Hacker News Reader

Comment by sschueller | Hacker News Reader

$17k for a drone with 10min flight time? Ok, it can carry up to 20lbs which is quite impressive. [1]

[1] http://store.freeflysystems.com/products/freefly-alta8

By the way, the flight time to lift ratio of LiIon battery powered drones are one of the reasons I know all those stories about Amazon using drones to deliver are absurd. The drones would have power for two or three deliveries at most. LiIon batteries don't have that long of a lifetime and they are expensive. Not gonna happen with today's technology.

Other than simplicity, is there any reason drones couldn't run on jetfuel or other combustible fuel? AFAIK it's been used for RC helicopters and planes for decades?

Sure, if you're an electrical engineer, you might want to stay with an all electric thing (and electric engines are flexible) -- but if one was serious about lifting packages, it would seem you'd want to explore other, well-tested options that give you more lift/flight time?

I mean, if we can deliver hellfires from drones, surely we can deliver some books?

None whatsoever --- there are lots. They work well, although engines are heavy, so they have to be big.

http://mydronelab.com/blog/gas-powered-quadcopter.html

I don't know whether they use direct coupling or motor-generators, though.

Model aircraft engines have a long and time-honoured history. My father was into them as a child in the 1950s (before radio --- you controlled them by swinging them round your head on wire harnesses!), and still has some of the engines. Beautiful, beautiful engineering; about the size of my thumb, which ran on some ghastly cocktail of exotic oils and volatile spirits.

Actually quad chopters don't scale very well. The reason they work so well as tiny drones is they are mechanically simple. No need to tilt the blades, just apply less power and almost instantly you get less lift on that side. However as you scale the drone it's blades need to get larger which causes more momentum which slows the response time. You can try and force the blades to slow down faster but the forces quickly get extreme, or change their tilt to generate less lift.

But, if your going with tilting blades just use a helicopter design which is mechanically simpler than 4 tilting blades.

The swashplate mechanism is more complex than a tilting mechanism, be it 4x tilting mechanisms.

You can get into a lot of different trade-offs.

EX: There are some options that don't involve a swashplate. (ZX-35041) considering it's 20$ it's probably junk. But, notice how the tail blade is pointing up.

You could build a tilting quad copter with 4 engines, but that's significantly more complex. You can also use one engine and get power out to the 4 blades but that also get's complicated. Still the point still stands, Quadcopter's need to get more complicated as you scale them and things like vibration make everything worse.

The example you posted has far more things in common with a multirotor than with a helicopter.

What amazes me at those little things is how little side-to-side drift they have, considering that's all depending on the CG being well centered.

A nice variable pitch quad: https://www.youtube.com/watch?v=Vy5Ky50eGJs

That's a nice find with the variable pitch quad, though it looks like it has serious issues with vibration. Thrust vectoring is another option, as is using 3 blades at 90 degrees from each other. Autogyro's are another design which might work.

However, in the end people have tried a lot of man sized flying designs and for vertical takeoff and hovering helicopters really are about as simple as it get's.

> though it looks like it has serious issues with vibration

Yeap, nailed it, apparently, vibrations are its biggest issue.

Last mile delivery is really expencive, if the drone uses 1$ worth of battery life per delivery that's still reasonable costs. It's getting everything else to work that's the problem.

Consider a golf course that can deliver cold beer out to the green via drones. 8 ice cold beers at ~10lb in a few minutes via cellphone app. A drone could probably do that 5 times on a single charge, and swapping battery's is easy. Charge 2$ extra per beer and it's both rather profitable and for now a unique experience.

Exactly. What is it worth to have a fleet of these deliver all packages in a multi block radius all within a few minutes? You can swap batteries quickly on the base truck (and probably automate how that swap is done) and move on to the next zone.

Saves massively on gas and people time (especially when self driving trucks are mainstream) and in theory are the main way you might deliver packages in a fully autonomous delivery setup.

Fixed wing aircraft are far more efficient, enough so that lipo batteries are pretty practical. Many drone delivery concepts I've seen are a hybrid fixed-wing with vertical take off and landing. Amazon appears to be an exception. The only image I have seen of their drones is a traditional multirotor, and they claim to be aiming at a ten mile range with a 5 pound payload.

DJI Spreading Wings S1000 is $1900, and can carry 15 lbs.

https://www.dji.com/spreading-wings-s1000/feature

It can do 30 minutes with a 2LB payload., or 15 minutes with 17LB http://freeflysystems.com/alta-8/dependable. 20LB means both fewer battery's and more work.

Also, I think the cost is more from how modular the thing is, it's very much a professional product which limits sales.

PS: Looking at lose curves, it should be able to do 1hour with 5lb and significantly more battery's. But you get more total flight time with fewer battery's that your hot swapping.

Even ignoring the inevitable progress of technology, 10min is still plenty of time for a crew with two of these and a stack of spare batteries to have continuous light until the stack runs out.

While one is in the air, the other is getting its battery packs swapped out.

look at the shot of the guys walking on the bridge, I don't see how you could do something like this in conventional cinematography for only 17k, I would expect this drone or something like it to become quite useful in the film industry for night shots where 10 min flight time is definitely not a limiter

Thanks! I missed the point the first time round, giving up watching a few minutes in. Truly beautiful, although I prefer the shifting shadows on the tree shots.

I think the flight time was lower due to the weight of the light and battery array?

Clearly there was some sort of sponsorship here, would be nice to see it explicitly mentioned though.

Reply on news.ycombinator.com