75
Cornell Rocketry Team Critical Design Review CORNELL ROCKETRY TEAM (CRT) CDR PRESENTATION

CORNELL ROCKETRY TEAM (CRT) CDR PRESENTATION · CORNELL ROCKETRY TEAM (CRT) CDR PRESENTATION. Cornell Rocketry Team Critical Design Review AIRFRAME. ... Center of Gravity (CG) = 59.JK

  • Upload
    others

  • View
    35

  • Download
    0

Embed Size (px)

Citation preview

Cornell Rocketry TeamCritical Design Review

CORNELL ROCKETRY TEAM (CRT)CDR PRESENTATION

Cornell Rocketry TeamCritical Design Review

AIRFRAME

Cornell Rocketry TeamCritical Design Review

LAUNCH VEHICLE DIMENSIONS

Total Length 103”

Airframe Tubing OD = 5.15” ID = 5.00”

Coupler OD = 4.998” ID = 4.815”

Motor Mount Tube OD = 3.098” ID = 3.000”

Fin/Bulkhead Thickness = 3/32” Centering Ring Thickness = 1/8”

1515 Delrin Rail Buttons

Cornell Rocketry TeamCritical Design Review

LAUNCH VEHICLE COMPONENTS

Cornell Rocketry TeamCritical Design Review

MOTOR SELECTION

75mm AeroTech L1520T

Provides sufficient thrust to reach target apogee.

Maximum G-forces experienced by LV low enough that DRS will remain secured and undamaged during takeoff

Cornell Rocketry TeamCritical Design Review

STABILITY

From tip of nose cone

Center of Gravity (CG) = 59.34”

Center of Pressure (CP) = 78.29”

Static Stability Margin = 78.29" − 59.34"

5.15"= 3.68 cal

Cornell Rocketry TeamCritical Design Review

Total mass = 37.3 lb

Thrust-to-weight-ratio = 9.1

Projected apogee = 5478 ft

Velocity off rod = 81.4 ft/s

Cornell Rocketry TeamCritical Design Review

MASS STATEMENT

Mass Margin

Maximum mass before projected apogee below 5,000 ft: 40.3

Percent difference: 8%

Section Mass (lb)

Booster (launch mass/spent motor) 22.5 18.6

Forward 14.8

Total (launch mass/spent motor) 37.3 33.4

Cornell Rocketry TeamCritical Design Review

RECOVERY SYSTEM

Event Description Location Redundancy

1 Launch ready Launch pad N/A

2 Nose cone ejection Apogee 4000 ft

3 Booster section separation Apogee +0.5s Apogee +1.5s

4 Booster & AV separation Apogee +1s 4000 ft

5 All main parachutes deploy 500 ft N/A

6 Landing Ground level N/A

Cornell Rocketry TeamCritical Design Review

RECOVERY SYSTEM

Cornell Rocketry TeamCritical Design Review

RECOVERY SYSTEM

Forward and booster sections will use ½” thick, 25’ long kevlar shock cords

Launch Vehicle Component Drogue Size (in) Main Size (in)

Forward Section 15 72

Booster Section 18 96

Launch Vehicle Component

Drogue Descent Velocity (ft/s)

Drogue Kinetic Energy (ft-lb)

Main Descent Velocity (ft/s)

Landing Kinetic Energy (ft-lb)

Forward Section 82.19 1553.77 16.76 64.63

Booster Section 76.74 1700.56 14.14 57.76

Cornell Rocketry TeamCritical Design Review

RECOVERY SYSTEM DRIFT DISTANCES

Launch Vehicle Component

0 mph wind 5 mph wind 10 mph wind 15 mph wind 20 mph wind

Forward Section 0 645.18 1290.36 1935.55 2580.73

Booster Section 0 715.99 1431.97 2147.96 2863.95

Cornell Rocketry TeamCritical Design Review

SUBSCALE RESULTS

Apogee: 3110 ft

Projected apogee: 3128 ft

1% difference from measured apogee

Coefficient of drag: 0.58

Cornell Rocketry TeamCritical Design Review

SUBSCALE GROUND TESTING

Cornell Rocketry TeamCritical Design Review

SUBSCALE RECOVERY

Cornell Rocketry TeamCritical Design Review

AIRFRAME PLANNED TESTS

Full-Scale Ground Test

Test ObjectiveValidate that the black powder charges are sufficient to break the shear pins and separate the sections and verify that the recovery system properly deploys parachutes.

