Measurement and Computation of Supersonic Flow in a Lobed … · 2013-08-30 · Measurement and...

NASA/TM--2002-211127

Measurement and Computation of Supersonic

Flow in a Lobed Diffuser-Mixer for TrappedVortex Combustors

Andreja Brankovic and Robert C. Ryder, Jr.

Flow Parametrics, LLC, Bear, Delaware

Robert C. Hendricks, Nan-Suey Liu, and John R. Gallagher

Glenn Research Center, Cleveland, Ohio

Dale T. Shouse and W. Melvyn Roquemore

Wright-Patterson Air Force Base, Dayton, Ohio

Clayton S. Cooper and David L. Burrus

General Electric Aircraft Engines, Evendale, Ohio

John A. Hendricks

Diligent Design, Toledo, Ohio

July 2002

https://ntrs.nasa.gov/search.jsp?R=20020080901 2020-04-08T04:50:06+00:00Z

The NASA STI Program Office... in Profile

Since its founding, NASA has been dedicated to

the advancement of aeronautics and spacescience. The NASA Scientific and Technical

Information (STI) Program Office plays a key part

in helping NASA maintain this important role.

The NASA STI Program Office is operated by

Langley Research Center, the Lead Center forNASA's scientific and technical information. The

NASA STI Program Office provides access to the

NASA STI Database, the largest collection of

aeronautical and space science STI in the world.

The Program Office is also NASA's institutional

mechanism for disseminating the results of its

research and development activities. These results

are published by NASA in the NASA STI Report

Series, which includes the following report types:

TECHNICAL PUBLICATION. Reports of

completed research or a major significant

phase of research that present the results of

NASA programs and include extensive data

or theoretical analysis. Includes compilations

of significant scientific and technical data and

information deemed to be of continuing

reference value. NASA's counterpart of peer-

reviewed formal professional papers but

has less stringent limitations on manuscript

length and extent of graphic presentations.

TECHNICAL MEMORANDUM. Scientific

and technical findings that are preliminary or

of specialized interest, e.g., quick release

reports, working papers, and bibliographiesthat contain minimal annotation. Does not

contain extensive analysis.

CONTRACTOR REPORT. Scientific and

technical findings by NASA-sponsored

contractors and grantees.

CONFERENCE PUBLICATION. Collected

papers from scientific and technical

conferences, symposia, seminars, or other

meetings sponsored or cosponsored byNASA.

SPECIAL PUBLICATION. Scientific,

technical, or historical information from

NASA programs, projects, and missions,

often concerned with subjects having

substantial public interest.

TECHNICAL TRANSLATION. English-

language translations of foreign scientific

and technical material pertinent to NASA'smission.

Specialized services that complement the STI

Program Office's diverse offerings include

creating custom thesauri, building customized

data bases, organizing and publishing research

results.., even providing videos.

For more information about the NASA STI

Program Office, see the following:

• Access the NASA STI Program Home Page

at http:llwww.sti.nasa.gov

• E-mail your question via the Intemet to

[email protected]

• Fax your question to the NASA Access

Help Desk at 301-621-0134

• Telephone the NASA Access Help Desk at301-621-0390

Write to:

NASA Access Help Desk

NASA Center for AeroSpace Information7121 Standard Drive

Hanover, MD 21076

NASA/TM--2002-211127

Measurement and Computation of Supersonic

Flow in a Lobed Diffuser-Mixer for TrappedVortex Combustors


Flow Parametrics, LLC, Bear, Delaware


Glenn Research Center, Cleveland, Ohio


Wright-Patterson Air Force Base, Dayton, Ohio


General Electric Aircraft Engines, Evendale, Ohio

John A. Hendricks

Diligent Design, Toledo, Ohio

Prepared for the

2001 19th International Congress on Instrumentation in

Aerospace Simulation Facilities (ICIASF 2001)

cosponsored by IEEE, AES, NASA Glenn, and OAI

Cleveland, Ohio, August 27-30, 2001

National Aeronautics and

Space Administration

Glenn Research Center

July 2002

NASA Center for Aerospace Information7121 Standard Drive

Hanover, MD 21076

Available from

National Technical Information Service

5285 Port Royal RoadSpringfield, VA 22100

Available electronically at h__GLTRS

Measurement and Computation of Supersonic Flow in a

Lobed Diffuser-Mixer for Trapped Vortex Combustors


Flow Parametrics, LLC

Bear, Delaware 19701


National Aeronautics and Space Administration

Glenn Research Center

Cleveland, Ohio 44135


Wright-Patterson Air Force Base

Dayton, Ohio 45433


General Electric Aircraft Engines

Evendale, Ohio 45215

John A. Hendricks

Diligent Design

Toledo, Ohio 43614

Summary

The trapped vortex combustor (TVC) pioneered by Air Force Research Laboratories (AFRL) is under

consideration as an alternative to conventional gas turbine combustors. The TVC has demonstrated

excellent operational characteristics such as high combustion efficiency, low NO x emissions, effective

flame stabilization, excellent high-altitude relight capability, and operation in the lean-bum or rich burn-

quick quench-lean bum (RQL) modes of combustion. It also has excellent potential for lowering the

engine combustor weight. This performance at low to moderate combustor mach numbers has stimulated

interest in its ability to operate at higher combustion mach number, and for aerospace, this implies

potentially higher flight mach numbers.

To this end, a lobed diffuser-mixer that enhances the fuel-air mixing in the TVC combustor core was

designed and evaluated, with special attention paid to the potential shock system entering the combustorcore.

For the present investigation, the lobed diffuser-mixer combustor rig is in a full annular configuration

featuring sixfold symmetry among the lobes, symmetry within each lobe, and plain parallel, symmetric

incident flow. During hardware cold-flow testing, significant discrepancies were found between com-

puted and measured values for the pitot-probe-averaged static pressure profiles at the lobe exit plane.

Computational fluid dynamics (CFD) simulations were initiated to determine whether the static pressure

probe was causing high local flow-field disturbances in the supersonic flow exiting the diffuser-mixer

and whether shock wave impingement on the pitot probe tip, pressure ports, or surface was the cause of

the discrepancies. Simulations were performed with and without the pitot probe present in the modeling.

A comparison of static pressure profiles without the probe showed that static pressure was off by

nearly a factor of 2 over much of the radial profile, even when taking into account potential axial

displacement of the probe by up to 0.25 in. (0.64 cm). Including the pitot probe in the CFD modeling and

NASA/TM--2001-211127 1

datainterpretationleadtogoodagreementbetweenmeasurementandprediction.Graphicalinspectionoftheresultsshowedthattheshockwavesimpingingontheprobesurfacewerehighlynonuniform,withstaticpressurevaryingcircumferentiallyamongthepressureportsbyover10percentin somecases.

Aspartof themeasurementmethodology,suchmeasurementsshouldberoutinelysupplementedwithCFDanalysesthatincludethepitotprobeaspartoftheflow-pathgeometry.

Introduction

The trapped vortex combustor (TVC) is being considered as an alternative to conventional gas turbine

combustors. The TVC pioneered by Air Force Research Laboratories (AFRL) (refs. 1 to 3) represents a

relatively simple and highly advantageous alternative to conventional swirl-stabilized combustors.

Hardware photographs, combustion flow-visualization images, and detailed performance plots for NO x

emissions, lean blowout, and combustion efficiency are presented in detail in references 1 and 2 and to

some extent in reference 3. Also discussed are fluid dynamics and combustion physics for a baseline

TVC that is operating under different fueling conditions at subsonic combustor core flows, presented with

the aid of computational fluid dynamics (CFD) images of the flow-field aerodynamics and combustion

process.

The outstanding performance at low to moderate combustor mach numbers has stimulated interest

in its ability to operate at higher combustion mach number, which for aerospace applications implies

potentially higher flight mach numbers.

At the lower combustor mach numbers the TVC has demonstrated excellent operational characteris-

tics such as high combustion efficiency, low NO x emissions, effective flame stabilization, excellenthigh-altitude relight capability, and operation in the lean-bum or rich bum-quick quench-lean bum

(RQL) modes of combustion. It also has excellent potential for lowering the engine combustor weight.

