Assessment of the Performance Potential for a Two-Pass Cross Flow Intercooler for Aero Engine Applications

ISABE-2013-1215

ASSESSMENT OF THE PERFORMANCE POTENTIAL FOR A TWO-PASS CROSS FLOW INTERCOOLER FOR AERO ENGINE APPLICATIONS

Xin Zhao Ph.D. student Chalmers University of Technology SE-412 96 Göteborg, Sweden [email protected]

Tomas Grönstedt Professor Chalmers University of Technology SE-412 96 Göteborg, Sweden [email protected]

Konstantinos G. Kyprianidis Rolls-Royce Defence Aerospace Bristol, United Kingdom [email protected]

Abstract h Heat transfer coefficient j Colburn j factor The performance potential for a K Loss coefficient for ducts two-pass cross flow intercooler has kloss Tube internal loss coefficient been estimated through an analysis ̇ Mass flow rate of a long range mission for a p Static pressure geared turbofan engine. The P Total pressure application of a set of CFD based T Temperature correlations allows the Density simultaneous coupled optimization Roughness of the tube wall of the intercooler conceptual Dynamic viscosity design parameters and the engine Polytropic efficiency design. The coolant air for the Thermal conductivity of air intercooler is ejected through a Re Reynolds number separate variable exhaust nozzle Pr Prantdl number which is used to optimize the Nu Nusselt number engine performance in cruise. By St Stanton number comparing the optimized intercooled BPR Bypass ratio geared engine with an optimized FPR Fan pressure ratio advanced non-intercooled geared HPC High pressure compressor engine, a reduction of 4.8% fuel IPC Intermediate pressure burn is observed. Compressor LPT Low pressure turbine

OPR Overall pressure ratio Nomenclature PR Pressure ratio a Major axis length of the SFC Specific fuel consumption elliptical tube SFN Specific thrust b Minor axis length of the TOC Top of climb elliptical tube NEWAC New Aero Engine Core Concepts D Diameter LEMCOTEC Low Emissions Core-Engine D Hydraulic diameter of any h Technologies internal passage f Friction factor

Introduction it is argued that it is beneficial to provide a relatively large Intercoolers have recently amount of intercooling at take-off received a considerable attention to allow for a compact engine as a means to improve aero engine design with a high OPR and a efficiency, in particular within reduced cooling flow need. In the European research projects cruise on the other hand, it is NEWAC and LEMCOTEC [1-8], and beneficial to reduce the within the US N+3 project [9]. intercooling by use of a variable Intercoolers have the potential to intercooler exhaust nozzle, improve engine SFC, ease the design indicated in the lower part of of an efficient turbine cooling Figure 1, to establish an optimum system by reducing compressor exit between intercooling and incurred temperatures and hence cooling air pressure losses. temperatures, as well as reducing NOx emissions. Intercooling may also Another design issue that has to be provide benefits by increasing the addressed is to decide where in the specific output of the engine core compression process the and therefore reduce total engine intercooling should be introduced. weight. An early introduction is favourable from a thermodynamic perspective Although the design of compact heat whereas a later introduction will exchangers is a mature field and a reduce the intercooler pressure wealth of design data exists [10], losses, weight and simplify its the availability of data directly integration by reducing the applicable to aero engine intercooler volume requirement. performance studies is quite limited. Aero engine performance is Finally, the choice of the strongly influenced by the weight intercooler configuration itself and volume of the intercooler and has a decisive impact on the the pressure losses incurred from performance that can be achieved by its integration as well as the an intercooled engine. It is argued design strategy employed. Here the here that there is a need for term design strategy refers to the performance estimates that sizing of the intercooler related realistically represent to the mission. Within this work intercoolers designed and optimized for aero engine applications.

Figure 1 Intercooled engine impression [11]

This will enable engine performance pressure shaft is limited by the in a more realistic way than can be fan tip speed. For the LPT, an achieved by assuming pressure loss increased rotational speed gives a levels or by applying simplified lower blade loading, higher correlations. efficiency and allows the use of fewer stages. Hence, the bypass The general configuration of the ratio can be ultra-high without an two-pass cross flow intercooler can excessive number of LPT stages. The be seen in Figure 2 below. Flow increased shaft speed also allows exiting an intermediate compressor for a smaller shaft diameter that or high speed booster enters the makes it possible to integrate the inflow duct through which it is high pressure compressor at a lower diffused. The flow then enters the radius. This increases last stage first stack of a tubular heat blade height which is particularly exchanger located downstream in the critical for intercooled engines, cooling flow direction, returns to which are expected to gain some of an upstream tubular heat exchanger their benefits from a substantially and then continues to an increased OPR. accelerating duct leading to the high pressure compressor entrance. Intercooler Modelling Bypass air flows over the external surfaces of the two tubular stacks The correlations of the two- to achieve the sought intercooling. pass cross flow intercooler are This paper extends the previous divided into a correlation for the work on this concept [4, 11] by internal hot side connecting ducts considering an involute spiral and tubes, as well as configuration for the intercooler characteristics for the external tubes. cold side staggered tubes. The general staggered tube CFD studies for the re-designed configuration, that is the cross- concept was used to obtain a set of section of the tube stack, can be correlations which were used to seen from Figure 3 and the general evaluate the performance of a high parameters for this tube bypass ratio intercooled geared arrangement are defined in turbofan. Table 1.

