Jump to content
Sign in to follow this  
cyber-flyer

Quadrocopter Vs Single Rotor

Recommended Posts

I've finally managed to put together X3D-Bl kit from http://www.xufo-shop.de/ and I must say I am impressed. The kit quality is very good (9 out of 10) and it was a pleasure to assemble the model from parts machined to a fraction of mm precision. I know it's expensive kit - but you get what you pay for.

The initial test flight was good but not very exciting. I was mentally comparing X-3D flight characteristics to TRex-450 which is similar weight/size heli. And it felt like X3D was half as responsive as Trex. Even if I used Advanced/Expert settings it still was too slow to change cyclic/yaw angles as compared to highly tuned variable pitch helicopter. But it felt quite lively on collective response. I played around with X-3D control parameters and after about couple of dozen of trials the X3D became much more quicker on cyclics. With the changed control parameters it feels within 20-30% of the T-Rex on collective/cyclic and probably about 50% of yaw. Which is pretty darn good for fixed pitch rotor. I was flying it on a windy day in front of my house and the wind didn't bother X3D. With little practice I was able to do very fast forward flights with 60-70 degree banking turns avoiding trees and stopping very quick when required. I felt confident flying it as aggressively as I would do with TRex. Obviously it's not up to par when negative pitch is required (rolls and loops). And the yaw is still not nearly as crisp - it feels like gyro on TRex is about 50% of the full gain. But forward flight is very capable, full power climb ups are also very quick - the darn thing has 2:1 thrust to weight ratio and it flies about 20 minutes on 2100 mAh battery. All things considered - thumbs up for the model as a possible FPV platform. And thanks to Kilrah for the recommendation!

Hope it helps, happy new year!

Share this post


Link to post
Share on other sites

Obviously this topic is worthless without pics or video! :D

Cheers,

Sander.

Share this post


Link to post
Share on other sites

Nice! I wouldn't have taken the risk of comparing it with a 3D heli, they're not really meant for the same purpose in the first place anyway ;)

Oh, I had forgotten to post my MK video here BTW:

WMV

Vimeo:

http://www.vimeo.com/460273

Share this post


Link to post
Share on other sites

All very good stuff! The comments about the dumbed-down cyclic and yaw sounds perfect for me. I'm looking for something that is a very tame rather than a 3D ship. For now, I'm patiently waiting for the prices to fall so that I can avoid the strain on my pocket book.

Share this post


Link to post
Share on other sites

There are bunch of videos of the X3D flying, eg:

If I get anybody to make a video of me racing X3D vs TRex, I'll post it. I still think TRex will win the race but quadrocopter will be a better camera ship. I'd still compare the two from point of view how accurately they can handle the pilot inputs - helps me to figure out what to expect from a bigger version of X3d ;)

Thanks for the video Kilrah, it looks good.

Edited by cyber-flyer

Share this post


Link to post
Share on other sites

Kilrah, I noticed that your video is making its way around the various forums. You may very well be responsible for a few more folks joining the fun. :)

Share this post


Link to post
Share on other sites

I've been searcing the web for bigger quadros.

What would happen if someone built a macine with four or eight AXI 53XX?

Share this post


Link to post
Share on other sites

Indeed, also got contacted by the MK's developers via Youtube not even 1 day after I posted the video, it's on their site too now :)

I didn't think there was anything that interesting there, but they seemed to like the "height record" hehe

As to scaling the thing up, I wouldn't see too many problems whatsoever. I guess the place where work would have to be done is the brushless controller as for example the MK's one are 10A continuous, 20A peak. For a bigger thing you'd most likely need more, so a solution would be to keep the current hardware and just replace the power stage with other FETS with lower resistance and better cooling...

Then sourcing bigger "pusher / puller" prop sets could be a bit of a challenge too.

The bigger ones I've found all used the usual ESC/motor/prop sets, but simply more of them. A guy made a copter with 12 of them, and can lift a big 5kg weight... OK only in stable hover flight. That of course requires modifying the flight controller's software.

Edited by Kilrah

Share this post


Link to post
Share on other sites

I would like to build something up to 2 meters in diameter and lift capacity of at least 4 kg.

Will the vehicle become more stable if I increase diameter?

Is it possible to use ordinary Jeti ESC for the motors?

Also, how about voltage? Could I use a 12 cell 20 000 mAh LiPo with the original avionics?

Lots of questions here, and I still have a thousand more to ask…..

Share this post


Link to post
Share on other sites

Larger size does usually increase stability, as less thrust difference is needed to compensate an incorrect attitude. Now what I don't know is if at some point you wouldn't get in trouble because of this, as the minimal practical throttle step would actually already be too much correction. Some experimentation needed, it's a factor of a lot of stuff.

No, you need to use the specifically designed ESC's. Common ones take a simple 50Hz PWM input and take ages to set the motor RPM to the desired values. The custom ESCs are updated via I2C at a 500Hz rate and can set the motor RPM in the range of a few milliseconds for small adjustments as required for stabilization.

I know some commercial ESC's that use Atmel AVR can be reflashed with a custom firmware that will give them the desired properties, haven't read much about it so you should maybe take some time there.

The MK's ESCs can hold up to 4 cells. For more it's either modification again or finding a suitable and "hackable" ESC.

Share this post


Link to post
Share on other sites

Good discussion - I also looked into building big multirotor heli.

In my view the speed of quadrocopter reaction is not mainly determined by the refresh rate of the ESC but due to inertia of the motor/prop combination.