Success Criteria All Launch Vehicle sections separate cleanly, and all parachutes deploy.

Testing Variable Functionality of the recovery system.

MethodologyThe ejection charges are loaded into the AV bay and the forward airframe sections. All sections are assembled and secured with shear pins or rivets. Each charge is ignited manually to observe separations.

Cornell Rocketry TeamCritical Design Review

AIRFRAME PLANNED TESTS

Full-Scale Launch

Test Objective Validate the overall Launch Vehicle, payload, and recovery designs.

Success CriteriaAll recovery systems function as designed, the vehicle is stable during launch, and the vehicle is fully recoverable.

Testing Variable Functionality of the Launch Vehicle as a whole and the Launch Vehicle apogee.

Methodology

The Launch Vehicle is constructed following the design planned in the CDR documentation. The sections are assembled and attached using shear pins and rivets. The Launch Vehicle is launched using an AeroTech L1520T motor. The performance of the recovery system, recorded apogee, and any unanticipated launch events are analyzed for future full-scale launches.

Cornell Rocketry TeamCritical Design Review

COMMUNICATIONS (COMMS)

Cornell Rocketry TeamCritical Design Review

SYSTEM REQUIREMENTS

Required Features:

Use GPS and radio modules to determine the position of the Launch Vehicle at all points during the flight

Incorporate multiple GPS/radio transmitters onboard the Launch Vehicle

Determine the location of all sections of the Launch Vehicle after they have landed using the SRB, Homemade GRB, and TRACER module

Utilize multiple backup systems in the case that any single system fails to perform as expected

Additional Expectations:

Obtain video of flight utilizing an on-board camera

Onboard camera is the Xiaomi Yi, controlled by a Arduino Pro Mini

Save and transmit all flight information for vehicle tracking, post-launch analysis, and redundancy

Utilize and Arduino Mega to read data from every sensor on the TRACER shield and write to SD module

Cornell Rocketry TeamCritical Design Review

ELECTRICAL COMPONENTS

Cornell Rocketry TeamCritical Design Review

HOMEMADE GPS RADIO BEACON (PRIMARY SYSTEM)

Cornell Rocketry TeamCritical Design Review

HOMEMADE GPS RADIO BEACON (PRIMARY SYSTEM)

GPS Radio beacon

Onboard GPS module to gather positional information

Utilizes LoRa radio for transmission, which will allow for direct communication with Ground Station

Run by ATMEGA328P microprocessor

Custom firmware allows for fine control over information transmitted as well as interval of transmission and frequency of transmission

Cornell Rocketry TeamCritical Design Review

TRACER MODULE ELECTRONICS (PRIMARY SYSTEM)

The TRACER Module Utilizes Arduino Mega connected to:

▪ GPS Module

▪ Adafruit 10DOF Sensor

▪ Xiaomi Yi

▪ SD Card Module

▪ LoRa Radio

Transmits data to Ground Station

Saves data to SD Card module

Xiaomi Yi assembly Includes Pro Mini to control camera state

Seated with lens facing out of nosecone to record flight

Cornell Rocketry TeamCritical Design Review

SIMPLE RADIO BEACON (REDUNDANT SYSTEM)

Simple Radio Beacon

Morse Transmission of HAM Radio Licensed Operator

Requires minimal power

Utilized for simple direction finding using fox-hunting techniques

Transmits at 100mW

Transmission occurs at a set interval of 10 seconds

Transmission contains no encoded information (only tone)

Cornell Rocketry TeamCritical Design Review

GROUND STATION GUI

Receiving Data

LoRa radios will receive incoming packets from TRACER Module and Homemade GRB

LoRa radios transfer contents of transmission to Arduino Uno, which interprets the data and streams the data to the Ground Station laptop

Displaying Data

The CRT GUI receives the data streamed from the Arduino Uno

GUI plots location of LV

Displays telemetry information including velocity and acceleration of sections

Displays separate tracking information for each section

Cornell Rocketry TeamCritical Design Review

DEPLOYABLE ROVER SYSTEM (DRS)

