Sports | Parachuting » Imam-William-Diego - Fully Automated Parachute Design

Source: http://www.doksinet Fully Automated Parachute Design Team -“T-Wolf” Colorado State University - Pueblo Senay T. Imam, Berry William, Scott Kennymore, Diego Ramirez Cesti, Richard J. Chartarro Junior Advisers: Dr. Huseiyn Sarper Date 4/7/2008 1 Source: http://www.doksinet Table of Content 1. Introduction: 2. Design Approach Mechanical design Electrical Design 3. Landing Estimation Analytical analysis Program Output 4. Conclusion and recommendation Annex 1 Annex 2 Reference 2 Source: http://www.doksinet 1. Introduction: Background The T-Wolf Parachute Lander is a Colorado state university – Pueblo project design sponsored by Colorado Space Grant Consortium in collaboration with CSU-Pueblo. The entire purpose of the project is to build a completely autonomous rover that can simulate an autonomous robot mission on Mars and a completely automated parachute Lander that holds this rover. This report, therefore, presents the concepts considered and the process used to

make our design decisions for the parachute lander. Objective: The T-Wolf parachute Lander team designed a balanced and aerodynamically stable parachute probe that would maintain structural, aerodynamic, and survive impact upon landing. The probe was designed to be launched with a balloon and caries it to about 1500 ft. At this altitude a microcontroller triggers a release mechanism to detach the parachute from the balloon. Once the release mechanism is triggered, information is sent to the ground that notifies the beginning of the descending process. The probe makes a free fall for a specific period of time (≈6 to 10 seconds), until the parachute adjusts is itself. After the probe is adjusted, it starts to descend at an acceptable terminal velocity until it reaches to a level of 2.5 ft from the ground Once the microcontroller detects a surface, it triggers another release mechanism to detach the probe from the parachute (canopy). Once the release mechanism is triggered, the probe

makes a free falling to the ground. This is basically designed to clear a path for the rover when it exits the probe. 3 Source: http://www.doksinet Project Schedule: January 18th: First Group Meeting January 28th: Allocating of tasks to each team member February 5th – 8th: Researches, Literature review, design proposal, brain storming ideas February 12th – 22nd: Analyzing the mechanical and electrical conceptual design March 26th: Ordering of Materials March 11th- April 31st: Beginning construction of the probe April 1st – April 4th – Testing different mechanism of the probe including the parachute 2. Project Approach The project were split into two sections, design concept and location prediction. To utilize time and specialties of team members, the design concept was further split into two subsections, mechanical and electrical. 2.1 Design concepts 2.11 Mechanical concept The mechanical design concepts are categorized into a. Mechanics of Ascent b. Probe shape c. Release

Mechanisms d. Parachute specification f. Material selection a. Mechanics of Ascent The probe is attached to 8in diameter high altitude helium balloon which delivers it to the specified height (1500 ft) for release. 4 Source: http://www.doksinet b. Probe Shape The shape and geometrical features of the probe is selected after careful consideration of the weight limit, its ability to settle down on its desirable position and withstand sock during the process of crash landing. The probe has a shape of geometrical pyramid made up of aluminum sheet, imbedded inside a semispherical shaped aluminum tubes. c. Release Mechanisms The release mechanism is based on the process of gun trap. In this design, Nichrome is used to trigger the gun trap. Nichrome is a nonmagnetic ally of nickel and chromium with a very high melting point of around 1400oC.1The gun trap is designed to be triggered on three different stages: releases the balloon from parachute, the parachute from the probe and finally

opens the doors of the probe. d. Parachute specification Based on the information collected from different literature reviews the parachute was decided to be a round parachute. e. material selection The material selection system for the entire project is based on weight, strength and price. The materials used in construction of the probe are aluminum sheet, large aluminum tubes, foam padding, a manner of attachment, such as bolts or rivets, microcontroller and gun trap assemblies. The pyramid-shaped probe is fabricated from sheet aluminum stock, and the “cage” assembles from the long aluminum tubes. Small booster cables are used to open the 1 Wikipedia – Nichrome 5 Source: http://www.doksinet bay doors upon trap detonation. The probe doors will be attached via two springloaded hinges 2.12 Mechanical Design Design: Autodesk Autocad and Autodesk 3DS Max software are used in developing the probe design in a 2D and 3D design respectively. Figure 1.1 2D Probe design

