It doesn’t look right to me. It looks like the batteries would need to be stored in the hubs next to the motors. It doesn’t look like there is enough volume in those hubs for a flight of two hours.
I realize the propeller speed can’t be judged from the video since the propellers exhibits a wagon wheel strobing effect. But I’d expect to hear the normal quadcopter sounds from the motors. The sound doesn’t match what I expected.
Near the end if the video a Mission Planner screen is displayed, The displayed flight doesn’t appear to match the flight shown in the video. The Mission Planner display shows a flight with weigh points. The quad in the video was relatively stationary and didn’t fly a weigh point mission. At least that’s my understanding of what was shown.
Skeptical for sure, but they have a 2 hour supposedly unedited version I’ll post below. There’s a comment on the above video with a reply that the whole thing weighs 2.5 lbs without batteries and 7.5 lbs AUW. Their own site boasts about the batteries stored in the pods, which are modular and can be configured as quad, hex, and octo frames.
As a happenstance, I fast forwarded toward the end of the 2 hour video and heard one of the guys state that it started to fail on a previous flight at “11 something volts.” That means they are most likely using 3S batteries, which have energy density enough to support ~50Ah for 5 lbs of a lithium chemistry battery. That means 20A continuous for 2 hours (down to 80% discharge), which seems plausible, but endurance would likely decrease rapidly with any significant payload.
EDIT: Another happenstance find (I’m not watching the whole thing!) At 23 minutes they discuss that it’s 16A continuous on 40Ah batteries, which exactly matches my 80% discharge assumption above. Again, plausible, and I don’t think they are faking this video.
Of note, the Mission Planner screenshot in the shorter video above looks like a still frame and shows disarmed, so that’s fairly misleading. But I’m glad to see ArduPilot in use on such a project, and I wish them success.
Thanks for sharing your thoughts about the project. As I’ve thought about it a bit more, I think you are correct in your assumption they aren’t faking the video.
Their spoke system of attaching the motors to the frame is interesting. I imagine this system saves weight compared with a conventional frame.
Any idea about how their batteries are wired? Does each motor have it’s own battery pack? Are the packs connected together?
It’s an interesting project but I have severe doubts about its practicality.
In a reply to one of the comments (on YouTube), they mention they use gears with the brushless motors.
I don’t recall seeing other quadcopters with geared brushless motors. I know some of the little quads use brushed motors with gears but I don’t recall seeing quads the size of this Modovolo quad with geared motors. I’m sure there are others, I just can’t think of any.
It’s an efficiency tradeoff. You can spin a smaller motor faster and gear it down to drive a bigger/higher pitched prop. And in so doing you’d have to carefully analyze the efficiency gains from the smaller motor (ostensibly operating in its peak efficiency RPM range) vs the losses and mechanical complexity/longevity of a gear train.
There’s also potentially a cost concern, where small, high RPM motors are cheap and plentiful, and larger, slower ones are expensive and possibly harder to source. If a gear train is more cost effective, it may be worth using, depending on its reliability.
Both can be exceedingly tough nuts to crack, and I bet you could write a master’s degree or doctorate thesis on the subject!
It looks very light, the frame is all 1mm carbon rod, I don’t know how they could manufacture something like that, it looks like it would need to be glued one rod at a time like matchsticks.
You can hear the geared motors so they are probably tiny, saving more weight.
Parrot machines, AR.Drone 1 and 2, used the same idea: geared BLDC motors. They also had relatively impressive flight times, but WiFi-based remote control at distances of about 50 meters at best in a stock configuration was kind of not OK.
While I’m still skeptical about how useful this design will be, I’m convinced their two hour flight claim is true. I look forward to seeing what they do with it.
I have been thinking about their design and I believe it flies I doubt how much it could actually carry, with motors that small your thermally limited with how much power you can put through them. So I would be very surprised if they can actually carry anything without overheating the motors.
It currently has about 64% battery-mass-fraction. Simple theory suggests that if the propulsion scheme is 50% efficient, they just need a battery with 196 Wh/kg specific energy, which is about right for modern 21700 cells. I do a lot with T-Motor MN4006 on 15" props, and that setup is right around 50% efficient, a little worse at light loading, a bit better at heavier loading.
What I would be concerned about is separate packs in the 4 pods. In any kind of forward speed, the aft motors will require a lot more power, and flight time is only as good as the 1st pack to go. At 28s into the indoor video, it looks like they might have balance leads that could help out with this - equalizing the packs in real time.
At that same time frame, it looks like maybe we’re looking at the brushless motor that drives the prop, but oddly, it seems like we can look straight through the motor to the prop beneath it, as if there was no stator. Probably I’m looking at something wrong. I suspect the motivation for a geared motor is to capitalize on the insane power-density of racing motors, if you allow them to reach super high RPMs. Power = torque*rotational speed. Torque is generally limited by the mass of copper and magnets. The current trend is to use large radius, thin motors with the coils and magnets at a large radius in order to get enough torque from the motor to be useful. They chose the geared approach instead. Giving up some efficiency to gearing losses, but saving a lot of weight. The gearing certainly adds a lot of noise.
Based on the pack weight (1.25 lb/pod) and advertised capacity (40Ah or 10Ar per pod), I’m Guessing they are using something like Samsung 50S cells in a 4S2P arrangement/pod. They also give a starting battery voltage of 16.8V, so definitely not 3S. They recommend landing at 2.75 v/cell, so 11V. If they switched from the 4 packs to a single pack made with Amprius SA08 packs, they could save a chunk of weight. Their batteries are currently around 2.27kg, but the SA08 single pack would be around 1.84 kg, with slightly higher capacity.
