End of Quarter 2
As the second quarter comes to a close, we’ll be taking a two week break from working on the E-Hawk plane for winter break. So, we thought this would be a great opportunity to give a brief update of the progress that we’ve made since the end of the first quarter. Happy Holidays from the E-Hawk Team!
Mariko
Through the second quarter, I continued to work on covering the plane structures with Oratex fabric. I had finished the elevators during the first quarter, so I started with the horizontal stabilizers. The curvature in the horizontal stabilizers made it challenging to achieve a smooth application, but I was able to remove most of the wrinkles by applying more heat and pressure.
Next, I covered the rudder. Due to its odd shape, I had to cover this piece using two separate sheets of Oratex, rather than doing a wrap-around as I had done with the previous structures.
Finally, I was able to finish covering the two ailerons. The length of these structures posed a challenge, but overall the edges were all straight and simple so they were fairly easy to cover.
Next quarter, I will begin by covering the vertical stabilizer. Since the vertical stabilizer is not a detachable structure and it is quite complicated, I am expecting it to take quite a bite more time than the previous structures.
Dylan
During the second quarter I put some finishing touches on the airspeed indicator and moved on to work on the altimeter and vertical speed indicator
After getting values to display onto TFT with the airspeed sensor, I moved onto to exploring potential ambient pressure sensors for measuring altitude. After some research, I discovered a library for an ambient pressure sensor we already had, and MPL3114A2. Using this library, I didn’t have to manually convert the binary data into numerical data. I was also excited that this sensor had built in altitude calculation. However, when tested, the readings were extremely off, so I programmed my altitude calculations using the measured pressure from the sensor. Doing research, I found an equation from the national weather service that I used. This equation required I have all my pressure in units of millibars, so I made sure to have a conversion function.
Then, I proceeded to add a rotary encoder. We would use this to set the QNH of our altimeter. During this process I encountered an issue where the program was not detecting all turns in the rotary encoder. To fix this, I added an interrupt function for the rotary encoder.
Next, it was time to attach a TFT to the setup and get data onto the display. I found that this required me to change the microprocessor I was using because the Node MCU did not have enough digital pins for the entire setup. We initially switched and got it working on an Arduino Nano, however after discussion, we decided to use Teensys for all of our microprocessors for uniformity sake.
With the TFT connected and displaying, I began testing our altimeter at locations around our school. I tested the altimeter in both Sullivan and Weinberg Parking Structure. I also tested it at an apartment building so we could test the accuracy at higher elevations. I used elevations measured from Google Earth to compare and got the hourly QNH from the Daniel K Inouye Airport. I discovered that the altimeter had a consistent offset of around +40ft. Once I accounted for the offset in the program, the altimeter consistently got the correct altitude with errors ranging only around 2 or 3 ft.
My next task was much simpler, programming the Vertical Speed Indicator (VSI). To do this, I simply calculated the change in altitude at four second intervals.
In the final weeks of the quarter, I finalized the display layout for our altimeter and VSI. I was also able to add a bar on the display as a visual indicator of relatively how fast your you are ascending or descending.
In the next quarter, I’m hoping to be able to test the accuracy of the VSI and then finalize the display for the horizontal airspeed indicator. I also need to attach these setups to perf boards which will be something personally new to me.
Jonah
After having displayed live motor controller data on a TFT display in the first quarter, we transitioned to polishing the layout and implementing a data logger.
One new feature is a set of red crosses that pass through the data when the can bus is not actively sending data. This makes it clear to the user that the readout is not updating. We will continue to make small tweaks to our layout as we progress, such as including the E-Hawk logo and multiple temperature readings from different locations in the battery pack.
As for a datalogger, the Teensy 4.1 has a built-in microSD slot that easily allows us to store data at regular intervals. The current perfboard set up has a button that allows us to begin and end readings, making it useful for when we do test flights and would like to log multiple runs without any sort of reset. The current code has a function which counts the number of files on the microSD and names the .csv file that it writes to accordingly so that each run is saved to a different .csv file. There is also a visual countdown bar for toggling writing to the microSD. Logged data points include all live readouts and a timestamp.
Data can be analyzed by removing the SD card, but we would prefer a USB connection so that a simple Processing script could make graphs from the logged data on the microSD without the card actually being removed. We will have to factor this into our design of the instrumentation panel.
We have also made a TX/RX connection between the Teensy 4.0 that will be interpreting altimeter and airspeed data and the Teensy 4.1 that receives and logs motor controller data. This will allow for data from the Teensy 4.0 to be included in the data logging.
Lastly, some progress has been made on the instrumentation panel. The shape has been modified to better fit the keel tube and we designed a collar to function as a third attachment point to the keel tube. A 3D-printed shroud with an integrated collar will eventually be attached to the instrumentation panel. Our next considerations in this regard will have to be designing circuitry that fits well within the space behind the panel.