Construction: The probe will have a square base, and will resemble a pyramid shape; with the help the foam, this will house the rover tightly to avoid any damage incurred by landing. The means of having the rover land upright consistently will be through three 30” diameter circular tubes forming a spherical shape around the rover bay unit. This will allow the center of gravity caused by the weight of the rover to settle on the base, no matter at what angle of impact. The bay doors will be operated via a pressure-fit gun trap, allowing the doors to open upon receiving signals from the onboard microcontroller. Two doors are employed in this system, in case one door cannot function due to an external force or obstruction. 6 Source: http://www.doksinet Systems: The parachute lander will employ a number of assemblies and systems to carry out its function. The entire vehicle will be fitted with an onboard microcontroller; its function is to properly engage various circuits and

processes designated by speed and height. It will send impulses to the gun traps, as well as transmit its position to a landbased computer system through a streaming connection or constant beacon Another system is that of the gun traps. Their purpose is to provide a constant link or connection via a silver-soldered cable, until a predetermined time. The gun trap is designed to have a pressure fit holding the assembly, and a small powder charge held in place by a cotton ball will separate the respective sides. After testing, the correct powder charge used in the assembly is 6.1 grains of IMR4831 smokeless gunpowder Assembly: Preliminary assembly of the rover entry vehicle started with fabricating the pyramid-shaped housing out of sheets of .032” 6061 aluminum sheet All edges are lapped and pressed, to increase rigidity of the unit. The rover bay is assembled on an 18.5” square, with approximately 10” of height The rover itself is 8” tall, so protective padding will keep the unit

in place during ascent and descent. A small box fabricated with 6061 aluminum sheet is used to shield the microcontroller on top of the pyramid bay. Spring-loaded hinges are attached to the top of the pyramid bay, to assist opening the bay doors. The initial design plan was to use the spring loaded hinges as the primary opening mechanism for the system, but upon actual assembly they were found to be insufficient at opening the doors effectively. To overcome this problem, two small bungee cables attached to the “cage” will be more than sufficient to fix the problem, at little cost to weight limit. Next, a small gun trap tunnel is also fabricated out of aluminum sheet, to shield the rover from the internal gun trap detonation. The gun traps were the next to be fabricated; using spent and deprimed .44 Remington 7 Source: http://www.doksinet Magnum cartridges and a piece of aluminum tubing stock at .45” diameter A lathe was used to machine holes for the primer exit from the

assembly, and threads and a shoulder were cut into the assembly to support and seal the system. Wire was silversoldered to the plugs on either end, to allow attachment to different mechanisms Assembling the traps themselves required model rocket fuses to be shielded with rubber shrink tubing. The fuses were then inserted into a drilled out primer hole in the cartridge, until approx. 15” of total primer is exposed It was then held in place by modeling clay, and .6CC’s of clear epoxy was injected into the cartridge to seal the primer hole and retain the fuse. After the epoxy hardened, the clay was removed and 61gr of smokeless powder was added to the cartridge, which is held next to the fuse by a small piece of cotton ball. The plug was machined to 425”, which allows a tight press-fit that can hold double the amount of actual weight of 15 pounds. The plug is inserted into the cartridge and the gun trap is ready to be installed onto the entry vehicle. One of the last stages of frame