Figure 3 Intercooler external side configuration

Figure 2 Two-pass cross flow intercooler impression

For a geared turbofan the LPT has a higher rotational speed than in the conventional turbofan, for which the rotational speed of the low Table 1 Tubes arrangement parameters

Outflow duct (Reynolds number is based on the outflow duct outlet condition. The correlation is valid in the range of 1,500,000 to 5,500,000. The largest deviation from CFD data in this correlation is 0.72%):

Figure 4 Involute spiral tubes configuration

An involute spiral tube configuration design is adopted for the tube stack. As seen from In the correlations above, is the Figure 4 this concept gives a good static pressure and is the total utilization of the space available. pressure, while 'in' and 'out' subscripts represent the inlet and Correlations outlet of the duct respectively; is the roughness of the tube wall, Internal Core Side (a value of 0.002 mm is assumed Inflow duct (Reynolds number is here). The hydraulic diameter of based on the inflow duct inlet the ducts and the tube are used to condition. Correlation is valid in calculate the Reynolds number in the range of 500,000 to 1,400,000. the corresponding correlations. The largest deviation from CFD data in this correlation is 0.81%): External Cold Side The Colburn j factor is used here

which is the dimensionless heat

transfer coefficient as defined in [10]:

Tubes (Pressure loss correlation from [12]. Heat transfer correlation from [13]): The Nusselt number, the Reynolds number and the related hydraulic

diameter used in the external cold side correlations are calculated

√ as:

√( )

Crossover duct (Reynolds number is based on the crossover duct inlet condition. Correlation is valid in ̇ the range of 100,000 to 350,000. The largest deviation from CFD data in this correlation is 0.78%):

5 where is the local heat transfer Engine modelling coefficient; is the hydraulic diameter of the flow passage; is The advanced non-intercooled geared engine is optimized by the thermal conductivity of air; is the surface area of the varying the bypass ratio, fan, IPC and HPC pressure ratio. The OPR is ellipses; is the minimum flow limited due to the turbine blade cross sectional area; is the total and disc cooling temperature, as flow length of the intercooler; ̇ manifested by a maximum compressor is the mass flow at the inlet and exit temperature of 950 K. The is the dynamic viscosity. engine take-off net thrust is set to be 65625 lbf which is considered The friction factor is defined suitable for the twin engine through [10]: aircraft model used for this study. Design point data and constraints common to the geared and the ̇ intercooled geared engine are

summarized in Table 2 below. The where is the pressure drop efficiencies in the table are polytropic. is the allowed through the intercooler, and are the fluid inlet and outlet maximum blade temperature. A hot day take-off condition is used to density, is the average of and evaluate the engine at the start of , is the ratio between the the mission analysis. The cooling minimum flow cross sectional area flow is then calculated based on and the intercooler frontal area . the model established in [14].

The correlations provided below are valid in the range of 10,000 < Re < Parameter Value 110,000. The largest deviation from CFD data is 0.90%. < 950 K < 1900 K

< 1210 K

93.5%

92.2%

92.5%

Furthermore, the additional weight 90.7% and nacelle diameter caused by the 91.4% intercooler has been considered. Titanium is assumed to be used as 93.25% intercooler material. The tube Net Thrust 65625 lbf thickness is calculated based on the pressure difference between the inner and outer side of the tube. A Table 2 Design point performance minimum tube thickness is assumed parameters (take-off) to be 0.2 mm. The nacelle line moves with the change in intercooler size, in order to keep The data presented in Table 2 are the external bypass flow Mach based on estimates on performance number lower than 0.6. levels achievable for an engine entering into service year 2020+. The optimization procedure establishes an engine that provides minimum fuel burn for a fixed

