Solar Powered RC Plane

Jon Kalow, Jun Kudo, James Nick Vines

 

---

Project Summary:

Our goal is to design and build a remote-controlled toy plane that can be recharged using solar panels mounted on the plane.

 

---

 Intro: 

Although solar aviation is nothing new, solar power aircraft are not commercially available.  We argue that an affordable RC plane that utilizes solar panels to extend its total flight time can be designed and built. 

 

Many existing RC planes run out of juice very quickly, while existing solar-powered planes are built for proof-of-concept rather than everyday use.  We think that there is an open space inbetween the two where a mid-range, solar powered, remote-controlled plane can be designed and produced for recreational use.

---

Background:

 

While there are several different solar-powered, remote control airplanes out there ranging from small hobbyist projects to NASA's 247-foot Helios, there are very few commercially available solutions that allow the recreational user to experience solar-powered flight.  

 

Scientists and engineers have been building unmanned, solar-powered planes since the mid 1970's for educational and proof-of-concept purposes.  Many of these planes have the goal of being able to fly indefinitely- while they may carry batteries onboard for nighttime flight, the hope is that they can fly independantly without external power sources.  While there has been talk of using these high-altitude, long-endurance planes for communication, observational, and various research purposes, most of these planes have little practical value.

 

We have been able to find quite a few smaller-scale projects done by hobbyists and students in a vein similar to ours.  However, many of these planes are not commercially viable, or are designed for purposes similar to the NASA planes.

 

 

Nasa's Helios Project

 

The Mikrosol

 

Good history on solar powered flight: http://www.asl.ethz.ch/research/asl/skysailor/History_of_Solar_Flight.pdf

Similar project with useful posts: http://www.solarfreaks.com/rc-solar-plane-t154.html

 

Intellectual Property:

 

US Patent 4415133: Solar Powered Aircraft. Link

                 -A special aircraft design meant to keep solar panels facing the sun. Not in conflict with this project. 

 

 

 

---

My Proposed Solution:

 

We would like to design, build, and fly a remote-controlled plane which derives a certain amount of its power from solar panels.  The plane should be robust, and have the necessary control systems to make it a commercially viable.  Enough power should be generated by the solar panels to extend the flight of the plane significantly, and allow the plane to be recharged out in the sun without having to swap in new batteries.

 

Requirements:

1. Requirements
   - The plane must derive at least 10% of its operating power from its solar panels 

          - The plane must be cheap and commercially viable (we define cheap as less than $200)

    2.  Constraints

          - The plane must fly and have some sort of radio-control mechanism which allows it to take off and land safely

          - The plane must be newbie friendly (relatively impact proof and slow flying)

---

 

First Prototype:

 

Brainstorms: 

For our first prototype, we plan on buying an existing RC plane and modifying it to use solar power.  This will involve researching solar panels, and designing the electronic system necessary to recharge the battery safely and easily with a minimum weight.  The primary focus of the first prototype is the design of system connecting the solar panels to the batteries.

 

RC Plane Criteria:  

The plane purchased must be cheap, light, low-powered, have a large planform for the placement of solar panels, and be beginner friendly.  The planes considered are compiled in the appendix.

 

We purchased the B29 Bomber 2-channel RC plane which has a wing span of 1050 mm, fuselage length of 710 mm, and a flight time of 5-6 minutes.

B29 Bomber RC Plane

 

The plane was put together and flown to test the general ability of the out-of-box plane.  However, our first flights with the plane ended in prompt crashes with various parts snapping off in the collisions.  The flight seems to fail to get enough lift even at full thrust.  Further testing is necessary.

 

The plane uses a 7.2 V rechargable battery source, which conveniently matches the output voltage of the solar panels we are purchasing.  The radio control system is two-channel, steering the plane using differential thrust between the 2 motors mounted on each wing.  We are planning on doing some investigation as to the power consumption of the motors, and the power capacity of the battery pack.