construction is bending 8ft 1” aluminum tubing into a 30” diameter, which is approximately 95 inches. A homemade bender was employed to fit the 3 tubes around a 30” diameter round metal disc to retain most of its bent qualities. Next, the 3 tubes were placed in a conduit bender to achieve the exact 30” diameter. Two of the aluminum tubes were now cut in half, and cut with an approx. 20” base, so the bottom could be fastened to the probe bay. The center tube spanning the unit horizontally has the other tubes welded to it, making a sufficient mounting point for the parachute and upper gun trap. The tubes had to be heat treated during welding, to avoid cracking the thin tubing. At this point, the foam insulation was cut and fitted to the inside of the bay, which is adhered to the siding by using commercial spray adhesive. The last piece of sheet metal fabrication involved the round parachute8 Source: http://www.doksinet mounting disc. A disc with approximately 6” had 16

notches cut into it at 225 degrees, to evenly distribute the parachute cords. A hole was cut into the center of it to allow for the mounting of the gun trap. The ends were slightly folded over, so commercial tape could be wrapped around the perimeter so the parachute cords were retained in the assembly. The last stage of assembly is attaching the parachute to the balloon A wire rill be run from the microcontroller to the upper gun trap, and the trap will be fastened to the balloon and a small hole in the top of the balloon will allow for one side of the trap to be placed through it, and retained by a piece of aluminum sheet and cardboard. These pieces will drop out once the unit is separated Operation: The entry vehicle assembly will be attached to its descent mechanism from the top of the “cage”. The descent mechanism used in this case is a 96” parachute designed for a fifteen-pound payload. The parachute will be fully deployed for the current challenge, since an ascent

mechanism is required to reach its proper altitude of 1500 feet. A large weather balloon attached to the top of the parachute will cause the unit to ascend. A positioned gun trap will cause the balloon to separate, and the parachute will fully deploy. Upon detection of the proper altitude (via an onboard altimeter/accelerometer), another gun trap mechanism will disengage the parachute from the rover bay at a designated height; in order to clear the landing area of a possible hazard caused by the parachute. At this point, the bay’s geometry will allow the unit to land upright on any surface at any angle of approach. After a designated time or sequence a gun trap, utilizing a cable with attached machine screws, will activate and cause the doors to open, thus allowing the rover to exit. 9 Source: http://www.doksinet 2.13 Electrical Concept 2.14 Electrical Design Materials used: For the main system: - Arduino micro controller board w/ atmega 168 chip - PC with arduino programming

software/drivers - Parrallax ultrasonic sensor - 433MHz wireless transciver. - Battery pack/connection. For balloon Release Timer: (schematic included) - .01 uF Ceramic Disc Capacitor - 4700 uF Electrolytic Capacitor - 1N914 Silicone Switching Diode - 555 Timer Integrated Circuit - 10Kohm carbon film resistor - 47 K ohm PCB Mount Micro Potentiometer - General Purpose PC Board - 9 Volt Battery Clip 10 Source: http://www.doksinet Figure 2 Overall electrical system Layout Figure 3: Time Mechanism 11 Source: http://www.doksinet Operations 1. Initiate launch 2. Time assent to determine altitude 3. Release balloon at appropriate altitude to begin descent 4. Detect proximity to ground to release parachute to avoid entanglement 5. Detect or time landing to open doors and release payload Control systems: Electrical work has been done primarily off of the Amtel Atmega based arduino platform. The complications of running a wire from the microcontroller up to the top

of the parachute to release the balloon have made it impractical. There fore so a more simple device using a standard 555 ic timer circuit to delay for a certain rise time is used and this at a designated time release the balloon from the craft and allow it to return to the surface. This device will delay a predetermined time and then initiate a small powder charge ejecting an interferance-fit coupling, effectively releasing the landing craft. A microcontroller/board is used as a central controlling system. It is programmed in the C programming language and written to the eeprom using software and usb drivers available on the internet. The probe uses an ultrasonic range-finder device to test proximity to the landing site in order to effectively release the parachute prior to landing to reduce the risk of the payload becoming entangled in the chords. It was initially hoped that a wireless link could be created between the landing craft and a notebook computer on the ground that would

