Flight Test I

 

 

Drag Polar / Power Required

 

For

 

The KOPP BD-4 N375JK “Miss Daisy”

 

 

 

 

 

 

 

 

 

 

 

 

By LT Kenneth G. Kopp

CO-Builder/Owner of the Kopp BD-4 “Miss Daisy”

 

 

Table of Contents

List of figures. 3

Introduction.. 4

Kopp BD-4. 5

Wings. 5

Fuselage. 5

Empennage. 6

Mission.. 6

Part I - Instrument Position Errors.. 7

DVpc Determination.. 9

DVpc  Flight Test.. 11

Part II – Drag Polar / Power Required.. 17

Flight Test Procedure. 18

Drag Polar / Power Required Flight Test.. 19

Flight Safety. 19

Data Analysis.. 27

Drag Polar.. 27

Power-required.. 34

Summary and Conclusions. 44

Appendix A.. 46

Appendix B.. 47

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

List of figures

 

Figure 1 Airspeed Position Error Plot.. 12

Figure 2 Vcas vs. Vias (mph) 13

Figure 3 Ps. 14

Figure 4 Hpc Plot.. 15

Figure 5 Propeller Thrust vs. Airframe Thrust Required.. 26

Figure 6 Drag Polar.. 28

Figure 7 Cd vs. Cl^2. 30

Figure 8 Calculated Drag Polar.. 31

Figure 9 Cl / Cd.. 33

Figure 10 THP vs Vtas (mph) 35

Figure 11 Normalized Power Required.. 37

Figure 12 Normalized Power Required (curve fit) 38

Figure 13 Power Curve. 40

Figure 14 Normalized Drag Polar.. 41

Figure 15 Prop efficiency.. 47

 

 

LIST OF TABLES

 

Table 1Kopp BD-4 Specifications. 6

Table 2 Calibrated Airspeed Data.. 11

Table 3 Crew and altitude assignments. 19

Table 4 Flight Responsibilities. 19

Table 5. 3000 ft Data.. 20

Table 6 Data Reduction for 3000 feet PA.. 22

Table 7 Error Corrections for 3000 ft.. 24

Table 8 Cdo and e. 30

Table 9. Max Lift / Drag.. 32

Table 10 Minimum Power-required.. 35

Table 11Cdo and e comparisons. 40

Table 12 Performance Table. 42

Table 13 Summary.. 44

Table 14 7500 ft Data and Reduction.. 46

Table 15 Instrument Corrections for 7500 ft.. 47

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Introduction

 

            The purpose of this report is to formally release findings and conclusions conducted during initial flight test of the experimental homebuilt Kopp BD-4 depicted on the title page.  This is the first of an extensive series of planned flight tests and reports to be generated while investigating the flight characteristics and handling qualities of this airplane.  Specific objectives include:

The test was conducted in two parts.  Part one consisted of instrument position error determination using the course-over-ground method.  Part two consisted of drag polar and power required data collection conducted at  different altitudes and gross weights.  Results of each test were tabulated and reduced using Microsoft Excel and MATLAB 5.03.  Plots of all pertinent data along with the data itself is included within this report. 

 

 

 

 

 

 

Kopp BD-4

 

The Kopp BD-4 is a single engine, 4 place, and IFR capable homebuilt airplane. It is equipped with a Lycoming O-360-A1A 180 HP horizontally opposed, direct drive, normally aspirated engine turning a 74” Hartzell 7666-2 constant speed propeller.

 

Wings

 

The BD-4 has a cantilever high wing with a 64-415 modified airfoil.  A plain flap of  71% span and cord of 15% MAC can deflect from 0°-30°.  Ailerons of the sealed configuration, also have a chord of 15% MAC and are deflected differentially by 1” diameter torque tubes.  The unique tubular spar and metal-to-metal bonding used in the wings kept costs of construction and maintenance low, weight light and construction easy.  Three components comprise the entire cantilever spar design; the center section and two slightly larger wing tubes which are all bolted together with four AN4 bolts ( not so jokingly referred to as Jesus bolts).

Fuselage

 

The all-metal fuselage was fabricated entirely of simple flat aluminum gussets and varying length angles of different dimension.  The entire assembly is bolted together “erector set style” using the highest quality AN hardware.  .020” 2024 T3 aluminum skin is bonded and blind riveted to the structure and together form a sturdy, dependable airframe rated to a limit load of +-6 g’s.

 

 

 

 

 

 

               

Empennage

 

The horizontal tail is of the “all-flying” variety found on many Piper airplanes.  The stabilator consists of a single tubular spar and several rib sections formed into a 63-009 airfoil.  The vertical tail is of similar construction. 

Table 1  below is a detailed listing of all Kopp BD-4 specifications..

Table 1Kopp BD-4 Specifications

Wing Span

25.6 ft

Cabin Width

42”

Wing Chord

4 ft

Cabin length

89”

Wing Area

102.33 ft2

Cabin height

41”

Aspect Ratio

6.4

Fuel Capacity

60 gal

Aileron Area

3.5 ft2

Elevator Def up

15°

Flap Area

8 ft2

Elevator Def down

6°

Flap Span

71%

Trim Tab Up

18°

Aileron Defl Up

25°

Trim Tab Down

10°

Aileron Defl Down

17°

Rudder Deflection

+- 25°

Length

21.4 ft

Flap Deflection

0°-30°

Horizontal Stab Span

7.3 ft

Max Gross Weight

2200

Horizontal Chord

3 ft

Empty Weight

1412 lbs

Horizontal Stab Area

21.9 ft2

Useful Load

788 lbs

Horizontal Stab AR

2.4

Wing Loading

21.5 lbs/ft2

Vertical Stab Area

12 ft2

Power Loading

12 lbs/BHP

 

Mission

 

            The Kopp BD-4’s designed mission is that of medium range cross-country cruiser and general recreational aircraft.  The main focus of these flight tests will be to determine how its performance aids or deters fulfillment of its designed mission.

