WiNG Research Member Roles

Douglas L. Wilson is the team leader on this project. Doug is a student pilot and has logged over 65 hours in Cessna 172 aircraft. His familiarity with the aircraft made him a logical choice for leader of the team. In addition to being the team leader, Doug serves as Chief Test Pilot for the project.

Abraham Gutierrez is the Chief Systems Engineer of the project. Abraham has undertaken the task of understanding the phugoid mode. This includes learning the mathematical models associated with the phugoid mode. Abraham has concentrated his efforts on the Matlab coding that will be used in the phugoid analysis.

Anh Huy Nguyen is the Chief Flight Engineer for the team. Anh Huy has studied the Cessna crash and probable causes of the accident. In addition to this duty, Anh Huy has been instrumental to the team by completing the report from The Design of Everyday Things. His report on how this applies to this information applies to this project immediately follows this section.

These are the designated roles for each team member. Additional tasks such as the writing of the weekly memos are written by each member on a rotating basis.

The Design of Everyday Things

In Donald A. Norman’s book, The Design of Everyday Things, he discusses the importance of creating guidelines for everyday objects for the use of consumers. Items like door handles, stove tops, and light switches must be designed to maximize consumer efficiency. Early forms of products often overlook the user’s necessity of basic instruction. For example, doors should be designed so that people can automatically locate where the door opens – by either pulling or pushing the door. If people are even a little confused about the door’s orientation, then its design is deemed faulty. Similarly, clear instructions should be made on the layout of the knobs of stove tops. Distinctions of the knobs should be made to correspond to the placement of the stove.

Some aspects of this project relate to the design principles of The Design of Everyday Things. Companies like Cessna and Mooney must take into account all design errors to ensure safety and efficiency for its consumers. In case of emergencies, the aircraft must be supplied with adequate safety features. For the pilot of the aircraft, the locations of these objects must be within reach. Similar measures have to be made for passengers of the aircraft as well. So designers of the airplanes need to study and design the switches and levers within a natural distance from the pilot and passengers.

To avoid confusion for the pilot in a Cessna or a Mooney aircraft, the control panel should be relatively simple to read. Throughout the years, aircraft companies have streamlined the control panels of their airplanes. They have grouped together similar instruments on the control panel, which reduced the difficulty for pilots reading the control panels. Glass cockpits (as shown if Figure 1) have been designed to make viewing all of the instruments even easier for pilots. Even complex objects should have the ability to have simple improvements made to them. Companies should acknowledge mistakes and improve on them.

New Glass Cockpit from a Cessna Mustang [1]

Cessna 172 cockpit circa 1958 [2]

Figure 1–Comparison between the new glass cockpits and the older style cockpits.

According to Norman, the instructions which outline the steps taken during an emergency should be clearly understood. Unnecessary words and phrases should be left out and only a succinct description should be included. For our case, the passengers during a crash should be told where the exits are and what to do during an incident.

The tasks of completing a complicated project can be made using the principles of design according to Norman. These are listed below.

Use both knowledge in the world and knowledge in the head.
Simplify the structure of tasks.
Make things visible: bridge the gulfs of Execution and Evaluation.
Get the mappings right.
Exploit the power of constraints, both natural and artificial.
Design for error.
When all else fails, standardize.

These principles have been incorporated in the project by researching the crash investigations online and from documents currently available. The project has been divided into sections where each team member is responsible for their part, but can also assists others. The research has been applied to creating virtual models of the aircraft, which have been modified from previous specifications of other planes. The errors in the project will be analyzed to determine what improvements need to be made for future assignments [3].

Crash Report

On May 5, 1998 at 9:30 a.m. (pacific), a Mooney 305 that was owned and operated by the pilot experienced an in-flight breakup while descending. According to the Federal Aviation Administration (FAA), the pilot spoke with Oakland Flight Watch Personnel while he was cruising at an altitude of 19000 feet. According to the FAA report, at 8:55 a.m. Flight Watch personnel asked the pilot if he was aware of the forecast for icing and asked the pilot if his plane had deicing equipment. The pilot replied that he did not have de-icing for the airplane and asked Air Traffic Control if he could turn around and land at Bakersfield Municipal Airport. Air Traffic Control gave the pilot permission to go back and land at Bakersfield Municipal Airport [4].

The timeline for the Mooney 305 crash started at 8.55 a.m. when the pilot was cruising at an altitude of 19000 feet when he decided to turn around and land the airplane at Bakersfield Municipal Airport. As he was heading back, the pilot descended to an altitude of 15000 feet in which he cruised there for about seven minutes due to air traffic. Air Traffic Control cleared the pilot to descend to an altitude of 2300 feet. At about 9:29 a.m. the pilot reported to Air Traffic Control that he was down 2300 feet, but after that report there was no further communication between the pilot and the Air Traffic Control Center [4].

