mikrokopter

Owen

Has anyone on this forum built a Mikrokopter
http://www.mikrokopter.com/ucwiki/
sorry it's in German but google does an ok job translating. I'm working on a design for my own mikrokopter but with a propeller chip instead of the atmega. I figured i would try to port the software to spin. I'm thinking I would build a 2" x2" propeller main board and a second board with sensors like gyro's, accelerometers, compass, altimeter, and GPS (maybe GPS would have it's own Board). I figured I would design a board and see if there was any other interest out there in the propeller community.

parts-man73

I want one!!!!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Brian

uController.com - home of SpinStudio

PropNIC - Add ethernet ability to your Propeller!

SD card Adapter

Gadgetman

Looks like a stripped-down version of an Silvelit X-UFO.
(I have a project where I'm going to make one of those fly autonomously using a Propeller. Unfortunately, it has been on the backburner for a while...)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Don't visit my new website...

Ale

They are

OzStamp

Hi

check this out.. they have these copters + some really super high speed DC motors as well..

http://www.gws.com.tw/english/english.htm

cheers Ron Mel OZ

stamped

Ron,

Looks like a nice piece of kit. I have thought about using a Prop with my cheap electric helicopter, but I doubt it could carry the weight unless I went the whole hog and fabricated up a board. I would be interested to see what you come up with.

simonl

Glad to see I'm not the only one who's thinking of doing a quadcopter with the propeller Maybe we should start a SIG and pool our resources?

I see that mikrokopter decided to design their own speed controller, as they say it needs a 100Hz update rate. Anyone know how to control a brushless motor from a PChip? (It's similar to a stepper motor, but with feedback, isn't it?)

BTW: Loads of quadrcopter info' at: wiki.xufo.net/wiki/Main_Page

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheers,

Simon

---

www.norfolkhelicopterclub.co.uk
You'll always have as many take-offs as landings, the trick is to be sure you can take-off again ;-)
BTW: I type as I'm thinking, so please don't take any offense at my writing style

Post Edited (simonl) : 1/4/2008 11:15:23 AM GMT

Gadgetman

The update rate shouldn't be a problem, at least...

I'll be using the '5 Axis of Freedom' Gyro/Accel from Sparkfun, which can run off 3.3V, weighs 2grams and is read with ADC. From what I've calculated, a single COG should be able to run ADC on 16 different sensors 100times/second and still have plenty of processing time to spare.(Assuming 8 bit accuracy. didn't look into anything else as the technology - resistors and caps - isn't that accurate)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Don't visit my new website...

OzStamp

HI Stamped.

What do you mean ?

Ron Mel oz..

Owen

I figured on using the same design for the motor controllers as the mikrokopter. the controllers have a ATmega 8bit chip that handles the motor and and an I2C connection to the propeller this greatly simplifies things from the propellar side. you can buy the raw pcb boards for the motor controllers directly from the mikrokopter page.

Brian L

I think simonl is right. We should start a SIG on this. Or better yet, a SIG to work an stand-alone autopilot / navigation system that could be added to any electric quad copter of decent size. There are advantages to separating the control system from the flying machine itself.

There are lots of really nice brushless motor controllers available for good prices, and sending control inputs to those controllers is a lot simpler than having the Prop chip control the motor directly. Standard motor controllers are controlled by the same square wave input pulses that control standard servos. Pulse width controls throttle setting - that's it. This way we simplify the software running on the prop - or at least that aspect of it - and it simplifies the board we have to make, and each person can select the size of motors and motor controllers he wants to use, depending on the beast he wants to build. We would really only be concerning ourselves with the autopilot & control system that accepts pulse train inputs from a standard RC receiver made for driving servos. Another plus is that we'd be taking advantage of all the development work that others have already done to maximize the efficiency of the available motor controllers, which is a pretty complicated subject. Motor controllers are certainly not all created equal, and if you've never flown an electric RC helicopter before, believe me - maximum efficiency in getting battery power to the motors is a REALLY BIG DEAL. it can make or break your battery duration, which translates directly into your maximum range (meaning traveling distance), which in turn translates directly into the maximum real-world mission-usefulness of any flying machine. Especially an electric one. (Just in case you wanted to do something with it other than hover it over your coffee table.)