relay data about the stages of the operation on the probe and the state of the processor; however, due to time restraints and a mistake made during ordering, it is excluded from this project. Despite such mistakes, the wireless link on the craft will be constructed; however the uplink on the ground will be missing. The ultrasonic device produces an analog output relative to the distance it detects which is read by the 12 Source: http://www.doksinet microcontroller which can make logical decisions based on its input. 3. Landing Estimation A communication signal is attached to the parachute that sends a signal when the parachute is released from the balloon and starts descending, however this signal does not provide any direction of descending or location of landing. What’s more adding GPS to the parachute was beyond the budget allocated for the project. Therefore the team decided to write a program using Matlab that inputs all the governing factors and predicts the velocity and

direction of descending and the possible landing location of the parachute. Although the program was tested to see its reliability of estimation for the velocity, the reliability of the direction and the estimation of the landing location haven’t been tasted yet. However, the results obtained from various tests have proven to have similar output to the results obtained from the program. Methodology: Using the inputs, the program computes the ascending and descending velocity, location of possible landing location and its range from the launching site. It also uses a Mont Carlo simulation system to determine the mean direction and speed of the wind. Furthermore, the program is designed to function using two basic unit system, English and Metric system. 3.1 Analytical analysis Following are the analytical analysis used in determining velocity, location and range landing of the probe: 1. ascending velocity: Lift force = Buoyant – Weight of the probe + Weight of the balloon Ascending

velocity = square root (8*Lift/ (Coefficient of lift density of air PiDiameter of balloon^2)) 2. Descending velocity 13 Source: http://www.doksinet Total Weight = Drag force Total Weight = weight of the probe + weight of the parachute Deriving the descending velocity from the weight Descending velocity = square root (8*Total weight/ (Coefficient of drag density of air PiDiameter of parachtue^2)) 3. Range Range = average wind speed *(ascending time + descending time); Program output ------------Location Estimator------------------English System ----Lift (lbs) = 120.282725 Ascending ---->Velocity (m/sec) = 10.870617 Ascending-----> time (sec) = 138.798933 Descending----> velocity (ft/sec) = 16.504768 WIND ---> Mean (ft/sec) = 8.334849 Time (sec) = 90.882829 Stdev = 1.435354 Range (ft) = 1914.362899 ANGLE ---> Mean (degree) = 165.006475 Stdev = 6.130701 Direction of wind ---> SOUTH EAST Programming code is attached at on Annex 2. 3. Conclusion and recommendation

The main purpose of the project is to build a robust parachute lander with a reasonable or affordable price. Having been able to achieve this target, the next level of the project should aim on upgrading the mechanical system, minimizing the size, weight, 14 Source: http://www.doksinet and the electrical assembly system and also adding more navigation analysis to make the system more robust. The team recommends that, materials such as GPS, Telemetry system, and more sophisticated balloon release system, which were need but not being able to be used due to budget and time constrain, should be available to enhance much more research, as the direction and the goal achieved so far are promising. We recommend that the project be continued next year in much more detailed design. Building upon our future experience, the future team could improve in by implementing 1. More sensors such as infra-red devices, speed of sound sensor, and aim to test the project in a very high altitude. 2.

Improve the probe design and make it more impact resistor 3. create more redundant devices that ensures the operation of the system 4. Pressurize the probe Annex 1 Team Picture 15 Source: http://www.doksinet Picture 1 Picture 2 16 Source: http://www.doksinet Picture 3 Picture 4 17 Source: http://www.doksinet Annex 2 : Programming Code function uplifting Cls = 0.5; %for spherical objects Cdp = 1.5; % for round parachtute %global lift force drag force disp(---Menu---); disp(1. Meteric System); disp(2. English system); disp(Choose the Unit System); choose= input( ); fid=fopen(Test result.txt,wt); fprintf(fid,------------Location Estimator--------------- ); if choose ==1 fprintf(fid,----Metric System ----- ); else fprintf(fid,----English System ----- ); end switch 1 case choose==1 Dia =input(Enter Diameter of the balloon: ); while Dia<=0 Dia =input(Enter Diameter of the balloon: ); end D air =input(Density of air: ); while D air<=0 D air=input(Density of air: ); end D