According to the report at about 9:30 am, an eyewitness, who was at Interstate Highway 5, reported hearing the sound of an engine at a high rpm and a loud “pop.” The eyewitness also observed debris falling down near Interstate Highway 5; in fact, the main wreckage of the Mooney 305 was found 330 yards east of Interstate Highway 5 [4].

Investigations

National Transportation Safety Board (NTSB) Report

According to the National Transportation Safety Board (NTSB), during the last 24 seconds of radar-recorded flight the airplane's rate of descent increased to 3,500 feet per minute, and over its southerly course, its ground speed increased to 240 knots. Shortly after this at 9:30 a.m., emergency locator beacon was recorded in the radar control facility. The report also indicates that the conditions in which the pilot was flying were light to moderate turbulence and light to moderate rime/clear icing conditions. Engineering analysis and testing by the airframe manufacturer predicted the onset of flutter to occur at 241 knots [4].

The Mooney Aircraft Corporation, manager of structures, examined the airplane under the supervision of the NTSB investigators. Following a review of the observed evidence, the company reported that its preliminary conclusion is the aircraft may have had a failure to the horizontal stabilizers caused by flutter, initiated after buckling of the horizontal stabilizers due to either excessive loading above the design allowable limits or due to previous damage. Also, the FAA airframe specialist from the Ft. Worth, Texas, Airplane Certification Office, examined the recovered wreckage. Furthermore, the specialist noted the there was no indication from the structural fractures of corrosion or fatigue failure being present. Furthermore, he also indicated that both horizontal tail surfaces and elevators left the airplane in the air and that the right side did a negative/positive cycle to come up and hit the vertical tail and rudder. In the follow up from the FAA's Airplane Certification Officer reported that the airplane lost the horizontal tail due to some type of induced flutter caused by exceeding the never to exceed velocity (Vne) [4].

Dr. Stearman’s Report

According to the Stearman report, significant indications of flutter were found in the wreckage. The most obvious indication was the loss of the control mass balance weights on all of the control surfaces. Also, the report indicated that it was quite evident from the observations of the wreckage that all of the aerodynamic control mass-balance weights attached to prevent flutter were torn from their supporting structures by oscillations sufficient to destroy their supporting strength. The report also indicates that significant indication of flutter was present but it seemed (from wreckage) that the aircraft did not exceed the actual flutter speed of the undamaged airframe. Instead the flutter was triggered by an overload that damaged the horizontal tail brought on by the aircraft maneuvering in an unstable phugoid mode well above the structural cruise speed in the presence of turbulence [5].

The report also stated that the reading of the radar indicated that the pilot was flying below the structure’s cruise speed and also below the never to exceed speed. The radar also indicated that oscillations of long periods started occurring during the descent phase of the flight. These findings are indications of phugoid mode continuously being excited by the turbulence and gust existing in the lower atmosphere. Furthermore, the continuous presence of the phugoid mode was a result of a malfunction autopilot that did not account for modifications made to the aircraft [5].

Mooney M20k Modifications

The motivation for our analysis and design comes from the report provided for us by Dr. Stearman in which he stated that the caused of the accident of the Mooney 305 was due to a malfunction control system during a phugoid mode oscillation. The malfunction was due to the modification made to the airframe of the Mooney M20k, in which 45 percent more horsepower was added in addition to an extra propeller blade. These modifications can bee seen on Table 1. This also shows how the aircraft performance changed.

Table 1: Mooney M20K and Mooney 305 Specifications [6].

Specifications Mooney M20K Mooney 305

Horsepower 210 HP 305 HP

Propeller 2 blades 3 blades

Max takeoff weight 2900 lbs 3200 lbs

Max landing weight 2900 lbs 3040 lbs

Empty weight 1860 lbs 2068 lbs

Max. useful load 1040 lbs 1132 lbs

Stall Speed 60 kts 61 kts

Service Ceiling 24000 ft 24000 ft

Max. Level Speed 198 kts 240 kts

Cruise Speed (kts) 188 kts 120 kts

Takeoff over obstacle 50ft 1500 ft 1200 ft

According to our findings the Mooney 305 was FAA certified, meaning that the changes would not affect the airplane’s performance. Unfortunately, the Mooney 305 was never tested to see if the control system would respond to the new changes. The purpose of our project is to see if, in fact, the control system failed during the Mooney 305 crash.

Phugoid Mode Analysis

The phugoid mode is the simplest oscillatory mode in aircraft dynamics; therefore, a simple mathematical model is given to identify the phugoid mode. This model is given below. First, it was assumed that the thrust of the airplane was equal to the drag, so the only effective forces acting on the aircraft are lift and gravity. Therefore, the analysis started with the conservation of energy.

Mathematical Model of the Phugoid Mode

The simple mathematical model for the phugoid mode started with the energy equation which is shown below [7].

