American Institute of Aeronautics and Astronautics

1

Experimental Investigation on Methane in Transcritical Conditions

R. Votta1, F. Battista2, A. Gianvito3, A. Smoraldi4 and V. Salvatore5 CIRA (Italian Aerospace Research Center), 81043 Capua (CE), Italy

M. Pizzarelli6, G. Leccese 7, and F. Nasuti8, Università di Roma “La Sapienza“, 00185 Rome, Italy

and S. Shark9, R. Feddema10 S. Meyer11

Purdue University, West Lafayette, Indiana, USA

The use of the methane as coolant in a regenerative liquid rocket engine (LRE) presents some difficulties since transcritical fluid dynamics operating conditions occur in the cooling channels. Transcritical conditions cause large fluid properties variation that may strongly influence the coolant performance. The HYPROB program is carried out by CIRA under contract by the Italian Ministry of Research with the main objective to improve National system and technology capabilities on liquid rocket engines for future space applications, with specific regard to LOX/LCH4 technology. Its main objective is to develop an test a LOX/LCH4 demonstrator. In order to match this objective a specific breadboard, the Methane Thermal Properties (MTP) breadboard has been manufactured and test. It is based on an electrical heating of a single representative cooling channel that has the aim to validate numerical methodologies and to improve the understanding of relevant physics of methane thermal properties in transcritical conditions. The paper presents the main results experimental test campaign that has been successfully performed at Maurice J. Zucrow Laboratories in Purdue University.

Nomenclature AR = Aspect Ratio BB = Breadboard CFD = Computational Fluid Dynamic FEM = Finite Element Method GCH4 = Gaseous Methane LCH4 = Liquid Methane LOX = Liquid Oxygen LRE = Liquid Rocket Engines 1 Research Engineer, Propulsion, Via Maiorise, 81043, Capua (CE) Italy, r.votta@cira.it, AIAA member, 2 Research Engineer, Propulsion, Via Maiorise, 81043, Capua (CE) Italy, f.battista@cira.it, AIAA member 3 Technician, Propulsion, Via Maiorise, 81043, Capua (CE) Italy, a.gianvito@cira.it 4 Research Engineer, Propulsion, Via Maiorise, 81043, Capua (CE) Italy, a.smoraldi@cira.it, AIAA member 5 Research Engineer, Propulsion, Via Maiorise, 81043, Capua (CE) Italy, AIAA member, v.salvatore@cira.it 6Assistant Professor, Dipartimento di Ingegneria Meccanica e Aerospaziale, marco.pizzarelli@uniroma1.it, AIAA member 7 Graduate Student, Dipartimento di Ingegneria Meccanica e Aerospaziale, giuseppeleccese87@gmail.com 8 Associate Professor, Dipartimento di Ingegneria Meccanica e Aerospaziale, francesco.nasuti@uniroma1.it, AIAA Associate Fellow. 9 Research Assistant, School of Aeronautics & Astronautics, steven.c.shark@gmail.com, AIAA Student Member. 10 Research Assistant, School of Mechanical Engineering, rick.feddema@nts.com, AIAA Student Member. 11 Managing Director, Zucrow Laboratories, meyerse@purdue.edu, AIAA Senior Member.

American Institute of Aeronautics and Astronautics

2

MTP = Methane Thermal Properties q = Heat Flux SAF = Safety Factor, ratio between real stress and allowable stress

I. Introduction he HYPROB1,2 program is carried out by CIRA under contract by the Italian Ministry of Research with the main objective to improve National system and technology capabilities on liquid rocket engines (LRE) for future

space applications, with specific regard to LOx/LCH4 technology. The HYPROB program strategic objectives and the overall development plan have been set in a preliminary

step, based on interactions with the institutional, industrial and scientific stakeholders. The program is structured in 3 main development lines, each corresponding to a specific implementation project:

• System: design and development of technology LRE demonstrators, including intermediate breadboards • Technology: R&T development in the areas of CFD combustion modeling, thermo-mechanical

modeling and materials, advanced optical diagnostics; • Experimental: testing capabilities for both basic physics and system-oriented (demonstrators)

experimentation. As far as the system line devoted to the LOx/LCH4 technology is concerned, a first implementation project has

been launched, called HYPROB BREAD, aimed at designing, manufacturing and testing a LRE demonstrator, of 3 ton of thrust, based on a regenerative cooling system using liquid methane as refrigerant.

The study logic implemented in the present project is based on the following drivers: • to make use of existing know-how and design solutions for critical items; • to design one or more intermediate breadboards which can enable the analysis and validation of the

most critical design solutions, in particular the design of the combustion chamber and its injection plate as well as the regenerative cooling.