 

 

Part I - Instrument Position Errors

 

            Prior to any serious flight test it is necessary to determine  accuracies of all measurements taken during data collection.  Determination of and documenting these accuracy leads to a more quantitative analysis of the airplanes performance and helps ensure more consistent and meaningful results.  Because the performance of the Kopp BD-4 falls well below the accepted speed threshold of M=.3, where effects of compressibility become significant, all calculations will assume incompressible flow.  The  instrumentation used for this test consists of the installed primary flight performance instruments, which include airspeed (a/s), altimeter (alt), vertical speed indicator (vsi) and the outside air temperature gauge (oat).  With the exception of OAT the most common source of error for the remaining instruments are those related to the static pressure port position, hence the term static position error is used.  Because the A/S, Alt and VSI indicators all operate by measuring  static ambient pressure via the static port, differences between actual ambient pressure and the pressure sensed by the instrument results in an error indicated by the individual instrument.  Unfortunately it is very difficult, if not impossible, to position the static port such that no errors in measurement of static ambient pressures, Ps, are introduced during all flight regimes.  Because the static port is usually located along the fuselage, sensed static pressure, P`s, will vary as fuselage boundary layer conditions vary.  Therefore; errors introduced because of static port position also vary with flow conditions.  For this reason the difference between Ps  and  P`s or DPs,  must be calculated throughout the airplanes complete range of airspeeds and configurations (flaps up and down, gear up and down, etc…).  Additionally each instrument has an internal error called instrument error that must be determined by the manufacturer or tested in a calibrated laboratory environment.  To achieve the highest degree of accuracy when reducing the data collected a correction factor for each error source must be applied to the recorded (indicated) data.  DPs is difficult to measure directly without additional costly equipment.  Instead it is much simpler to determine the position error of a single indicator, such as A/S, and mathematically relate this to the others to determine their errors as well.  Once individual indicator position errors are determined, DPs can be solved analytically. The two primary indicator correction factors are DVpc and DHpc for the A/S and Altimeters respectively. Once these factors are obtained the following sequence is used to correct the data:

Vi                     Indicated airspeed (as read on the gauge)

+DVic               instrument correction (from lab)

=Vic                       Indicated corrected airspeed

+DVpc                   static position correction

= Vcal                    Calibrated Airspeed

+ DVcomp          compressibility correction (for M>.3)

= Ve                       Equivalent airspeed

/                correction for density at flight condition with reference to standard sea level density

= VĄ                      True Airspeed, actual flight speed[1]

For altitude corrections:

Hi                           indicated pressure altitude (set to 29.92)

+DHic               instrument correction (from lab)

= Hic                Indicated Pressure Altitude corrected for inst. error.

+DHpc              Position Error correction for static position error.

= Hc                 Calibrated Pressure Altitude.

 

 

DVpc Determination

 

 

                The ground-course method was chosen to determine DVpc for the Kopp BD-4 due to its simplicity and because it requires no additional support equipment unlike other commonly used methods such as the tower-fly-by and trailing-bomb techniques. The ground-course flight procedure and data reduction is outlined below:

 

 

 

 

 

 

 

 

 

 

 

 

DVpc  Flight Test

 

Flight test to determine DVpc was conducted 11 November, 1999 over a 4.17 statute mile section of hwy 101 between Salinas and Soledad, California.  During preflight chart study two bridges located at prominent intersections along the course to be flown were chosen as start and stop points for each run. Each run was conducted in the clean configuration (flaps up).  A Magellan 3000XL handheld GPS was used to refine the charted distance during flight.  The test was conducted under severe clear VFR (visual flight rules) conditions at 10:00 a.m.. Surface winds reported from Salinas automated weather service at the time of test were 120° at 4 knots. The flight was conducted single piloted and flown according to the procedures outlined previously.  After each run a button-hook maneuver was executed to reverse course to arrive on altitude and  airspeed prior to the start point of the next run.  If the start point was reached prior to attainment of the target altitude and airspeed the run was aborted and another button-hook performed.  Table 2 below lists all data and results of this test.

Table 2 Calibrated Airspeed Data

Date

Date

 

 

Weight

Distance

ρ sea

P1000

ρ 1000ft

sigma

 

 

 

 

11/11/99

11/11/99

 

 

1800

4.17

0.00237

2040.9

0.00229

0.96567

 

 

 

 

Run 1

 

 

 

 

Run 2

 

 

 

 

Data Reduction

 

 

Vias(mph)

time (sec)

alt

oat

Vias(mph)

time (sec)

alt

oat

V1g

V2g

Vtas(mph)

Ve(mph)

Vias ave

DVpc

162

90.77

1000

59

160

97.57

1005

59

165.39

153.86

159.62

156.86

161.00

-4.14

155

94.81

1010

59

155

101.32

1000

59

158.34

148.16

153.25

150.60

155.00

-4.40

148

99.18

1000

59

148

106.17

990

59

151.36

141.40

146.38

143.84

148.00

-4.16

142

103.56

1020

59

140

111.51

1000

59

144.96

134.62

139.79

137.37

141.00

-3.63

138

107.53

1000

59

138

111

1000

59

139.61

135.24

137.43

135.05

138.00

-2.95

129

115.82

1000

59

129

118.93

1000

59

129.61

126.23

127.92

125.71

129.00

-3.29

115

129.85

1000

59

115

129.36

1000

59

115.61

116.05

115.83

113.82

115.00

-1.18

103

142.85

1020

59

103

141.17

1010

59

105.09

106.34

105.71

103.88

103.00

0.88

96

151.20

990

59

96

149.37

1010

59

99.29

100.50

99.89

98.16

96.00

2.16

80

173.69

1000

59

80

173.19

1000

59

86.43

86.68

86.55

85.06

80.00

5.06

75

188.64

1000

59

75

179.6

1000

59

79.58

83.59

81.58

80.17

75.00

5.17

 

            DVpc obtained in the last column of table 1  along with a fifth order polynomial curve fit of the data is plotted below in figure 1.

Figure 1 Airspeed Position Error Plot

 

mph

 

 

 

Using MATLAB to determine the coefficients of the polynomial fit, the equation for DVpc(Vias) is:

This equation is used to analytically determine  DVpc for all values of Vi throughout the clean configuration envelope.

For  use in the cockpit a more useful tool is a plot of Vcas vs. Vias as shown below in figure 2.

Figure 2 Vcas vs. Vias (mph)

 

To obtain an analytic solution for Vcas at any Vias a first order polynomial fit was used to generate the following equation:

With this equation for DVpc it is possible to determine DHpc  and DPs from the following relationships derived in the NPS,  AA4323 Flight Test Engineering Class notes:

 , where DVic and DVpc must be in ft/sec.

and

Close examination of the above two equations reveals  DPs  to be insensitive to altitude and atmospheric conditions while DHpc is affected by changes in altitude, as would be expected in an altimeter!  Therefore, DHpc must be determined for each altitude at which a flight test was conducted. 

By substituting the 5th order polynomial fit for DVpc into the above two equations and iterating throughout the full range of airspeeds (Vi), the following plots and equations result:

Figure 3 Ps

 

Where DPs can be represented by:

using a 3rd order polynomial fit of the plot in figure 3 above.

 

Iterating altitude from sea level to 10,000 pressure altitude results in DHpc curves as shown below in figure 4.

Figure 4 Hpc Plot

 

Where the DHpc is also represented also by a 3rd order polynomial fit equal to:

@ 1000 ft Hi

 

            Otherwise for the general solution :

 DHpcstd,Vic)=

where 

and

 

In truth the values DVpc, DHpc and DPs for this airplane also include the instrumentation errors as data is not available for the errors of the instruments themselves.  The important point however, is to determine the total system error and apply a correction to any further data collected. 