T+V=K, (, V=mgh, and K=constant) (1)

The thrust of the airplane is neglected and the mass (m) is assumed to be constant. Then the energy equation is differentiated with respect to time, and this operation results in the following.

m(+gh)=K (2)

=0 (3)

Another assumption is assumed to make the analysis simpler in which the equation for the lift coefficient (Cl), density () and the platform area (S) are assumed to be constant.

(4)

Another equation that is used in our mathematical approach was Newton’s second’s law which is shown below.

F=ma (5)

F=m=L-mg (6)

(7)

Once the conservation of energy and Newton’s second law are used to analyze the Mooney 305, the equations are combined to come out with a mathematical model that governs the behavior of the airplane with respect to height. The equation that comes out by combining Newton’s second law and the conservation energy is shown below:

(8)

After a model for height is found, a mathematical model for velocity is obtained by differentiating twice the equation of motion as shown below.

(9)

Once this equation is obtained, is substituted into the equation in order to have an equation only with respect to velocity. This is shown below.

(10)

In order to linearize this equation, small perturbations were taken into account to make the following equation show the approach we that was taken to obtain the model that shows the phugoid mode.

(11)

(12)

(13)

(14)

(15)

From these steps, a velocity model and a height model are obtained.

(16)

(17)

In order to add an external force to the equations, the formula uses a further manipulation which is shown below.

(F is the external force) (18) (19)

Once the external force is added to the equation, the derived the model looks as it is shown below.

(20)

(21)

This mathematical approach is obtained from previous studies already conducted to analyze the simple phugoid mode. The approach outlined above is for a simple, undamped phugoid mode oscillation occurring to an airplane. Although the mathematical model will be modified later, it will assist us in fully understanding how a phugoid mode works. The task will be to create a Matlab program that can demonstrate that in fact the phugoid mode caused the accident in Bakersfield, California.

Autopilot/Control System

In order to analyze the control system of the Mooney 305 a transfer function is being modeled using the equations of motion of the aircraft. The simple model will consist of velocity and acceleration inputs that will be obtained from flight simulator using the X-Plane software. Once we obtain the inputs, the control system will consist of three things: a vertical/directional gyro, a control servo, and an aircraft dynamics box. The components will control the stabilization of the aircraft (also known as the autopilot). The output expected from the control systems are those of a Mooney 305. The task will be to produce a control system that will stabilize an airplane during a phugoid mode.

Cessna 172H Study

After observing the aircraft during the entire flight, it is important to notice the effects the flaps have on the aircraft. The part on the trailing edge of the wing is called a flap. During takeoff and landing the flaps were deployed to vary the amount of force produced by the wing. The reasons for including flaps on an aircraft design are to change the airfoil shape and increase the wing area. Pivoting the trailing edge of the flap downward increases the airfoil camber and lift. Also, the large aft-projected area of the flap increases the total drag on the aircraft, which allows the airplane to slow down during landing [8].

Cessna Crash Background

The main focus of the Cessna investigation is the crash of the Cessna 172H model aircraft on August 12, 2000 in Peachtree City, Georgia. The incident occurred at 0920 eastern daylight time at the Falcon Field airport and resulted in the fatality of the pilot and two passengers as well as significant airplane damage. The meteorological conditions recorded that day at 1053 eastern daylight time were clear skies and calm winds with 10 miles of visibility.

Falcon Field consists of just one runway, which is 5,220 feet long with a field elevation of 808 feet. The airplane model flown that day was Griffin Flight School’s (Georgia) Cessna 172H, N1749F and was last inspected on August 8, 2000 [9]. This model consisted of a Continental Motors O-300-D, high-wing single engine with 145 horsepower. The table below was made showing some of the specifications of the aircraft [10].

Table 2 – Cessna 172H Specifications [10]

A list of the weights and balances of the aircraft and its passengers are shown below.

Table 3 – Weights of Aircraft and Passengers [8]

These values conclude that the aircraft were close to max gross weight upon takeoff.

The time of departure from the airport was at 0915 and the airplane stayed in the traffic pattern after takeoff. Following this, touch-and-go landings were made on runway 31.

Timeline of the Cessna 172H Crash

The picture below illustrates the steps involved in the crash at Falcon Field.

Figure 2 – Timeline of the Cessna 172H Crash [11]

The first point shows where the pilot started his descent for a touch-and-go landing. The aircraft was at an altitude of 1000 feet above ground level before turning downwind to the base. He was abeam to the runway and should have been going roughly 85 knots, with 10 degree flaps, at a descent rate of 500 feet per minute (fpm). At the second point the pilot started his final approach towards the runway, traveling at 75 knots with a descent rate of 500 fpm. The third location illustrates where the pilot started his final approach to the runway. The aircraft should have used 30 and 40 degree flaps with a 3 degree glide path at 65 knots. The aircraft flared at touchdown at point four and accelerated for takeoff up to point five. Hopefully, the flaps were raised at this point. According to an operator witness, the accident occurred when the pilot tried to climb out. Other witnesses mentioned that when the pilot tried to conduct a climb, the aircraft did not gain any altitude. Point six shows that the pilot made a left turn into the crosswind, where the aircraft should have been traveling at 550 fpm. A witness stated that the wing appeared to be perpendicular to the ground. However, the aircraft sustained problems and seemed to stall before crashing at point seven [9].