 

 

B29 RC Plane:  Airplane fully assembled

 

B29 RC Plane:  Underbody with battery visible

 

B29 RC Plane:  Two-channel radio unit

 

B29 RC Plane:  Unsuccessful test flight

 

B29 RC Plane:   Planned layout for solar outfitting

 

Solar Panels:

We are looking at using thin, flexible solar panels offered by Powerfilm.  Powerfilm makes ultralight, flexible panels specifically for use in applications like RC planes.  Even the non-aircraft panels are extremely light and plenty flexible enough for our purposes.

 

We purchased 6 of the RC7.2-75 10.6''x3.5'' strips, which offer a decent power output at an extremely light weight, in a convenient & flexible package.

 

Solar Panel: RC7.2-75

 

Solar Panel:  Ultra-thin film design

 

-7.2 V Output

-100 mA operating current

-10.6 in x 3.5 in

-0.008 in thick

-0.21 oz weight

 

Other Considerations: 

Circuit Probing

Since we plan on using the B29's remote control system in our future prototypes, we must analyze the circuit onboard the plane and understand how the RC input fundamentally controls the motors.  This step is critical for any future modifications/changes to the batteries and/or motors.  We hope to be able to draw up a simple schematic representing the onboard controls.

 

Battery charging circuity

It is necessary that the plane have both regulation and protection circuits for the battery charging system.  We plan on developing the system to charge the batteries from the solar panels, with an emphasis on simplicity.  We need to control the flow of current from the solar panels to the battery with a diode (to prevent battery drainage to the panels), and make sure our system does not interfere with the existing operation of the airplane.

 

Battery Research

The B29 comes with a heavy NiMH Battery pack for its energy source.  Research should be done if lighter but equally powerful batteries can be affordably purchased.  (Look into LiPO batteries)

 

Some measurements will be taken on the current battery pack- determining the battery's internal resistance and charging characterisitics will be important in learning how much power it will take to recharge the pack in flight.

 

Tests on solar panels

We plan on characterizing the output voltage and current from the solar panels when they come in.  The linearity of output voltage and variation of power output with solar energy will be important for both getting our first prototype to work, as well as the feasibility of our overall project. 

 

Testing and Theory:

We took measurements on the battery, plane power consumption, and solar panels to determine the characteristics of the current setup, and to create a theoretical energy model for the final design.

 

 

Testing current draw from one motor at full throttle

 

Battery Information
Battery voltage:	7.2	V
Storage Capacity	650	mA-h
Energy Capacity	4.68	W-h
Cell Resistance	10.5	Mohms
Motors
Number	4	motors
Full throttle voltage	6.3	V
Current draw	1.7	A
Power consumption	10.71	W
Total power consumption	42.84	W
Max flight time (theoretical)	6.554621849	minutes
Actual drainage time (power diminishes to zero)	8	minutes
Wall socket charger
Voltage	7.2	V
Current	250	mA
Power	1.8	W
Charging time	2.6	hours
Solar Panels
Number	4	panels
Voltage	7.2	V
Max current	100	mA
Power generated	0.72	W
Total power	2.88	W
Charging time	1.625	hours
Integrated Plane (first prototype)
Power Consumption	42.84	W
Power Generation (maximum)	2.88	W
Net power consumption	39.96	W
Flight time	7.027027027	minutes
Increase in flight time	7.21%	percent increase
Two-motor plane, 6 panels (theoretical)
Power Consumption	21.42	W
Power Generation (maximum)	4.32	W
Net power consumption	17.1	W
Flight time without panels	13.1092437	minutes
Flight time with panels	16.42105263	minutes
Increase in flight time	25.26%	percent increase

 

Results:

 

The first prototype was successfully wired and solar panels attached.  The plane runs with the solar panels attached, which very slowly charge the battery under sunlight.  The 4 7.2 V panels are connected in parallel to charge the battery and supplement extra power during flight.