To preserve these characteristics when designing a new combustor for operation in a higher mach

number flow regime, special attention must be paid to the combustor diffusion system and to understand-

ing the physics of the potential shock system entering the combustor core. This was accomplished by

evaluating the flow through the lobed diffuser-mixer into the TVC annulus, which enhances the fuel-air

mixing in the TVC combustor core.

For the present investigation, the combustor rig is in a full annular configuration as opposed to the

planar configuration of references 1 to 3. The computer-aided design (CAD) solid model of the full

annular combustor with a lobed diffuser-mixer is shown in figure 1 (a), with a view looking downstream.

The experimental combustor geometry consists of a short, straight inlet pipe, lobed transition duct,

trapped vortex annulus with fuel and air injector ports facing upstream, and a circular combustor duct

downstream of the core. This design features sixfold symmetry so that CFD analyses modeled only

a 60 ° sector to capture all the relevant flow physics, assuming the incident flow is also symmetric.

During the evaluation stage, significant discrepancies were found between the computed and the

measured average static pressure profiles at the lobe exit plane. The present work was initiated during

hardware cold-flow testing to determine whether the static pressure probe was causing high local flow-

field disturbances in the supersonic flow exiting the diffuser-mixer and whether shock wave impingement

on the pitot probe tip, pressure ports, or surface was the cause of the discrepancies between computationsand measurements.

CFD simulations of air flow in the lobed duct system were performed with and without the pitot

probe present in the modeling. The pitot probe configuration was introduced into the computational

model through a full CAD solid model representation and was inserted and traversed consistently with

the experimental sampling method.

A comparison of experimental and CFD static pressure profiles without including the probe in the

model showed that the CFD static pressure was off by nearly a factor of 2 over much of the radial profile,

even when taking into account potential axial displacement of the probe by up to 0.25 in. (0.64 cm).

NASA/TM--2001- 211127 2

Lobe valley

Lobeddiffuser-mixer

Plenuminlet

7 Trapped vortexannulus

(a)

Lobe peak -_ Axial forwardfacing fuelinjectors

Lobed

10-in. (25.4-cm)diameter

Figure 1.--Trapped vortex combustor (TVC) with lobed diffuser-mixer.Fueling concept features axial forward and rearward facing injectors.(a) CAD solid model. (b) Test section schematic.

The pitot probe was then included in the CFD modeling and data interpretation, and good agreement

between measurement and prediction was achieved. Graphical inspection of the results showed that

the shock waves impinging on the probe surface were highly nonuniform, with static pressure varying

circumferentially among the pressure ports by over 10 percent in some cases.

The CFD simulations demonstrated that pitot probe pressure measurements in supersonic flow

regions must be interpreted with care because of significant local flow-field disturbances and their impact

on the measured result. Averaging the four static pressures clocked 90 ° about the pitot probe introduces

significant errors and is inadequate as opposed to individual sampling. Also, probe incidence angle

and sting strain require consideration. As part of the measurement methodology, such measurements

should be routinely supplemented with CFD analyses that include the pitot probe as part of the flow-path

geometry.

The appendix presents supplementary results for flow through a lobed diffuser-mixer that illustrate

the relevant fluid dynamics effects of a pitot probe in the flow path.

NASA/TM--2001-211127 3

Lobed Diffuser-Mixer Cold-Flow Test

The cold-flow test hardware consisted of the upstream nozzle and the lobed diffuser-mixer that pro-

jected into a pipe that was 10 in. (25.4 cm) in diameter and 39 in. (100 cm) long. The TVC cavity annulus

was not part of the test, and the pipe exhausted to the atmosphere (fig. (lb)). The working fluid was air.

CFD Codes and Conditions

The flow solver selected for this work was the commercially available, second-generation version of

the NASA Glenn Research Center (GRC) National Combustion Code (NCC) called FPVortex that was

developed by Flow Parametrics, LLC (Bear, Delaware). The approach of Ryder and McDivitt (ref. 4) is

followed to model the test rig hardware as installed on the test stand. In this analysis, the pitot probe was

incorporated in the existing lobed duct CAD solid model, and new computational grids were generated for

each of the six probe positions modeled.

The resulting grid is transferred to an unstructured geometry data base and read in by the flow solver.

Intermediate results data bases are created and can be viewed automatically in graphics postprocessing

packages. The physics-based modeling for the aerodynamic simulations includes the two-equation k-emodel of turbulence with wall functions.

The inlet boundary condition consists of the subsonic airflow conditions (velocity and mass flow

rate), which feed the inlet plenum. The inlet static pressure for these cases was 50 psia (345 kPaa). The

exit boundary condition simulated the laboratory pressure condition; the flow at the pipe exit was vented

into the room at a nominal atmospheric pressure of 14.7 psia (101 kPaa). At high flow rates, the flow

chokes at the minimum-area location at the exit and forms a shock pattern, which extends well into the

downstream duct, eventually dissipates, and changes back to subsonic flow.

High-Speed Flow Structure

A CFD simulation of the aerodynamics of flow through the lobed duct diffuser-mixer was run to gain

an understanding of the main flow features in the device. Mach number contour plots of the flow from

the lobe into the large-diameter pipe are shown in figure 2, illustrating the flow complexity in that region.

A supersonic jet core penetrates well into the combustor core and remains supersonic for a length of

about five lobe heights downstream of the lobe exit plane. The supersonic core transforms from a lobe-

shaped cross section to a circular shape over a length of about two lobe heights as it moves downstream.

High-speed, discrete cellular structures are clearly present, forming a shock train with five to six distinct

cells. Clearly visible at the lobed duct exit plane are regions of high-velocity and high-pressure gradients

as the flow expands from subsonic to supersonic. This is the region in which pressure profiles were taken

using a pitot pressure probe constructed for use in this test rig.

Pitot Pressure Probe

The performance characteristics of the lobed duct were measured to assess the effectiveness of the

design. Measurements included pressure loss and radial profiles of total and static pressure at the lobed

duct exit plane. To accomplish this, a pitot probe with multiple static openings (ref. 5) was designed and

mated to a traversing mechanism attached to the outer duct. The probe was moved radially in line with

either the lobe peaks or troughs (which were clocked at 30°). A schematic diagram of the probe is shown

in figure 3. The frontal portion of the probe had a 0.125-in. (0.3175-cm) diameter and was turned to face

upstream into the flow. Using pressure transducers, the pitot tube directly measures total pressure PT at

NASA/TM--2001- 211127 4

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_i_i_#iiiii/i//i_%_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_i_i_i_i_i_i_i_i_i_i_i_i•

(a)

(b)

,:+:+:+:+:+:+:+::+ :+:+:+:+:+:+:+:+:+:+

....: _:ii_ ::ii_ ......... Flow exit

/ _-1 Z Supersonic jet core flow Isocontours Mach number,with shock train (cellular M

.........................................................................structures) _ Cyan _ 0.94.................................................................................................. Green _ 1.12:: .........................................:::::: /

_ii / Yellow 1.32

/ Red _ 1.50

Sonicsu,ac between cyan and /_ _ %

/

green contours J

Figure 2.--Supersonic flow structure downstream of the lobed diffuser-mixer exit plane.(a) Overall length of the supersonic core flow. (b) Detail of the cellular structure in the core.

the probe tip and static pressure PS at four ports clocked at 90 ° about the circumference and 0.50 in.(1.27 cm) downstream of the tip. Accuracy issues include providing tubing with adequate frequency

response, calibration, and pitch-yaw sensitivity to the flow direction.

In the supersonic flow case, the accuracy of pitot tubes is more questionable mainly because of the

presence of shock waves that may intersect the probe normally or at oblique angles and also because of

high-velocity gradients and stability of small-scale flow features, which may easily be disturbed by the

presence of the probe.