Performance of the Cessna 172H

A plot of the rate of climb versus the altitude was created from values taken from the Cessna 172 and Skyhawk Owner’s Manual and is shown below.

Figure 3 – Rate of Climb vs. Altitude for Cessna Aircraft

The linear fit was made and the top line represents the values for a 2000 pound Cessna aircraft, while the bottom line shows the values for a 2300 pound Cessna aircraft. The extracted point shows the maximum values that could be obtained for ideal conditions. The rate of climb is approximately 575 fpm and the density altitude is slightly less than 2000 feet [10].

Investigations of the Cessna 172H Crash

According to NTSB reports of the incident, the crash site was located approximately 300 yards southwest from the end of runway 31. The airplane remained in a nose-down position and its nose and engine was “displaced aft into the cabin area.” Further examination indicated that the wings received chord-wise crush damage aft to the wing forward spars, the left door was disconnected to the fuselage, and a propeller blade was bent and scratched. Also the tail was separated at the bulkhead of the frame. However, no component failures or mechanical malfunctions of the engine assembly were discovered. The flap actuator of the airplane was working correctly while the flap settings of the trailing edge wing were placed at 40 degrees. In contrast, the manufacturer suggests that flaps settings be set at 0 degrees for takeoff. The examination of the crash reported that the debris was littered locally around the plane [9].

Outside investigation indicated that upon impact, the tachometer showed a 1500 RPM reading. If this was the correct engine output, the aircraft would not be able to maintain a climb attitude.

The results from the crash investigations led to two conclusions of the causes of the crash of the Cessna 172H aircraft. Either the pilot tried to perform a touch-and-go landing and left the flaps down at 40 degrees, or the engine malfunctioned during take-off that resulted in a partial or total power loss.

Flight Testing

This section of the report discusses the actual flight testing of a Cessna 172N model aircraft. This test was performed at the Georgetown Municipal Airport on February 22, 2004.

Background of Flight Test

Originally this design project was to be conducted only using flight simulators. However, the decision was made to perform flight testing with an actual Cessna 172 aircraft. The reason for performing the test was to gain actual flight performance data on the Cessna 172. This data could then be used to compare against simulated data.

Selection of the Test Site

For the purposes of our investigation, we wished to select a test site that accurately depicted the Peachtree City Falcon Field Airport (KFFC). Falcon Field is a class E airport. This means that the aircraft using this airport communicate with each other using a common frequency rather than with a control tower. Falcon Field has a field elevation of 808 feet mean sea level. The dimensions of the runway are 5220 feet long by 100 feet wide and is made of asphalt [12].

We consulted the San Antonio Sectional Aeronautical Chart and searched for a suitable airport near Austin, Texas that we may use for testing. Georgetown Municipal Airport (KGTU) appeared to be the most logical choice as a test site. The airport has a field elevation of 790 feet mean sea level. Additionally, Georgetown Municipal Airport is a class E airport. The dimensions of the primary runway are 5000 feet long by 100 feet wide, and the runway is made of asphalt [12]. Because the two runways are very similar, this site appeared to suit our needs. Table 4 shows a comparison of the fields.

Table 4 – Comparison between Peachtree City – Falcon Field (KFFC) and the Georgetown Municipal Airport (KGTU)

	Falcon Field	Georgetown Municipal
Class of Airspace	E	E
Elevation (in feet msl)	808	790
Field Length (in feet)	5220	5000
Field Width (in feet)	100	100
Runway surface	Asphalt	Asphalt

After selecting Georgetown as our test site, we needed to inquire about the possibility of renting a Cessna 172H model aircraft. The Cessna Flight School representative at Georgetown is Wright Aviation. When we contacted them, we were informed that they had a Cessna 172N model that we could use for the purposes of conducting the flight tests. The primary difference between the two models is that the 172H model has a 145 horsepower engine while the 172N model has a 160 horsepower engine. Nevertheless, the two aircraft possess many of the same key features. They have essentially the same maximum gross takeoff weight, dimensions, and fuel capacity. The other feature that was appealing about the 172N model was that it had a 40 degree flap setting. Table 5 below illustrates the comparison between the two aircraft. The decision was made that the aircraft were substantially similar enough so that the flight testing could be performed.