Cornell Rocketry TeamCritical Design Review

DRS OVERVIEW

Two-wheeled rover

Lead Screw Mechanism (LSM) for deployment

Cornell Rocketry TeamCritical Design Review

ROVER

Two primary wheels

6.50” x 3.82” x 2.48” chassis

Stabilizing wheel

Solar panel housing

Cornell Rocketry TeamCritical Design Review

PRIMARY WHEELS

Premade rubber tire with deep treads

Premade wheel shell with grooves for the tire

3D-printed wheel insert and cap to contain nut, collar, and axle

Cornell Rocketry TeamCritical Design Review

AXLES AND MOTORS

Hex axle for stronger connections

Gear system to protect servo motor

Bearing to reduce friction

Collars for axial constraint

Cornell Rocketry TeamCritical Design Review

CHASSIS

Aluminum panel shell for strength

3D-printed body to house electronics

Rubber sealant for waterproofing

Cornell Rocketry TeamCritical Design Review

CHASSIS SHELL

Six shell pieces bolted together

⅛” thick panels

¼” thick edges for screws

Cornell Rocketry TeamCritical Design Review

CHASSIS BODY

Three sections to allow for maintenance and disassembly

3D-printed for ease of manufacturing and mass reduction

Cornell Rocketry TeamCritical Design Review

CHASSIS BODY

Securely houses axles, electronics, and wiring

Cornell Rocketry TeamCritical Design Review

STABILIZING WHEEL

Prevents chassis from spinning in place

Folds under rover when stored, deployed by a spring-loaded hinge

Cornell Rocketry TeamCritical Design Review

SOLAR PANEL ACTUATION

Solar panels held with faces touching

Unfolded with a servo motor after rover stops

Cornell Rocketry TeamCritical Design Review

BOTTOM SOLAR PANEL HOUSING

Bottom solar panel held face up on the top shell of the chassis

Rectangular hole with protrusions to allow for sharp angles

Cornell Rocketry TeamCritical Design Review

TOP SOLAR PANEL HOUSING

3D printed slot connected to the servo motor

Cornell Rocketry TeamCritical Design Review

LEAD SCREW MECHANISM (LSM)

Deploys rover by turning lead screw

Secured via mounting screws

Guide shafts removed due to overconstraint

Cornell Rocketry TeamCritical Design Review

MOTOR COMPLIANCE ASSEMBLY (MCA)

Square nuts prevent nuts from spinning in place

Forward bulkhead moves in plane to prevent load on lead screw

Cornell Rocketry TeamCritical Design Review

MOTOR COMPLIANCE ASSEMBLY (MCA)

Ball bearings to reduce friction for the forward bulkhead

Lead screw coupled to DC motor

Coupling constrained by aft bulkhead

Cornell Rocketry TeamCritical Design Review

SLEDS

Sit between the wheels

Constrained by tire treads

Ball bearings to reduce friction

Fall off after rover is deployed

Cornell Rocketry TeamCritical Design Review

AIRFRAME INTEGRATION

Aft end shear pinned to AV forward airframe

Forward end attached to forward airframe using mounting screws

Bulkhead cap tethered to AV Bay

Cornell Rocketry TeamCritical Design Review

SUBSCALE PROTOTYPE

Former design using fixed LSM, guide shafts and moving bulkheads

Chassis and wheels both entirely 3D-printed

Cornell Rocketry TeamCritical Design Review

SUBSCALE PROTOTYPE

Added tether to prevent personnel hazard

Unable to deploy due to overconstraint

Cornell Rocketry TeamCritical Design Review

DRS TESTING - MECHANICAL

Centrifuge Test

Impact Test

Splash Test

Solar Panel Deployment Test

Drawbar Pull Test

Ground Terrain Test

Slope Test

Reorientation Test

Cornell Rocketry TeamCritical Design Review

ELECTRICAL & SOFTWARE (E&S)

Cornell Rocketry TeamCritical Design Review

POWER

The Power distribution will supply 6.0V to the motors of the rover

The Power distribution will supply 5.0V to the microcontroller of the rover

2 step-down circuits using basic L78XX series regulator

Second circuit is modified to supply 6.0V

Fuse in circuits limits current to 200mA and 2.4A respectively (use 80mA and 1A fuse)

Cornell Rocketry TeamCritical Design Review

MOTORS

Pololu 99:1 25Dx54L DC motor for linear actuator

▪ Updated from subscale (increased RPM, adjusted torque requirements)

▪ 97 RPM

▪ 13.13 lb-in stall torque (2.2 min required)

Power supply of 6.0V

Schematic includes flyback diodes to prevent current spikes, decoupling capacitors to reduce noise

Board has wider trace widths to avoid burnout in the connections

Cornell Rocketry TeamCritical Design Review

MOTORS

Two Pololu Power HD Continuous Rotation servos for wheel motors

Also updated from subscale

71 RPM

5.8 lb-in stall torque (1.5 min required)

Power supply of 6.0V

Similar decoupling capacitors and flyback diodes used

Trace widths increased as well

Cornell Rocketry TeamCritical Design Review

WIRELESS COMMUNICATION

Requirements:

Must be able to transmit at least one mile.