He=input(Desnity of Helium: ); while D He<=0 D He=input(Desnity of Helium: ); end Mass Probe=input(Probe Mass: ); while Mass Probe <=0 Mass Probe=input(Probe Mass: ); end H=input(Max height of release: ); while H<=0 H=input(Max height of release: ); end gravity=9.81; %meter/second^2 case choose==2 Dia =input(Enter Diameter of the balloon: ); D air =input(Density of air: ); D He=input(Desnity of Helium: ); Mass Probe=input(Probe Mass: ); H=input(Max height of release: ); gravity=32.17; %feet/second^2 otherwise end % inorder for lift to take place "force up = force down" % and the enternal pressure has to be lighter than the outside/atmospheric % pressure % force up --> lift % force down ---> drag % lift = Bouyant force upward - gravitational force 18 Source: http://www.doksinet % Bouyant force = weight of balloon and probe plus the weight of helium gas % To find the mass of the empty balloon volume = pi*(Dia^3)/6; Mass Bal=D He*volume; Bouyant= D

air*volumegravity; %to compute for the vertical lift velocity %Lift = cross sectional area of the ballon times density times %velocity times the coefficient of drag Lift = Bouyant - ((Mass Probe + Mass Bal)*gravity); if choose ==1 fprintf( Lift (N) = %f , Lift); fprintf(fid, Lift (N) = %f , Lift); else fprintf( Lift (lbs) = %f , Lift); fprintf(fid, Lift (lbs) = %f , Lift); end Velo y = (8*Lift/(ClsD airpiDia^2))^(1/2); if choose==1 fprintf(Velocity (m/sec) = %f , Velo y); fprintf(fid,Velocity (m/sec) = %f , Velo y); else fprintf(ascending --->Velocity (fts/sec) = %f , Velo y); fprintf(fid,ascending ---->Velocity (m/sec) = %f , Velo y); end acc up = Lift/(Mass Probe + Mass Bal); %fprintf(Upward acc until the terminal velocity ---> (ft/sec^2)= %f ,acc up); time up1 = Velo y/acc up; Y = 0.5*acc uptime up1^2; time up2 = (H-Y)/Velo y; time up = time up1+time up2; if time up>0 fprintf(ascending----> time (sec) = %f , time up); fprintf(fid,ascending-----> time (sec) =

%f , time up); else disp(Error); end % the free fall of the aprachute which is equal to gravity*time = the % terminal velocity will give us the time and lenght the parachute covers % to stablize it self. in this case since the parachute is open in air the % distance come out to be about 5 feet and the time was about 0.6 seconds % and therefore it is ignored from the computation % Weight W implies to the weight of the parachute and probe Dia par = input(Diameter of the parachute: ); V par = Dia par^3*pi/12; W pra = V par*D airgravity; W = W pra + Mass Probe*gravity; vel par = sqrt(2*W/(Cdp(piDia par^2/4)D air)); %Terminal velocity time down= H/vel par; if choose ==1 fprintf(Decending----> velocity (m/sec) = %f Time (sec) = %f ,vel par,time down); fprintf(fid,Decending----> velocity (m/sec) = %f Time (sec) = %f ,vel par,time down); else 19 Source: http://www.doksinet fprintf(Decending----> velocity (ft/sec) = %f Time (sec) = %f ,vel par,time down);