6 mission and fixed aircraft. The and the inner diameter of the mission length is 6800 km. Initial internal bypass duct, has the cruise altitude is 35000 ft and limitation of keeping the external final cruise altitude is 39000 ft. bypass flow Mach number, the A trade factor is established to nacelle maximum diameter and the estimate the full fuel burn saving internal bypass diffusion loss potential for a scalable aircraft. down. However, an increased tube length gives larger frontal area With the optimization of the which in turn gives a lower flow intercooler parameters, the mission Mach number through the intercooler study has more design freedom and and hence a lower intercooler is more complicated than analysing external side pressure loss. the advanced non-intercooled engine. The basic intercooler Basically, the total heat transfer parameters to be determined are the area is determined by the single tube diameter, the tube length, the tube size and the number of tubes. number of rows and number of The number of tubes is equal to the columns. number of tube rows (circumferential direction) The intercooler coolant mass flow multiplied by the number of tube is being controlled by a separate columns (axial direction). The variable geometry exhaust nozzle. distribution between the number of The exhaust nozzle area is hence rows and the number of columns included as part of the plays an important role in the optimization parameters. Closing intercooler performance. Generally, the nozzle reduces intercooler with a given single tube size and external Mach numbers and related heat transfer area, fewer columns pressure losses, both over the are desirable for decreasing the intercooler tubes and in the external side pressure loss. A high upstream intercooler diffuser. The radius installation of the closing of the nozzle also intercooler gives more rows and decreases transferred heat, which fewer columns but also increases leads to an increased combustor the nacelle diameter. inlet temperature and a reduced fuel flow need. The increased The design parameters determining compressor power requirement the basic intercooler parameters results in an increased turbine are given in Table 3. Here, is the inlet temperature and also a major axis length of the inside of somewhat increased temperature in the elliptical tube, is the hub the core exhaust nozzle. In total diameter of the intercooler matrix, this process reduces the thermal is the diameter of the shroud of efficiency but this is outweighed the intercooler and Mtube is the flow by the strong reduction in Mach number inside the tubes. M irreversibilities generated in the tube and are input to the design intercooler. process whereas and are

obtained from the conceptual design The tube diameter is controlled by of the engine. the major axis length of the tube, since the aspect ratio is fixed. As The conceptual design process used the size of the tube is decreasing, here, apart from the intercooler the total perimeter length of the design has been outlined in [11] tubes increases for a given cross and is described in more detail in sectional area which leads to an [15, 16]. The weight calculation is increase of the wetted area. integrated in the optimization

process and changes for every new The tube length, which is set by the radial position of the diffuser

7 engine/aircraft mission being engine, as seen in Table 4. For the evaluated. fixed aircraft mission this integrated to a 3.2% fuel burn Basic intercooler Design benefit. The SFC reduction is due parameters parameters to a higher thermal efficiency Tube diameter resulting from a higher OPR and a reduced cooling flow need. Tube length

Number of Tube Mtube rows Advanced Intercooled non-IC Number of Tube Mtube columns OPR 61 81 Table 3 Intercooler design TOC OPR 76 101 parameters Cruise OPR 58 75 BPR 15.42 16.74 Fan PR 1.40 1.44 Results and Discussions IPC PR 4.91 4.55 HPC PR 7.72 13.78 For the advanced non- HPC last 19.4 13.6 intercooled geared turbofan engine, blade height the optimal OPR (take-off) is found (mm) to be 61, whereas for the Take-off HPC 941.9 888.4 intercooled engine an OPR of 81 is exit obtained. For the intercooled temperature geared turbofan, a further (K) reduction in mission fuel burn can HPT Cooling 0.19 0.14 be achieved for an even higher OPR. bleed ratio However, the minimum compressor Take-off SFC 6.28 6.17 blade height constraint of the HPC (mg/Ns) then becomes active. A rule of TOC SFC 13.07 12.93 thumb is that the compressor blade (mg/Ns) height should not be lower than Cruise SFC 13.10 12.68 12 mm and the hub tip ratio should (mg/Ns) be kept lower than 0.92, otherwise, Cruise 0.514 0.524 the tip clearance loss and end wall boundary layer losses of the HPC Cruise 0.823 0.810 will deteriorate its efficiency. (mm) NA 30.7 Tube NA 0.37 In order to reduce the number of Length (m) design parameters, the flow Mach Rows NA 11 number inside the tubes (Mtube) is Columns NA 20 fixed to be 0.07. This was done Intercooler NA 309 after a number of preliminary Weight (kg) optimization studies revealed that Engine 7005 6705 the optimizer always produced the Weight (kg) lowest internal tube Mach number Nacelle 3.47 3.41 possible. Lower values lead to Diameter(m) designs that could not be fitted Mission fuel 30425 29440 within the available space burn (kg) Base -3.2% constraints. Attempts to increase the external nacelle diameter to Table 4 Optimal engine accommodate a larger intercooler configurations resulted in an increase in mission fuel burn. In addition, the intercooler also contributes to a reduced engine A cruise SFC benefit of 3.2% is core size through an increase in observed for the intercooled the specific power of the core,

resulting in an estimated 300 kg over the external intercooler weight reduction. The nacelle surface, which is associated with a diameter is only slightly reduced relatively high pressure loss, is for the intercooled engine reduced. This will also add to the contributing only marginally to the improvement of the thermal reduced fuel burn. efficiency.