The data comparisons shown in figures 4 and 5 illustrate pitot probe accuracy issues for supersonic

flow. In figure 4, radial profiles of total pressure are shown from the duct centerline to the lobe valley

(fig. 4(a)) and from the duct centerline to the lobe peak (fig. 4(b)). Because the total pressure is measured

at the probe tip, there is minimal local flow disturbance, and excellent agreement between experimental

data and CFD predictions can be

seen. The relevant profiles are at

axial position Z, which represents

the plane in which the radial pro-files are measured. Additional

curves sampled from the CFDsimulations are shown at down-

stream locations, establishingthe trend in axial variation of total

pressure. Thus, for total pressure it

appears that in spite of the potential

for shock waves and other high-

gradient flow structures, the pitot

probe is adequate for capturing

the quantitative total pressure

profiles for the lobed diffuser-

mixer at high-speed operation.

2.00 ir

)-1.75 in. (4.445 cm)

/// _4_

/

L_Four static pressureposts, clocked at 90°0.50 in. (1.27 cm)from probe tip

(5.08 cm)

/- 0.125-in. (0.3175-cm) diameter/ Entire probe tip

r- 0.25-in. (0.635-cm) diameter/ Support post

Figure 3.--Pitot pressure probe used to measure radial profiles indownstream pipe.

NASA/TM--2001-211127 5

55 55

50

L4513.

g= 40

o_ 35

30

250.00

ICenterline

Axial position',_ "_ _/ n

from exit plane, 'fl/\!'/

,n. cm iJt-- CFD prediction 0.000 0.000 _ o_

..... CFD prediction .083 .211 !_! -_

....... CFD prediction .166 .422 _i _-

..........................CFD prediction .250 .635

• Experimental data at exit plane....... _........ I ..... (a)

0.50 1.00 I

IRadial position, Y, in.

Lobe valley

5o _ _.,:_,

45 Axial position _,,S_from exit plane,

40 AZ,in. cm_

35 CFD prediction 0.000 0.000CFD prediction .083 .211

....... CFD prediction .166 .42230 '. .........................CFD prediction .250 .635

• Experimental data at exit plane (b)25 ' ' ' _ ' ' ' ' _ ' ' ' '

0.00 0.25 0.50 I 0.75

I Radial position, Y, in. ICenterline Lobe peak

Figure 4.--Comparisons of experimental and CFD simulation static pressure PT profiles for baseline casewithout pitot probe in model. (a) Profiles from duct centerline to lobe valley. (b) Profiles from duct centerlineto lobe peak, 30 ° off axis.

40

D.

n 30

"-1if)

"_ 20Q.

o

NlO

Axial position Axial position

from exit plane, from exit plane,AZ, AZ,

in. cm50_ in. cm 50 -

CFD prediction 0.000 0.000 -- CFD prediction 0.000 0.000CFD prediction .083 .211 ..... CFD prediction .083 .211

....... CFD prediction .166 .422 ....... CFD prediction .166 .422

..........................CFD prediction .250 .635 ..........................CFD prediction .250 .635

• Experimental data at exit plane • Experimental data at exit plane

Centerline

40

D.

n 30

_ ..... _ ..... _.... (a)

0.00 0.50 1.00 I

I Radial position, Y, in. ILobe valley Centerline

\

0.00 0.25 0.50 I 0.75

I Radial position, Y, in. ILobe peak

Figure 5.--Comparisons of experimental and CFD simulation static pressure Ps profiles for baseline casewithout pitot probe in model. Static pressure ports are located 0.50 in. (1.27 cm) downstream of probe tip.(a) Profiles from duct centerline to lobe valley. (b) Profiles from duct centerline to lobe peak, 30 ° off axis.

NASA/TM--2001- 211127 6

The major problem with the correlation between the experimental and CFD data is in the inter-

pretation of the radial profiles of static pressure, measured using the same pitot probe. Comparisons of

experimental data and CFD simulations are shown in figure 5 and demonstrate discrepancies of a factor

of 2 or more between measured and computed values of PS, even at the duct centerline. This was alsothe case even for CFD simulations taken at axial positions further downstream in the duct, corresponding

to the axial location of the static pressure ports 0.50 in. (1.27 cm) downstream of the probe tip, as seen in

figure 3.

Initial attempts to resolve these discrepancies included rechecking the pressure transducer calibration

technique and values as well as the simulation boundary conditions. However, no significant errors were

discovered. To further investigate these discrepancies, it was decided to proceed with a more detailed

CFD study and to include the presence of the pitot probe in the flow simulations.

Lobe and Pitot Flow Description

Because of the large discrepancy in static pressure profiles between experimental data and CFD simu-

lations observed and discussed in the previous section, the CFD analysis of the duct was repeated with the

pitot probe included in the computational model.

The results of these simulations are shown graphically through contour plots in figures 6 and 7,

which illustrate the flow-field distortion in the vicinity of the probe tip and the static pressure ports

caused by the pitot probe in the presence of the high-velocity and high-pressure gradients. Axial cross

sections of three of the six simulations are shown in figure 6, for static pressure and mach number. The

baseline flow without the probe is shown at the top, whereas flows with the probe at radial positions

0.60 in. (1.5 cm) and 1.30 in. (3.3 cm) are shown immediately below. The flow region shown represents

only the near field of the lobed duct exit plane and probe tip. The overall computational model used in

the updated simulations was the same as that used to generate figure 2.

It is observed in figure 6 that the pitot probe tip is relatively unaffected by the shock formed at the

lobed duct exit plane lip, explaining why the total pressure profiles are in good agreement with the

data, even with the nearby shock system. The shock instead impinges on the probe stem and wraps

around it, which results in large pressure and mach number gradients axially down the stem of the probe.

(Circumferential variations in pressure could have been monitored if the four clocked static ports had

individual readouts. If individual stings or a wedge probe were used, the flow angle could have been

monitored; however, the probe used in this analysis had a common port, providing a uniform static

pressure.)

The effect of the shock system is further illustrated in figure 7, which shows axial cuts of the flow

field in the plane of the pitot probe pressure ports 0.50 in. (1.27 cm) downstream of the probe tip. In

addition to the axial gradient along the stem noted above, there is a strong circumferential gradient of

static pressure and mach number about the probe, which differs substantially from one probe position to

the next. The impact of the probe on the local flow field is clearly visible in these figures. At radial posi-

tion Y = 0.60 in. (1.5 cm), the probe attracts the low static pressure region of flow in the lobe, causing a

deformation of the pressure field. At radial position Y = 1.30 in. (3.3 cm) offthe centerline, the probe

distorts the flow issuing from the probe valley, forcing a type of bifurcation substantially from one probe

position to the next. The impact of the probe on the local flow field is seen clearly in the flow and in the

shifting of the low-pressure region inward toward the duct centerline.

These flow simulations strongly suggest that the shock wave impingement, high-velocity and high-

pressure gradient flow, and the local flow disturbance due to the pitot pressure probe cumulatively con-

tribute to the discrepancies observed in the computed and measured radial profiles of static pressure

shown in figure 5. Such discrepancies would even be found for the case of uniform flows if one were

seeking pressure distributions downstream of the probe (fig. 8). These pitot probe accuracy issues do

not arise in low-speed, incompressible flows because of the absence of compressibility effects.

NASA/TM--2001-211127 7

Ps,psia

15.0013.63

:::::::12.6711.5o10.33

N 9.178.00

I>" 1.5 1.5

O O•_ 1.0 -_ 1.0U]o 8o_ .5 o_ .5

-o 0 -o 0

I I I I I I I I I I I I I I I I I I I _n- 6 7 8 9 lv 6 7 8 9 10

Axial position, Z, in. Axial position, Z, in.

probe

iiiiiilliiiiiiii iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii i ,_ 2.0 .__ 2.0 _ M

>-" 1.5 >-" 1.5 _ _ 1.80_ 1.50

o o•-_ 1.0 Probe at -_ 1.0 1.20"_ u_ 0.90

.5 Y = 0.60 in. .5 O.6O0.30

-5 0 -5 0 0.00

n" 6 7 8 9 10 6 7 8 9 10


,_ 2.0.m

>-" 1.5t-

O•_ 1.0U]OQ. .5

-o 0

rr

,_ 2.0

>-" 1.5

........" Probe at

__ _ Y =1.30 in.-_'_" 1.0.5

-5 0

6 7 8 9 10 6 7 8 9 10


Static pressure Mach number

Figure 6.--Flow-field distortions in static pressure Ps and mach number M distributions for three of six CFDsimulations. Top row: baseline flow field without probe. Middle row: probe at 0.60 in. (1.52 cm) off duct

centerline. Bottom row: probe at 1.30 in. (3.30 cm) off duct centerline.