This approach has been defined in order to proceed step by step, from the understanding of the basic physical phenomena i.e. combustion and heat transfer, and then to validate design and analysis methodologies by simple breadboards. The final step is to apply such consolidated basis of design and analysis in order to design correctly all subsystems of the regenerative thrust chamber.

Figure 1: Study logic

Following the above reported study logic (Figure 1), four products shall be built in the frame of the project; three technological breadboard and one demonstrator:

• The first breadboard is designed in order to investigate methane thermal behavior in heating process (MTP breadboard);

• The single injector breadboards (SubScale calorimetric BreadBoard, SSBB) whose aim is to investigate combustion and heat transfer phenomena in a sub-scale modular calorimetric and heat sink thrust chambers;

T

American Institute of Aeronautics and Astronautics

3

• A third breadboard is foreseen, namely a Full Scale calorimetric BreadBoard (FSBB), water cooled chamber, whose aim is to allow the demo injection head qualification;

• Finally the demonstrator, that is a regenerative cooled chamber LOX/LCH4 of 30-kN-class of thrust, as a the first step through the design of expander engine thrust chamber.

The present paper relies on the MTP breadboard and, in parrticular, design aspects and experimental campaign characterization.

II. Methane as coolant If methane is considered as coolant in the regenerative cooling circuit of a turbopump-fed rocket engine, it will

typically enter the cooling channels with a supercritical pressure and a subcritical temperature. As methane is heated up, due to the entering heat from hot-gas, its temperature passes through the critical value and therefore its behavior can be referred to as that of a “transcritical” fluid flow2. This flow is substantially different from a “supercritical” fluid flow (where pressure and temperature are far larger than their critical values) as many important thermodynamic variables (such as specific heat at constant pressure, thermal conductivity and speed of sound) exhibit a peak value in the vicinity of the critical temperature; in fact, the large change of the fluid properties of a transcritical fluid can greatly influence heat transfer.

Figure 2 Methane pressure-temperature-density state diagram

Figure 3 Methane pressure-temperature-density state diagram

Methane thermodynamic behavior is presented in Figure 2 on a pressure-temperature state diagram. In this

diagram the critical point (identified by pressure pc = 45.98 bar and temperature Tc = 190.53 K) is clearly visible, because above that pressure the phase-change no longer occurs and density variation with temperature, although strong, is continuous: density ranges from 450 kg/m3 in the low-temperature liquid-like region to 20 kg/m3 in the high-temperature gas-like region. Others thermodynamic variables change dramatically with temperature in the case of supercritical pressure, such as specific heat at constant pressure, speed of sound, viscosity and thermal conductivity. For example, specific heat at constant pressure is presented as a function of temperature and for various pressures in Errore. L'origine riferimento non è stata trovata.. It is evident that specific heat tends to an infinite value at the critical point and that it presents a peak value in case of supercritical pressure. In particular, as pressure increases, the specific heat peak value and occurs at increasing temperatures (this peak occurs at the so-called “pseudo-critical temperature”). The specific heat peak almost vanishes in case of very high pressure. Hence, strong specific heat variations are expected in case of transcritical temperature, the variations being larger as the pressure gets closer to the critical value. On the contrary, mild specific heat variations are expected both in case of temperature larger than the pseudo-critical value and in case of pressure far larger than its critical value. In case of subcritical pressure, the specific heat maximum value is located at the saturated-liquid point while the minimum value is located at the saturated-vapor point. The strong specific heat discontinuity at the saturation temperature is due to the different thermal behavior of the liquid and the vapor. The discontinuity disappears at supercritical pressure where phase change does not occur.

American Institute of Aeronautics and Astronautics

4

A. MTP Breadboard description and main design issues The MTP breadboard is a test article conceived for the study of the thermal characteristics of the methane as a

coolant in the pressure-temperature range interesting for the HYPROB-DEMO2 combustion chamber, both for design validation and collect experimental data to share in the scientific community. It perfectly matches the development logic of the HYPROB-BREAD project and it is the first step towards the development of the final demonstrator. Its concept is based on the electrical heating of a conductive material that transfers heat fluxes, similar to those experienced by methane in the regenerative cooling chamber of a rocket, to a channel (dimensions comparable with the HYPROB demonstrator2) in which methane flows at high pressure. The concept has to be correctly shaped in order to ‘drive’ heat to the channel wall.

