The next step in the project was fabricating the blades. It wasn’t until now that I noticed a problem with my PCB design. Knowing that I would have to somehow determine the angular position and angular velocity of the blades in flight in order to properly time the flashing of the LEDs, I had incorporated a hall effect sensor near the outside edge of the PCB. I had assumed I could mount a permanent magnet somewhere on the body of the helicopter in such a place that it would be detected by the hall sensor once per revolution. The problem with this was that my helicopter’s flybar is located underneath the rotor blades, and the permanent magnet would be struck by the flybar if it were positioned underneath the spot where the PCB swings past.
The solution I came up with was to remotely mount the sensor closer in near the root of the blades. To do so, I cut some “long nuts” out of some 1/4″ aluminum stock. The blade grip bolts each screw into these long nuts, and the hall sensors are super glued to the bottom of the long nuts. This locates the sensor lower and nearer to the fulcrum of the flybar, which allows the magnet to be mounted in a place where it is not in danger of being struck by the flybar.
The picture above shows the sensor mounted on the long nut. Right below it the magnet is mounted to a small piece of acrylic which is bolted to the frame of the helicopter. Both the sensor and the magnet are mounted using CA glue.
The PCBs were mounted to the blades using double-sided foam mounting tape. Wiring up the blades was an incredibly tedious procedure. I needed 1/16″ wide copper tape, and the wiring wouldn’t have been so bad if I had been able to find some of it. But the narrowest stuff I could get my hands on was 1/4″ wide, meaning every strip I made had to be sliced lengthwise four times. I ended up doing this with a razor blade and straight edge. Ultimately, each blade took about 10 hours of work to wire up. It was incredibly satisfying to see all 33 lights lit up simultaneously when a blade was finished. (The 33rd light is one out at the tip that is connected directly to Vcc with a current-limiting resistor. This is meant to always be on, even if something goes wrong with the microcontroller, so that the helicopter never goes completely dark in the middle of a night flight.)
I found a 180mAh 2S LiPo battery pack on HobbyKing for under $2. I only needed one cell to power each blade, so I carefully dissected the pack into its two individual cells. If you do this, use extreme caution. Shorting or otherwise damaging a lithium polymer battery can cause it to violently explode into flames! You’re better off buying two individual cells, it’s much safer that way. Even in that case, you must still be extremely cautious when soldering to the battery. A solder bridge can literally be deadly. Below is a picture of the separated cells with Deans connectors soldered onto them.
You can see in the picture that one of the cells is already attached to a blade. The cells were attached by using double-sided scotch tape between the cell and the blade and then shrinking some 2″ layflat clear PVC heatshrink around the cell and blade.
I almost forgot to mention — the Molex connectors that I took so long to pick out ended up being a very poor choice. The male and female parts snapped together so firmly that everytime I tried to separate them I feared I might tear the female part off of the board. This led me to purchase some Deans micro connectors (the red ones that each have one protruding lead) for the batteries. These connectors are very firm, yet easy to connect and disconnect without excessive force.
You’re probably wondering how much damage I did in terms of weight gains on the blades.
Those are some heavy blades for a little 450-sized heli to be swinging around… Sure hope it flies!