 

 

 

 

 

 

 

 

 

 

 

Part II – Drag Polar / Power Required

 

 

 

In this section the findings from flight testing to determine the drag polar and power required for the Kopp BD-4 will be reported.  Determination of an accurate drag polar enables the calculation of many vital performance metrics such as (L/D)max, Cdo, and the Oswald Span Efficiency Factor, e.  Determination of power required curves provides both a graphical and analytic method for determination of other important parameters such as maximum range and endurance airspeed and minimum power required. 

An aircraft drag polar is simply a plot of lift vs. drag or Cl vs. Cd for a particular configuration and is a measure of the aerodynamic efficiency of the complete aircraft independent of the installed propulsion source (to the extent the propulsion configuration contributes to added aircraft drag).  Lift is the easiest quantity to determine since in level flight the total lift equals the gross weight of the aircraft at that instant in time.  Drag however is quite difficult to measure with any accuracy in-flight without  additional equipment.  Fortunately, using proprietary software on loan from Hartzell Propeller Inc.  determination of the 7666-2 blade efficiency and thrust generated was greatly simplified. To use this software the relationship of Vtas for given engine power settings, which is a combination of manifold pressure and propeller rpm, must be determined. Altitude and temperature are important inputs as well aircraft weight at the time of testing.  This requires an accurate fuel-burn chart for calculation of  weight as a function of time for each data point. The Lycoming O-360 operator’s manual includes both engine power  and fuel-burn charts for use in computing the necessary information.  Copies of these charts can be found in the appendix.    Determination of power required as a function of airspeed is obtained using the data collected for the drag polar.

 

Flight Test Procedure

 

There are two basic methods for drag-polar data collection, the constant-airspeed and constant-altitude method.  In the constant airspeed method the pilot flies a designated airspeed while power is adjusted as required to maintain altitude at that airspeed.  Power, Hi, Ti and Vi,  are recorded after all parameters have stabilized. This process is repeated over the full range of airspeeds in the designated configuration.  Conversely the constant-altitude method requires the pilot to establish an altitude, set  power to a desired level and adjust pitch attitude to maintain  altitude. Once stabilized, the information is recorded and the next power setting adjusted.  This process is repeated throughout a range of power settings.  The constant-altitude method is beneficial at higher airspeeds because a power schedule can be developed and tabulated during pre-flight planning for organized data collection during flight. Additionally it is much easier to set a specified power setting and record the resulting airspeed than it is to fly an exact airspeed and determine the power setting.  This is largely due to the size and scaling of the manifold pressure and rpm instrument face used for power determination.  Because priory information of minimum power required for this particular airframe is not known, it is necessary to revert to the constant-airspeed method at lower airspeeds. To help ensure accurate data two runs are conducted for each airspeed / power combination.  The data is averaged over both runs to arrive at one set of data for each altitude and weight of interest.

 

 

 

Drag Polar / Power Required Flight Test

 

                This test was conducted in the Kopp BD-4 on 27 July, 2000 departing from Monterey Peninsula Airport (MRY) at 10:11 am.  Conditions at take-off were:

Wind:   290/8

Alt:       30.04

Sky Clear

Rwy:    28R

 

It was determined during pre-flight planning that two separate runs would be made at different gross weights and altitudes.  Crew  and altitude assignments were as follows:

Table 3 Crew and altitude assignments

 

Crew

Altitude

Gross Weight (approx)

LT Ken Kopp / Maj. Jim Hawkins

3000 ft

1950 lbs

LT Ken Kopp / LT Anthony Fortesque

7500 ft

2150 lbs

 

 

 

 

 

 

 

            The test area was restricted to the Salinas Valley from Salinas to 15 miles South East of King City.  Crew coordination and a thorough test procedures briefing preceded each flight.  Data collection sheets were developed, printed and discussed in detail prior to flight as well. Specific responsibilities were delegated as follows:

Table 4 Flight Responsibilities

Responsibility

Pilot at the Controls

Pilot Not at the Controls

Flight Safety

Primary

Secondary

Airwork

Primary

 

Test Procedure

 

Primary

Data Recording

 

Primary

Communications

Primary

Secondary

Navigation

Secondary

Primary

Visual Lookout

Secondary

Primary

Emergencies

Primary

Secondary

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ATC flight following was utilized to the maximum extent possible to aid in collision

avoidance.  King City and Salinas Muni were designated primary diverts in the event an emergency due to mechanical failure or weather occurred.  

 

                Each pilot was responsible for one run.  At the completion of a run a control swap was accomplished and the second run completed.  The results of the test conducted at 3000 feet pressure altitude are listed in table 5 below. 7000 feet data is included in the appendix.

Table 5. 3000 ft Data

 

Flight Test Data Sheet

 

 

 Power Required

 

 

 

 

N375JK Kopp BD-4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

st fuel

G W

Wind

Alt

Temp

Srt T

T/O T

C pwr

lvl T

trans pwr

C BR

GW

A br

Delt Time

final gw

Ave GW

Ts

33

1991.4

290/8

30.04

16

10:00

10:11

160

10:16

164

13

1984

7.8

63

1928.34

1956

50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Run #1

Ken

 

 

 

         Test Data

 

 

 

 

Run #2

Jim

 

 

         Test Data

 

 