The main parameter calculations are based on the following inlet conditions well-matched with existing test facility capabilities:

• Mass flow ranging from 20 to 60 g/s • Inlet pressure 150 bar or 60 bar • Inlet temperature 120 K • Power up to 20 kW in order to have heat fluxes similar to the cylindrical part of the combustion chamber. The behavior of the methane along the cooling channels has been described by means of a fluid-dynamic model

and a thermo-structural one described hereinafter. As mentioned above, the heat flux q in the breadboard is generated by an electrical heater and used as input for

the thermo-structural analysis on the solid part of the breadboard. The thermal field on the wall will be used as boundary condition for the subsequent thermo-fluid-dynamic analysis on the methane along the cooling channel.

A first approach taken into account is an engineering one and the used software is EcosimPro4. This approach is very useful since fast calculations are available and in design phase several channels configurations (i.e., channel aspect ratio, length, etc.) have been taken into account, in this way a test matrix has been defined for the detailed calculations. In this first part of the work a very simplified model have been used (Figure 4), with the specific objective to simulate only the behavior of methane and the overall parameters.

Figure 4 EcosimPro MTP schematic view.

In order to consolidate the design of the MTP BB detailed CFD and FEM simulations have been performed. For what concerns CFD analysis a simple model of the channel has been preliminarily used, the details of the model has been reported hereinafter. An important consideration has to be made about the thermal insulation of the model. On one side it is important in order to avoid losses of heat power throughout lateral conductor walls, on the other is important because the adiabatic condition applied at the channel wall is similar to the symmetry condition of the channel walls in a real rocket.

Figure 5 Main elements of MTP BB architecture

American Institute of Aeronautics and Astronautics

5

Figure 6: A slice of the grid.

The number of cells, the subdivision between fluid and solid part and the wall grid spacing are reported in the

following table 5:

Table 5 Main features of the grid. Total  elements   422415  Fluid  elements   138719  Solid  elements   283696  Wall/interface   0.001  mm  

Several simulations have been performed at beginning of the activities in order to be confident with both the

numerical tool (Fluent code) and the particular phenomenology (supercritical flow). Several real gas models have been analyzed based on the evolution of the Van Der Walls equation of state (Soave-Redlich-Kwong, Peng-Robinson) and on NIST5 database.

The final simulations have been performed with the NIST database5 (REFPROP7.0) that, in addition to the other models, can take into account for the thermal and transport properties, and showing a more stable behaviour at higher power inputs also. The power input has been considered varying from 1 kW to 10 kW.

The following figures 11 and 12 show the pressure drop along the cooling channel and the average temperature of the fluid at the outlet versus the power input at the bottom compared with the results of the EcosimPro.

Figure 7: Pressure drop in function of the power supplied at the bottom.

0

100000

200000

300000

400000

500000

600000

700000

800000

0 2000 4000 6000 8000 10000 12000

Power [kW]

Pres

sure

Dro

p [P

a]

EcosimPro

Fluent

z

y

Solid/copper

American Institute of Aeronautics and Astronautics

6

The pressure at the outlet is set at a fixed value (5340000 Pa) as a boundary condition, while the pressure at inlet increases with the power, because the methane heats up and its mean velocity raises.

Figure 8: Average temperature of the fluid at the outlet against the power supplied at the bottom. Simulations have been performed with and without a surface roughness of 10-5 m. Figure 9 is a comparison of

the pressure drop in both the situations.

Figure 9: Comparison of pressure drop with and without surface roughness of 10-5 m. As it can be expected the value of pressure drop corresponding to a surface roughness of 10-5 m is greater than

the values calculated during the simulations with a smooth channel. Another effect of the roughness is the increase of the heat exchange at wall. This gives an increase of the

regenerative/refrigerant effect of the methane and so a lower wall temperature. In the Table 6 we can see (case 3 and 4) the different temperature obtained by means a FEM tool with two different convective heat transfer. Higher values of hconv gives lower values of temperature. We can see also a comparison between CFD and FEM (case 2 and 4) with the same value of hconv.

Table 6:Effect of roughness on heat transfer and comparison with a FEM analysis.

Case ID Tool Power (W)

Roughness (m) T bottom (k)

h_conv (kW/m2/K)

1 CFD 7000 10-5 246.8 86156.4 2 CFD 9000 10-5 289.1 71233.3 3 FEM 9000 --- 428.2 41536.0 4 FEM 9000 --- 302.1 71233.3

0

50

100

150

200

250

0 2000 4000 6000 8000 10000 12000

Power [kW]

T ex

it [K

]

EcosimPro

Fluent

0

100000

200000

300000

400000

500000

600000

700000

800000

0 2000 4000 6000 8000 10000 12000

Power [kW]

Pres

sure

Dro

p [P

a]

Fluent rough=0.0

Fluent Rough=0.01 mm

American Institute of Aeronautics and Astronautics

7

Figure 10: Effect of roughness on temperature.