Must be able to accurately transmit data at distance

Modules:

Both Ground and LV use Xbee Pro 900

Ground module uses elevated Yagi Antenna

LV module uses a half dipole antenna

Cornell Rocketry TeamCritical Design Review

WIRELESS COMMUNICATIONS LOGIC

Uses two step authentication:

▪ Initial packet

▪ Trigger Packet

Cornell Rocketry TeamCritical Design Review

CONTROLS (LSM)

LSM is controlled using ATMega328p

Operates at 5.0V and 16MHz

FTDI chip used for fast programming and debugging

RFM69 Wireless Transceiver used for communication between LSM and rover Controls modules

Level shifting is required for communication between the ATMega and the transceiver

Integrates with LSM motor controller and Wireless Communication system

Will run LSM for predetermined amount of time to ensure that the rover is out of the LV, then transmits a “go” packet to the rover Controls module

Cornell Rocketry TeamCritical Design Review

CONTROLS (LSM)

Cornell Rocketry TeamCritical Design Review

CONTROLS (ROVER)

Rover uses ATMega328p as onboard computer

Operates at 5.0V and 16MHz

Includes RFM69 Transceiver

To receive signal for rover activation

Same level shifter setup as LSM Controls module

Uses same FTDI programming interfaces as LSM Controls module

Has interfaces with PWM pins for communicating with the rover’s servos

Has I2C interface to receive distance information from time of flight distance sensors

Upon receiving “go” signal from LSM, will begin navigation algorithm and attempt to travel 5 ft from the LV

Cornell Rocketry TeamCritical Design Review

CONTROLS (ROVER)

Cornell Rocketry TeamCritical Design Review

OBJECT AVOIDANCE

Two Time of Flight Distance Sensors

Mounted to front of chassis

I2C Communication

Rover remains idle until signal is received from LSM

System runs through state machine until timer ends

Cornell Rocketry TeamCritical Design Review

SOLAR PANELS

Must be able to expand in area size

Foldable model using 2 panels

Cannot exceed output of 5.0V

Required to protect microcontroller

Open Circuit Voltage of 4.9V

Voltage information shown on 7-pin display

Must be able to fit within chassis area (2.3” x 1.9”)

Cornell Rocketry TeamCritical Design Review

TESTING

DRS Rover Sensor Test

DRS Rover Ground Test

DRS Wheel Motor Torque Test

Accuracy of Wireless Communications Test

Wireless Communications Range Test

Cornell Rocketry TeamCritical Design Review

INDEPENDENT TEST AND VALIDATION (INTEV)

Cornell Rocketry TeamCritical Design Review

INTEV MISSION REQUIREMENTS

All components on board the LV shall be capable of completing the mission

Appropriate testing procedures shall be developed to produce valuable data through reliable and repeatable testing of components

Long-term, general-use testing devices shall be built to validate models and predictions

Cornell Rocketry TeamCritical Design Review

PLANNED TESTING

General Testing

Centrifuge: Verifies onboard components can withstand up to and including 8 g’s of constant acceleration (launch acceleration load expected from motor curve)

Shock Test Rig (STR): Verifies onboard components can withstand up to 75 g’s of impulse acceleration

Subsystem Specific Testing

Motor Test Rig (MTR): Characterizes actual motor torque/RPM, allows verification of expected specifications

Parachute Test Rig (PTR): Characterizes parachute drag force in various recovery configurations, allows verification of Launch Vehicle recovery system

Cornell Rocketry TeamCritical Design Review

REQUIREMENTS VERIFICATION – COMPLETED, SEPTEMBER

Section 2 (Launch Vehicle)

The Launch Vehicle will be designed to be recoverable and reusable. Reusable is defined as being able to launch again on the same day without repairs or modifications