MP

RPM

IAS

PA

OAT

Time

Ind HP

BR

GW

MP

RPM

IAS

PA

OAT

TIme

Ind HP

GW

26.5

2700

154

2995

69

2.0

163

13

1981

26.5

2700

154

3000

69

1.0

163

1927.0

26

2600

150

2995

68

2.0

159

13

1978

26

2600

150

3000

69

1.0

159

1928.4

25.5

2550

148

3000

68

1.0

152

11.5

1977

25.5

2550

146

3000

69

2.0

152

1929.9

25

2500

143

2995

68

1.0

146

11.5

1976

25

2500

142

3010

69

1.0

146

1932.5

24.5

2450

137

3000

66

2.0

140

11.5

1973

24.5

2450

140

3020

69

1.0

140

1933.8

24

2400

135

2998

67

2.0

132

8.8

1971

24

2400

135

3020

69

1.0

132

1935.1

23.5

2350

131

3000

68

1.0

124

8.8

1970

23.5

2350

131

3020

69

1.0

124

1936.1

23

2300

129

3000

68

1.0

120

8.8

1969

23

2300

129

3000

69

1.0

120

1937.1

22.5

2250

122

3000

68

1.0

114

7.5

1968

22.5

2250

125

3000

69

1.0

114

1938.1

22

2200

119

3000

68

1.0

108

7.5

1967

22

2200

122

3020

69

1.0

108

1939.0

21.5

2200

115

3000

70

2.0

104

7.5

1966

21.5

2200

120

3000

69

1.0

104

1939.8

21

2200

116

3000

69

2.0

100

6.3

1964

21

2200

119

2990

69

1.0

100

1940.7

20.5

2200

113

3000

69

2.0

96

6.3

1963

20.5

2200

115

3000

69

1.0

96

1941.4

20

2200

112

2995

69

1.0

92

6.3

1962

20

2200

110

3000

69

1.0

92

1942.1

19.5

2200

109

2995

69

1.0

89

6.3

1961

19.5

2200

104

3010

69

1.0

89

1942.8

19

2200

104

3000

69

2.0

86

6.3

1960

19

2200

95

3050

69

1.0

86

1943.5

18.5

2200

100

2995

69

1.0

82

5.5

1959

18.5

2200

94

3000

69

1.0

82

1944.2

18

2200

94

3000

69

3.0

78

5.5

1957

18

2200

90

2970

69

2.0

78

1944.9

18

2200

90

2995

68

2.0

78

5.5

1956

17

2200

88

3000

69

1.0

72

1946.1

17.5

2200

85

3000

68

2.0

74

5.5

1955

19

2200

76

3050

69

1.0

86

1946.7

17

2200

78

3000

69

1.0

72

5.5

1954

18

2200

70

2925

69

2.0

78

1947.4

17.5

2200

74

3050

68

3.0

74

5.5

1952

 

 

 

 

 

 

 

 

18

2200

70

3000

70

6.0

78

5.5

1949

 

 

 

 

 

 

 

 

 

 

 

The top row of the table consists of starting weight, starting fuel, basic weather information, engine start time, take off time, climb power and level off time.  These values are used to determine the fuel burned during t/o, climb and transit to the working area in order to calculate an accurate test start weight.  The first six columns for each run are recorded data; manifold pressure, propeller rpm, indicated airspeed, pressure altitude, outside air temperature and the elapsed time between power changes.  Engine power was determined through use of the manufacturers supplied engine power chart provided in the appendix.  With engine HP recorded the fuel-burn chart, also included in the appendix, was entered and the corresponding value placed on the data sheet.  A running reduction in aircraft gross weight was calculated according to the following relationship:

 

The excel spreadsheet shown in table 6 automatically links raw  data from table 5 and calculates results for export to MATLAB for further analysis and plotting.

 

 

 

 

 

 

 

 

Table 6 Data Reduction for 3000 feet PA

 

 

 

Reduced Data for 3000 ft and 1950 lbs

 

 

 

 

 

 

 

 

 1

3

4

10 

11 

12 

13 

14 

MP

RPM

IAS

IAS

Ave IAS

AveGW

Vcas

Vtas

M

Ind HP

Corr HP

n

THP

Thrust

26.5

2700

154

154

154

1954.03

149.54

159.26

0.207

163

159.6

0.761

121.5

286

26

2600

150

150

150

1953.29

146.05

155.54

0.202

159

155.7

0.771

120.0

297

25.5

2550

148

146

147

1953.38

143.43

152.75

0.199

152

148.8

0.774

115.2

293

25

2500

143

142

142.5

1954.03

139.50

148.57

0.193

146

143.0

0.773

110.5

295

24.5

2450

137

140

138.5

1953.38

136.00

144.85

0.188

140

137.1

0.773

106.0

295

24

2400

135

135

135

1953.03

132.95

141.59

0.184

132

129.2

0.776

100.3

286

23.5

2350

131

131

131

1953.03

129.45

137.87

0.179

124

121.4

0.778

94.5

277

23

2300

129

129

129

1953.03

127.71

136.01

0.177

120

117.5

0.777

91.3

269

22.5

2250

122

125

123.5

1953.10

122.90

130.89

0.170

114

111.6

0.772

86.2

262

22

2200

119

122

120.5

1953.10

120.28

128.10

0.167

108

105.7

0.773

81.7

252

21.5

2200

115

120

117.5

1952.68

117.66

125.31

0.163

104

101.8

0.771

78.5

253

21

2200

116

119

117.5

1952.39

117.66

125.31

0.163

100

97.9

0.778

76.2

242

20.5

2200

113

115

114

1952.03

114.61

122.06

0.159

96

94.0

0.783

73.6

230

20

2200

112

110

111

1952.03

111.99

119.27

0.155

92

90.1

0.784

70.6

223

19.5

2200

109

104

106.5

1952.03

108.06

115.08

0.150

89

87.1

0.781

68.1

222

19

2200

104

95

99.5

1951.68

101.94

108.57

0.141

86

84.2

0.772

65.0

225

18.5

2200

100

94

97

1951.72

99.76

106.25

0.138

82

80.3

0.771

61.9

219

18

2200

94

90

92

1951.10

95.39

101.60

0.132

78

76.4

0.776

59.3

217

18

2200

90

90

90

1951.10

93.65

99.74

0.130

78

76.4

0.763

58.3

219

17.5

2200

85

 

85

1950.79

89.28

95.09

0.124

74

72.4

0.758

54.9

216

17

2200

78

 

78

1950.79

83.17

88.58

0.115

72

70.5

0.742

52.3

223

17.5

2200

74

 

74

1952.35

79.67

84.86

0.110

74

72.4

0.728

52.7

235

18

2200

70

70

70

1948.61

76.18

81.13

0.106

78

76.4

0.708

54.1

252

 

Atmospheric Data

 

                Temp

s sound

          CAS Curve Fit

 

 

 

 

 

 

sigma

rstd

P3000

Ts

T

a

slope

intercept

 

 

 

 

 

 

0.88161

0.00237

1896.7

48.3

69

1127.33

0.87

15.05

 

 

 

 

 

 

Columns 1-4 are linked cells and are self explanatory as is column 5.  Column 6 is the average gross weight of the aircraft at the time each data point was collected. Vcas in column 7 is derived from the curve fit data at the bottom of the table.  The curve fit was obtained as discussed in the previous section. Vtas and Mach number are calculated for entry into the propeller thrust and efficiency software. Indicated HP was obtained from the engine chart . Because the Lycoming O-360-A1A is normally aspirated (fancy way of saying it uses a carburetor), the power generated is a function of the ratio of standard atmospheric temperature (at a specific pressure altitude) and the inlet temperature.  For our purpose we will assume the OAT measurement of static ambient temperature is equal to the inlet temperature.  In fact this may not be a reasonable assumption as inlet air passes many very hot engine components prior to fuel-air atomization.  In effect this ratio is a measure of combustion efficiency and is calculated according to the following relationship:

 

     ,  where Ts and Ti are given in °F.

If the altimeter had zero instrument error and it was known the static source was also error free, Ts would correspond to the standard temperature found in the atmosphere tables for the Hi (indicated pressure altitude) flown.  However, as discussed in the previous section it is necessary to apply error corrections to Hi  in order to obtain the true  pressure altitude, Hc,  flown at each test point.  Once Hc is obtained the standard atmosphere tables can be used to retrieve the actual Ts and Ps  through interpolation.  With Ts determined for each data point a more accurate calculation of HPcorrected can be obtained.  Ps is used to calculate actual air density at Hc and Ti through use of the equation of state for a perfect gas as shown below.