Figure 10 shows the temperature at the bottom and at the interface between solid and fluid for several values of

roughness and power input. The FEM simulations have been conducted in the entire solid domain with a coupling at the solid/fluid interface obtained through the convective heat transfer coefficient.

In the following some pictures are reported for a qualitative analysis of the results. The case of 9 kW power input is considered. From the slices of planes at constant x and constant y respectively Figure 6 we can see the thermal stratification especially in solid part and the beginning of temperature stratification in the fluid, which is not completely developed. The high temperature flow region remains close to the hot wall and there is an evident non uniform thermal distribution.

It has to be said that boundary conditions for the external upper and the lateral parts of the solid are adiabatic. While this hypothesis can be considered acceptable for the lateral part because in the real layout they should be periodic boundaries, it could be non-realistic for the upper part. At the time of writing of these notes simulations with a radiative upper boundary is planned in order to verify whether a stronger stratification of the fluid temperature will appear.

Figure 11: Temperature contour plot along a plane at x constant at about the end of the cooling channel.

0

100

200

300

400

500

600

700

800

0.000 0.002 0.004 0.006 0.008 0.010 0.012

roughness [mm]

T [K

]

T bottom 7kW

T wall 7kW

T bottom 9 kW

T bottom 9 kW FEM

American Institute of Aeronautics and Astronautics

8

Figure 12: Temperature contour plot along a plane at y constant in the half of the computational domain (not

in scale).

The specific heat at constant pressure at the exit section along z and along the x axis is also reported. The peaks that can be observed are reached near the critical value the temperature (Tc= 190.564 K). From the horizontal asymptote, in Figure 13, it is possible to deduce the existence of a thermal barrier inside the fluid. The consequence of this behaviour is the lower part of the cooling channel heats while the upper part does not.

Figure 13: Specific heat at constant pressure profile along z-direction.

A comparison of the average values of temperature, pressure and constant-pressure specific heat (cp) between the

CFD results and the EcosimPro ones is reported on the following figures. A general good agreement can be appreciated on the overall parameters.

Figure 14: Temperature profile along x direction.

American Institute of Aeronautics and Astronautics

9

Figure 15: Pressure profile along the x direction.

Figure 16: Specific heat at constant pressure profile along the x direction.

It is worth to note that the behaviour of cp is very sensitive to small temperature variations; moreover a more

detailed ECOSIMPRO model shall be used for the experiment rebuilding. Based on the input of the CFD analysis some thermo-structural calculation have been made in order to optimize

the shape of the BB and drive correctly heat fluxes to the channel without structure overheating and on finally to verify the design and cycle life.

As above mentioned, static structural analyses have been performed applying the steady state temperature distribution. Figure 17 and Figure 18 illustrate respectively principal stress S1 and principal strain ε1 contour plots for the case tc=1 mm, Glidcop Al-15. Maximum values for both principal stresses and strains are encountered in the cooling channel where high thermal gradients are present. Those values are then used to evaluate the maximum number of tests the MTP could withstand.

American Institute of Aeronautics and Astronautics

10

Figure 17: first principal total thermal and mechanical strain – Glidcop Al 15

Figure 18: first principal total thermal and mechanical strain – Glidcop Al 15

The number of tests without failure of the breadboard for these configurations (considering a safety factor equal

to 10 for fatigue life prediction) are 39. Figure 19 illustrates the contour plot of the inverse of the safety factor SAF, where

strengthmaterialloaddesignSAF

__

=

Ultimate strength failure is expected when

1≥SAF

Figure 19: Inverse of safety factor contour plot - Glidcop Al 15

High values are encountered near the MTP basis; however, those values could be ignored since clamped

boundary conditions are too conservative. In order to increase the fatigue life of the component, it is necessary to increase the close-out thickness since it

causes a stiffening effect on the cooling channel and, then, reduces the maximum strain. Figure 19 illustrates the contour plot of the inverse of the safety factor SAF respectively for Glidcop Al 15 with a

close-out thickness equal to 2 mm.

American Institute of Aeronautics and Astronautics

11

Figure 20: Inverse of safety factor contour plot - Glidcop Al 15 close-out: 2 mm

The number of tests without failure of the breadboard for these configurations increases to 48. The trade-off analysis has shown that for the examined configurations, the maximum number of cycles with

neither fatigue failure nor ultimate strength failure of the component could be reached by adopting a MTP breadboard configuration with Glidcop Al 15 and a close-out thickness equal to 5 mm.

Figure 21: Number of cycles vs close-ot thickness

As shown in Figure 21 the number of tests to failure does not change significantly from 5 mm to 10 mm close-

out thickness. As a consequence a close-out thickness equal to 5 mm has been chosen in order to mimimize the weight of the MTP and to reproduce realistic geometries of thrust chambers.