NASA/TM--2001- 211127 8

" 1.5

>_-

1.oStatic

pressure __ .5

=50¢r

Circumferential position, X, in.

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_" 1.5>,.-

iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_ 1.0_iiiiiiiiiiiiiiiiiiiiiiiiiiNiiiiiiiiiiiiiiiiiiiiiiiii__ ..=

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii__: 8_ .5

0,,,i .... 11,3-

0 1 -1


'- 1.5>2

1.0O

o_ .5

0 1 -1

psiaiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii

1°00........................................................_ 13.83

..........................................12.6711.5010._3

.¥_ s.ooI I I I I I I I I I I I I I I I I I I

0 1


" 1.5>_-

,_ 1.0Mach

number 8_..5

=5

tr 0 , , , ,

m"

" 1.5>_-

1.o

8_ .5

ii!i!iiiiiiiii_" ._

' ' 1 ¢r


Baseline flow (no probe)

ml


Probe at Y = 0.60 in.

>-".1.5

1.oU')

°.5

.... I .... m .... I .... m 1'3" 0

0

1.801.50

: ::::::::::: 1,20

0.90P: 0.60

N 0.300.00

" , , , i .... i .... i , , i

1 -1 0 1


Probe at Y = 1.30 in.

Figure 7.--Axial slices in plane of pitot-static pressure ports, 0.50 in. (1.27 cm) downstream of probe tip, for three

of six CFD simulations. Top row: static pressure Ps distributions. Bottom row: mach number M distributions.

Figure 8.--Pitot probe and support shock structureinteraction.

NASA/TM--2001-211127 9

With this knowledge, CFD simulations with the pitot probe in six unique positions were run, post-

processed, and compared again with the radial profiles of static pressure.

Comparison of Experimental Data With CFD Predictions

With the new understanding gained from the analysis of the flow-field distortions caused by the

pitot probe, the six CFD simulations that included the probe were further sampled and compared with the

experimental data.

Total Pressure Profile Results

Although predicted total pressure profiles were already considered to be in good agreement with

experimental data, the results were replotted using the probe tip as the sampling point. These results are

shown in figure 9 in the form of a radial total pressure profile aligned tangentially with the lobe valley at

its highest radius. Both the predictions and experimental data show a relatively flat radial profile across

the duct at a level of about 50 psia (345 kPaa) (inlet pressure specification). Note that discrete data points

and simulation points are now being plotted because the presence of the probe affects the flow field dif-

ferently at each unique location. Thus, with the probe causing large local disturbances, it is no longer

assumed that line profiles can be extracted from a single CFD simulation and be compared with data.

For the simulation with the probe at a radial position ofY = 1.30 in. (3.3 cm), the CFD predictions

fall off slightly from the experimental value. It was observed in the CFD line profiles in figure 4 that there

was a rapid dropoff in the total pressure at positions further axially downstream. This pressure drop can

be appreciated to some extent in the contour plots in figures 6 and 7, which show the pitot probe facing

into the lobe wall boundary layer, with high-speed flow on the lower probe surface and low-speed flow

on the upper surface. Under this condition, which features extreme pressure and velocity gradients, any

variance in either the radial or axial positioning of the probe can result in large variation in the sampled

60

50

o_ 40¢-

13.

g' 30

£1.

-_ 20

lO

o

D

• Experimental data[] CFD simulation

+LI:.J:I+t,:I,; tJl:Jt,lt_ t } t : J t |

0.00 0.25 0.50 0.75 1.00 1.25 1.50

I Radial position, Y, in. ICenterline Lobe valley

Figure 9.--Comparison of experimental and CFD simulationtotal pressure PT profiles with pitot probe in model. CFDdata points sampled at probe tip.

NASA/TM--2001- 211127 10

pressure.Thus,it isconcludedthatbecauseof theproximityoftheprobetothelobeexitplaneanditspositionwithintheductboundarylayer,thisparticulardatapointissubjecttohigheruncertaintythandatapointstakenwiththeprobeinthefree-streampositionsawayfromtheductwall.

Static Pressure Profile Results

The key results of comparisons between the static pressure measurements and CFD analysis appear in

figure 10, which presents radial profiles of measured and predicted static pressures that include the pres-

ence of the pitot probe. The overall agreement between experimental data and CFD simulations is good

and is much better than that observed without including the probe (fig. 5). Once again, discrete experi-

mental data points and discrete CFD simulation points are compared because the probe affects the local

flow field differently at each probe position. At the duct centerline (see fig. 10), excellent agreement

between measurement and prediction is observed. This agreement was expected, because the CFD com-

putational model, taken to be a symmetric 60 ° sector, models the flow blockage from a pipe (in this case,

a pitot pressure probe) situated at the centerline.

As the probe is moved radially outboard, there is a discrepancy in the comparison at a radial position

of Y = 0.20 in. (0.5 cm) off the duct centerline. The interpretation of this result is that with a symmetric

model of the duct, insertion of a single probe is the physical equivalent of inserting six probes into the

domain and accounts for the same percentage of flow blockage in that area. The symmetry modeling

condition thus forces an unrealistic flow blockage to occur in the duct, with resulting questionable to poor

agreement in the results. Further outboard, the agreement between predictions and measurements is very

good. Although the flow model is still symmetric, as the probe moves deeper into the lobed duct valley,

the interaction with the actual probe and modeled symmetry probes becomes negligible, and the probe

correctly models the local flow disturbances. This agreement extends to the furthest radial point measured

16

14

_12

g= 8

o_ 6O

4

t []

o[2

i

• Experimental data[] CFD simulation

, _ [, _ tJ I_ l_lt J _ :l_ tJ t ]J t_J _J : LJ:t

0.00 0.25 0.50 0.75 1.00 1.25 1.50

I Radial position, Y, in. ICenterline Lobe valley

Figure 10.--Comparison of experimental and CFD simulation

static pressure Ps profiles with pitot probe in model. CFDdata points taken 0.50 in. (1.27 cm) downstream of probetip and circumferentially averaged.

NASA/TM--2001-211127 11

topositionY = 1.30in. (3.3cm),wherethestaticpressureportsareexposedto acomplexflowfieldrang-ingfromlow-speedsubsonicflowontheportsfacingtheouterductwalltohigh-speedsupersonicflowataboutmach1.8onthepressureportsfacingtheductcenterline.

Modelingofthepitotprobeforthissupersonicflowappearstobenecessarytointerprettheexperimentalpressuremeasurements.Althoughprobe-inducedflowinterferenceisanecessityinmanycomplexandhigh-speedapplicationsinwhichopticalandothernonintrusivemeasurementsmaybeapplied,thelocalflowdisturbancesobservedinthesesupersonicaerodynamicmeasurementswereverylargeandcouldnoteasilybeinterpretedwithoutthebenefitoftheCFDmodeling.Forgoodagreementacrosstheentireductcrosssection,it is likelynecessarytomodelthefull 360° of the duct to avoid the

probe blockage effect described above.

Effective Area Analysis

For use in flight, the lobed duct must have a low area discharge coefficient (ACD) loss. Analysisof the CAD solid model of the lobed duct indicates that the cross-sectional area is equal to 2.748 in. 2

(17.73 cm 2) along the 0.25-in. (0.64-cm) straight length just upstream of the lobe exit plane. The lobed

duct area is equal to 3.08 in. 2 (19.87 cm 2) at the lobe exit plane and includes the area increase due to a

0.09-in.- (0.24-cm-) radius lip along the edge of the lobe exit.