Table 5 – Comparison between the Cessna 172H model and the Cessna 172N model

	Cessna 172H [10]	Cessna 172N [13]
Max Gross Weight (in lbs)	2300	2300
Usable Fuel Capacity (in gal)	40	40
Max Flap Setting (in deg)	40	40
Engine	0-300	0-320
Engine Horsepower	145	160
Propeller Type	Cruise	Climb

Flight Test Plan

Once the test site and aircraft were determined, the flight test profile was created. Douglas Wilson and Dr. Ron O. Stearman discussed the potential flight test to be considered. The flight test was being conducted to determine the effects that flap settings have on climb rate.

Initial consideration was given to flight testing from takeoff with different flap settings. Cessna only recommends takeoff with zero degrees of flaps for takeoff in the 172N model aircraft. Furthermore, Cessna does not provide performance charts detailing the climb performance of the aircraft for flap settings other than zero degrees.

The possibility exists that the Falcon Field accident may have been caused by a takeoff with the 40 degree flap settings. Because of this fact and the manufacturer’s recommendations, we elected to perform our rate of climb tests at altitude rather than at ground level. The decision to test at altitude was reached mutually by Dr. Stearman and Douglas Wilson.

The next issue regarding the flight test was who would pilot the test aircraft. The test was to determine the operational limits of the aircraft, so the decision was made to have a flight instructor at Georgetown fly the aircraft for the purposes of testing. Douglas Wilson contacted Wright Aviation to verify that an instructor could perform this test. Alan Weaver, Douglas Wilson’s flight instructor, agreed to fly the aircraft.

The actual test flight profile called for the rate of climb data on the aircraft. In order to obtain the data, the test called for the pilot to climb to a designated altitude and perform a climb at a designated airspeed and flap setting. This profile was to be done at two different altitudes, two different weights, five different flap settings, and four different airspeeds. This meant 80 data points were to be collected during the test.

In order to provide for post flight analysis of the test, the decision was made to video tape all phases of the testing procedure. The plan called for a video tape to be secured in the rear passenger area of the aircraft. This camera would then be pointed at the instruments during the test.

Flight Test SP4-E101

Test flight SP4-E101 was conducted at the Georgetown Municipal Airport on Saturday, February 21, 2004, from 8:00am to 2:00pm (1400Z to 2000Z). Present at the time of testing was Alan Weaver (test pilot), Dr. Ron O. Stearman, and Douglas L. Wilson. (An explanation of the test flight numbering scheme is found in Appendix 2 of this report.)

The temperature at the field was measured at 61˚F by the Automated Weather Observing System (AWOS) located at the airport. The dew point was recorded at 48˚F. The sky conditions were broken clouds at 800 feet AGL (above ground level). The barometric pressure was recorded as 30.04 in Hg (inches of mercury) adjusted for sea level. The winds were from the southeast between four and six knots. The pressure altitude at the field was determined to be 630 feet mean sea level.

The aircraft used in testing was the Cessna 172N model with a registration number of N739ZY. This aircraft is owned by the Wright Corporation and is based at the Georgetown Municipal Airport. The aircraft gross has an empty weight of 1469.10 pounds. The center of gravity for the aircraft is located at 38.97 inches. The useful load of the aircraft is 830.90 pounds. Appendix 1 of this report shows the weight data on this aircraft.

The preflight checklist was performed by Alan Weaver while Dr. Stearman and Douglas Wilson set up the video recording device. The Cessna 172N aircraft has a fuel capacity of 43 gallons (40 gallons usable), and the fuel tanks were full prior to takeoff for the test. The occupants for the first test were Alan Weaver (seated in the pilot’s chair) and Douglas Wilson (located in the co-pilot’s chair).

The weights and balances on the aircraft were performed. From this data, the aircraft was determined to be within its weight restrictions for flight. Table 6 below shows the determination prior to takeoff. The center of gravity for this aircraft at this weight qualified the aircraft for operation in the normal category.

Table 6 –Weight determination for the first test flight

	Quantity	Total Weight
Plane Empty Weight	1	1469.10 pounds
Usable Fuel (6 lbs/gal)	40 gallons	240.00 pounds
Pilot	1	190.00 pounds
Co-Pilot	1	193.00 pounds
Gear	1	10.00 pounds
TOTAL		2102.10 pounds

Cessna 739ZY departed Runway 18 from Georgetown at approximately 9:20 am (1320Z). Alan Weaver maneuvered the aircraft to the practice area and climbed to 3000 feet msl. Flight data was collected for the climb rate for all five flap settings at airspeeds of 60, 65, 70, and 80 knots indicated airspeed (KIAS). For the 40 degree flap setting the 80 KIAS airspeed was substituted by 55 KIAS due to lack of climb performance at that airspeed and setting.

After all of the data was collected, the aircraft returned to the Georgetown Municipal Airport, where the occupants departed the aircraft. A debriefing occurred at the Wright Aviation facilities. Following the debriefing, Alan Weaver and Douglas Wilson return to the aircraft and proceeded to collect the second set of data.