 

 

Prototype 1:  B29 RC plane with solar panels attached

 

A diode in the electronics bay prevents the battery from discharging over the solar panels.  This is a Schottky diode, which has a smaller forward voltage drop than other diodes:

 

Prototype 1:  Circuit schematic for solar panel system

 

Prototype 1:  Additional electronics onboard B29 Bomber including Schottky Diode

 

Second Prototype:

Brainstorms:

For our second prototype,  we plan on fabricating a new RC plane from scratch material with the intention of constructing a plane to better fit our needs.  We want this plane to be lighter and powered only by two motors while having enough wing span for the placement of solar panels.  For this prototype, we focus on material selection, plane sizing (dimensioning of all the parts), airfoil selection for the main wings and tail wings, and motor/battery selection.  As for the control mechanism, we plan on simply utilizing the same radio control system used in the B29.

 

 

Plane Sizing: 

General How Big Should it be?

The actual size of the plane will need to be chosen and an appropriate motor paired with this size. Some equations that will be useful in approximating the parameters of the plane are located here. These equations will help us in designing the plane.

 

We looked at two major factors:  Power loading and wing loading. 

We looked into the power loading of "typical" RC planes and found that while advanced aerobatic trainers need about 50 watts per pound, 10-30 watts per pound is about average for "intermediate" electric rc planes.  Since we planned to use two out of the four motors from the B29 plane, we estimated a propeller efficiency, set a target power loading of 30 Watts/lb and calculated a target weight of roughly .5 lbs or 226 grams.

For the wing loading, we set a "stall speed" or a minimum speed of the plane from which we calculated the size of the wing.  As a guide, for model rc planes, if the wing loading is under 10 ounces per square foot, it is suitable for slow flying while from 10-20 ounches per square foot, the plane handles like a "intermediate trainer."  Since we want a reasonably slow park flyer with large wings, we chose the stall speed to be 12 ft/s or roughly 8 mph.  We estimate the maximum coefficient of lift to be around 1.4.  This gives us an estimated wing loading of around 4.4 ounces per square foot (well under the target 10) and with the target weight of .5 lbs, a planform area of 262 square inches.    

 

Main Wing Sizing:      

We pick the aspect ratio (wingspan squared divided by the planform area) from empirical data of previous planes (somewhere between 6 and 8 - a higher value gives lower drag but is usually heavier). 

As for the taper ratios, readings suggest ratios between 0.5 and 0.8.  Tapering of a wing is used to change the spanwise lift distribution - how much of the lift occurs at what spanwise location.  Ideally, we want the lift to be spread from tip to tip in the shape of an ellipse (elliptical lift distribution).  Also tapering is favorably for structural reasons because the root chord is longer and as a result deeper; this provides greater leverage for handling wing bending moments.  However, we found that many successful homebuilts have untapered wing configurations due to the simplicity of construction.  For this reason, for this prototype, we choose to NOT use any tapers in any of the wings.

 

Balancing:

For stability, the center of gravity of the entire plane must be located at roughly 25% of the mean aerodynamic chord (MAC) (this is where the center of pressure of the airfoil is).

 

Tail Wing Sizing:

Airplanes have tails for one purpose - to make stabilizing moments.  Aspect ratios were chosen using empirical data.  The sizing of the tail wings depends solely on the distance of the MAC of the main wing to the MAC of the tail wings. 

 

Airfoil selection:  

We need an airfoil for low reynolds number flight, preferably high lift for low speeds.  General airfoil classes include symmetrical airfoils, semi-symmetrical airfoils, "modified" flat-bottom airfoils.  Symmetrical airfoils and semi-symmetrical airfoils seem to be used for aerobatic planes (i.e. planes that can fly upside down and do crazy loops).  Flat-bottom airfoils are used for for aircraft that are willing to make the compromise of having more drag in exchange for slow fight/high lift capabilities.  Flat-bottom airfoils tend to be found on "trainer class" RC planes.  Airfoils seen used in model planes include:  NACA 2415, Clark Y (flat-bottom), Great Plane PT-40 trainer airfoil (used on the Great Plane Perfect Trainer), NACA 6412, FX63137.  DAT files for various airfoils can be found at the UIUC Airfoil Coordinates Database .  

 

For the selection of airfoils, Mark Drela's XFOIL  will be used to analyze the lift and drag coefficients as a function of angle of attack.