Computing the duct effective area in compressible, transonic flow introduces complications in the

data reduction process, particularly those involving measurements taken with the pitot probe. Two

experimental cold-flow measurements of the lobed duct parameter AC D were performed. The first

used an aluminum prototype lobed duct on an incompressible water table and gave AC D = 2.35 in. 2

(15.16 cm2). The second, a compressible blowdown air test, used the stainless steel experimental lobe2 2

only, discharged into the 10-in. (25.4-cm.) duct, and gave AC D = 2.57 in. (16.58 cm ). For the lobe-TVC2 2

assembly, a hot-flow (reacting) test resulted in a value of 2.90 in. (18.71 cm ). The experimental AC D

is calculated from the measured mass flow rate and the pressure differential between the plenum and the

duct static pressure DP (fig. l(b)). The density is evaluated at the upstream plenum conditions.2 2

CFD simulations produced a value ofAC D = 3.05 in. (19.68 cm ) when corrected for compressibil-2 2

ity effects but produced only 1.65 in. (10.64 cm ) when not corrected. With compressibility effects taken

into account, the lobed duct appears to have a very high discharge coefficient CD of approximately 0.99,

which is an acceptable value for use in flight hardware applications.

Concluding Remarks

This report described an experimental and computational investigation aimed at understanding

diffuser-mixer exit profiles for a lobed duct design that will potentially be used for higher mach number

combustion evaluations of the trapped vortex combustor (TVC). Experimental hardware was described by

using CAD solid models to show the major flow features present in this diffuser-mixer combustor system.

Initial CFD simulations were compared with experimental data and showed good agreement with total

pressure measurements but poor agreement with static pressure measurements, all of which were taken

with a single simple pitot pressure probe sampling one sector of the lobe configuration. Even when con-

sidering axial and radial variations in positioning of either the probe or the CFD sampling plane, checking

the pressure transducer calibrations, and verifying the CFD boundary conditions, poor agreementremained.

The CFD modeling methodology was modified to take into account the presence of the pitot probe,

and the simulations were rerun. Flow-field contours of static pressure and mach number in the vicinity

of the probe verified that the probe caused large, local disturbances. The graphics also showed that large

NASA/TM--2001- 211127 12

axial,radial,andtangentialvelocityandpressuregradientsnearthestaticpressureportswereactingontheprobeandwerereflectedintheexperimentaldatapoints.

Withtheprobeincludedin theCFDmodel,sixnewsimulationswererunwiththeprobemovedthroughthemodelasin theexperiment.Theseresultswereveryencouragingin thattheestablishedgoodagreementbetweenmeasuredandcomputedvaluesfortotalpressurewasrepeatedunderthenewmethodology,andstaticpressureprofileswerealsonowingoodagreement.

Measurementstakenataradialposition0.20in.(0.5cm)off theductcenterlinewereinpooragree-mentwithpredictions.ThisresultwasattributedtothesymmetrymodelingconditionassumedfortheCFDsimulations:atapositionclosetothecenterline,theeffectof anetworkof sixprobeswastocauseaflowblockage(modelingartifact).Thiseffectwasnegligibleatlargerradialpositionswheregoodagreementbetweendataandpredictionswasachieved.Especiallyimpressivewasagreementataradialpositionwheretheprobestraddledtheboundarylaye_shearlayercomingoff thelobevalley.At thisposition,theprobeuppersurfacefacedlow-speed(subsonic)flowconditions,whereastheprobelowersurfacefacedtheductcenterlineandexperiencedanattachedshockwavewithaspeedexceedingmach1.7.CircumferentialaveragingoftheCFDpredictedstaticpressuresatthefourpressureportsresultedinexcellentagreementwithdata.

It isconcludedthatwiththeCFDsimulationresults,thepressuremeasurementsin thelobeddiffuser-mixeraremoreaccuratelyinterpreted,andthepressureperformanceoftheductisbetterunderstoodsothatdesignimprovementof thelobeddiffuser-mixersystemcanproceedwithmoreconfidence.

NASA/TM--2001-211127 13

Appendix--High-Speed Aerodynamic Results for Flow Through a Lobed

Diffuser-Mixer Duct Issuing Into a Large Diameter Circular Pipe

Herein are supplementary charts from the CFD study of the effects of a pitot pressure probe in the

supersonic flow path. These CFD results provided the initial proof that the pressure probe was interacting

with the flow and suggest that at least the probe tip should be modeled in any duct flow cases involving

supersonic and transonic flows. In many cases the support sting also requires modeling as bow shocks,

shock interaction loadings, and probe angle of attack can interfere with the measurements and cause

distortions (see fig. 8).

Each of the supplemental figures is titled and largely self-explanatory when viewed as supplementary

to the body of the report. Consequently, only a brief statement relative to each figure will be given.

• Figure 11 gives the geometric parameters associated with the test section (see fig. 1).

• Figures 12 and 13 provide more details of the local shock patterns associated with the lobed

diffuser-mixer configuration and provide additional details supporting figure 2.

• Figures 14 and 15 show the static pressure contours at the exit plane and 0.5 in. downstream at the

four-hole static location on the pitot probe (see also fig. 7).

• Figures 16 and 17 show the total pressure contours at the exit plane and 0.5 in. downstream at the

four-hole static location on the pitot probe.

• Figures 18 and 19 give the local mach number contours at the exit plane and 0.5 in. downstream at

the four-hole static location on the pitot probe (see also fig. 7).

• Figure 20 represents the axial variations in static pressure along the lobed diffuser-mixer valley (see

figs. 1 and 11) without the pitot probe in the flow field.

• Figure 21 represents the axial variations in static pressure along the lobed diffuser-mixer valley (see

figs. 1 and 11) with the pitot probe at the centerline of the flow field.

• Figure 22 gives the static pressure contours for the pitot probe at 0.75 in. from the flow centerline

(compare with fig. 6).

• Figure 23 represents the axial variations in total pressure along the lobed diffuser-mixer valley (see


• Figure 24 represents the axial variations in total pressure along the lobed diffuser-mixer valley (see


• Figure 25 gives the total pressure contours for the pitot probe at 0.75 in. from the flow centerline.

• Figure 26 represents the axial variations in mach number along the lobed diffuser-mixer valley (see


• Figure 27 represents the axial variations in mach number along the lobed diffuser-mixer valley (see


• Figure 28 gives mach number contours for the pitot probe at 0.75 in. from the flow centerline (com-

pare with fig. 6).

• Figure 29 gives axial slices in static pressure for the lobed diffuser-mixer (see figs. 1 and 11) with-

out the pitot probe in the flow field to 1.5 in. downstream of the exit plane (see also fig. 7).

• Figure 30 gives axial slices in total pressure for the lobed diffuser-mixer (see figs. 1 and 11) without

the pitot probe in the flow field to 1.5 in. downstream of the exit plane.

• Figure 31 gives axial slices in mach number for the lobed diffuser-mixer (see figs. 1 and 11) without

the pitot probe in the flow field to 1.5 in. downstream of the exit plane (see also fig. 7).

• Figure 32 gives axial slices in static pressure for the lobed diffuser-mixer (see figs. 1 and 11) with

the pitot probe in the flow field to 1.5 in. downstream of the exit plane (see also fig. 7).

• Figure 33 gives axial slices in total pressure for the lobed diffuser-mixer (see figs. 1 and 11) with the

pitot probe in the flow field to 1.5 in. downstream of the exit plane.

• Figure 34 gives axial slices in mach number for the lobed diffuser-mixer (see figs. 1 and 11) with

the pitot probe at the flow centerline to 1.5 in. downstream of the exit plane (see also fig. 7).

NASA/TM--2001-211127 15

• Figure35givesaxialslicesinstaticpressureforthelobeddiffuser-mixer(seefigs.1and11)withthepitotprobein thelobevalleyto 1.5in.downstreamoftheexitplane(seealsofigure7).

• Figure36givesaxialslicesin totalpressureforthelobeddiffuser-mixer(seefigs.1and11)withthepitotprobein thelobevalleyto 1.5in.downstreamoftheexitplane.

• Figure37givesaxialslicesinmachnumberforthelobeddiffuser-mixer(seefigs.1and11)withthepitotprobein thelobevalleyto 1.5in.downstreamoftheexitplane(seealsofig.7).