At the end of the design process the MTP BB design has been consolidated and has been manufactured Figure 22

Figure 22: MTP BB

Number of tests vs close-out thickness

0

20

40

60

80

100

120

0 2 4 6 8 10 12close-out thickness (mm)

Num

ber o

f tes

ts

American Institute of Aeronautics and Astronautics

12

A. Electrical system and monitoring systems overview The MTP BB Electric System is basically composed by five functional subsystems, here below reported with

acronyms: • MTP BB and Heating Elements (HE) • Data Acquisition System and Temperature Controller (DAQ) • Heating Elements Power Control System (HE-PCS) • Test Engineer Facility Management System (TE-FMS) • Harness, Diagnostic and Protection System (HAR)

The HAR and the HE-PCS shall be installed into a electric cabinet designed specifically for the required test

hardware and the testing environmental conditions. The next schematic (

Figure 23) represents the MTP BB Electric System general conceptual design:

MTP BB ELECTRIC SYSTEM

HE DAQ TE-FMS HE-PCS HAR

Figure 23: Electric System conceptual design

As a consequence of available electric power to be shared upon a relatively small extended surface, a highest density power source in the range of 60 W/cm2 is required for the present application.

The state of the art of the high wattage electric heating elements is represented by Nickel-Chrome wire wrapped Heating Cartridges (Figure 24). Because of the high dielectric insulation outer sheath, this kind of electric heater can be directly inserted into the metallic unit to be heated, without interposing of any Thermo dielectric Refractory between the electric heating source and the load.

Figure 24: Nickel-Chrome wire wrapped Heating Cartridge

As a consequence of the overall expected dimensions of the MTP BB and in compliance with the standard of

cartridges available on the market the Heating Cartridges adapt for the present application, they have the following relevant features.

• Outer Diameter in the range of 6.5 mm;

American Institute of Aeronautics and Astronautics

13

• Cartridges body length in the range of 300 mm or two set of 150 mm in order to reduce the length of the metal of the MTP BB to be drilled. In case of two set long 150 mm, one set will be inserted into one face of the MTP BB and the other one into the other face.

• Maximum wattage of each Heating element in the range of 1.0 – 5.0 kW, with respect to state of the art of the technology that allows power density up to 60 W/cm2;

• Cartridges are equipped with embedded Temperature sensors adapt for the testing temperature range of the MTP BB;

In order to suitably exploit the experimental test campaign, the MTP breadboard has been instrumented to acquire data to be compared with numerical results provided by numerical method used in the design process that are the same that has been used in the Demonstrator design.

The experimental set-up foresees the use of thermocouples to measure temperatures in different points of the test article, deriving also an estimation of the heat flux, and strain gauges to acquire measurement of deformation to be compared with the estimation of thermo-structural analysis.

The expected temperatures to be measured are lower than 800 K therefore thermocouples type K will be used. The thermocouple KMTSS-M050U-300 model has been selected. It provides accurate and stable values and very good performances at high temperatures. The probe diameter is 0.50 mm therefore the hole in the structure can be less than 1mm thus minimizing the effects on the test article concerning the structural integrity. The probe junction is ungrounded and provides good response time and good reaction to electrical noise and ground loop. The output voltage to be measured by the acquisition system is in the range 0-100 mV. Other thermocouples have been installed on the top of the test article to verify the cooling performances of the methane channel. Moreover some sensors has been installed on the side of the test article below the thermal protective layer; the measurement of these TCs will be extrapolated to the vertical axis below the channel (see temperature contours reported in Figure 25) and then used to derive the value of heat flux close to the channel itself avoiding the need to make holes in the structure in this area which could compromise the integrity of the test article itself.

In synthesis, temperature measurement has been carried out in the following points: • Four different sections along x-axis equipped with 3 TCs each installed at different positions to derive

also an estimation of heat flux; these thermocouples will be installed through two holes made in the structure from the bottom and embedded in the MTP breadboard;

• Four sections equipped with 2 TCs each on the lateral side of the test article; • 4 TCs will be installed on the top of the test article

Figure 25: Temperature distribution within the test article and TC positioning

The foreseen instrumentation layout is shown in the following Figure 26.

American Institute of Aeronautics and Astronautics

14

Figure 26: Surface thermocouples positioning with visible PT measurements at inlet and outlet of the channel

Figure 27: Embedded Thermocouples position (embedded thermocuples in magenta)

III. Test Campaign Tests were performed in the High Pressure Lab at Purdue University’s Maurice Zucrow Laboratories. Zucrow

Labs is situated on 24 acres, remote from campus, near the end of the Purdue University Airport runway. The High Pressure Lab houses two large test cells, a control room, and office space for students and support staff. The test cells have reinforced-concrete, blast-resistant walls measuring 20 ft wide × 25 ft deep × 12 ft high. The experiment was conducted adjacent to the 10k stand in the rocket test cell. An aerial photo of the lab is shown in Figure 31.