No refueling occurred prior to conducting the second test. Based on gage readings taken prior to the flight, an estimated 7.5 gallons of fuel was used during the initial data collection period. This would adjust the gross takeoff weight to 2057 pounds.

The aircraft proceeded to the practice area and climbed to an altitude of 2000 feet msl where an identical test was performed. The aircraft returned to the field and landed without incident. The estimated loss of fuel for this set of data collection was 45 pounds.

The flight crew returned to the Wright aviation facilities and a debriefing occurred with the crew and Dr. Stearman. The decision was then made to repeat the tests at a second weight. Dr. Stearman elected to join the flight crew as an additional passenger. He was positioned behind the pilot in the rear passenger seat. A weight check was performed to estimate the total weight of the aircraft as tested. Table 7 below shows the weight calculation.

Table 7 –Weight determination for the second tested weight

	Quantity	Total Weight
Plane Empty Weight	1	1469.10 pounds
Usable Fuel (6 lbs/gal)	25 gallons	150.00 pounds
Pilot	1	190.00 pounds
Co-Pilot	1	193.00 pounds
Rear Passenger	1	215.00 pounds
Gear	1	10.00 pounds
TOTAL		2227.10 pounds

The aircraft was loaded, and the aircraft proceeded to Runway 18. Following takeoff, the aircraft was maneuvered to the practice area. The aircraft climbed to an altitude of 2000 feet msl. The same series of tests as before were conducted for this weight and altitude.

A decision was made in the aircraft to add an additional test while at this altitude. The test called for the pilot to reduce the engine output to 2000 RPM. The pilot was then instructed to determine if, at any velocity achievable, any positive rate of climb could be obtained. This test was run with all five flap settings. The rate of climb was noted during this test.

At this point, the original flight plan called for the test to be repeated at this weight, but at a different altitude. Because the cloud ceiling was lowering, the decision was made to take a smaller sample of data at the 3000 foot msl altitude. Only two flap settings were tested. Data was collected for climbs at both 0 and 20 degrees of flaps. Following this data collection, the aircraft returned to Georgetown where Douglas Wilson landed without incident.

This concluded flight test SP4-E101. The data was then taken for further analysis.

Results of Testing

The test data collected consisted of the amount of time required for the aircraft to climb from the starting altitude to a preset altitude (generally 200-300 feet above the starting altitude). In order to compute the rate of climb, delta h (change in altitude) versus delta t (change in time) was computed. The results for the first test at 2100 pound gross takeoff weight are recorded in Figure 4 below.

Figure 4 –Calculated performance of the Cessna 172N as calculated from Test Flight SP4-E101

The data for the same altitude is represented in Figure 5 below. In this figure, the gross weight of the aircraft was increased. This figure shows the data for the aircraft at nearly the same weight as the aircraft involved in the accident at Falcon Field.

A copy of the flight test sheets is contained in Appendix 3 of this report. The Appendix includes the weather observations as well as the written documents from the rate of climb tests.

Figure 5 – Calculated performance of the Cessna 172N as calculated from Test Flight SP4-E101 (Max Gross).

Data Conversion

The aircraft tested in Test Flight SP4-E101 was a 160 horsepower engine. However, the aircraft involved in the accident had a 145 horsepower engine. This means that a data conversion needed to be performed in order to access the climb performance of the Cessna 172H model. This conversion was performed using the following equation [14]:

(22)

In this equation the RC represents the rate of climb and ΔP represents the change in power (in horsepower) from one engine to the other. The weight in pounds is included in the equation as W. By using this equation, this conversion constant can be determined in order to estimate the performance characteristics of the Cessna 172H. This function is performed below.

ft/min (23)

Applying this conversion to the previous figure, the derived performance can be calculated. Figure 6 below shows the estimated performance.

Figure 6 –Derived performance of the Cessna 172H

Figure 6 above suggests that the Cessna 172H model could not have climbed from the runway to the altitude prior to the crash in the allotted time. Additional testing will be performed through either actual flight tests or simulated flight tests in order to determine if these figures are correct.

Flight Simulation

The Flight simulation was used to recreate the aircraft flight profiles.. The flight tests were conducted using the X-Plane software. Tests were conducted in the Flight Simulation laboratory located in the WRW Laboratories at the University of Texas at Austin. The tests were conducted on the Macintosh computer. Additional flight testing was conducted at the house of Douglas Wilson. In that case, test were performed using a Hewlett Packard Pentium III laptop computer.