A trade factor was established to The minimum mission fuel burn for take into account that the reduced the intercooled engine is fuel burn can be used to reduce the remarkably insensitive to BPR aircraft operational empty weight variation as is indicated by and take-off weight. The trade Table 6 below. From comparing the factor also takes into account that cruise point SFC and engine size, the reduced take-off weight allows the 14.5 BPR engine benefits from for a reduced aircraft drag and a lower nacelle drag and engine reduced engine thrust. The value of weight while the higher BPR engine the trade factor was estimated at has a greater SFC benefit. 1.5 which gives a net fuel burn reduction of 4.8%. OPR 81 81 81 Take- Top of Mid BPR 14.50 15.50 16.74 off Climb Cruise Fan PR 1.50 1.48 1.44 OPR 81 101 75 IPC PR 4.70 4.59 4.55 Core Flow 87 72 59 (mm) 29.4 29.0 30.7 Temperature Number of 22 21 20 Drop (K) tube Core 3.7% 5.0% 4.6% columns Pressure Tube 0.38 0.39 0.37 loss Length (m) Coolant flow 14.5% 10.9% 4.0% Nacelle 3.30 3.34 3.41 Pressure Diameter(m) loss Engine 6385 6438 6705 Coolant mass 111 38 16 Weight (kg) flow (kg/s) Cruise SFC 12.88 12.78 12.68 Altitude(ft) 0 35000 39000 (mg/Ns) Mach number 0 0.81 0.81 0.524 0.525 0.524 Net Thrust 66040 15191 8351 0.797 0.802 0.810 (lbf) 0.449 0.531 0.524 Mission 29555 29449 29440 Fuel burn 0 0.757 0.810 Table 6 Intercooled configurations Table 5 Intercooler operation for a parametric BPR variation (the conditions (Optimal engine last column is the optimal engine) configuration)

From Table 5 above it can be seen From BPR 15.5 to 16.74, there is a that the take-off point demands the LPT stage increase due to that the highest amount of intercooling. For stage loading in the LPT has the top-of-climb and cruise points, reached its limit. An additional the heat transfer can be reduced by LPT stage gives a step increase in controlling the separate nozzle to weight and hence a step increase in reduce the coolant mass flow. With fuel burn. However, the reduced FPR a lower coolant mass flow, less implies reduced specific thrust and heat rejection from the core to the hence an improved propulsive external bypass leads to a higher efficiency. Additionally, as the thermal efficiency and propulsive BPR is increasing, the fan size efficiency. Additionally, the flow increases as well as the nacelle

9 line. This actually gives more compressor exit temperature which space for the intercooler gives a reduced cooling flow need. installation and then the Potentially the reduced compressor intercooler is installed in a exit temperature could also be used higher radial position. As to reduce NOx emissions. mentioned before, a higher radial position gives fewer columns which are desired for reducing the Acknowledgement intercooler pressure losses. The tube length has been reduced as This work is financially supported well due to the increased radial by the E.U. under the “LEMCOTEC – position of the intercooler Low Emissions Core-Engine installation. This leads to a Technologies”, a Collaborative reduction in the internal side Project co-funded by the European pressure loss. Commission within the Seventh Framework Programme (2007-2013) The combined effect offsets the under the Grant Agreement n° weight penalty from the additional 283216. LPT stage. The optimal engine is then defined at BPR 16.74. A We also would like to thank Richard further increase in BPR will result Avellán and Anders Lundbladh at GKN in an increase in fuel burn due to Aerospace, and Andrew Rolt at that the nacelle drag starts to Rolls-Royce for valuable dominate. discussions.

