I mounted the plywood base with its test sting and tripod to the top of the rack of my brother-in-law's old pickup truck, and added another piece of plywood to extend the test surface beyond the front of the rack to avoid turbulence from the windshield. Driving to a test sight with the Blade Flyer in the back of my brother-in-law's old pick-up truck has been counter-productive. The two times he has been off work from the LA County Fire Department and it was not windy, I've had to terminate the test after mounting the Blade Flyer on the sting, a video camera on the tripod and connecting a safety harness, so I can ride on the back of the truck to activate the Blade Flyer and stabilize it after transition (the servo-released roller locked the upper hub to the lower, but the vertical stabilizer was too small to overcome the ring bearing resistance, so I removed the vertical).

I terminated the first test, because I thought inconsistent camber behavior was caused by bright sunlight interfering with the photo detector. It turned out to be a loose receiver power connector I was touching when I covered the photo detector. I replaced the connector.

I terminated the second test because engine thrust was erratic. The internal receiver that controls the engines and roller release was shot, so I replaced it, but that did not solve the problem. I could not isolate the problem to a specific circuit, so I re-soldered all the wires on the transmitter printed circuit board (PCB) and the PCB between the battery, engines and transmitter. When that didn't work, I removed all of the electronics from the hub and installed all the transmitter PCB wires in the PCB holes designed for them instead of soldering them to the face of the pads on the reverse side as the manufacturer did to get it to fit in its box.

That worked until I reinstalled the PCBs, pot, switch, LCD PCB and receiver back in the hub. Maneuvering the transmitter PCB into its mounts sorta fixed the problem. At least I can get full power from the engines.

Next time I'll follow the truck to the test site with the Blade Flyer in my van.

Third time is the charm on Monday, 2010 August 2, around 7 AM PST on 80th Street West, north of Avenue K, Quartz Hill, California with the truck owner and driver, Jim Swift, my brother-in-law. Plywood added to front to extend base well over cab for smooth airflow.

Sony MHS-PM1 HD video camera on tripod. Safety strap and harness on clearance from Lowe's for 75% off. My bike helmet. The sting, instrumentation and safety support described in previous pages.








Field Test 1 - Neglected to un-weight scale when enabling it and re-enabling it so it would display the total weight of the model and pedistal, but that mattered little, because the wind speed and weight meters cannot be seen well in the bright light.
Field Test 2 - Released model support, but blade lift is too little, actuator delay is too great, camber response amplitude is too little, or some combination thereof to affect roll. Gimbal installation was too tight to allow roll gimble to be rotated to advance its actuation to compensate for system delay. Lost two weights from one blade that balance the blade in pitch, so moved one to it from the other blade.
Field Test 3 - Increased camber potentiometer.

Field Test 4 - Stopped configuration is good except retreating blade has too much friction and inertia to flip in conjunction with its engine.
Field Test 5 - All transitions occurred at about 28 mph.

Spy Camera view

Field Test End
Conclusions

• The self-leveling intended by having the blades camber to counter roll failed due to excessive vehicle mass, low blade lift, control input lag or some combination thereof.
• Blade lift-drag was insufficient to overcome blade axle friction.
• Engines pivoted well using bearings and telephone chord untanglers as commutators.
• Even without split flaps on the trailing edges of the vanes, the engine vanes demonstrated that the engines could be made to rotate with the thrust vector down to compensate for retreating blade lift loss.
• $300 is insufficient for a model of the size and sophistication needed to fully demonstrate no loss of lift during transition.