X-plane

The flight simulation package that was used for testing wass the X-plane software. The simulators in WRW 210 already had this program installed and were configured so that the corresponding control apparatuses could be used for virtual flight testing. X-plane comes with a number of standard aircraft, which can be adjusted or modified for a specific use. All of the aircraft react realistically because X-plane views the structures (horizontal stabilizer, vertical stabilizer, propeller, etc.) of an aircraft as geometric shapes that are broken into small elements. Then, finite element analyses are made to calculate the forces on each piece. With the forces and accelerations, velocities and positions are determined by integration.

The velocity vectors of the elements are determined from the linear and angular velocities, in addition to the longitudinal, lateral, and vertical arms of the elements. To maintain the correct physics of the models, the downwash, propwash, and induced angle of attack are also considered. The downwash is calculated for each coefficient of lift by considering the aspect ratio, taper ratio, sweep of the wing, and the horizontal and vertical distance of the horizontal stabilizer from the wing. X-plane calculates the propwash that is required for momentum to be conserved. This is done observing the area of each propeller, the thrust of each propeller, and the local air density. Dynamic pressures are found for all of the elements based on temperature, air speed, altitude, wing sweep, and propwash. Using the coefficients determined in X-plane with the dynamic pressure, the total force of the aircraft can be determined. From this the linear acceleration and angular acceleration can be determined by dividing the forces by the mass of the aircraft.

The Plane-Maker option in X-plane was used to create all of the models of this project. Also, World-Maker was used to add the scenery to the airports where the airplanes were flown [15].

Simulator Validation

The data from flight test SP4-D102 was used to confirm the validity of the X-Plane software. The flight test will be repeated using Cessna N463QE which has been created in X-Plane. This aircraft is a Cessna 172N model which will be used primarily for the verification flights.

The validation phase of the simulation is vital to the design project. The simulator must be able to reproduce actual flight if the team is to analyze aircraft design and performance. Without this task being accomplished, the reliability of any findings this team makes will be questionable.

Accident Recreation

Following the validation of the software, Cessna N463QD will be used to recreate the Falcon Field accident. Tests will be conducted to access the capabilities of climb out using 40 degrees of flaps. Additional tests will be performed to simulate engine failures shortly after takeoff, and we will attempt to recreate the most probable accident scenario.

The Mooney Rocket will also be investigated. The flight testing will be performed by employing the autopilot and recording the change in altitude versus time. This data will then be exported to Excel to be used in a Matlab program. This data will be processed to determine if the phugoid mode is present.

Cost Analysis

The only expenses associated with the project are related to the actual flight testing. The costs include the aircraft rental and instructor fees. A proposal has been made for a second test flight using a Cessna 172 with a 145 horsepower engine. If this is performed, the cost of the second flight should be comparable with the first. Table 8 below shows the monetary outlay for the project.

Table 8 – This table shows the cost estimates for the project.

Test Flight #1	$ 472.00
Proposed Test Flight #2	$472.00
TOTAL	$944.00

All simulated flights will be conducted at the WRW Laboratories. As such, these test flights have no out-of-pocket expense associated with them.

Time Analysis

This project has been separated into three separate phases. Each phase requires completion of the previous phase in order to be started. Table 9 highlights the time line of the project.

Phase one contained three separate parts. This included the primary investigation into the events and actions that lead to the two accidents in question. Additionally, because the phugoid mode is suspected in the Mooney accident, research into this mode needed to be conducted. This required learning how to analyze the output data to perceive if the phugoid mode exists. The final part of phase 1 included the completion of the test models to be used in the X-Plane software. This phase has been completed.

Phase two has been split into two areas. Phase 2A was included after the project was started. This phase involves the actual flight testing of the Cessna 172 aircraft. Phase 2A should be completed by March 15, 2004.

Phase 2B calls for the simulated flight testing of all vehicles. This includes three series of validation flights. Once the simulator use has been verified, the accident recreation can begin. Phase 2B is set to be completed by April 15.

Phase 3 of the project will involve the final analysis of the data. Included in this will be the completion if the final report and final recommendations. This phase is set to be completed by May 5, 2004.

Table 9 – Time line of the project

TASK	Completion
Phase 1	February 20, 2004
Phase 2A	March 12, 2004 - PROJECTED
Phase 2B	April 15, 2004 - PROJECTED
Phase 3	May 5, 2004 - PROJECTED

Conclusion

Flight testing is an integral phase of the design process, but testing can be very expensive. The expenses can be measured not only in dollars and cents but also in human lives. Flight simulation offers a way to limit expenses.

Flight simulators have become better representations of flight in the past decade. Many pilots, both civilian and military, use simulators for training and practice. The question still remains, however, if these simulations are accurate enough to assess possible safety issues and stability and control concerns.

WiNG Research is currently trying to answer this question. Flight testing has already been conducted in actual aircraft to be used for comparison in the flight simulations. These flight simulation tests shall occur within the next week. WiNG Research will continue this work for several weeks. A final report on this topic will be available in early May.