. Density is a key factor in accurately determining the lift and drag coefficients CL & CD used extensively in performance analysis.

Through application of the equations derived in the previous sections and development of yet another spreadsheet the following corrections and values were obtained in table 7.

 

 

 

 

 

Table 7 Error Corrections for 3000 ft

 

3000ft

rho std

sigstd

gama

ao

 

 

 

 

0.0022

0.918

1.4

1116.3

 

 

 

Ave Ias

Ave Alt

D Vpc

D Hpc

Hc

Ts

Ps

rho

154

2997.50

-2.44

-27.55

2969.95

47.23

1899.00

0.0020885

150

2997.50

-2.40

-26.34

2971.16

47.22

1898.92

0.0020884

147

3000.00

-2.34

-25.18

2974.82

47.21

1898.66

0.0020881

142.5

3002.50

-2.20

-23.00

2979.50

47.19

1898.34

0.0020878

138.5

3010.00

-2.03

-20.60

2989.40

47.16

1897.64

0.002087

135

3009.00

-1.84

-18.18

2990.82

47.15

1897.54

0.0020869

131

3010.00

-1.57

-15.06

2994.94

47.14

1897.25

0.0020866

129

3000.00

-1.42

-13.39

2986.61

47.17

1897.84

0.0020872

123.5

3000.00

-0.94

-8.48

2991.52

47.15

1897.49

0.0020869

120.5

3010.00

-0.64

-5.65

3004.35

47.10

1896.60

0.0020859

117.5

3000.00

-0.32

-2.74

2997.26

47.13

1897.09

0.0020864

117.5

2995.00

-0.32

-2.74

2992.26

47.15

1897.44

0.0020868

114

3000.00

0.08

0.70

3000.70

47.12

1896.85

0.0020862

111

2997.50

0.45

3.69

3001.19

47.12

1896.82

0.0020861

106.5

3002.50

1.05

8.21

3010.71

47.08

1896.15

0.0020854

99.5

3025.00

2.11

15.30

3040.30

46.98

1894.08

0.0020831

97

2997.50

2.52

17.88

3015.38

47.07

1895.82

0.002085

92

2985.00

3.45

23.14

3008.14

47.09

1896.33

0.0020856

90

2997.50

3.85

25.31

3022.81

47.04

1895.30

0.0020845

85

3025.00

4.99

30.98

3055.98

46.92

1892.98

0.0020819

78

2962.50

6.99

39.76

3002.26

47.11

1896.74

0.002086

74

3050.00

8.41

45.38

3095.38

46.78

1890.22

0.0020789

70

3000.00

10.09

51.54

3051.54

46.94

1893.29

0.0020822

 

 

 

 

 

Interpolation Values for 3000 ft

 

 

 

 

 

A

Ts

Ps

 

 

 

 

 

2500

509.77

1931.9

 

 

 

 

 

3500

506.21

1861.9

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Referring back to table 6, HPcorrected can now more accurately be solved with the calculated values of Ts  listed above. 

With Mach number, altitude, rpm, HPcorrected and Ti, Hartzell’s propeller software is used to generate the efficiency and thrust generated by the propeller.  The software contains proprietary performance maps of the 7666-2 blade generated through many hours of ground and flight testing and assumes a .4 blockage factor in thrust calculations. Because Lycoming engines are of the direct drive category, meaning engine crankshaft and propeller are connected directly and turns at the same rate, HPcorrected is equivalent to the more commonly used term Shaft Horsepower or SHP.  Engines equipped with reduction drives must account for less than 100% power delivery to the propeller in calculations of SHP.

With SHP determined and software generated propeller efficiency in hand, it is now possible to determine the actual power delivered to the airframe by the engine / propeller combination.  This power is referred to as Thrust Horsepower or THP and is determined by the simple relationship shown below:

  , where η is propeller efficiency and is precisely how values in column 13 of table 6 were determined.  Column 14 lists the thrust calculated by the Hartzell program as a function of Vtas.  The software generated thrust is determined based on a standard fuselage blockage factor of approximately .4. This results in thrust values while  representative of the actual thrust developed by the propeller they are higher than the thrust required by the airframe to maintain level flight.  This is an important distinction if the power required data generated is to remain independent of the propulsive system installed.  The thrust (drag) required by the airframe is calculated then from the equation below.

 

 

 

 

A plot of the propeller generated thrust and airframe required thrust clearly illustrates the  effect of the inclusion of blockage factor during thrust calculation. 

 

Figure 5 Propeller Thrust vs. Airframe Thrust Required

 

As may be expected as velocity increases the effect of fuselage blockage becomes greater as highlighted in the plot.  All remaining calculations will be based on the airframe required thrust calculated according to the equation described on the previous page.

 

 

 

 

 

 

Data Analysis

 

Drag Polar

 

            As discussed previously the drag polar is a measure of an airframes aerodynamic efficiency independent of the installed propulsion system and is generated by plotting Lift vs. Drag.  Because total lift and drag are cumbersome terms it is common to plot the drag polar as a function of the non-dimensional lift and drag coefficients, Cl and Cd according to the following relationships:

 

 , where V∞ is Vtas is ft/sec and S is the total aircraft wing area .

and

 

, recalling basic Newtonian physics;  in level unaccelerated flight the airplane is in equilibrium, therefore:

  and        shown pictorially below.

                                  

Thus,  lift = weight and drag = thrust.  Substitution of these equalities leads to:

 and . 

The drag polar can now be plotted from our tabulated data obtained during flight test and as shown in figure 6 below.

Figure 6 Drag Polar

 

Another method for determining the drag coefficient is provided by finite wing theory, which predicts Cd to be equal to the sum of parasite and induced drag given by the following equation.

induced drag

 

parasite drag

 
,  where Cdo is the zero lift drag coefficient, e is  the Oswald Span Efficiency Factor and accounts for non-elliptical shaped wing planforms and the contribution to parasite drag due to lift[3].  AR is the wing aspect ratio calculated as:

 ,  where b is  wing span and S is  total wing area.

 

Determination of  an accurate Cdo is an important step in the design process of all aircraft.  Parasite drag is the major source of drag at high speeds and therefore must be determined early in the design phase for accurate engine sizing, weight determination and performance calculations.  For an airplane still on the design table, detailed methods have been developed by aerospace manufactures to estimate Cdo through elaborate accounting  of parasite drag of all wetted components.  Thankfully the benefit of a fully functional airplane allows us to solve for Cdo and e directly from flight test data.  This also provides a means to double check the theory as an academic exercise.  Recalling the finite wing theory equation for Cd as

,  notice that CD is linear in CL2 !

Cd and Cl have been determined previously from flight test data  and plotted in figure 5 as a drag polar.  All that remains is to plot CD as a function of CL2 and determine the slope and intercept for e and Cdo respectively.  CD vs CL2 is shown in figure 7.