The Launch Vehicle will have a maximum of four independent sections

The Launch Vehicle will be limited to a single stage

Section 4 (Payload)

Each team will choose one design experiment option from the following list

Section 5 (Safety)

Each team must identify a student safety officer who will be responsible for all items in section 5.3

During test flights, teams will abide by the rules and guidance of the local rocketry club’s RSO (ongoing)

Teams will abide by all rules set forth by the FAA (ongoing)

Cornell Rocketry TeamCritical Design Review

REQUIREMENTS VERIFICATION – COMPLETED, OCTOBER

Section 2 (Launch Vehicle)

All teams will successfully launch and recover a subscale model of their rocket prior to CDR

The subscale model should resemble and perform as similarly as possible to the full-scale model, however, the full-scale will not be used as the subscale model

The subscale model will carry an altimeter capable of reporting the model’s apogee altitude

The Launch Vehicle will not utilize forward canards

The Launch Vehicle will not utilize forward firing motors

The Launch Vehicle will not utilize motors that expel titanium sponges (Sparky, Skidmark, MetalStorm, etc.)

The Launch Vehicle will not utilize hybrid motors

The Launch Vehicle will not utilize a cluster of motors

The Launch Vehicle will not utilize friction fitting for motors

Cornell Rocketry TeamCritical Design Review

REQUIREMENTS VERIFICATION – COMPLETED, NOVEMBER

Section 3 (Recovery System)

The recovery system electrical circuits will be completely independent of any payload electrical circuits.

All recovery electronics will be powered by commercially available batteries

The recovery system will contain redundant, commercially available altimeters

Removable shear pins will be used for both the main parachute compartment and the drogue parachute compartment

An electronic tracking device will be installed in the Launch Vehicle and will transmit the position of the tethered vehicle or any independent section to a ground receiver

Any rocket section, or payload component, which lands untethered to the Launch Vehicle, will also carry an active electronic tracking device

Section 4 (Payload)

Teams will design a custom rover that will deploy from the internal structure of the Launch Vehicle

Cornell Rocketry TeamCritical Design Review

REQUIREMENTS VERIFICATION – JANUARY

Section 2 (Launch Vehicle)

Each altimeter will be armed by a dedicated arming switch that is accessible from the exterior of the rocket airframe when the rocket is in the launch configuration on the launch pad

Each altimeter will have a dedicated power supply

Each arming switch will be capable of being locked in the ON position for launch (i.e. cannot be disarmed due to flight forces)

The Launch Vehicle will be capable of being launched by a standard 12-volt direct current firing system. The firing system will be provided by the NASA-designated Range Services Provider

The Launch Vehicle will require no external circuitry or special ground support equipment to initiate launch (other than what is provided by Range Services)

The Launch Vehicle will use a commercially available solid motor propulsion system using ammonium perchlorate composite propellant (APCP) which is approved and certified by the National Association of Rocketry (NAR), Tripoli Rocketry Association (TRA), and/or the Canadian Association of Rocketry (CAR)

Cornell Rocketry TeamCritical Design Review

REQUIREMENTS VERIFICATION – JANUARY

Section 2 (Launch Vehicle, continued)

The total impulse provided by a College and/or University Launch Vehicle will not exceed 5,120 Newton-seconds (L-class)

Any structural protuberance on the rocket will be located aft of the burnout center of gravity

Section 3 (Recovery System)

The recovery system altimeters will be physically located in a separate compartment within the vehicle from any other radio frequency transmitting device and/or magnetic wave producing device

The recovery system electronics will be shielded from all onboard transmitting devices, to avoid inadvertent excitation of the recovery system electronics

The recovery system electronics will be shielded from all onboard devices which may generate magnetic waves (such as generators, solenoid valves, and Tesla coils) to avoid inadvertent excitation of the recovery system

The recovery system electronics will be shielded from any other onboard devices which may adversely affect the proper operation of the recovery system electronics

Cornell Rocketry TeamCritical Design Review

REQUIREMENTS VERIFICATION – FEBRUARY (FULL SCALE)

Section 2 (Launch Vehicle)

The vehicle will deliver the payload to an apogee altitude of 5,280 ft above ground level (AGL)

The vehicle will carry one commercially available, barometric altimeter for recording the official altitude used in determining the altitude award winner