The High Pressure Lab data acquisition and control systems provide remote control of all test operations in the test cells. The rocket test cell system has 512 channels of 16 bit A/D at 4 kHz per channel including signal conditioning for voltage and 192 thermocouple inputs. There are 128 channels of digital outputs for on/off valve control and 32 channels of ± 10 volt or 4 to 20 milliamp analog outputs for control valve and electronic pressure regulator control. The data acquisition and control system utilizes National Instruments hardware and is programmed using the LabView graphical programming language. All test operations are handled by the computer control system and are monitored and controlled by lab personnel in the Control Room.

American Institute of Aeronautics and Astronautics

15

Figure 28: Aerial Photo and Layout of the Maurice Zucrow Laboratories High Pressure Lab

B. Experimental set-up Operation of the MTP breadboard experiment required three principal things:

• Control of temperature and mass flow rate of liquid methane into the breadboard • Control of the pressure inside the breadboard • Control of the cartridge heaters supplying the thermal heat input to the breadboard

Liquid methane was produced in a condenser located in the test cell adjacent to the experiment. Gaseous

methane from gas cylinders was filtered and condensed in a heat exchanger chilled with liquid nitrogen. The liquid methane temperature was controlled by controlling the pressure at which the liquid nitrogen in the heat exchanger boiled. The maximum operating pressure of the liquid methane system is 340 bar. A cavitating venturi, with associated pressure and temperature instrumentation, was installed upstream of the inlet to the breadboard to facilitate mass flow rate control and measurement. The throats of the venturis utilized for the various tests were sized to assure cavitating operation at all test conditions. In this manner, the liquid methane mass flow rate was de-coupled from the operating pressure and pressure drop of the breadboard. The schematic plumbing diagram of the experiment, simplified for clarity, is shown in Figure 32. In addition to conditioning the temperature of the liquid methane in the tank, the tube supplying liquid methane to the breadboard was also jacketed with liquid nitrogen to provide the desired temperature entering the experiment.

Control of the pressure within the MTP breadboard was achieved by placing an actuated needle valve in the discharge tubing downstream of the experiment. An additional tube, connected to the outlet of the control valve, vented the methane from the experiment in a safe location. The required open position of the valve was estimated prior to each test. During the test, the valve was adjusted to achieve to desired pressure condition inside the breadboard at steady-state conditions. A photograph of the fully assembled experiment is shown in Figure 33.

The cartridge heaters providing the thermal heat input to the breadboard operated on 240 VAC single phase at 60 Hz. Each heater operated at 2 kW full power during the tests. Only six of the ten heaters were turned on during the tests utilizing 12 kW of heat input. The current draw for each heater was measured and recorded to verify the desired power was achieved. Temperatures within each cartridge heater as well as within the breadboard were monitored and would have triggered a test abort if safe limits had been exceeded. Four inches of insulation were placed between the breadboard and its structural support. Two inches of insulation surrounded the sides and greater than one inch covered the top, limiting convective heat transfer to the surrounding air in the test cell.

Bulk LN2

H2O2 StorageCleaning

Natural Gas Heat Exchanger

Bulk LOx

GN2 Storage

Bulk GH2 Storage

Test Cells

Annex

High Pressure Air Storage

Control Room and Cubicles

Cryogenic Conditioning Site

Electrical Buildup

Control Room

Cubicles & Office Space

Test Cells

Mechanical Buildup

Containment Court

American Institute of Aeronautics and Astronautics

16

A purge flow of dry nitrogen gas was maintained through the experiment at all times when the MTP breadboard was at greater than ambient temperature to prevent oxidation of the cooling channel walls. Typical test sequences involved heating the breadboard to the anticipated experiment steady-state temperature while flowing nitrogen at low pressure. Liquid methane flow would then be initiated, checking off the flow of nitrogen. With the cartridge heaters on at the required power condition, methane flow would be maintained and the outlet control valve would be adjusted until steady-state operating conditions were achieved. Tests were concluded by turning off the heater power followed by terminating the methane flow after the breadboard temperature dropped below 400K. Nitrogen purge was maintained until the breadboard temperature dropped below 300K.