NASA/TM--2001-211127 16

z>>

I00

b_

--4

(a)

Flow

t !

Red _ 1.50

(C) ."

Figure 11.--Test section geometric parameters and three-dimensional cellular shock structure downstream as lobe expands into straight pipe. (a) Lobeddiffuser-mixer geometry. (b) Overall view. (c) Local details of cellular shock structure.

>

I00

b_

-.J

_ :::::-:i:i:-:::i:_ :i:i:-:i_i:-:i:i:- i:::_ _i:i:-:13-:i:i: :i:i:-:i:i/_:i:i:-:i:?-::_i/_i:i_-:i_i:_:i:::_:i:/:18-:_:i: _I:I/LI:I_-:I_S_:::I:_i:i_-_i:/_:_:i:LI::/_I::: _:I:_:-::8-:_:I_-_I:SLI:I_-_I:I/:i:/

(a)

Lobe

/

/i

//

z._ Shock train rapidlytransforms from lobe

shape to circular shape

Isocontour Mach number

.. "/-- Shock rapidly reverts M" ! to circular shape/ Yellow 1.32

1.50Red

_iiil iiiiiii ::_.........

Figure 12,--Local shock pattern associated with lobed diffuser-mixer configuration, including shape transition and shock train dissipation, (a) Overall view,

(b) Local details,

>

I00

bJ

-...J

(a)

Flow direction

L Supersonic jet coreflow with shock train J

/!

Isocontours Mach number,M

Cyan _ 0.94Green _ 1.12Yellow 1.32

Red _ 1.50

\\

\

Sonic surface between \cyan and green contours---_

(b)

" "__"_ Fully supersonic core flow _. -" _"with discrete structures _ " _"

Figure 1&--Cellular shock structure in near field of lobe exit. (a) Overall view. (b) Local details.

,_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiTi_

,iii..............................................................ii_,_

Figure 14.--Distortion of static pressure Ps field due to pitot probe, shown at lobe exit plane Z = 0.00 in. (flushwith tip of pitot probe). (a) Baseline CFD solution without probe. (b) CFD solution with probe along centerline.

(c) CFD solution with probe offset at radius of 0.75 in. in lobe valley.

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii__:¸¸_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_

%iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii?__iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii___iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_:_

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii?_`%iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii__

`_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii?_`_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii....

Ps, _'_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii''_ m, _"_!iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_'_ m, _iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_::psia _iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii::: psia ................................. psia :_iiiiiii.......................................iiiiiiii_::30.0 ::_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii.... 30.0

N 26.0 :_iiiiiiiiiiiiii_ N 26.0 N 30.0i:i:i:i:i:i:i:i:i: 'i:i:i:i:i:i:i:i:i i:i:i:i:i:i:i:i:i: 26.0

iiiiiiiiiiiiiiiiii22.0 iiiiiiiiiiiiiiiiii22.0 iiiiiiiiiiiiiiiiii22.0

iiiiiiiiiiiiiiiiii18.0 iiiiiiiiiiiiiiiiii18.0 iiiiiiiiiiiiiiiiii18.0iiiiiiiiiiiiiiiiii14.0 iiiiiiiiiiiiiiiiii14.0 iiiiiiiiiiiiiiiiii14.0......... .........

N lo.o N lo.o N lo.o6.0 6.0 6.0

(a) (b) (c)

Figure 1&--Distortion of static pressure Ps field due to pitot probe, shown at 0.50 in. downstream of lobe exitplane. All contours taken at 0.50 in. downstream of lobe exit plane, coincident with location of static pressure

ports on probe. (a) Baseline CFD solution without probe. (b) CFD solution with probe aligned along centerline.

(c) CFD solution with probe in lobe valley, offset at radius of 0.75 in.

NASA/TM--2001- 211127 20

Niiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii

iiiiiiiiiiiiiiiiii

@

PT,psia48.0

42.5

37.0

31.5

26.0

20.5

15.0

PT, ......... PT,psia psia48.0 48.0

N 42.5 N 42.5.................. [!i!i!i!i!i!i!i!i!_ 37.0 ..................37.0.................. iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii31.5 iiiiiiiiiiiiiiiiii31.5iiiiiiiiiiiiiiiiii26.0 iiiiiiiiiiiiiiiiii26.0..................20.5 ..................@ 150 @ 205

15.0

(a) (b) (c)

Figure 16.--Total pressure PT contours with and without pitot probe, shown at lobe exit plane Z = 0.00 in. (flushwith tip of pitot probe). No total pressure flow-field distortions observed at this plane. (a) Baseline CFD solution

without probe. (b) CFD solution with probe aligned along centerline. (c) CFD solution with probe in lobe valley.

PT, PT, PT,psia psia psia

48.0 _ 48.048.0 _ 42.5 42.542.5 _: _

37.0 37.0 37.0

31.5 31.5 31.526.0 26.0 26.0

...................................................... 20.5@ 205 @ 205 @ 15015.0 15.0

(a) (b) (c)

Figure 17.--Total pressure PT contours with and without pitot probe, shown in plane of static pressureports 0.50 in. downstream of lobe exit plane, coincident with the location of static pressure ports on probe.

(a) Baseline CFD solution without probe. (b) CFD solution with probe aligned along centerline. (c) CFD

solution with probe in lobe valley, offset at radius of 0.75 in.

NASA/TM--2001-211127 21

z>>

I00

b_

--...1

b_b_

_.

• _ __0 0

.... 0

_oo_o_

_o___o-_

Q.

_o_

_P

_._____o°_-_b_o_2

t,_ 0 "_

_o_oo-,_._._o__._o_o

0"

0

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_0000_

_ NiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiNbg_bg_g

_ NiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiNbg_bg_g

"11

_-_.o_"-n-_

o _

gY_g'_

-N'_0 0

gg__<g'_

_-gg-_.

_'_

_, __-_.--. ¢-- __

_._

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_0000_

0_0_0_0

"I

_. NiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiN0000 _

0_0_0_0

o_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_

__ --,__: 0 r3"l 0 r3"l 0 r3"l 0

__,.o

>

I00

b_

-...J

b_

Inlet

(a) 0 10 20

Axial position, Z, in.

Exitpv

(b) 6 7 8 9


Figure 20.--Axial variation in static pressure Ps along lobed diffuser-mixer valley, without pitot probe. (a) Overall view, showing multiple cellularshock structures. (b) Detail in vicinity of exit plane.

z>>

I00

b_

-...J

b_

._5

0,m

Inlet _r

go

Ps,psia

50.00_ 42.83

35.67

28.50

21.33

...................14.177.00

(a) -5

1

.m

=-O

_0OD_

_5

tr

0 5 10 15 20


Exit

iv

Figure 21 .--Axial variation in static pressure Ps along lobed diffuser-mixer valley with pitot probe (0.125-in. diameter; head, 1.5 in. long) alongduct centerline. Static taps located 0.50 in. behind tip of probe. (a) Overall view, showing shock interaction with probe. (b) Detail of pressure

variation along probe tube.

z>>

I00

b_

-...J

Inlet

._ 5>__

O

o)

..oy

-O

g0

Exit

L_h

Ps,psia30.0

_ 26.0

22.0

18.0

14.0

10.06.0

(a)

f1

10


20

0

(b)

/ /

I

/ 86 / 7/// / Axial position, Z, in.

/ /

Probe tip _/ /'-Static taps

Figure 22.--Axial variation in static pressure Ps along lobed diffuser-mixer valley with pitot probe (0.125-in. diameter; head, 1.50 in. long) alongduct centerline. Static taps located 0.50 in. behind tip of probe. (a) Overall view, showing shock interaction with probe. (b) Detail of pressurevariation along probe tube.

z>>

I00

b_

"-4

._ 5

0

Inlet _D.

_5g:o

(a) 10


20

Exit

pv

PT,psia48.0

42.5

37.0

31.5

26.0

20.5

15.0

(b) 6 7 8 9


Figure 23.--Axial variation in total pressure PT along lobed diffuser-mixer valley without pitot probe. (a) Overall view showing axial variation ofpressure losses as jet issues into large pipe. (b) Detail in vicinity of exit plane.