The Launch Vehicle will be capable of being prepared for flight at the launch site within three hours of the time the Federal Aviation Administration flight waiver opens

The Launch Vehicle will be capable of remaining in launch-ready configuration at the pad for a minimum of 1 hour without losing the functionality of any critical on-board components

The Launch Vehicle will have a minimum static stability margin of 2.0 at the point of rail exit. Rail exit is defined at the point where the forward rail button loses contact with the rail

The Launch Vehicle will accelerate to a minimum velocity of 52 fps at rail exit

All teams will successfully launch and recover their full-scale rocket prior to FRR in its final flight configuration. The rocket flown at FRR must be the same rocket to be flown on launch day

Cornell Rocketry TeamCritical Design Review

REQUIREMENTS VERIFICATION – FEBRUARY (FULL SCALE)

Section 2 (Launch Vehicle, Continued) The vehicle and recovery system will have functioned as designed

If the payload changes the external surfaces of the rocket (such as with camera housings or external probes) or manages the total energy of the vehicle, those systems will be active during the full-scale demonstration flight

The full-scale motor does not have to be flown during the full-scale test flight. However, it is recommended that the full-scale motor be used to demonstrate full flight readiness and altitude verification. If the full-scale motor is not flown during the full-scale flight, it is desired that the motor simulates, as closely as possible, the predicted maximum velocity and maximum acceleration of the launch day flight

The vehicle must be flown in its fully ballasted configuration during the full-scale test flight. Fully ballasted refers to the same amount of ballast that will be flown during the launch day flight. Additional ballast may not be added without a re-flight of the full-scale Launch Vehicle

After successfully completing the full-scale demonstration flight, the Launch Vehicle or any of its components will not be modified without the concurrence of the NASA Range Safety Officer (RSO)

Full scale flights must be completed by the start of FRRs (March 6th, 2018)

The Launch Vehicle will not exceed Mach 1 at any point during flight

Vehicle ballast will not exceed 10% of the total weight of the rocket

Cornell Rocketry TeamCritical Design Review

REQUIREMENTS VERIFICATION – FEBRUARY (FULL SCALE)

Section 3 (Recovery System)

The Launch Vehicle will stage the deployment of its recovery devices, where a drogue parachute is deployed at apogee and a main parachute is deployed at a lower altitude

Each team must perform a successful ground ejection test for both the drogue and main parachutes. This must be done prior to the initial subscale and full-scale launches

At landing, each independent sections of the Launch Vehicle will have a maximum kinetic energy of 75 ft-lbf

Recovery area will be limited to a 2500 ft radius from the launch pads

The recovery system electronics will not be adversely affected by any other on-board electronic devices during flight (from launch until landing)

Section 4 (Payload)

At landing, the team will remotely activate a trigger to deploy the rover from the rocket

After deployment, the rover will autonomously move at least 5 ft. (in any direction) from the Launch Vehicle

Once the rover has reached its final destination, it will deploy a set of foldable solar cell panels

Cornell Rocketry TeamCritical Design Review

PROJECT PLAN

Cornell Rocketry TeamCritical Design Review

EDUCATIONAL ENGAGEMENT

Completed Events Event Date Grades 5-9 All Grades

Ithaca Math Circle #1 10/14/2017 2 15

Carl Sagan Walk 10/14/2017 11 17

Spooky Science Day 10/29/2017 72 80

Ithaca Math Circle #2 11/05/2017 4 17

Ithaca Math Circle #3 11/17/2017 1 10

Ithaca Math Circle #4 12/10/2017 0 12

Total 90 151

Cornell Rocketry TeamCritical Design Review

BUDGET - EXPENSES

Subteam Projected Expenses

Airframe $3,375.35

Communications (Comms) $903.90

Deployable Rover System (DRS) $417.22

Electrical & Software (E&S) $1,432.20

Independent Test and Validation (INTEV) $1,690.44

Business $12,450.00

Expense Summary

Total Projected Expenses $20,269.11

Total Projected On-Pad Costs $3,693.99

Cornell Rocketry TeamCritical Design Review

BUDGET - INCOME

Source of Funding Contribution

Cornell University Organizations $12,000.00

Corporations $5,250.00

Gifts in Kind $1,345.00

Fundraising $1,849.00

Budget Summary

Income $20,444.00

Expenses $20,269.11

Total Surplus/Deficit +$174.89