Figure 29: Simplified Schematic for Methane Propellant and Test Conditions Control

Figure 30: Breadboard installed in Purdue facility

IV. Main Results and Discussion The following Table 1 shows the test matrix performed. A variation of pressure ranging from 150 to 60 bar and a

variation of mass flow from 15 to 25 g/s have been achieved. All tests relies on impressed power equal to 12 kW except case 17 in which the power impressed is equal to 20 kW. Moreover, two tests (ID.14 and 15) have been performed without power source in order to hydraulically characterize the channel using two value of mass flow (i.e., 15 and 20) for an exit pressure of about 80 bar.

400 bar Methane

7 bar Liquid

Nitrogen

T

PT

340bar

LCH4

T

PT

T

PT

CH4 Vent Valve

CH4 Supply Isolation Valve

CH4 Pressure Regulator

Filter

Liquid Nitrogen Valve

LCH4 Isolation Valve

LCH4Cavitating

Venturi

MTPBreadboard

Liquid Nitrogen Backpressure Control Valve

LCH4Run Valve

Breadboard Methane

Backpressure Control Valve

Liquid Nitrogen Jacketed LCH4 Run Line

LiquidNitrogenJacketed

LCH4 Tank

Venturi Inlet Temperature

Venturi Inlet Pressure

Breadboard Inlet & Outlet Temperatures

Breadboard Inlet & Outlet Pressures

American Institute of Aeronautics and Astronautics

17

Table 1 : MTP Test Matrix Case ID Exit Pressure

(bar) mdot (g/s)

Power Impressed (kW)

1 150 20 12 2 150 25 12 3 150 15 12 4 120 15 12 5 120 20 12 6 120 25 12 7 100 15 12 8 100 20 12 9 100 25 12 10 80 20 12 11 80 15 12 12 80 25 12 13 80 10 12 14 80 15 0 15 80 20 0 16 60 20 12 17 80 20 20

Success criteria are hereinafter reported: • Stationary conditions for 15 seconds • Variation of maximum 5% of Inlet nominal pressure • Variation of maximum 5% of Inlet nominal temperature • Variation of maximum 5% of Inlet nominal mass flow • Oscillation of maximum 1% around the actual inlet pressure value • Oscillation of maximum 1% around the actual inlet temperature value • Oscillation of maximum 1% around the actual inlet mass flow value • No overheating on control thermocouples Test 10 results are presented here as a reference case since the flow conditions (pressure, mass flow, etc) are

similar to the 30 kN LOX-CH4 demostrator2 cooling channels. Figure 31 shows the mass flow recorded during the test. At t=100 s the mass flow is regulated to the nominal

value, at t=180 s the mass flow is about 20.5 g/s and minor oscillations are noticeable.

Figure 31: MTP test 10 mass flow vs time

American Institute of Aeronautics and Astronautics

18

Also inlet and outlet temperature (see Figure 32 and Figure 33) fulfill the success criteria. The inlet temperature (green curve in Figure 32) recorded during the test is about 140 K, while the inlet pressure is about 90 bar (green curve in Figure 33); therefore, the methane is in liquid supercritical conditions. The outlet temperature is about 240 K (red curve in Figure 32Errore. L'origine riferimento non è stata trovata.), while outlet pressure is 80 bar (red curve in Figure 33). From the inlet to the outlet pressure still remains supercritical, while the temperature passes from supercritical to subcritical value, therefore the supercritical transition of the methane has been achieved.

Figure 32: MTP test #10 inlet and outlet temperature

Figure 33: MTP test #10 inlet and outlet pressure

Also the embedded thermocouples (referring to Figure 34) have reached steady conditions allowing a correct

evaluation of the heat flux and thermal field inside the solid part of the breadboard. The measurments performed in the solid part reach the steady contitions in about 150 seconds (see Figure 34)

while, as expected, the fluidic part (Figure 32 and Figure 33) in about 50 seconds.

Figure 34: MTP test #10 embedded termocouples

Finally, being the voltage constant (see Figure 35) is possible to have the correct estimation of power impressed and the values are reported in Table 1.

American Institute of Aeronautics and Astronautics

19

Figure 35: MTP test #10 – Current in the cartridges

Figure 36 shows the outlet temperature for the performed test cases with respect the mass flow. All the test cases exhibit an outlet temperature higher than critical value, therefore all the experiments had successfully accomplished the supercritical transition, since the outlet pressure (see Figure 37) is higher than the critical value.

Figure 36:MTP Outlet Temperature vs mass flow Figure 37:MTP Outlet Pressure vs mass flow

The next Figure 38 shows, in a syntetic fashion, that for all the test selected the supercritical transition occurs.

American Institute of Aeronautics and Astronautics

20

Figure 38: MTP Outlet Pressure/temperature diagram

Figure 39 and Figure 40 report, respectively, temperature increase and pressure drop along the channel for the performed test cases with respect the mass flow. Of course, as the mass flow raises, temperature decreases and pressure drops increases. Temperature increase of methane, for each mass flow, is higher as the pressure level rises, while pressure drop at each mass flow decreases; this is due to the cp behavior of the methane with pressure.

