Replies to This Discussion

DeveloperPermalink Reply by Doug Weibel on January 2, 2010 at 7:44pm

Quadcopters typically use regular props - except that they use 2 designed for clockwise rotation and 2 for counterclockwise rotation.

Size of props and size of motors depends on the size of the quadcopter. You may find quadcopter airframe info easier at one of the quadcopter threads on rcgroups.

• ▶ Reply

Permalink Reply by James Pistorino on January 3, 2010 at 10:07am

Thanks. I will check it out.

In order to avoid re-inventing the wheel, are you aware of any automatic parachute systems for model planes/helis, etc.? I assume that these exist but I haven't been able to find one. Something that deploys pneumatically is what I am looking for.

Thanks again.
James

• ▶ Reply

Permalink Reply by Martin Seven on January 3, 2010 at 12:05pm

It's easy to reinvent the wheel :-)

You can buy a rocket parachute (or make one yourself), place it in paper cup inside of another paper cup with a gram of gunpowder around an electric igniter beneath it and connect that to a pin on Arduino through a power transistor. It'll weigh around 50 g, cost around $1 (minus the parachute) and will save you hundreds one day.

• ▶ Reply

ModeratorPermalink Reply by Gary Mortimer on January 3, 2010 at 10:22am

Far be it from me to turn you away from Arduino, but this Quad http://www.rcgroups.com/forums/showthread.php?t=1158234 seems to be very promising!!

• ▶ Reply

Permalink Reply by Martin Seven on January 3, 2010 at 11:42am

1) By far the most popular props are EPP1045. They are cheap, available everywhere, and they already come in matched pusher/puller variants. A quad will need two pushers and two pullers. They are however easy to break (but then again they're cheap to replace too) and usually need balancing before use (otherwise they vibrate a little).
A better but more expensive propeller is the carbon-fiber Graupner (left, right). Stronger, lighter, better machining.
An extreme would be the German Microdrone prop - huge, light, strong and costs about $150 a piece :-)

2) The bigger the prop, the more thrust it produces at a given speed. The lower the (rotational) speed of the prop, the less noise and vibrations you get and, depending on the motor, the better mileage you get. The pitch is also important - for a quad you want to use a low-pitch prop (called "slow fly" in stores). The lower the pitch, the finer the control you get per change in RPM. Prop sizes are in the name: EPP1045 means 10" prop with 4.5" pitch.

3) You want a motor that's light (because you don't wanna lug a pound of cast iron around). Its kV number (RPMs per volt) should be low, around 900, because the ESC's voltage control granularity is fixed and therefore the lower the kV, the finer control over the RPMs you get. Also lower kV means more torque, which is needed to drive large props. This motor is a good winner.

The ESC is chosen with three factors: ability to deliver enough power to the motors (watch for lipo cell counts/voltage and current), its control interface (Arduino can do both PWM and I2C; PWM ESCs are everywhere, I2C ESCs are pricy) and its update rate (the faster you can update the RPMs, the better you can control your rig). This one is a reasonable start.

And by the way, (at about 150 bucks for the above selection) the motors/ESCs are by far the cheapest parts :-) you'll kill the most of your cash in radio and inertial measurement gear (although this might save you a buck or two).

(btw, the prop/motor/ESC selection is based on a sub-1kg 640mm quad)

• ▶ Reply

Permalink Reply by James Pistorino on January 4, 2010 at 6:07pm

Thanks for this great information.

Exactly what I was looking for and I already ordered some of the Graupner props.

At this point, I just want to get a working proof of concept and then work on an actual prototype.

Thanks again,
James

• ▶ Reply

Permalink Reply by James Pistorino on January 4, 2010 at 11:44pm

Yet another question. Four of the Turnigy motors and ESCs you recommended come out to more than 1.1lbs. Is each motor is driven by its own LiPo battery or is one used for all? I see the 1300mAh 4S 30C batteries (which seems to be a common size), weigh ~200 grams. Thus, 4 would add another 800 grams which is about 1.75lbs. Thus, the motors, ESCs, and batteries would come to ~3lbs. Is that right?

In terms of a parachute, I was thinking that rather than crashing all the time, if you had an automatic parachute that was pneumatically deployed you could connect it to an accelerometer and deploy it on command or when it detects it is in too great a fall, etc.? I am wondering if anyone has made something like this already.

