1672-9897（2014）06-0073-07doi：10.11729／syltlx20140026

超聲速流動燃燒氣體比熱比的集成傳感測量技術(shù)

王鐵進(jìn)1，陳 軍2
（1.中國航天空氣動力技術(shù)研究院，北京 100074； 2.北京大學(xué)工學(xué)院力學(xué)與工程科學(xué)系湍流與復(fù)雜系統(tǒng)國家重點實驗，北京 100871）

介紹了為測量燃燒氣體的比熱比而設(shè)計的包含了皮托管和激光光束跟蹤系統(tǒng)的集成測量系統(tǒng)。該系統(tǒng)通過測量壓力比和穿過皮托管前正激波的激光折射角來獲得比熱比，從而免使用復(fù)雜而昂貴的光相色譜儀器。激光光束折射角通過兩臺CCD攝像頭和移動空間坐標(biāo)測量系統(tǒng)（MSc MS）來準(zhǔn)確記錄激光光束折射角。密度比通過Gladstone-Dale關(guān)系和Snell律來獲得。由此，通過所獲得的跨激波的壓力比和密度比可以定量確定比熱比。這一技術(shù)可有效降低比熱比測量時間。本文對激光光束跟蹤系統(tǒng)和皮托管所帶來的不確定性進(jìn)行了分析，結(jié)果表明在總溫不超過1000K時系統(tǒng)不確定性低于5.5%。可行性分析表明該技術(shù)可以實現(xiàn)，但仍有部分技術(shù)問題需要在實施前解決。

比熱比；燃燒氣體；超聲速流動；皮托管；激光

O35，TJ530

：A

WangTJ，ChenJ.Anintegratedsensortechnologyformeasurementsofspecificheatratioofcombustiongasesinsupersonicflow.JournalofExperimentsinFluidMechanics，2014，28（6）：73-79.王鐵進(jìn)，陳軍.超聲速流動燃燒氣體比熱比集成傳感測量技術(shù).實驗流體力學(xué)，2014，28（6）：73-79.

date：2014-03-15；Revised date：2014-09-02

National Nature Science Fund（11172006，10572004，90716008），MOST 973 Project（2009CB724100）

Chen Jun，E-mail：Jun＠pku.edu.cn