 

Figure 7 Cd vs. Cl^2

 

 

Data for both the 3000 ft 1950 lb and 7500ft 2130 lbs case converge nicely showing that values of e and Cdo are relatively insensitive to atmospheric and weight changes .    Using MATLAB to determine the slope and intercept of these two curves results in:

 

Altitude

Oswald Span Efficiency

Cdo

3000 ft

.7039

0.0440

7500 ft

.7289

0.0433

Table 8 Cdo and e

 

 

with e and Cdo for each altitude Cd is recalculated according to the finite wing equation and a new drag polar plotted based on the theory as shown below in figure 8.

Figure 8 Calculated Drag Polar

 

 

Clearly the theory does an excellent job of determining the drag coefficient from calculated values of Cdo and e.

            Recalling the free-body diagram depicted on page 26, from the equilibrium sum of forces during level unaccelerated flight where lift=weight and drag=thrust we can derive an important relationship as follows:

    and  where , where

 

Therefore:

  , Thus

,  from this equation it is apparent that minimum thrust required to maintain level unaccelerated flight occurs when the ratio Cl/Cd is a maximum!

Because values for Cl and Cd have been calculated from the data it is a simple matter of choosing the largest value of the resulting ratio.  Using MATLAB to carry out this calculation results in values of:

Table 9. Max Lift / Drag

Altitude / Weight

Max Cl/Cd

Min Thrust Required

3000 ft / 1950 lbs

8.955

217.87 lbs

7500 ft / 2130 lbs

9.1957

229.63 lbs

 

Referring back to table 6 for the reduced data at 3000 ft shows the recorded minimum thrust (highlighted in red) to be 222 lbs, thus providing excellent correlation with the theoretical value predicted in table 9 above.  Another useful means for determination of  (L/D)max is simply graphing the ratio and noting the value at which point a maximum occurs as depicted in figure 9 on the next page. 

For a given airspeed Cl must increase with gross weight to maintain lift necessary for level flight.  Concurrently as Cl increases Cd increases as a function of Cl2, the net result of which is that (L/D)max remains constant but the airspeed corresponding to (L/D)max increases for an increasing gross weight.  The .24 difference in (L/D)max calculated in table 9 may be attributed to instrument and other errors introduced during data collection.

 

Figure 9 Cl / Cd

 

 

Additionally, the point at which a line drawn from the origin and tangent to the drag polar curve intersects is also the point of (L/D)max. 

With the drag polar plot accurately determined many performance factors can now be analyzed as will be discussed in subsequent sections.

 

 

 

 

 

Power-required

 

            For propeller driven aircraft determination of power-required for level unaccelerated flight constitutes a critical step toward performance measurement, prediction and optimization.  The values obtained in this analysis directly impact the operational procedures used by the pilot to fly a particular mission profile. For example, power-required data provides the means to determine maximum range airspeed, a critical performance factor for flights over open water where distances between airfields may be great.  Minimum power-required leads to maximum endurance airspeed. Critical for missions requiring long loiter times, such as patrol and search operations.  The combination of power-required and power available data leads to performance calculations for climb rate, maximum airspeed, altitude performance effects and others.  From a pilots perspective power-required determination is the single most important flight test data for real-time operational use.

                In actuality,  power-required data was compiled when the thrust horsepower THP was determined while deriving the drag polar.  All that remains is to plot THP vs. Vtas for a graphical representation of this airplanes power-required as shown in figure 10  on the next page.

 

 

 

 

 

 

 

 

 

Figure 10 THP vs Vtas (mph)

 

 

 

The high altitude, high weight configuration results in a shift of the power-required curve up and to the right.  As altitude increases the minimum power-required increases while (L/D)max  remains constant.  The effect of increased weight is higher minimum power-required and at higher velocity.  Minimum power-required to maintain level unaccelerated flight is simply  the lowest point of each curve.  The values determined from curve fit are:

Table 10 Minimum Power-required

Altitude

Minimum Thrust Horsepower-required

3500 ft 1950 lbs

52.385 HP

7500 ft 2130 lbs

60.590 HP

 

Although these plots are useful in determining the performance parameters mentioned  at the beginning of this section it would be necessary to derive a curve for each altitude and weight of interest due to their dependence on these parameters as shown above.  To alleviate this tedious dilemma, development of normalized power-required vs. normalized airspeed reduces the data such that the resultant curve is independent of weight and altitude.  In effect, one curve fits all.[4] Once this curve is generated a simple algorithm is used to extrapolate the data and a table of performance parameters for specific weights and altitudes is generated. The normalized parameters Piw & Viw   are power and calibrated (M=.3 and below) airspeed corrected to standard weight sea level conditions.  Where standard weight, Ws is defined as the maximum gross take-off weight for propeller driven aircraft.[5]  For the Kopp BD-4 Ws = 2,200 lbs. Conversion of  Pr & Vtas to their normalized values requires the following equations:

 , where Wt is the test weight.

for  Piw  recall

  and therefore

simplifying the above equation results in:

The normalized power required curve can now be plotted and performance parameters determined directly from the plot as shown below on figure  11.

Figure 11 Normalized Power Required

 

Ideally each curve for different weights and altitudes should converge into a single normalized curve.  Unaccounted errors in propeller efficiency, oat instruments, MP and rpm can account for the slightly inconsistent results.  Based on purely qualitative analysis of the conduct of the flight test it was determined the data recorded during the 3000 ft  1950 lbs gross weight run is the more accurate and will therefore be used to derive a 5th order polynomial fit to provide a working equation by which to calculate the remaining parameters.  MATLAB was used to generate the curve fit as follows:

 

Iterating the above equation for a range of Viw from 70 -160 mph results in the curve fit plot of normalized power required shown in figure 12  below.

Figure 12 Normalized Power Required (curve fit)

 

 

max endurance airspeed

 

 

 

 

The next step is to determine the maximum range and endurance normalized airspeeds.  This can be done graphically by drawing a line from the origin to a point of tangency along the Piw curve, the intersection of which corresponds to the maximum range normalized airspeed.  Maximum endurance airspeed occurs at the minimum power required for level unaccelerated flight and is easily determined graphically by selecting the lowest point on the curve or analytically by differentiating the curve fit normalized power curve, equating to zero and solving for velocity.  For propeller driven aircraft maximum range and endurance airspeed occur at (Cl/Cd)max and (Cl3/2/Cd)max.  For use of the normalized curve normalized values of Cl and Cd must be calculated.  Cl  is calculated using Ws and sea level density, ρ∞, throughout the speed range of interest.  Cd is calculated as done previously using the finite wing theory equation for total airplane drag: .  To ensure this method is valid values of Cdo and e can be determined by plotting the linear equation:

PiwViw = A1Viw4 + B1 , called the Power Curve, and compared the values determined from the drag polar.  By calculating values for the slope and intercept as A1 and B1 respectively and solving for Cdo and e according to the following relationships:

  and    total Cd can be determined.