=>>

I00

b_

-...J

b_-...J

Inlet

.__/5>.-

O

¢9O

b,_ D.w03

_3

Exit

PT,psia

50.00_!_!_,_!_!_!_!_!_':43.42

36.83

30.25

23.67

.......... 1 7.08_ 10.60

(a) --5

1

,m

>.-=-O

OD.

r_

0 5 10 15 20


/(b) 6 7 / _ 8 9

/ Axial position, Z, in.

Probe tip ---/ _ Static taps

Figure 24.--Axial variation in total pressure PT along lobed diffuser-mixer valley with pitot probe (0.125-in. diameter; head, 1.5 in. long) alongduct centerline. Static taps located 0.50 in. behind tip of probe. (a) Overall view of pressure losses as jet issues into large pipe. (b) Detailalong probe surface, near exit plane.

=>>

I00

b_

-...J

Inlet

,__Z5

O.m

¢9

O

"O

co0

(a) 0 10


20

Exit

Oo

wmwm

PT,psia48.0

42.5

37.0

31.5

26.0

20.5

15.0

,m

O

e9OO.

-8_5

rr 0/ /

(b) 6 / 7 / 8/ /

/ / Axial position, Z, in.

Probe tip g Z-Static taps

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii

9 101

Figure 25.--Axial variation in total pressure PT along lobed diffuser-mixer valley with pitot probe (0.125-in.diameter; head, 1.5 in. long) offset atradius of 0.75 in. Static taps located 0.50 in. behind tip of probe. (a) Overall view of distortion in pressure losses as jet issues into large pipe.(b) Detail along probe surface, near exit plane.

>

I00

bJ

-...J

bJ

Inlet

M

1.80_ 1.50

1.20

0.90

0.60

0.300.00

(a)

,m

=-O

U)O

_3c_£g

0 10 20


Exit

(b) 6 7 8 9


Figure 26.--Axial variation in mach number M along lobed diffuser-mixer valley without pitot probe. (a) Overall view showing axial dissipationof shocks as jet issues into large pipe. (b) Detail in vicinity of exit plane.

z>>

I00

b_

-..J

0

Inlet

(a) -51 0 5

M

1.80

_ 1.50

1.20 .__

0.90 >-"=-

0.60 ._o

1.a00.00 c_

10 15 20


Exitb.

(b) 6 / / 7 ti 8 9" " Axial position, Z, in.

Probe tip J Z'-Static taps

Figure 27.--Axial variation in mach number M along the lobed diffuser-mixer valley with pitot probe (0.125-in. diameter; head, 1.5 in. long) alongduct centerline. Static taps located 0.50 in. behind tip of probe. (a) Overall view of shock structure as jet issues into large pipe. (b) Detail of

variation along probe surface near exit plane.

z>D_

>

I00

b_

-...J

._ 5

0

Inlet _ o

_5_o

Exit

(a) 0 10 20


M

1.50

1.25

1.00

0.75

0.50

0.25

0.00

.m

iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iii_i_

:,_ _iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiii!_!__

_ iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiii_!!i__. -iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiiiiiiii_!iii_

iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_ii_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_!!i!i_

_r 0 _iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iii!iiiiiiii_/ /

-I I I I I _ I I/I I I I I I I I I I I I I I I(b) 6 l 7 / 8 9 101

/ // / Axial position, Z, in.

Probe tip _ /'-- Static taps

Figure 28.--Axial variation in mach number M along lobed diffuser-mixer valley with pitot probe (0.125-in. diameter; head, 1.5 in. long) offsetat radius of 0.75 in. Static taps located 0.50 in. behind tip of probe• (a) Overall view of distortion of shock pattern as jet issues into large pipe.(b) Detail of variation along probe surface near exit plane•

z>>

I00

..,q ®

Ps,psia30.0

26.022.018.0

14.010.06.0

Z = 6.90 in. (-0.25 in.)

jiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii t

_!iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii!!i_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii _

Z = 7.15 in.

(0.0 in., exit plane)

iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiI_ _i

Z = 7.40 in. (+0.25 in.)

"_iiiii iiii_':_

Z = 7.65 in. (+0.50 in.)

L_O

::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

iiiiiiiiiiiiiiii iiiiii iiiiiiiiiiiiiiiiii iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii

Z = 7.90 in. (+0.75 in.)

::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

Z = 8.15 in. (+1.00 in.)

___

::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

iiii

:_iiiiiii_iiiiii::

Z = 8.40 in. (+1.25 in.) Z = 8.65 in. (+1.50 in.)

Figure 29.--Axial slices of static pressure Ps contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.Without pitot probe.

>

I00

..,q

PT,psia48.042.5

37.0

31.5

26.0

20.515.0

Z = 6.90 in. (-0.25 in.) Z = 7.15 in.


Z = 7.40 in. (+0.25 in.) Z = 7.65 in. (+0.50 in.)

L_L_

.... iiiii iiiiii ?

Z = 7.90 in. (+0.75 in.) Z = 8.15 in. (+1.00 in.) Z = 8.40 in. (+1.25 in.) Z = 8.65 in. (+1.50 in.)

Figure 30.--Axial slices of total pressure PT contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.Without pitot probe.

>

I00

..,q

M

1.50

1.25

1.00

0.75

0.50

0.25

0.00

Z = 6.90 in. (-0.25 in.) Z = 7.15 in.


Z = 7.40 in. (+0.25 in.) Z = 7.65 in. (+0.50 in.)


Figure 31 .--Axial slices of mach number M contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.

Without pitot probe.

z>>

I00

b_

-...J

Ps,psia30.026.0

22.018.014.0

10.06.0

Z = 6.90 in. (-0.25 in.)

!iii

Z = 7.15 in.

(0.0 in., exit plane)Z = 7.40 in. (+0.25 in.) Z = 7.65 in. (+0.50 in.)

ii!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!iiiii!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!iiii!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!i!iiilililililililililililililililill_ ilililililililililililililililiiii7

iiiI!i


Figure 32.--Axial slices of static pressure Ps contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.With pitot probe along duct centerline.

>

I00

..,q

PT,psia48.042.5

37.0

31.5

26.0

20.515.0

Z = 6.90 in. (-0.25 in.) Z = 7.15 in.


Z = 7.40 in. (+0.25 in.) Z = 7.65 in. (+0.50 in.)

L_

Z = 7.90 in. (+0.75 in.)

:_:_i_i_i_i_i_i_::_i_i_i_!_¸¸ :_............................

Z = 8.15 in. (+1.00 in.) Z = 8.40 in. (+1.25 in.) Z = 8.65 in. (+1.50 in.)

Figure 33.--Axial slices of total pressure PT contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.With pitot probe along duct centerline.

>

I00

..,q ®M

1.50

1.25

1.00

0.75

0.50

0.25

0.00

Z = 6.90 in. (-0.25 in.) Z = 7.15 in.


Z = 7.40 in. (+0.25 in.) Z = 7.65 in. (+0.50 in.)

L_


Figure 34.--Axial slices of mach number M contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.

With pitot probe along duct centerline.

=>>

I00

..,q

Ps,psia30.026.0

iiiiiiiii

22.0

18.014.0

10.06.0

Z = 6.90 in. (-0.25 in.)

i!iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_

'_iiii iiii_''_

Z = 7.15 in.

(0.0 in., exit plane)Z = 7.40 in. (+0.25 in.)

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_%_i_i_i_i_i_i_i_i_i_i_i_i_i_i_i_iiiiiiiiiiiiii_¸ %iiiiiiiii_i_i_i_i_i_i_i_i_i_i_i_i_i_i_i_i_i_i_i__:

_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiiii_Iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_':_

Z = 7.65 in. (+0.50 in.)

L_aOO

:'_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_':::::_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii'_

:iiiiiiiiiiiiiiiiiiiiiiii_i_i_i_i_i_i_i_i_i_i_i_i_i_i_i_iiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiii

Z = 7.90 in. (+0.75 in.) Z = 8.15 in. (+1.00 in.)