Thanks again for your time.
James

• ▶ Reply

Permalink Reply by Martin Seven on January 5, 2010 at 1:36am

One battery for all. You need to make sure the battery can handle the load, however:

These motors have a recommended battery of 3s1p, that is 11.1 V. A 2200 mAh battery with 30C discharge rate can safely output 2200*30 mA = 66 A of current. Each of these motors has a max. constant current of 14 A, so 4 of them will eat 4*14 A = 56 A. Therefore this bat can easily handle the load. Total weight somewhere around 1.12 lbs.

The 3s1p code means "three cells in series, one parallel block". More cells in series = higher voltage (and with higher voltage comes greater efficiency and less current flow), more cells in parallel = higher capacity and more amps. So for example a 3s2p pack has two parallel blocks of three cells connected in series.

A 3s 2200 mAh pack weighs about 25% less than a 4s 2200 mAh pack (because it has one cell less), but it needs to output more amps to do the same work. A 4s pack outputs less amperage, but at the same time more work is needed because of the larger weight. I'm wondering if there's a benchmark or a graph that shows what's better (flight time-wise) - higher voltage, or lighter packs.

I'm planning to build a parachute device probably the same as you :-) an independent controller with its own backup power source and a triaxial accelerometer connected through an overvoltage safety to the main flight computer for arming or disarming. If the acc detects free fall for a specific amount of time, it would trigger a pyro with a parachute canister and save the day. It's going to be possible to turn it off, for example when flying low, doing acrobatics or when it would be unsafe to use pyros.

I'm sure government aerospace has already made something like this :-)

• ▶ Reply

Permalink Reply by James Pistorino on January 5, 2010 at 10:43am

Thanks again. You are really helping me get going with this thing.

I know it sounds odd but I am only planning on powering my quad for around 5 minutes so it sounds like many of the batteries will work for me. I will go ahead and order the remaining stuff and get started.

In terms of the parachute, pyro seems like it adds difficulties. I was thinking of using Co2 cartridges because:
1) they are cheap;
2) they are readily available; and
3) they do not present the same safety issues as pyro.

Thanks again.

James

• ▶ Reply

Permalink Reply by Martin Seven on January 5, 2010 at 11:22am

Go for the smallest (=lightest) battery that can provide you with reasonable current (so it wouldn't overheat) and you'll be fine. The real current will be much lower than the maximum specified for the motor, for example a 1.2 kg quad will hover with as little as 12 A total (that's 3 A per motor).

As for the parachute, sure, whatever suits you better. It will be much heavier though. And safety perceived will be a myth, cryoburns are a seriously nasty thing :-) but you're right that it's more readily available with no R&D needed, and therefore better if you want to get off the ground fast.

May I ask what's the intended purpose of the quad? 5 minutes doesn't sound like much...

• ▶ Reply

Permalink Reply by James Pistorino on January 5, 2010 at 12:59pm

On the parachute, if you know of anything that could electrically release the Co2 cartridge under control of an arduino that would be great.

I am trying to make a very stripped down version of a quad that only goes straight up and down in level flight. It goes straight up and down taking video the whole way. That's it. The only operator control I am thinking of is to specify the altitude it goes to, and start and land. Everything else is predetermined or under autonomous control.

The parachute seems like a good idea from a safety perspective if power or communication is lost somehow,

• ▶ Reply

Permalink Reply by Martin Seven on January 5, 2010 at 1:56pm

I'd use a small CO₂ cylinder with some kind of mini butterfly valve on it (a valve with a lever that needs 1/4 of a revolution to fully open), attach a servo directly to the lever (so that the servo's axis goes through the lever axis) and wire it up to the Arduino and use its Servo library to work it. I'd raid a paintball shop to get that. The output would lead to a tube with a paper piston and the chute.

That's pretty much why I advocate a small charge instead of plumbing and servos. It's way too complex and prone to failure. The gas cylinder has enough weight per sq inch of its cross section to punch a hole through a roof tile, the cylinder could lose pressure over time and render the device useless, the servo could strip its gears, the pull rod could snap, the valve could fail, the plumbing could loosen over time and pull away when the valve opens...

Compare it to the essential simplicity of a pyro - you apply current to a wire and that's it, a sub-gram charge instantly changes itself into a liter of gas and pushes the chute out of the tube. A simple rocket motor igniter and 1/4 teaspoon of gunpowder (in every gunshop for about 20 bucks a pound) wrapped in tissue paper work great. Just don't use too much paper, it could create an ember that could damage your chute.

• ▶ Reply