Using XFOIL, we were able to specify a Reynold's number, a Mach number and normalized coordinates of airfoils to get graphs of the coefficient of lift as a function of angle of attack.  In these graphs we look for a high maximum coefficient of lift and easy stalling characteristics.  We selected the NACA2415 airfoil for the main wings (later changed to the thinner NACA2412) and the NACA0009 (symmetrical airfoil) for the tail wings.

 

In terms of manufacturing, we have access to a foam cutter that can cut tapered wings with specified cross sectional areas with variable washout angles.

 

 

Wing Placement:

Monoplane wing configurations are classified into three categories - high wing, low wing, mid wing.  High wing placement is when the wing is located above the fuselage and is usually supported with wires or struts of some kind.  The high wing configuration generally is considered the most stable and steady and is usually used for "trainer class" RC planes.

 

Material Research:

It is of the utmost importance that our plane can be manufactured from a lightweight material, in addition to being sturdy enough to withstand the dangers present in test flight.  We note that the B29 RC plane is made of relatively brittle material (as depicted in our first crash landing).  We have decided on using EPP foam (expanded polyproplyene) for the wing and body material.  EPP foam is a much more durable material than the usual styrofoam, maintaining a light structure while being flexible enough to absorb impact. 

 

The EPP foam can be cut using hot wire for the airfoils.  As the foam lacks rigidity, cavities will be necessary to maintain a support structure of wood, metal wire, or carbon fiber.  It is undetermined as to how easy it would be to create the plane body- it could be wire cut to an extent, but if the foam is flexible to the point of being impossible to machine or shape, a more standard styrofoam body with EPP padding may be necessary.

 

The EPP foam was purchased at: http://www.flyingfoam.com/foamsheet.html

 

 

 

Sample Airfoils in production

 

 

 

Solar Panel Considerations:

We plan on sticking with the flexible 7.2V panels, as they are well-suited to our purpose and exploring too many options may have a hefty price tag.

 

Battery selection and circuitry:

Most of the RC airplanes we are looking at use 3.7V lithium batteries, while the flexible solar panels put out 7.2V.  Even with the panels in parallel, a linear regulator will be necessary as to not fry the battery.  Fortunately, at currents under 1.5 A that we will be working at, cheap 3-pin integrated circuit components are available that can easily provide voltage regulation for our purposes.

 

Simple linear regular IC:

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=S-816A37AMC-BAMT2G-ND

 

Batteries are unfortunately prone to damage if one attempts to apply an excessive voltage to a charged battery, or excessive current to a discharged battery.  We will need to come up with a simple protection circuit to prevent the battery from being overcharged or damaged, and to prevent the battery from discharging to the solar panels.   Since none of us are electrical engineers, we may need to look for guidance before moving too far forward.

 

We have purchased a 7.2V lithium-polymer pack which comes with a regulating PCB.  Since lithium batteries are easily prone to damage under excessive charging or discharging, a printed circuit board is built onto the cells which limits the current flow in and out of the cells.  This will be extremely helpful for protecting the cell in our circumstances.

 

Battery:

http://www.all-battery.com/74v875mahli-polybatterypack-2.aspx

 

Motor Research:

We would like to find lightweight, efficient engines that can give us the best deal in terms of thrust-to-weight ratio.  We never really got around to upgrading the motors.

 

Control mechanisms:

Ailerons for roll, elevator for pitch, control for variable thrust, rudder control for yaw.  Although advanced control systems could be used, we simply stuck with the control mechanism of differential thrust from the B29 bomber.  Due to the control system present on the B29 that we may borrow, differential thrust may be necessary.

 

 

 

Preliminary Design:

 

Preliminary Design Calculations and dimensioning:

here 

 

Prototype 2:  A preliminary design in solidworks, with all the major structures, motor, and batteries, and appropriate weights has been developed to help locate our center of mass and balance the plane for the final design calculations

 

 

Secondary Design:

 

Secondary Design Calculations and dimensioning:

 

 

 

Prototype 2:  Secondary design in solidworks complete with electronics bay, solar panels, battery and motors.