This idea does of course add certain complications to the Prop software that would not be an issue when you're designing an autopilot for only one specific flying machine. For example: small aircraft, of whatever type, are able to suddenly roll themselves upside-down and crash extremely fast compared to a much larger aircraft of the same type. Generally speaking, the bigger it is, the easier it is to fly, because bigger means it's more sluggish on the controls, and that's especially true of helicopters. The difference in handling characteristics can be shocking. Therefore, when your machine is in a 10 degree bank and the autopilot needs to send a control correction to the motors, the nature of that control correction would be different in a tiny copter than in a really big copter. The software needs to offer the user with some very quick and easy way to adjust certain control system settings to suit the size, and the particular handling quirks of the aircraft it's installed in. But this is only a software complication, and not an electrical one.

I'm not suggesting I'm as qualified as the rest of you guys to design this thing because I'm still very new to the prop chip. I'm actually an industrial tool designer and not that much of an electronics engineer. But I do have a pilot's license and I do know something about flying machines, so I thought I'd throw in my two cents about why I think this project would benefit by separating itself from being tied to any one specific size (or performance class) of quad copter. Although I DO think we would have to tie ourselves to the quad copter concept, and not expect any one autopilot system to be adaptable to other aircraft types, like conventional tail-rotor helicopters or fixed wing airplanes.

Or course, separate projects to automate the flight of those other machines would be nice too, but right now I think the quad copter sounds like a better starting point, because it's probably the easiest machine for a computer to stabilize and land without crashing using just accelerometers, GPS, compass, barometer, (if it's sensitive enough to be useful) and a downward facing Ping sensor for inputs. Humans find fixed wing airplanes easier to fly & land, but as with driving a car, vision is more badly needed to land them except under tightly controlled runway conditions.

stamped

OzStamp said...
HI Stamped.

What do you mean ?

Ron Mel oz..

Sorry Ron. For some reason I thought it was your post. Let me rephrase... Owen, I would be interested in seeing what you come up with..

rjo_

I'm interested...

I'm not a aero-dynamicist but I have a lot of opinions about how this sort of thing could be done.

I would lobby for a co-axial(counter-rotating) design...

I realize that this is somewhat political... but I would remind everyone that Zhukovski did his work in Czarist Russia, many years before the revolution, and the idea survived. So if we promote the concept as being essentially a-political and patent rending, we should be fine.

Besides, if RadioShack sells only co-axial designs, it can't be such a bad idea...

At Christmas time, we could get the guts of a 4 copter platform at RadioShack for $26x4. With two channels available?

Rich

rjo_

http://www.spaceref.com/news/viewpr.html?pid=4973

Brian L

You're right about co-axial beign a good way to build a helicopter. They handle very nice. I was thinking the simpler quadcopter where it's just four motors with four airplane propellers, with two rotating clockwise and two CCW. That way it's as cheap as possible to build from scratch in any size you want, with no special hardware to make or buy. Separating the motor controllers from the autopilot/control system allows further savings by using brushed motors and controllers if the builder is willing to sacrifice some efficiency. If the builder wants to go all out, he can choose what voltage he wants the motors to run on and spend more for higher voltage stuff.

jazzed

HI. Nice thread.

I've been creating a system with first goal of autonomous operation on a small standard tail-rotor helicopter.
Started with a counter rotating design, but the dual DC motor control mechanism had various issues not easy to
overcome with over-the counter parts like motor-mind or the dual-motor controller.

If you're interested in the design and development log look at this RSS feed:
http://123.writeboard.com/abb0cbb24f2e6cff5/feed/d69403e2673e611d4cbd3fad6fd1788e

Brian L

Very interesting page jazzed. I'm going back there to read it all as soon as I have more time. It could be you're right to use the more conventional type of helicopter since it's aerodynamics and handling characteristics are already well understood and there's no question about it's ability to fly well under gusty wind conditions. I was thinking quad-copter because the mechanical simplicity is a no-brainer and keeps it more in the realm of just being an electronics project where the builder does not have to learn all about cyclic/collective pitch control mechanisms and how to solve the potential vibration problems they can throw at the uninitiated heli builder. Ultimately though, your approach will surely have the greatest potential to fly the best, but it requires builders to become helicopter mechanics, and I suspect a lot of guys on this forum would rather not have to do that. Now you've got me thinking about whether a single control board design could be software-adjusted to work in either kind of copter. Then we'd really have something a lot of people might want to build. I already have a tail rotor heli, so I may just want to duplicate what you're doing.

stevenmess2004

Most of the code could be the same if it was properly broken up into different objects and methods. The trick is to completely separate the part of the code that determines what roll, pitch and heading corrections are necessary from the code that actually outputs these changes to the motors. Its not that hard if you think about it when you start to code and lay out your objects, but, if you don't think about it until the project is nearly finished it will be a real pain in the neck.