Figure 39: MTP Temperature increase vs mass flow Figure 40: MTP Pressure drops vs mass flow

In order to better evaluate pressure drops along the channel, an analysis of the pressure losses in the fluidic

connections has been performed. Figure 41 shows the fluidic connection installed on the MTP, while Figure 42 the drawings. Pressure and temperature measurements are taken in the section indicated in Figure 42, therefore the measured pressure drops include the channel contribution and the drops that occur in the both fluidic connections. These values are mainly due by the area variations (area 1 and 2 in Figure 42).

American Institute of Aeronautics and Astronautics

21

A rough evaluation of the contribution to the total pressure drops of the fluidic interfaces can be done by means of the following equation (1):

(1) β coefficient ranges between 0.82 and 0.95. Density at inlet and outolet can be easlily derived once known

temperature and pressure from experimental data using NIST web database5. Velocity can be derived from the continuity equation in the area 1 and 2 in Figure 42:

(2) In the inlet fluidic interfaces, for all the performed tests, the fuid density order of magnitude is about 400 kg/m3

and the order of magnitude of pressure drop is ~ 0.5 bar. In the outlet section the fluid density is about 100 kg/ m3 and the pressure drop ~ 2 bar.

Figure 41: MTP Fluidic connection and taps Figure 42: MTP Fluidic connection drawings For a better pressure drop evalutation inside the cooling channel a detailed CFD analysis is needed in order to

simulate channel and fluidic interfaces geometry discountinuities. As already told, one of the goals of the present experimental activity is to validate the simplified tool used for

system analysis of the HYPROB-DEMO, therefore (see Figure 43) a more detailed EcosimPro4 model has been set-up. The channel has been modeled together with the junctions and the working fluid (azure part in Figure 43); the grey parts in Figure 43 represent the internal wall thickness, while the MTP massive part has been considered in a single node with the same MTP mass (about 4.5 kg). Also the thermal insulation and the heaters have been modeled.

American Institute of Aeronautics and Astronautics

22

Figure 43: MTP EcosimPro model used for the rebuilding

A good agreement (see Figure 44 and Figure 45) between EcosimPro model and experimental results has been found, therefore a preliminary validation has been accomplished in terms of ∆T and ∆p.

A rebuilding of the all test cases by means of EcosimPro and a deep CFD analysis is an on-going activity and the results will be presented in future work.

Figure 44: Test case #8 pressure behaviour: experimental results vs EcosimPro

Figure 45: Test case #8 temperature behaviour: experimental results vs EcosimPro

American Institute of Aeronautics and Astronautics

23

V. Conclusion The paper presents the results of the design, manufacturing and testing activities on CIRA MTP breadboard

made in the framework of the HYPROB BREAD project. The experimental test campaign has been successfully performed at Maurice J. Zucrow Laboratories in Purdue University. All the test cases respect the success criteria imposed and the supercritical transition of the methane has been achieved; data have been successfully collected for the analysis. A first comparison with engineering models shows a good agreement. A complete rebuilding by means of EcosimPro and CFD of all the performed tests is currently on going and will be object of a forthcoming work.

VI. Acknowledgements This work has been carried out within the HYPROB program, funded by the Italian Ministry of University and Research (MIUR) whose financial support is much appreciated. The authors want to thank all CIRA HYPROB-BREAD team for the cooperation at the realization of this work.

References 1 Borrelli, S., de Matteis, P., Ferrigno, F., Schettino, A., D’Aversa, E. and Biagioni, M. “The HYPROB Program: Mastering KEY Technologies, Design and Testing Capabilities for Space Transportation Rocket Propulsion Evolution” 63th International Astronautical Congress, 1-5 Oct. 2012 Naples, Italy. 2 Salvatore, V., Battista, F., De Matteis, P., Rudnykh M,, Arione, L., Ceccarelli, F., 2013, Recent Progress On The Development Of A LOx/LCh4 Rocket Engine Demonstrator In The Framework Of The Italian Hyprob Program. 64rd International Astronautical Congress, IAC-13,C4,3,4,x19439 3 Pizzarelli, M., Nasuti, F., and Onofri, M., “CFD Analysis of Transcritical Methane in Rocket Engine Cooling Channels,” The Journal of Supercritical Fluids, Vol. 62, 2012, pp. 79-87. 4 Moral, J., “ESPSS Simulation Platform” Space Propulsion 2010, San Sebastian, Spain. 5 http://webbook.nist.gov/chemistry/