When I was in high school my friends and I built all sorts of flying things,
including hovercraft. We used Cox .049 model airplane gas engines with a 4
bladed prop and ducting to direct the air under the hovercraft. These
motors were powerful enough to lift the hovercraft about an inch, high
enough it to be used outdoors with out a skirt.
The very low friction of the hovercrafts caused them to fly much faster then
kids on bikes and get lost when a second motor was used to propel them.
Instead we had adjustable slots in the back of the duct work that allowed
air to leak out for propulsion. Similar slots in the sides were used for
We never did manage to control the flight of these thing.
----- Original Message -----
Sent: Tuesday, October 31, 2000 11:36 AM
Subject: RE: [SeattleRobotics] Robot hovercraft project...Help!
> Tony Brenke <trbrenke@...> said:
> >if you are making the lift motor your self I suggest a duct fan type.
> >the air propelled/lost ratio is better.
> OK, gang, this is what I originally told them on their first inquiry:
> >Well, if you are determined to do a hovercraft, you should at least
> >find a successful hovercraft model to modify or learn from. Several
> >years ago Toys-R-Us had a simple RC hovercraft that would be a good
> >start, and hobby shops catering to RC aircraft builders might also
> >have something. The Toys-R-Us job was *very* light weight, which
> >should give you a clue about what you're up against.
> >If you can't find a working model hovercraft, I recommend that you go
> >to a framing shop and ask for scraps of foamcore or gator board.
> >These are super-light, easy to cut & glue material, yet quite rigid.
> >Start simple with a 12" disk of foamcore with a hole in the middle a
> >little larger than the size of the propellor, and with a tube made by
> >wrapping a strip of manilla folder to fit the hole. This tube will
> >serve as the duct. You'll also need some sort of material for a
> >skirt. I would just use a skirt of 1-1/2" masking tape around the
> >perimeter of the disk. Mount the motor in the tube using a spider (an
> >"X") made from sticks of balsa. Put a piece of light-weight grill
> >over the intake (plastic mesh, say) to keep debris from getting
> >sucked into the intake.
> >At this stage, you'll want to feed the motor's power leads straight
> >up to about six feet of zip cord to their power source. With this
> >setup you *may* get your little craft to hover. This is not the final
> >version, just a way to learn the bare principles, and as such should
> >be accomplished as *quickly and cheaply as possible*. I can't
> >emphasize that enough. If you can't accomplish at least this much
> >within the next seven days, you seriously need to reconsider what
> >it's going to take.
> >Supposing your first attempt does have enough lift to hover on ground
> >effect, you should next start piling nickels on the disk as test
> >weights. Each nickel weighs about 5 grams. Once you've piled enough
> >nickles on that it still barely hovers, that number of nickels times
> >five is the number of grams payload this thing will have.
> >Unless you start with a very hefty & efficient motor/propellor combo,
> >something from Astroflight, say, cobalt magnet motors with gear
> >reducers (these are in the $150 cost range *each*), you will be
> >disappointed with how few nickels you'll be able to lift. There are
> >cheaper ways than Astroflight, but they require subtlety. And,
> >definitely don't use cheap Mabuchi motors; they are horribly
> >inefficient, so you want the best motors they can afford. Remember: a
> >big, slow-turning propellor is better than a small, fast-turning
> >propellor in this application.
> >You'll also notice that the whole hovercraft wants to start spinning
> >in reaction to the propellor. The second iteration might then use
> >four motors & propellors, two turning clockwise ("pusher" props) &
> >two regular ("tractor") counterclockwise props to eliminate the net
> >spinning torque. You'll need two pusher and two tractor propellors,
> >or it won't work at all.
> >Now, assuming you've gotten this far, you need to think about light
> >weight batteries to replace the power tether. And then *more* motors
> >and propellors for steering, and perhaps micro servos to aim them,
> >and a BASIC stamp or OOPIC or some such AND some high-current
> >transistors, and maybe even a piezo rate gyro to stabilize this whole
> >Anyway, it's quite a job. You can do it, but you need to succeed in
> >tiny little increments starting yesterday.
> > * If you don't have a hovercraft hovering within thirty days, the
> >project will fail.
> > * If you hover but can't control it within 60 days, the project
> >will fail.
> > * If in 90 days you can do all of the above, but the signals from
> >your sensors are still being wiped out by electrical noise from all
> >these brute-force motors (and they will be at first until you learn
> >how to wire & shield against noise), the project will fail.
> >But if you can make steady, weekly progress against all these
> >adversities you might just do something very interesting.
> >I'm willing to answer questions such as you may have, though my
> >answers will only be general & aimed at first principles. Besides a
> >Robotics Engineering degree, I also studied aeronautics &
> >astronautics for a time in college, so I have a few textbooks around
> >here out of which I could dig helpful equations and graphs now and
> >again. But I insist you do real experiments with whatever you can get
> >your hands on, otherwise no amount of theory will help. Paper doodles
> >and wishful thinking will not get this project done. You need to know
> >what you are talking about, and by actually building stuff, taking
> >things in little steps & correcting errors as they crop up, you
> >*will* know.
> >One final word: the types of motors and propellors you'll need will
> >easily be strong enough to slice fingers off. Be safe. Make *sure*
> >that motor mount is strong enough, wear goggles while testing, and,
> >for goodness sake, keep your fingers, clothes, puppy tails, etc. out
> >of the way of the propellors!
> >Good luck. You'll have a ball with this one.
> To unsubscribe from this group, send an email to: