P16007: Motor-Assisted Wheelchair

Detailed Design

Table of Contents

Team Vision for Detailed Design Phase

During this phase we sought to:

During this phase we managed to:

Budget Increase

Last phase it was determined that the budget was something which definitely needed to be reviewed as it the project did not seem feasible at the set value of $500.

During the first week of the phase (over Thanksgiving break) the following proposal was written and sent to Dr. DeBartolo requesting review of the budgeted amount on Monday 11/30/15.

Budget Review

On Tuesday 12/1/15 a meeting was held and the budget reviewed along with supporting feasibility of the design described in the remaining parts. A budget increase was approved and the final bill of materials requested to be sent for review when all outstanding parts were looked at.

Mechanical System Design

Motor/Gear Box

At the moment, the feasibility for our decided motor is ultimately dictated by the fact that similar products such as the Smartdrive MX 2 are using similar rating equipment (a 250 Watt motor at 36 Volts) at a lighter weight point and achieving their goal. As such, we are confident that choosing a motor within the same ballpark will at least achieve the baseline goal, if not meet all our demands.

The motor selected needs to be able to provide enough torque and speed. We have looked at HUB motors and regular DC brushed motors. The HUB motors were rejected by the team because it would be difficult to find or make a smaller wheel for them. HUB motors generally built for bicycle wheels, which are a lot larger than the wheel size we want to use. Also generally they were made and sold from overseas, which means the shipping and the vendors would be unreliable.

Other DC brushed motors that we have looked at were generally too weak in terms of RPM or torque for our purpose. The motors we found with appropriate specifications from a company called Ampflow weren't meant for continuous applications. Their technical support said that they have a custom motor that can meet our specifications, expect that it would cost $459 for just the motor.

After analyzing multiple motors, the model we chose is Teknic CPM-MCVC-3432D-RLN. It is a brushless motor, so the lifespan is better than DC brushed motors. The continuous torque for the motor is 120.9 oz*in, and by using a 7:1 gearbox, the amount of torque will be increased to 846.3 oz*in, which is more than the calculated torque needed to keep the wheelchair from rolling down an incline of 6 degrees. The maximum RPM of the motor is also reduced from 1200 to 171.4 due to the gearbox. A wheel of 8 inch diameter rotating at 171.4 RPM has about linear velocity of 4.1 mph.

Energy Analysis/Torque Requirements

The decision regarding the gearbox has not been an easy one. The project requires a system that is simple and inexpensive but rugged, durable, quiet, and efficient. Finding the balance between these factors is an ongoing process. Currently we have gotten solid quotes from Designatronics Inc and CGI. Both of these companies quotes significantly exceed the amount originally budgeted for the gearbox.


The prints shown below are in order that the parts will be constructed in to mitigate the risk of parts not fitting in final assembly.

Toplink Access Part 1

Toplink Access Part 2

Toplink Access Part 3

Toplink Access Assembly

Motor Access Plate 1

Motor Access Plate 2

Gearbox Bracket

Battery Box Support

Welded Assembly


Right View


Left View

Short CAD Demonstration

There will be a 12-pin female connector and a 12-pin male connector that separates the wires between the permanent user control attachment from the detachable system (motor, battery, and the wheel).

To show how connectors look like, here is the 12-pin Female Connector:



Regarding the omniwheel, only one major option really presented itself, the 8” version provided by Vex Robotics. Vex Robotics parts are typically used in applications ranging up to 100-120 pounds of total weight. Considering that our system will only be bearing ~30 pounds, we believe it to be structurally stable. This also comes as relatively easy to buy and replace should the part wear out in the lifetime of use. At 8 inches in diameter, to achieve a speed of 5 miles per hour, we would need approximately 210 rpm. At 6 inches, this number jumps to 280 rpm. As such, we believe that 8 inches is the smaller end of functional diameters that we could allow in design.