Again ratios Cl/Cd and Cl3/2/Cd can be plotted.  The maximum of each is located and the value of Cl corresponding to each is used to solve for the normalized airspeeds in question.  The desired results are calibrated maximum range and endurance airspeed. With this, the total position error term DVpc can be subtracted to retrieve indicated airspeed as seen by the pilot  in flight. To illustrate this method maximum range airspeed will be determined in this fashion. 

 

 

 

The power curve is plotted below:

 

Figure 13 Power Curve

 

from the plot and equations listed on the previous page Cdo and e are calculated and compared to the values determined from the drag polar in table 11 below :

Parameters

Drag Polar

Power Curve

Cdo

0.0440

0.0425

e

0.7031

0.6507

Table 11Cdo and e comparisons

 

These values show good correlation to those calculated via the drag polar.

The normalized drag polar plotted below is used to determine normalized (Cl/Cd)max

and corresponding value of Cl .

 

Figure 14 Normalized Drag Polar

 

Cl at (Cl/Cd)max is .744.  Solving for Viw  with this value by

 

which corresponds nicely to the tangent drawn to the normalized power required curve in figure 11.  Because Vcas for max range is a function of (L/D)max and remains constant with changes in altitude, variations in Vcas for max range and endurance are dependent on weight changes only. Therefore, maximum range Vcas for any weight is given by:

 

   therefore the maximum range airspeed at the first test condition weight of 1950 lbs = 100.23 mph calibrated.

Maximum normalized endurance airspeed is determined by locating the airspeed corresponding to the minimum power required. Referring back to figure 11 for the normalized power required curve reveals 85.5 mph calibrated normalized airspeed gives maximum endurance performance in the Kopp BD-4.  Applying the equation above:

Vcas endurance= 80.49 mph @ 1950 lbs.

The real utility in this approach is the ease in which a performance table can be constructed within a spreadsheet program as displayed below.

 

Table 12 Performance Table

 

 

 

       

 

 

 

 

 

 

  

Airspeed       

Correction 

 

 

 

Vias

65.0

70.0

75.0

80.0

85.0

90.0

95.0

Vcas

71.6

76.0

80.3

84.7

89.0

93.4

97.7

Vias

100.0

105.0

110.0

115.0

120.0

125.0

130.0

Vcas

102.1

106.4

110.8

115.1

119.5

123.8

128.2

Vias

135.0

140.0

145.0

150.0

155.0

160.0

165.0

Vcas

132.5

136.9

141.2

145.6

149.9

154.3

158.6

Vias

170.0

175.0

180.0

 

 

 

 

Vcas

163.0

167.3

171.7

 

 

 

 

 

 

 

  Vcas

 

 

 

 

G Weight

1600

1700

1800

1900

2000

2100

2200

Range

90.8

93.6

96.3

98.9

101.5

104.0

106.5

 

 

 

 

 

 

 

 

Endurance

72.9

75.2

77.3

79.5

81.5

83.5

85.5

 

 

 

  Vias

 

 

 

 

G. Weight

1600

1700

1800

1900

2000

2100

2200

Range

87.1

90.3

93.4

96.4

99.4

102.3

105.1

 

 

 

 

 

 

 

 

Endurance

66.5

69.1

71.6

74.0

76.4

78.7

81.0

 

 

The normalized power transformation has utility in an ability to predict the minimum power required for various altitudes and gross weights and can be used to determine the maximum sustainable altitude as illustrated below.

 

To determine if the Kopp BD-4 can maintain level flight at an altitude of 15,000 feet at max gross weight (2200lbs) use the following relationships:

 

 

σ  @ 15K feet = .6313 therefore;

 

 

Consultation of the power chart and propeller efficiency software reveals a THPavailable equal to only 71.09 hp.  Therefore, at max gross weight the Kopp BD-4 will not  be capable of maintaining an altitude of 15,000 feet.  A gross weight of 2,000 lbs requires 70.99 hp at 15,000 ft, which should be attainable for standard day conditions. It will be interesting to investigate this prediction during flight test!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Summary and Conclusions

 

                In this test key instrument calibration correction factors were developed and applied, thereby providing means for more accurate data analysis in subsequent testing.  Applying FAA standards for certified aircraft for a maximum DVpc 0f  +- 6 mph shows  the Kopp BD-4 to be skirting the limit throughout most of its airspeed range.  Further testing of the instrumentation and aircraft are warranted in this matter. Also, an attempt to obtain manufacturer calibration errors may improve the results as well.  Development of the drag polar, power required curves and Cl/Cd provided valuable information immediately usable by the pilot (that’s me!).  A summary table below highlights many of the details resulting from this effort.  Further testing will provide even greater insight into the performance and characteristics of this versatile little airplane.

 

Table 13 Summary

Altitude / Weight

Max Cl/Cd

Min Thrust Required

3000 ft / 1950 lbs

8.8235

217.87 lbs

7500 ft / 2130 lbs

9.0329

229.63 lbs

Parameters

Drag Polar

Power Curve

Cdo

0.0440

0.0425

e

0.7031

0.6507

Altitude

Minimum Thrust Horsepower Required

3500 ft 1950 lbs

52.33 HP

7500 ft 2130 lbs

60.59 HP

Standardized

59.16 HP

Kopp BD-4 Summary Table

Test conducted 27 July, 2000

Data to be added upon further testing

 

Finally, recalling the specific objectives listed in the introduction:

Clearly all objectives were met during this flight test.  Flight Test II will investigate power-available and excess power.  The combination of both sets of data will enable a vast array of performance calculations to be made.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Appendix A

 

Table 14 7500 ft Data and Reduction

 

Reduced Data for 7500 ft and 2130 lbs

 

 

 

 

 

 

 

 

 

 

 

run 1

run 2

 

 

 

 

 

 

 

 

 

 

MP

RPM

IAS

IAS

Ave IAS

Ave GW

Vcas

Vtas

M

HP

Corr HP

n

THP

Thrust

22.7

2700

129

134

131.5

2128.38

129.78

150.39

0.197

142

138.0

0.754

104.1

259

22.7

2600

130

131

130.5

2125.77

128.91

149.37

0.196

139

135.1

0.764

103.2

267

22.7

2500

126

126

126

2123.28

124.98

144.82

0.190

134

130.2

0.76

99.0

272

22.7

2450

124

125

124.5

2122.28

123.67

143.30

0.188

132

128.3

0.761

97.6

265

22.7

2400

123

124

123.5

2121.28

122.80

142.29

0.187

131

127.3

0.763

97.1

259

22.7

2300

121

123

122

2120.29

121.49

140.77

0.185

126

122.4

0.758

92.8

247

22.5

2250

120

120

120

2117.79

119.74

138.75

0.182

124

120.5

0.75

90.4

245

22

2200

117

116

116.5

2116.52

116.68

135.21

0.177

118

114.7

0.746

85.5

239

21.5

2200

115

114

114.5

2115.67

114.94

133.18

0.175

113

109.8

0.752

82.6

232

21

2200

109

110

109.5

2114.39

110.57

128.12

0.168

108

104.9

0.751

78.8

230

20

2200

103

104

103.5

2113.12

105.33

122.05

0.160

102

99.1

0.749

74.2

229

19

2200

99

 

