Integrated System Build & Test with Customer Demo
Table of Contents
After expediting our tests during the integrated system build and test phase, our team was back on track to completing the working bicycle power meter by ImagineRIT. Our team had completed and finalized our subsystem designs after testing and were continuing to test the integrated systems. As our team moves closer to the project deadline we are feeling more and more confident that our customer will be excited about our working bicycle power meter.
Team Vision for System Level Demo with Customer
The team vision for this phase, the integrated system build and test with customer demo phase, is to have our device fully functional and working as expected for our customer. Our team is continuing to run tests with the integrated systems to ensure that they work as expected when they are incorporated into one system as a whole. Our team also plans to perform a fully functional test of the power meter by mounting it to a bicycle and having a rider apply some force to it and verify our power output values. The fully functional test will be compared to the values given by another manufactured bicycle power meter.
Test Results Summary
During the previous phase our MSD team was able to get both the accelerometer to read the RPM as well as the strain gauge to measure strain on the crankarm with sufficient accuracy for our particular application. The goal during this phase was to combine the function of both of these devices into one program code that would collect information using the Bluno Nano microcontroller and send this information, via bluetooth communication, to the smartphone app to process the information and display the power output and calories burned by the rider.
Our team continued to have a large focus on developing the app to function as we expected. With our team putting a large focus into the smartphone app design we were able to make a lot of progress towards the final product. The app is designed to allow for the rider to select calibrate which will calibrate the y-intercept for the slope of the strain gauges. The rider can also select to start a new ride which will cause the app to begin taking input data from the accelerometer and the strain gauges as the rider starts pedaling. Once the strain gauges have been calibrated and the rider selects to start a new ride, the app will then display the rider's power output, the rider's number of calories burned, and the battery level as seen from the incoming data as the rider pedals. The following image shows the app as it has been designed to function as mentioned:
The image on the left shows the app as it can be selected from the home screen of an iPhone. Once the app has been selected from the home screen of the smartphone it will then open up to display the image shown above on the right. The image on the right shows the buttons that can be pressed to calibrate the strain gauges as well as the button to be pressed which will start the new ride. Once the new ride button is pressed the data will start to be acquired from the power meter. As the data is acquired the instant power values will be displayed in the P-Data area below the heading "Instant Power" in red. Simultaneously the values for the amount of calories burned will be displayed in the C-Data area below the heading "Calories Burned" in green. It can also be seen in the image on the right that the smartphone app will also display the battery voltage level. The reason for adding this was due to one of our customer requirements. Other power meters on the market do not have a feature to let the user know when the battery is low which our device incorporates to meet this customer requirement.
Our MSD team initially planned to mount all of the components to the crank arms by screwing them directly into the crankarm. Our team had prepared for this by making sure that our strain gauge amplifier circuit PCB's had holes in the boards so that they could be screwed into the crankarm. However, our team came up with a better way to mount our components to the crankarm.
Our team decided to use velcro to attach the components to the crankarm. Both sides of the velcro strips had an adhesive on the backs of them. One side of the velcro was attached to the crankarm while the other half of the velcro was placed on the backs of the components. Using velcro was a better alrternative for our team because now it allows us to easily remove any components to make any changes to them as necessary and it also gives our team a lot of variability in where we can place our components on the crankarm. The following images show how components were placed on the crankarm using the velcro strips:
The image on the left shows both the bluno microcontroller and one of the strain gauge amplifier PCB velcroed to the non-drive side of the crankset. The image on the right shows all of the components fully mounted onto the crankset. Also the image on the right shows the plug in the center of the spindle with the accelerometer being held in the center of the plug to keep it at the center of rotation.
After all of the components for the bicycle power meter were placed on the crankset the finalized code could be written for the microcontroller. The Bluno microcontroller code was written to acquire data from both the accelerometer as well as from the two sets of strain gauge data. The code for the microcontroller works by first waiting for the smartphone to send a signal to the microcontroller once a button is pressed through the smartphone app. Depending on which button is pressed on the smartphone app and is received by the microcontroller will determine which function the microcontroller will perform.
Once the smartphone is connected through bluetooth to the microcontroller and the calibrate button is pressed, the microcontroller will then acquire the voltage signals coming from the strain gauge amplifier circuit while no force is being applied to the crankarm. The microcontroller then takes the voltage from the strain gauges and sets that voltage to the y-intercept of the slope of our voltage to weight relationship line. Once the bicycle power meter has been calibrated the microcontroller than waits for another signal to be sent from the smartphone.
If the "Start Ride" button is pressed this will then send a signal to the microcontroller to tell it to start acquiring data from both the accelerometer and the strain gauges. The microcontroller acquires data by sampling both the accelerometer data and the strain gauge data in 8 specified regions within one full revolution of the crankset. The microcontroller calculates the torque being applied to the crankarm in each of the eight regions. The microcontroller also measures the time that it takes for the crankset to rotate between each of the eight specified regions to calculate the angular velocity.
After the crankset makes a full revolution the microcontroller then averages all of the eight values of the angular velocity and the torque. Once the average torque and angular velocities have been calculated the power is then calculated by multiplying these two values together and the power value is then sent back to the smartphone and displayed on the app.
As our team is moving closer to the finalized design and working product it was time to create the finalized housing to protect all of the components on the crankset from being damaged. Our team decided to 3D print the housing that would cover all of the components. The housing was 3D printed to allow for two halves that will be placed on both sides of the crankarm and screwed together. Separate housing designs were created for both the crankarm on the drive side and the crankarm on the non-drive side of the bicycle. The housing on the non-drive side of the bicycle was designed to be longer due to the crankarm containing both the strain gauge amplifier PCB as well as the bluno microcontroller board. The housing on the drive side of the bicycle was designed to be shorted due to the fact that it only needs to cover the strain gauge amplifier PCB on that crankarm.
The cutout of the 3D printed housing section on the left of the image is a cutout to place a switch for the battery.
Risk and Problem Tracking
Continuing on from the previous phase our MSD team decided to review our risk items for the bicycle power meter. As our team comes closer to the completion of this project and moves closer to a working product for our customer, all of the risk items were revisited to see if the likelihood or severity of any of these risk items had changed. The following image shows the updated risk assessment:
Most of the risk items shown above have decreased in likelihood as our device nears completion because our power meter is very close to functioning as it is expected to during ImagineRIT. Some of the notable risk items that have changed are item numbers 1, 2, 3, and 4. The likelihood of item number 1, the power meter not being able to communicate with the smartphone, occurring was reduced from a 2 down to a 1 because our team is very confident that they will communicate because this communication is how our team has been conducting our testing. Also the likelihood of item number 2, the power source being unable to last for 10 hours, has been increased from a 2 to a 3 due to the battery issues that our team was facing however this item is not very severe to our project since the batteries can very easily be changed.
Our team also decided to revisit the problem tracking document for this project. Our team updated the problem tracking document as shown below:
No new problems had been identified from last phase into this phase. However the battery issue that was noted during last phase had been resolved during this phase and is shown as item number 3.
Functional Demo Materials
The following link will show the MSD Technical Paper.
Plans for next phase
Plans for next phase for our MSD team are as follows:
- Finalize all schematics and documents
- Finish and finalize MSD technical paper
- Resolve any outstanding issues with the bicycle power meter before handing it off to our customer