public/Subsystem Design Documents/omniwheel.jpg

Included with the omniwheel would be the purchasing of a proprietary Versa hub also supplied by Vex Robotics. This would allow the wheel to attach to the output shaft of the motor gearbox assembly.

public/Subsystem Design Documents/versa hub.jpg

Battery Casing & Cable Glands

Battery Cases

The battery choices were changed, so must the battery case. Similar idea to finding a battery case in Subsystem Design, the battery case has to be reasonably light weight but also satisfy the required volume and strength to house 2 battery packs with total weight of 10 lbs. The case is also expected to be durable to withstand constant shearing that occurs from wheelchair movement.

public/Detailed Design Documents/Battery Casing.jpg

Image from Amazon.com

Battery Case specifications:

1. Product name: 290mm x 210mm x 100mm Waterproof Plastic Enclosure Case DIY Junction Box

2. Product manufacturer: Amico

3. Product vendor: Amazon

4. Cost: $24 as of presented time (subjected to change because Amazon fluctuates price)

5. External dimensions: 290 x 210 x 100 mm / 11.4" x 8.3" x 3.9"(L*W*H)

6. Internal dimensions: 280 x 200 x 93 mm / 11" x 7.9" x 3.7"(L*W*H)

7. Thickness: 5 mm

8. Weight: 756 g / 1.67 lb

9. Material: Plastic

Certain modification can be done according to requirements; such as drilling, painting, punching, silk-screen printing, and etc.

Cable Glands

To install wires from the battery cases inside out safely, cable glands are needed. The case will be be drilled and threaded to fit the cable glands size. 5 Input-Output pairs of wires are requested, so 5 cable glands will be needed.

public/Detailed Design Documents/Cable Gland.jpg

Image from Amazon.com

Cable Gland specifications:

1. Product name: 5 Pcs Waterproof PG11 Black Plastic Cable Glands Joints

2. Product manufacturer: uxcell

3. Product vendor: Amazon

4. Cost: $4.05 as of presented time (subjected to change because Amazon fluctuates price)

5. Used for cable radius range: 5-10 mm /0.19-0.39"

6. Thread radius: 17.6 mm / 0.69"

7. Total Size : 23.5 x 37 mm / 0.92" x 1.45" (Max. D*H)

8. Color: Black

9. Weight: 32 g / 0.07 lb for 5 pcs

10. Material: Plastic

Fully Installed Battery Case

public/Detailed Design Documents/Installed Battery Case.jpg

Image from directindustry.com

public/Detailed Design Documents/Cable Gland Animation.jpg

Image from Lapp Tannehill on youtube.com

Electrical System Design


Things looking for when selecting the battery:

1. Cost: Minimum as possible

2. Voltage: 24V or 36V

3. Max Discharge: 4 Amps or greater

4. Lifetime for our application: 2 hours or more

5. Lightweight: 15 pounds or less

6. Rechargeable

We compared the option of doing Lead Acid Batteries compared to the Rechargeable Battery we think is best to see what the weight difference would be. The cost for the lead acid batteries would be $95.98 and the Nickel Hydride Batteries would cost $359.90. We wanted our design to have a weight limit less than 30 lbs. Below we compared the two with the heaviest pieces that would at the back of the wheel chair.

public/Detailed%20Design%20Documents/24V NiMH Battery Combo Pack.jpg public/Detailed%20Design%20Documents/Lead Acid Battery Option.png

public/Detailed%20Design%20Documents/Scenario 1 Battery Comparison.jpg public/Detailed%20Design%20Documents/Scenario 2 Battery Comparison.jpg



From comparing the two we decided we will use two 12V Nickel Hydride Batteries in Series.

The batteries max discharge is 13A well above what we want. It will cost a total of $359.90 for the two batteries and the battery charger plus $18.50 for shipping. The total weight will be 11 lbs. Lastly it will run for 2 hours of lifetime.


The Teknic motor has a built-in motor controller, so we can connect the power source, or two 12V battery packs in our case, directly to the motor. Teknic provides a software for doing initial setups for the motor.