Steven

Owen

I think one board could be used for both platforms right? the sensors would be universal. as i see it it's mostly an issue of how to talk to the motor drivers which the mikrokoper solves with an I2c connection, ultimately a board could have many universal io pins to accommodate different types of flying machines. As I see it the big difference between quadcopters and standard helicopters would be in the software, I still think a small light semi universal control board could be designed with out to much trouble.

As a separate thought I was under the impression that a quadcopter would be more stable than a standard helicopter am I correct in thinking this? Stability was the main factor in my interest in quadcopters.

Owen

simonl

I've never flown one, but my gut feeling is that a quadrocopter has similar stability to a standard copter (and a coaxial is more stable than both).

I think Brain L is correct: a quadrocopter is mechanically much simpler to build, and I believe it's a simpler control mechanism / maths model too - bear in mind that most copters these days have a 120deg CCPM control set-up).

Anyways: Steven is correct - if we are careful with our object design, there should be no reason not to create a core controller for any physical set-up; the differences will be in the motor / servo commands, and potentially some feedback requirement (?)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheers,

Simon

---

www.norfolkhelicopterclub.co.uk
You'll always have as many take-offs as landings, the trick is to be sure you can take-off again ;-)
BTW: I type as I'm thinking, so please don't take any offense at my writing style

Brian L

With the quad copter it's easy to eliminate vibration problems because airplane props can be balanced pretty well with a 10 dollar balancer from a hobby shop. With traditional helicopters the rotor blades are so long that static balancing alone is not always good enough, and it still vibrates when rotating. You have to measure the weight of each blade at each end separately so that not only is the weight of all the blades the same, but the center of gravity of each blade is the exact same distance from the center of the rotor head. It can be a pain. I like the idea of a machine that eliminates that problem for builders who don't want to deal with it. The quad also offers the opportunity to deal with crash damage easy because replacing airplane props is cheap and quick, and that's likely to be the only part that usually gets broken. Conventional heli's always break control linkages and other stuff that's time consuming and expensive to fix. The quad also offers the widest possible airframe structure, and gives you places to mount the roll-sensing accelerometers as far as possible from the center of the machine to maximize their sensitivity by providing greater output signal at any given nDEG/per second of roll rate.

How it would handle if flown manually is anyones guess because the only ones I've ever heard of had artificial stability added by uControllers. For this thing I was thinking that even manual flight would actually be just a matter of the pilot directing the autopilot on where to go, and not really direct manual control. Even if it's a real squirrel to handle manually, if it's the easiest one to write software for, the uControllers quick reaction time could still make it the most successful flier, and the best first machine for this kind of project.

With the quad and four fixed pitch propellers, control corrections are delayed by the fact that you have to wait for the motors to change their RPM before anything happens. Conventional heli's simply change cyclic pitch control while rotor speed stays the same and the reaction to control inputs is instant. For that reason I think quad's would be less able to quickly react to gusty wind conditions and would never match the aerobatic capabilities of regular heli's, but in calm winds I think they'd do fine. Stability in no wind gusts is a much different thing than having the ability to combat a strong gust.

I'm sure simonl already knows most of this stuff, but for anyone who doesn't already own a model helicopter: they have a lot of complications you may not want to deal with.

Post Edited (Brian L) : 1/8/2008 8:13:43 PM GMT

Brian L

Another thought:

Adapting to different kinds of machines is another reason to separate the motor controllers from the board. Then the outputs can either drive servos on a conventional heli, or motor controllers on the Quad. I think stevenmess2004 is right. If it's given proper forethought, even most of the code could probably be reused on different copter types.

simonl

Well, I just had to share this one with y'all!: www.rcmovie.de/view_video.php?viewkey=44048868b5a620f05678

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheers,

Simon

---

www.norfolkhelicopterclub.co.uk
You'll always have as many take-offs as landings, the trick is to be sure you can take-off again ;-)
BTW: I type as I'm thinking, so please don't take any offense at my writing style

jazzed

Happy landing Simon