fprintf(fid,Decending----> velocity (ft/sec) = %f Time (sec) = %f ,vel par,time down); end %Horizontal speed of the %input the upper wind speed, lower wind speed, and the mode U Wind = input(The upper wind Speed: ); L Wind = input(The lower wind Speed: ); M Wind = input (The mode wind Speed: ); disp(wind direction angles in DEGREE); disp(North ---> 0 or 360); disp(South ---> 180); disp(East ---> 90); disp(West ---> 270); %the user can input any range within this values U Ang = input(The Upper wind direction angle: ); L Ang = input(The Lower wind direction angle: ); M Ang = input(The Mode wind direction angle: ); mean w= (U Wind+L Wind+M Wind)/3; stdev w=sqrt((U Wind^2 + L Wind^2 + M Wind^2 - U Wind*L Wind - U WindM Wind - M Wind*L Wind)/18); %fprintf(WIND ---> mean = %f standard deviation = %f , mean w,stdev w); mean a= (U Ang+L Ang+M Ang)/3; stdev a=sqrt((U Ang^2 + L Ang^2 + M Ang^2 - U Ang*L Ang - U AngM Ang M AngL Ang)/18); %fprintf(ANGLE----> mean = %f

standard deviation = %f , mean a,stdev a); % Number of iteration for the simulation of the wind speed N=input(Number of iteration: ); wind=zeros(N,2); for ii=1:N Z=randn; wind(ii,1)=stdev w*Z+mean w; wind(ii,2)=stdev a*Z+mean a; end %average wind speed within a given duration Totalspeed=wind(:,1); %disp(Totalspeed); Total wind=sum(Totalspeed); W MEAN=Total wind/N; Sum std w=0; for ii=1:N Sum std w=Sum std w + Totalspeed(ii,1)^2; end W STDEV=sqrt((N*Sum std w - Total wind^2)/(N(N-1))); fprintf(WIND ---> Mean = %f Stdev = %f ,W MEAN,W STDEV); fprintf(fid,WIND ---> Mean = %f Stdev = %f ,W MEAN,W STDEV); %FINDING RANGE Range = W MEAN*(time up+time down); if choose==1 fprintf( Range (m) = %f ,Range); fprintf(fid, Range (m) = %f ,Range); else fprintf( Range (ft) = %f ,Range); fprintf(fid, Range (ft) = %f ,Range); 20 Source: http://www.doksinet end %average wind direction angle within a given duraiton Total angle=wind(:,2); %disp(Total angle); Angle sum=sum(Total angle); A

MEAN=Angle sum/N; Ang std=0; for ii=1:N Ang std=Ang std + Total angle(ii,1)^2; end A STDEV=sqrt((N*Ang std - Angle sum^2)/(N(N-1))); fprintf(ANGLE ---> Mean = %f Stdev = %f ,A MEAN,A STDEV); fprintf(fid,ANGLE ---> Mean = %f Stdev = %f ,A MEAN,A STDEV); switch 1 case A MEAN==0|A MEAN==360 fprintf(direction ---> NORTH ); fprintf(fid,direction ---> NORTH ); case A MEAN>0&A MEAN<90 fprintf(direction ---> NORTH EAST ); fprintf(fid,direction ---> NORTH EAST ); case A MEAN<360&A MEAN>270 fprintf(direction---> NORTH WEST ); fprintf(fid,direction---> NORTH WEST ); case A MEAN==90 fprintf(direction ---> EAST ); fprintf(fid,direction ---> EAST ); case A MEAN>90&A MEAN<180 fprintf(direction ---> SOUTH EAST ); fprintf(fid,direction ---> SOUTH EAST ); case A MEAN>180&A MEAN<270 fprintf(direction ---> SOUTH WEST ); fprintf(fid,direction ---> SOUTH WEST ); case A MEAN == 180 fprintf(direction ---> SOUTH );

fprintf(fid,direction ---> SOUTH ); case A MEAN == 270 fprintf(direction ---> WEST ); fprintf(fid,direction ---> WEST ); otherwise end Reference: 1. Wikipedia, the free encyclopedia, “ Nichrome” 2. Opening Shock and Shape of the Drag-vs-Time Curve, Jean Potvin, Physics Department, Saint Louis University, St. Louis MO 3. Hemisphere Parachute Design (for parafauna) http://www.stwikipediaorg/~anthony/kites/parafauna/chute design/ 4. Balloon lift calculations with different gases http://www.madsciorg/posts/archives/1998-08/900965216Phqhtml 21