 

Results: 

 

The second prototype was successfully constructed and wired up with the motors, battery, and solar panels. 

Prototype 2:  Constructed plane - Top View

 

Prototype 2:  Constructed plane - Side View

 

Prototype 2:  Constructed plane - Electronics Bay

 

 

 

 

 

---

 

Final Results:

 

Price Analysis

 

 

Bomber: $40

Solar Panels (4): $112

Battery: $19

Foam: $25

 

 

Other components (not purchased):

Balsa support struts (3)

Wiring

Solder

5-minute epoxy

Acrylic cover plate

Packing tape

Schottky diode

 

 

Total cost: $196 plus paraphernalia

 

 

Conclusion

 

 

Design Goals:

1) The plane must derive at least 10% of its operating power from its solar panels.

- The first iteration of the final design consumed 21 W of power at full throttle and generated 2.88 W of power from the solar panels.  This is 13.7% of power consumption- however, the plane could barely fly at full throttle.

- The final working plane consumed 43 W of power at full throttle and generated 2.88 W of power from the solar panels.    This is only 6.7% of power consumption- however the plane could be flown to some degree at partial throttle

- For most purposes, this design goal was not met

 

 

 2) The plane must be cheap and commercially viable (we define cheap as less than $200)

-  Less than $200 was spent on this project

- A number of smaller components were used in this project as well- however in a production setting, the components could be bought in bulk, and it would not be necessary to buy a ready-to-fly RC plane.

- This design goal was met

 

 

Design Constraints

1) The plane must fly and have some sort of radio-control mechanism which allows it to take off and land safely

- The plane could take off and land by remote control.  Landing gears were not implemented, as it was found that landing RC planes in a conventional manner is extremely difficult and for our skill level it is better to crash gently on grass. 

- As the plane could take off and fly by RC control, and hit grass while remaining structurally sound, this design constraint was met.

2) The plane must be newbie friendly (relatively impact proof and slow flying)

-The plane was difficult to fly, but did fly better than the ready-to-fly unit that we bought

-Differential thrust, while the cheapest, was determined to be a poor method of airplane control

-The plane was relatively impact proof, and held up much better over multiple flights than the first prototype (which had a 100% failure rate during flight).  Crashes involving breakage did occur, but this is largely unavoidable with beginner RC aircraft pilots.

-This design constraint was met

 

 

Conclusion

While not all design goals were met, a working solar RC plane was built and a good deal was learnt in the process that could be applied to a better solar RC plane.  Some of the considerations:

Control methods

Differential thrust was determined to be a poor method of control.  A plane has 3 axes of rotation that are necessary for control – pitch, roll, and yaw.  Using differential thrust affects all 3 of these at once, making it extremely difficult to pull out of a turn or adjust to wind gusts.  This was likely the worst aspect of the final plane, and was unfortunately limited by the control system taken from the first prototype.  Ultimately, differential thrust was an ineffective method of flight control.

Airfoil and plane design

The final plane was well-balanced, had sufficient lift with 4 motors, and could fly in a stable straight line.  It is not determined whether a different control system would have allowed better stability while turning.

 

 

Motor Efficiency

More efficient (brushless) motors would have been preferred to improve airplane efficiency - brushed DC motors mounted on the airfoil were an inefficient use of power

Materials

The expanded polypropylene foam used worked very well for most purposes.  It was light, durable, and relatively stiff.  It did not have a great surface finish, but this was easily remedied by coating the airfoils with lightweight tape.

 

 

Battery Selection and circuitry

The lithium-polymer battery worked great in the new airplane.  It was lighter, and had twice the capacity of the prototype plane’s battery.  In addition, a protection circuit onboard limited concerns about overcharging or excessive power draw.  Connecting the panels in parallel with a diode provided an easy and lightweight way to both charge the battery and help drive the motors.

Solar Panels

The solar panels used possessed an excellent weight and form factor, but we could not get them to produce as much power as they were rated to.

 

 

Future Considerations:

 

Air-Worthiness

                The two main problems that interfere with the air-worthiness of this design are the structural strength and the controllability.