Here is a simple wiring diagram from Teknic's website.


The motor has 6 inputs, besides the ones for power, and 2 outputs. The 6 inputs are divided into 3 sets: Input A, Input B, and Enable. The software allows us to tell the motor how to use the inputs coming from A and B. The Enable is for determining whether the motor should be enabled or not, and it will be connected directly to the battery packs because the motor reads any signal from 5 to 24V as logic high. When the Enable signal becomes low, it will take the motor 10mS to read the disable signal. Then the motor will The configuration chosen is to have A determining the direction, and B determining the speed based on the duty cycle of a PWM signal that is generated by a PWM controller.

The PWM controller will be powered directly from the battery packs, and it can output PWM signals with magnitude the same as the voltage supplied to the device. The PWM controller has a knob/dim switch that is a potentiometer, and by turning the knob, the duty cycle changes, which will affect the RPM of the motor. The frequency of the PWM signal is 12kHz, and it is in the acceptable range of 20Hz and 30kHz by the motor. Here is a timing diagram from the motor's user manual.


PWM controller:

public/Detailed%20Design%20Documents/PWM Controller_2.jpg

public/Detailed%20Design%20Documents/PWM Controller_3.jpg

The two output signals are HLFB+ and HLFB-. HLFB draws about 9mA of current from a source when the motor is running. That source is external, and we decided to have it coming out from the battery packs. HLFB- is connected to ground. A pull-up resistor is attached to HLFB+ in series and a buzzer is attached to HLFB+ and HLFB- in parallel. When the motor enters an emergency shutdown, the HLFB stops drawing current, causing the buzzer to go off. The pull-up resistor will be 2.4kOhms, so there will be very little voltage that is below the buzzer's operational voltage going into the buzzer. Here is the waveform from the motor user manual.


When the motor is operating at peak torque 403.on-in (2.85 N-m) and max speed of 1200 RPM, the max power drawn will be 360W, using the follow equation:

Power (kW) = Torque (N.m) x Speed (RPM) / 9.5488

Since the supplied voltage is 24V, the max current will be 360/24= 15A. We are planning to use 14 AWG wire for the power supply, so the peak torque will be limited to 2.7 N-m, which changes the max current to 14.17A.

The power going into the entire system will be controlled by two switches. The first one is an emergency switch, and the second one is an on/off rocker switch. The emergency switch has DC current rating of 20A and DC voltage rating of 14V, so it will be place in between and in series with the two 12V battery packs. The rocker switch has DC current rating of 20A and DC voltage rating of 24V. Therefore, this switch is placed in series and in between the batteries to the rest of the system. This rocker switch will have a protective cover to prevent accidental bumping.

Emergency Switch:



On/Off Switch:



Switch Cover:



Feedback to the User

In order for our customer to know what her battery level is at this Battery Monitor will be installed. It costs $12.95 and works for 24V applications. We plan to place it on one the arm rests of the wheelchair.

public/Detailed%20Design%20Documents/24V Battery Monitor.png


In case of the motor not working when it is on, there will be a buzzer attached that will alert the user. It is rated to work at 60 to 80db which is equivalent to background music to a food blender. The graph below shows how the voltage will change the sound level for this buzzer.

public/Detailed%20Design%20Documents/Buzzer.jpg public/Detailed%20Design%20Documents/Sound Level vs Voltage Graph.jpg


Wiring Diagram


Updated Bill of Material (BOM)


Test Plans


Battery Test Plans


Motor/PWM/Switches/Buzzer Test Plans

Engineering Requirements

Engineering Requirement Test Plans

Design Flowcharts


Risk Assessment


Design Review Materials

Home | Planning & Execution | Imagine RIT

Problem Definition | Systems Design | Subsystem Design | Preliminary Detailed Design | Detailed Design

Build & Test Prep | Subsystem Build & Test | Integrated System Build & Test | Integrated System Build & Test with Customer Demo | Customer Handoff & Final Project Documentation