99

2111.69

101.40

117.50

0.154

95

92.3

0.745

68.8

229

18.5

2200

90

 

90

2110.98

93.54

108.39

0.142

92

89.4

0.723

64.6

238

19.5

2200

85

 

85

2109.91

89.17

103.33

0.136

95

92.3

0.702

64.8

245

19

2200

80

76

78

2108.83

83.06

96.25

0.126

92

89.4

0.677

60.5

252

18.5

2200

75

 

75

2107.05

80.44

93.21

0.122

92

89.4

0.677

60.5

252

21

2200

70

 

70

2122.42

76.08

88.15

0.116

108

104.9

0.624

65.5

279

 

Atmospheric Data

 

                Temp

Sonic Speed

         CAS Curve Fit

 

 

 

 

 

 

sigma

rstd

p7500

Ts

T

a

slope

intercept

 

 

 

 

 

 

0.74477

0.00237

1602.3

32

60

1117.6976

0.8733

14.9441

 

 

 

 

 

 

 

Flight Test Data Sheet

7500ft

 

 Power Required

 

 

 

 

N375JK Kopp BD-4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

GW

fuel

Wind

Alt

T

Srt T

T/O T

C pwr

lvl T

trans pwr

C br

GW

A br

Delt Time

final gw

Ave GW

Ts

2168

56

290/8

30.04

17

12:30

12:39

160

12:53

164

13

2147

8

66

2089.0

2118

46

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Run #1

Ken

 

         Test Data

 

 

 

 

 

 

Run #2

Ant

 

         Test Data

 

 

 

MP

RPM

IAS

PA

OAT

Time

HP

BR

GW

MP

RPM

IAS

PA

OAT

TIme

HP

GW

22.7

2700

129

7500

60

0.0

142

11.5

2147.4

22.7

2700

134

7500

60

10.0

142

2109.4

22.7

2600

130

7500

60

3.0

139

11.5

2143.5

22.7

2600

131

7500

60

1.0

139

2108.1

22.7

2500

126

7500

60

3.0

134

8.8

2140.5

22.7

2500

126

7500

60

2.0

134

2106.1

22.7

2450

124

7500

60

1.0

132

8.8

2139.5

22.7

2450

125

7500

60

1.0

132

2105.1

22.7

2400

123

7500

60

1.0

131

8.8

2138.5

22.7

2400

124

7520

60

1.0

131

2104.1

22.7

2300

121

7500

60

1.0

126

8.8

2137.5

22.7

2300

123

7500

60

1.0

126

2103.1

22.5

2250

120

7500

60

3.0

124

8.8

2134.5

22.5

2250

120

7500

60

2.0

124

2101.1

22

2200

117

7500

60

2.0

118

7.5

2132.8

22

2200

116

7500

60

1.0

118

2100.2

21.5

2200

115

7500

60

1.0

113

7.5

2131.9

21.5

2200

114

7500

60

1.0

113

2099.4

21

2200

109

7500

60

1.0

108

7.5

2131.1

21

2200

110

7520

60

2.0

108

2097.7

20

2200

103

7500

60

1.0

102

7.5

2130.2

20

2200

104

7500

60

2.0

102

2096.0

19

2200

99

7500

60

1.0

95

6.3

2129.5

19

2200

85

7520

60

3.0

95

2093.9

18.5

2200

90

7500

60

1.0

92

6.3

2128.8

18.5

2200

80

7540

60

1.0

92

2093.1

19.5

2200

85

7480

60

1.0

99

6.3

2128.1

19

2200

76

7500

60

2.0

95

2091.7

19

2200

80

7500

60

1.0

95

6.3

2127.4

19.5

2200

70

7400

60

2.0

99

2090.3

18.5

2200

75

7500

60

1.0

92

6.3

2126.7

20.5

2200

65

7500

60

4.0

105

2087.4

21

2200

70

7500

60

5.0

108

7.5

2122.4

 

 

 

 

 

 

 

 

Appendix B

 

 

 

Table 15 Instrument Corrections for 7500 ft

7500ft

rho std

sigstd

gama

ao

 

 

 

 

 

0.0019

0.80042

1.4

1116.288

 

 

 

 

Ave IAS

Ave Alt

Delt Vpc

delt Hpc

Hc

Ts

Ps

rho

rhostd

131.50

7500.00

-1.61

-17.74

7482.26

31.14

1603.68

0.001794

0.001899

130.50

7500.00

-1.53

-16.80

7483.20

31.14

1603.62

0.001794

0.001898

126.00

7500.00

-1.17

-12.34

7487.66

31.12

1603.35

0.001794

0.001898

124.50

7500.00

-1.03

-10.78

7489.22

31.12

1603.26

0.001794

0.001898

123.50

7510.00

-0.94

-9.72

7500.28

31.08

1602.58

0.001793

0.001897

122.00

7500.00

-0.79

-8.11

7491.89

31.11

1603.09

0.001794

0.001898

120.00

7500.00

-0.59

-5.92

7494.08

31.10

1602.96

0.001793

0.001898

116.50

7500.00

-0.21

-2.02

7497.98

31.09

1602.72

0.001793

0.001898

114.50

7500.00

0.02

0.24

7500.24

31.08

1602.59

0.001793

0.001897

109.50

7510.00

0.65

5.95

7515.95

31.02

1601.63

0.001792

0.001897

103.50

7500.00

1.49

12.88

7512.88

31.03

1601.81

0.001792

0.001897

99.00

7510.00

2.19

18.13

7528.13

30.98

1600.88

0.001791

0.001896

90.00

7520.00

3.85

29.02

7549.02

30.91

1599.61

0.00179

0.001895

85.00

7490.00

4.99

35.52

7525.52

30.99

1601.04

0.001791

0.001896

78.00

7450.00

6.99

45.59

7495.59

31.10

1602.87

0.001793

0.001898

75.00

7500.00

8.03

50.37

7550.37

30.90

1599.53

0.00179

0.001895

70.00

7500.00

10.09

59.09

7559.09

30.87

1599.00

0.001789

0.001894

 

 

 

 

 

Interpolation Values for 7000 ft

 

 

 

 

 

 

A

Ts

Ps

rhostd

 

 

 

 

 

7000

493.73

1633.1

0.00193

 

 

 

 

 

8000

490.17

1572.1

0.00187

 

Figure 15 Prop efficiency

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



[1] AA4323 Class Notes

[2] AA4323 Flight Test Engineering Class Notes

[3] Anderson, “Introduction To Flight”, Ch.6

[4] AA4323 Flight Test Engineering Class Notes

[5] AA4323 Flight Test Engineering Class Notes