 

                Structural: The design of the fuselage had a thin walled section where the radio was housed. This was the location of the majority of the cracks even with an acrylic support. This could be improved by moving the radio to a different section that doesn't experience high stresses, such as closer to the tail or more towards the front. The second change that would help is the wing fuselage interface. Right now it is held together by epoxy. This works well most of the time but, as seen in the flight trial video, can give out. A possible fix is to use an even stronger insert in place of the balsa wood. This would help the plane withstand rough landings.

 

                Controllability: This plane used differential thrust, as shown in the video, which doesn't provide much control. The fix for this would be to use a more complex system that adds some or all of the following: elevators on the horizontal tail, rudder on the vertical tail, and ailerons on the main wings. The addition of elevators and ailerons would give the best control since that would then allow pitch and roll to be precisely controlled. The cons of this would be the added weight and complexity.

 

Power

                The solar panels currently cover most of the plane surface area, but there is still a decent amount in other areas such as the leading edge, the tail wing, and the fuselage. The addition of panels to each of these areas would require a design check to ensure the plane is still stable.

 

Miscellaneous:

 

 

 

Appendix:

RC Planes we looked at:

 

Product:  Cessna 210 RTF

Price:  $119.99

Weight:  18 g

Dimensions:  15" wingspan, 12-3/4" long

Power supply:  Micro 3.7V 70mAh Li-Po battery pack

Additional Comments:  "Simply install the AA batteries, charge the flight battery (which takes about 15-20 minutes) and you’re ready to fly! Enjoy up to ten minutes of flight time with each charge."   "this is an extremely lightweight indoor flyer and is unbelievably maneuverable."  "can only be flown outdoors when there is little to no wind."  READY  TO  FLY

 

Product: Vapor RTF  Micro RC Plane

Price:  $129.99

Weight:  15 g

Dimensions:  14-3/4" wingspan, 15-1/4" long

Power Supply:  3.7V 70 mAh LiPo battery with charger

Additional Comments:  Indoor plane or outdoor on calm day.  Super light weight, and slow. READY  TO  FLY

 

Product:  Mini Super Cub RTF

Price:  $99.99

Weight:  7 oz ~= 200g

Dimensions:  31-3/4" wingspan, 21-1/2" long

Power Supply:  2 cell 7.4V 300 mAh Li-Po battery with 2 cell DC Li-Po charger and AC adapter

Additional Comments:  "Durable and lightweight Z-Foam constructed fuselage makes repairs easy."   READY  TO  FLY

 

Product: EasyStar RTF  Electric Flyer

Price:  $189.99

Weight: 24 oz ~= 680 g

Dimensions:   54" wingspan, 35" long, 370 sq. in. wing area

Power Supply:  PC11X7F X-Cell 7 Cell 1100 mAh Folded NiMH Pack with Connector

Additional Comments:  Nick approves of this one and so does that random hobbyist READY  TO  FLY

 

Might want to look at cheaper, 2 channel trainer planes like:

http://www.hobbytron.com/Super-Sonic-RC-Electric-Plane.html

http://www.nitroplanes.com/yellowbeertf.html

 

We decide to purchase a 2-channel trainer plane due to the giant price difference.  Two channel planes can be purchased from amazon for roughly $30 - $40.  

Possible 2 channel trainer planes:

Product:  Air Hogs Nano Hawk  

Price:  ~ $30

Weight:  10.4 oz

Dimensions  7.5 x 10.5 x 11 inches

Additional Comments:  indoor flyer   -  READY TO FLY - seems to get good reviews everywhere - jon k claims it is too small.  nick agrees.  jun likes the plane.

 

 

Solar panel considerations:

RC7.2-75

http://www.powerfilmsolar.com/products/custom/index.php?voltages=ALL_PRODUCTS&model=RC7.2-75

 

Powerfilm's line of thin, flexible panels:

http://www.powerfilmsolar.com/products/custom/?voltages=ALL_PRODUCTS

 

Each RC7.2-75 panel sells for around $18 to $30 each, so price will be a concern on a larger plane.

 

---