References

1. Garmin Ltd. Home Web Page, “Garmin G1000,” URL: http://www.buygarmin.com/products/g1000/ [cited 5 March 2004].

2. Aircraft Shopper Online (Consumer Web Page), “1958 Cessna 172,” URL: http://www.aso.com/i.aso/AircraftView.jsp?aircraft_id=79716 [cited 5 March 2004].

3. Norman, Donald A., the Design of Everyday Things, Basic Books, New York, 2002.

4. National Transportation Safety Board, “LAX98FA154,” URL: http://www.ntsb.gov/ntsb/brief.asp?ev_id=20001211X10119&key=1 [cited March 4, 2004].

5. Stearman, Ronald, “Opinion Statements Concerning the Impact of the Rocket 305 Conversion on the Donald T. Michael Accident near Bakersfield, California on May 5,1998,” undated.

6. Rocket Engineering Home Web Page, “Specifications,” URL: http://www.rocketengineering.com/rock_spec.html [cited 5 March 2004].

7. Nelson Robert C., Flight Stability and Automatic Control, WCB McGraw-Hill, St. Louis, Missouri, 1998.

8. Hull, David G., “Introduction to Airplane Flight Mechanics,” Department of Aerospace Engineering and Engineering Mechanics, The University of Texas at Austin, Austin, Texas, Spring 2003.

9. National Transportation Safety Board, “ATL00FA079,” URL: http://www.ntsb.gov/ntsb/brief2.asp?ev_id=20001212x21645&ntsbno=atl00fa079&akey=1 [cited March 4, 2004].

10. Cessna Aircraft Company, “Cessna Model 172 and Skyhawk Owner’s Manual,” Wichita, Kansas, 1984.

11. Peachtree City –Falcon Field Airport Information Web Page, “Airport Information,” URL: http://www.kffc.org/airport.htm [cited 5 March 2004].

12. The Pilot and FBO Flight Planning Guide – AOPA’s Airport Directory, 2001 – 2002 edition, 2001, pp. 3-140, 3-504.

13. Cessna Aircraft Company, “Cessna 1978 Skyhawk: Cessna Model 172N – Pilot’s Operating Handbook,” Cessna, Wichita, Kansas, 1977, pp. inner cover.

14. Smith, H.C. “Skip,” The Illustrated Guide to Aerodynamics, 2^nd edition, McGraw-Hill, New York, 1992, pp. 120-121.

15. TheBeltsBro., Inc., “Description,” X-Plane, http://www.x-plane.com/descrip.html [cited March 4, 2004].

Appendix 1 – N739ZY Weights

Below is shown the aircraft weight and balance on the Cessna 172N aircraft that was used for Flight Test SP4-E101.

Figure 7 – Weights and Balance sheet for Cessna N739ZY

Appendix 2 – Flight Test Number Designation

This appendix provides for the flight test numbering that has been used in this report. This numbering scheme should be used for any future flight tests.

Characters 1 and 2 –These two characters identify the session in which the test was performed.

SP – 463Q Spring session

SU – 463Q Summer Session

FA – 463Q Fall Session

Character 3 – This digit represents the year in which the test was conducted.

4 – 2004

5 – 2005

6 – 2006

7 – 2007

8 – 2008

9 – 2009

Character 4 – This character follows the dash. The character identifies the aircraft used in the flight test.

A – Mooney Rocket 305

B – Cessna 172 SP

C – Cessna 172 SP – modified for X-Plane to include 40 degree flaps

D – Cessna 172 H

E – Cessna 172 N

F – Cessna 172 E

Character 5 – This character describes the test flight.

1 – Flight performed in an actual aircraft

2 – Flight performed in a simulator

Characters 6 and 7 – These characters represent the sequence number of the flight. The sequence numbers all start will 01. For instance the first test flight of the Mooney will start 01 and the first test flight of the Cessna 172 SP will be 01.

01 – First test flight

02 – Second test flight

03 – Third test flight

Example: SP4-E101

This means that the flight test was performed by the 463Q class during the Spring session. The test was conducted in 2004. The aircraft used in the flight was a Cessna 172 N aircraft. The test was performed under actual flight conditions. The flight was the first test flight conducted.

Appendix 3 – SP4-E101 Flight Data

The following pages contain copies of the actual recorded data for Test Flight SP4-E101.

Figure 8– This shows the weather readings recorded from AWOS.

Figure 9 – This is data recorded at the 2100 pounds and 2000 feet agl.

Figure 10 – This is data recorded at the 2100 pounds and 3000 feet agl.

Figure 11 – This is data recorded at the 2200+ pounds and 2000 feet agl.

Figure 12 – This is data recorded at the 2200+ pounds and 3000 feet agl.

The University of Texas at Austin

March 5, 2004

Appendix 1 – N739ZY Weights

Appendix 2 – Flight Test Number Designation

Appendix 3 – SP4-E101 Flight Data