Conclusions References

In general the two-pass cross flow [1] Kyprianidis, K., Rolt, A. M., intercooler has been seen to give a and Grönstedt, T., ‟Multi- relatively small internal pressure Disciplinary Analysis of A Geared loss. A relatively high external Fan Intercooled Core Aero-Engine‟. pressure loss has been observed, Proceedings of ASME Turbo Expo but this could be limited quite 2013, San Antonio, Texas, USA, effectively in cruise by the use of GT2013-95474, June 2013. the variable exhaust nozzle. The transferred heat with this [2] Xu, L., Kyprianidis, K., and intercooler concept is relatively Grönstedt, T., ‟Optimization Study modest. It is however sufficient to of an Intercooled Recuperated Aero- enable the higher OPR which is Engine‟. AIAA Journal of Propulsion restricted by the compressor exit and Power, Vol. 29, No.2, pp.424- temperatures for the advanced non- 432, 2013. intercooled geared engine. An OPR of 75 was achieved in cruise [3] Kyprianidis, K., Grönstedt, T., compared to an OPR of 55 for the Ogaji, S., Pilidis, P., and Singh, advanced non-intercooled engine. R., ‟Assessment of Future Aero- The intercooled engine OPR is Engine Designs with Intercooled and actually higher if the auxiliary Intercooled Recuperated Cores‟, nozzle area is kept at its design ASME Journal of Engineering for Gas value, but there is a net gain in Turbines and Power, 133(1), fuel burn reduction from reducing doi:10.1115/1.4001982 January 2011. its area. The intercooling also allowed for a somewhat smaller [4] Xu, L. and Grönstedt, T., engine giving a marginal benefit in „Design and Analysis of an nacelle drag reduction and a Intercooled Turbofan Engine’. ASME somewhat larger effect due to a Journal of Engineering for Gas reduced weight. A larger benefit is Turbines and Power, Vol. 132, No. obtained from the reduced 11, doi:10.1115/1.4000857 Nov 2010.

[5] Grönstedt, T. and Kyprianidis, [11] Xu, L., ‟Analysis and K., „Optimizing the Operation of Evaluation of Innovatiove Aero the Intercooled Turbofan Engine’. Engine Core Concepts‟, Ph.D Thesis, Proceedings of ASME Turbo Expo Chalmers University of Technology, 2010: Power for Land, Sea and Air, Sweden, 2011. Glasgow, UK, ASME GT2010-22519, June 2010. [12]Haaland, S. E., „Simple and Explicit Formulas For the Friction [6] Rolt, A. M. and Baker, N. J., Factor in Turbulent Pipe Flow’. „Intercooled Turbofan Engine Design J.Fluids Eng., 105(1), pp. 89-90. and Technology Research in the EU 1983. Framework 6 NEWAC Programme’. The 19th International Symposium on Air [13] Gnielinski, V., „New Equations Breathing Engines, Montreal, for Heat and Mass Transfer in Canada, ISABE-2009-1278, 2009. Turbulent Pipe and Channel Flow’. International Chemical Engineering, [7] Rolt, A. M. and Kyprianidis, 16, pp. 359-368. 1976. K., ‘Assessment of New Aeroengine Core Concepts and Technologies in [14] Wilcock, R.C., Young, J.B., The EU Framework 6 NEWAC and Horlock, J.H., „The Effect of Programme’. 27th International Turbine Blade Cooling on the Cycle Congress of The Aeronautical Efficiency of Gas Turbine Power Sciences, Nice, France, ICAS 2010- Cycles‟. Journal of Engineering for 4.6.3, Sep 2010. Gas Turbines and Power, 127, Jan, pp. 109-120, 2005 [8] Lundbladh, A. and Sjunnesson, [15] Grönstedt, T., Au, D., A., „Heat Exchanger Weight and Kyprianidis, K. and Ogaji S., Efficiency Impact on Jet Engine „Low-Pressure System Component Transport Applications’. 16th Advancements and Its Influence International Symposium on Air on Future Turbofan Engine Breathing Engines, Cleveland, Ohio, Emissions‟, GT2009-60201, ASME ISABE-2003-1122, Sep 2003. Turbo Expo 2009, Florida,

Orlando, USA, 2009.

[9] Bruner, S., Baber, S., Harris, C., Caldwell, N.,Keding, P., [16] Kyprianidis, K., Au, D., Rahrig, K., Pho, Luck., and Ogaji, S. and Grönstedt, T., Wlezian, N., „NASA N+3 Subsonic „Low Pressure System Component Fixed Wing Silent Efficient Low- Advancements and its Impact on Emissions Commercial Transport Future Turbofan Engine (SELECT) Vehicle Study‟, NASA/CR- Emissions‟, ISABE-2009-1276, 2010-216798, Nov, 2010. Montreal, Canada, 2009.

[10] Kays, W. M., and London, A. L., Compact Heat Exchanger, 2nd ed. McGraw-Hill, ISBN 07-033391-2. 1964.