:::::::::_:i_i_i_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_i_i_i:i:_:::::::::::

iiiiiiiiiiiiiiiiiiiiii '

Z = 8.40 in. (+1.25 in.)

:iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii

Z = 8.65 in. (+1.50 in.)

Figure 35.--Axial slices of static pressure Ps contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.With pitot probe offset 0.75 in. from duct centerline.

>

I00

..,q

PT,psia48.042.5

37.0

31.526.0

20.515.0

Z = 6.90 in. (-0.25 in.) Z = 7.15 in.

(0.0 in., exit plane)Z = 7.40 in. (+0.25 in.) Z = 7.65 in. (+0.50 in.)

iii iiii iii iii..... iiii iii_'''_iiiiiii_iiiiii_iiiiiiiiiii!iiii


Figure 36.--Axial slices of total pressure PT contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.With pitot probe offset 0.75 in. from duct centerline.

>

I00

..,q

0

M

1.501.25

1.000.750.50

0.250.00

Z = 6.90 in. (-0.25 in.) Z = 7.15 in.


""' i i iiiiiiiiiiiiiiiiiiiiiiiii..................................!i1ii iiiiiiiiiiiiiiiiiiiiii

,l!!!iiiiiiiiiiiiiiiiiiiiiiiiii iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii!!! ,

Z = 7.40 in. (+0.25 in.) Z = 7.65 in. (+0.50 in.)

ii!_!_!____iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_____!iiI _iiiii!_!_i__i_!!!_!iiiii_

_i_!!_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iii_


Figure 37.--Axial slices of mach number M contour through lobed diffuser-mixer from 0.25 in. upstream to 1.50 in. downstream of exit plane.With pitot probe offset 0.75 in. from duct centerline.

References

1. Roquemore, W.M., et al.: Trapped Vortex Combustor Concept for Gas Turbine Engines. AIAA Paper

2001 0483, 2001.

2. Burrus, D.L., et ah: Performance Assessment of a Prototype Trapped Vortex Combustor Concept for

Gas Turbine Application. ASME Paper 2001 GT 0087, 2001.

3. Hendricks, R.C., et ah: Experimental and Computational Study of Trapped Vortex Combustor Sector

Rig With High-Speed Diffuser Flow. Int. J. Rotat. Mach., vol. 7, no. 6, 2001, pp. 375 385.

4. Ryder, Robert C., Jr.; and McDivitt, Timothy: Application of the National Combustion Code Towards

Industrial Gas Fired Heaters. AIAA Paper 2000 0456, 2000.

5. Beckwith, Thomas G.; Marangoni, Roy D.; and Lienhard V, John H.: Mechanical Measurements. Fifth

ed., Addison-Wesley, Reading MA, 1993.

NASA/TM--2001-211127 41

Form ApprovedREPORT DOCUMENTATION PAGEOMB No. 0704-0188

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,gathering and maintainingthe data needed, and completing and reviewingthe collection of information. Send comments regarding this burden estimate or any other aspect of thiscollection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 JeffersonDavis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188),Washington, DC 20503.

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED

July 2002 Technical Memorandum

5. FUNDING NUMBERS4. TITLE AND SUBTITLE

Measurement and Computation of Supersonic Flow in a Lobed Diffuser-Mixer

for Trapped Vortex Combustors

6. AUTHOR(S)

Andreja Brankovic, Robert C. Ryder, Jr., Robert C. Hendricks, Nan-Suey Liu,

John R. Gallagher, Dale T. Shouse, W. Melvyn Roquemore, Clayton S. Cooper,

David L. Burrus, and John A. Hendricks

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)


John H. Glenn Research Center at Lewis Field

Cleveland, Ohio 44135-3191

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)


Washington, DC 20546-0001

WU-910-30-11-00

8. PERFORMING ORGANIZATIONREPORT NUMBER

E-12861

10. SPONSORING/MONITORINGAGENCY REPORT NUMBER

NASA TM--2002-211127

11. SUPPLEMENTARY NOTES

Prepared for the 2001 19th International Congress on Instrumentation in Aerospace Simulation Facilities (ICIASF 2001) cosponsored by IEEE, AES,

NASA Glenn, and OAI, Cleveland, Ohio, August 27 30, 2001. Andreja Braxtkovic and Robert C. Ryder, Jr., Flow Paxametrics, LLC, Bear, Delaware

19701; Robert C. Hendricks, Nax>Suey Liu, and John R. Gallagher, NASA Glenn Research Center; Dale T. Shouse andW. Melvyn Roquemore, Wright-

Patterson Air Force Base, Dayton, Ohio 45433; Clayton S. Cooper and David L. Burrus, General Electric Aircraft Engines, Evendale, Ohio 45215; and

John A. Hendricks, Diligent Design, Toledo, Ohio 43614. Responsible person, R.C. Hendricks, organization code 5000, 216_77_507.

12a. DISTRIBUTION/AVAILABILITY STATEMENT

Unclassified - Unlimited

Subject Categories: 07, 28, and 34 Distribution: Nonstandard

Available electronically at http://gltrs.grc.nasa.gov/GITRS

This publication is available from the NASA Center for AeroSpace Information, 301-621-0390.

12b. DISTRIBUTION CODE

13. ABSTRACT (Maximum 200 words)

The trapped vortex combustor (TVC) pioneered by Air Force Research Laboratories (AFRL) is under consideration as an altemafive to conventional gas turbine

combustors. The TVC has demonstrated excellent operational characteristics such as high combustion efficiency, low NO x emissions, effective flame stabilization,excellent high-altitude relight capability, and operation in the lean-burn or rich burn-quick quench-lean burn (RQL) modes of combustion. It also has excellent

potential for lowering the engine combustor weight. This performance at low to moderate combustor mach numbers has stimulated interest in its ability to operate

at higher combustion mach number, and for aerospace, this implies potentially higher flight mach numbers. To this end, a lobed diffuser-mixer that enhances the

fuel-air mixing in the TVC combustor core was designed and evaluated, wi/Ja special attention paid to the potential shock system entering the combnstor core. For

the present investigation, the lobed diffuser-mixer combustor rig is in a full annular configuration featuring sixfold symmetry among the lobes, symmetxy withineach lobe, and plain parallel, symme/xic incident flow'. During hardware cold-flow testing, significant discrepancies were found between computed and measured

values for the pitot-probe-averaged static pressure profiles at the lobe exit plane. Computational fluid dynamics (CFD) simulations were initiated to determine

whether the static pressure probe was causing high local flow-field disturbances in the supersonic flow exiting the diffuser-mixer and whe/Jaer shock wave

impingement on the pitot probe tip, pressure ports, or surface was the cause of the discrepancies. Simulations were performed with and wi/Jaout the pitot probe

present in the modeling. A comparison of static pressure profiles without the probe showed that static pressure was off by nearly a factor of 2 over much of the

radial profile, even when taking into account potential axial displacement of the probe by up to 0.25 in. (0.64 cm). Including the pitot probe in the CFD modeling

and data interpretation lead to good agreement between measurement and prediction. Graphical inspection of the results showed that the shock waves impinging

on the probe surface were highly nonuniform, with static pressure varying circumferentially among the pressure ports by over 10 percent in some cases. As part

of the measurement me/Jaodology, such measurements should be routinely supplemented wi/Ja CFD analyses/Nat include the pitot probe as part of the flow-path

geometry.

14. SUBJECT TERMS

Combustor; Lobe mixer; Supersonic flow; Pitot tube

17. SECURITY CLASSIFICATIONOF REPORT

Unclassified

NSN 7540-01-280-5500

15. NUMBER OF PAGES

4716. PRICE CODE

18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACTOF THIS PAGE OF ABSTRACT

Unclassified Unclassified

Standard Form 298 (Rev. 2-89)

Prescribed by ANSI Std. Z39-18298-102

Measurement and Computation of Supersonic Flow in a Lobed … · 2013-08-30 · Measurement and...

Documents

Transcript of Measurement and Computation of Supersonic Flow in a Lobed … · 2013-08-30 · Measurement and...