To be more correct max electrical power divided by the moment on inertia (motor + prop) is the factor to consider. And there the moment of inertia of a prop is probably the determining factor. Let's consider quadrocopter which is twice as large as the existing model.

Twice as large prop will be probably 4 to 8 times more heavier (just a guess) and have 4*8 = 32 larger inertia moment (I ~ mr^2). This tells me that power to the prop have to be 30 times as much to make the response on the same scale. Taking the argument a bit further - the RPM of the bigger prop has to be the same as the small one. And this is where the problem will show up as the blade tips can not become supersonic. To make a bigger model one can either increase the number of smaller props or use slower spinning big ones. The latter will put a limit on the aircraft responsiveness. Just my 2c.

Edited by cyber-flyer

Share this post


Link to post
Share on other sites
In my view the speed of quadrocopter reaction is not mainly determined by the refresh rate of the ESC but due to inertia of the motor/prop combination.

I've been talking today with one of the developers of the NG UAVP, and we talked about that for a moment. From the combined results of the 3 main teams, it does have a lot of importance. They found 100Hz do be the minimum for good performance on 50cm-size machines, with noticeable improvement until 500Hz. On smaller machines like the 35cm X3D 1kHz is even better (and used). Now I don't know if they've tried dissociating the control loop and ESC update rate, for example running the control loop at 1kHz but updating the motors only at 200Hz and if there's a difference.

Of course inertia is a factor, but how I said above the thing is that during flight the adjustments done by the stabilization algorithms are small ones. For example on the 15-250 allowed throttle range, the change in setpoint in a single cycle of a 500Hz controller isn't likely to exceed like 5 steps... that's a very small change in RPM, which is easy to reach in a very short time without hitting the power system limits.

Share this post


Link to post
Share on other sites
I would like to build something up to 2 meters in diameter and lift capacity of at least 4 kg.

I saw some information on a scaled-up version with more than 4kg lift. I recall it used eight motors. I don't remember much beyond that and sadly I don't have a link. It might be something that was discussed in the thread over at rcgroups, so just read through that 81 page (and growing) discussion. :)

EDIT: Found it. It has TWELVE motors and can lift 5kg. http://forum.xufo.net/bb/viewtopic.php?t=7251

Edited by Mr.RC-Cam
Found link

Share this post


Link to post
Share on other sites
I've been talking today with one of the developers of the NG UAVP, and we talked about that for a moment. From the combined results of the 3 main teams, it does have a lot of importance. They found 100Hz do be the minimum for good performance on 50cm-size machines, with noticeable improvement until 500Hz. On smaller machines like the 35cm X3D 1kHz is even better (and used). Now I don't know if they've tried dissociating the control loop and ESC update rate, for example running the control loop at 1kHz but updating the motors only at 200Hz and if there's a difference.

Interesting, thanks for the info. This does make things easier for scaling up.

Edited by cyber-flyer

Share this post


Link to post
Share on other sites

Thank you Mr.RC-CAM! That link was quite informing.

Kilrah, I might have to take a closer look at your machine too.

Now, the next question is, am I able to build something like this?

I have to dig deeper into this to learn more about the quadros, so now you know what kind of websites I'll be looking for this weekend ;)

Share this post


Link to post
Share on other sites
Thank you Mr.RC-CAM! That link was quite informing.

Kilrah, I might have to take a closer look at your machine too.

Now, the next question is, am I able to build something like this?

I have to dig deeper into this to learn more about the quadros, so now you know what kind of websites I'll be looking for this weekend ;)

I would start at this one http://www.rcgroups.com/forums/showthread.php?t=768115 (22 pages), follow by http://www.rcgroups.com/forums/showthread.php?t=716870 (82 pages), there are many other webs to look at but this two are imo the most representative on a single topic as they go through the entire evolution of some guys from starting point till having the UFOs on air. If you can read all those pages you will get a good idea about how these machines work and the best way to do it for you.

Hope it helps.

Share this post


Link to post
Share on other sites

Elossam: Thank you my friend. Very nice links there. I will read them asap.

How do you guys set up your RC?

Do you fly like a normal helicopter with cyclic and collective input?

How about some kind of idle up?

(maybe I’ll find some answers in the links you gave me) :)

Share this post


Link to post
Share on other sites

Nooooo never. TX in standard airplane mode only, no mixings. All the stuff is done on the onboard flight controller.

I can add the original sources of info too, the official forums of the 3 main developer teams:

http://www.mikrokopter.de/

http://www.uavp.eu

http://forum.xufo.net/bb/

Share this post


Link to post
Share on other sites

cFe7,

Building a mikrokopter is quite straightforward if you have reasonable electronic skills. Whilst completed pcb's may now be available I had to hand assemble the five surface mount pcb's. It was the first time I had ever attempted an smd project of this magnitude but with patience, a small soldering iron, good head mounted magnifying glasses I succeeded with only one board not working first time.

The second step is loading the software and testing the pcb's. For various reasons this gave me some trouble but again with patience and posting requests for help on RCGroups and the Mikrokopter forum I managed it.

Making the frame and mounting all the parts is the final step - relatively straightforward.

Good luck, Peter

Share this post


Link to post
Share on other sites

I just went with the ready-made boards. Didn't want to bother with that and I already have enough occasions to do SMD soldering ;)

Making the frame, assembling and testing, about 4hrs.

Share this post


Link to post
Share on other sites

Hmmm….you guys convinced me. Seems like I have to start building asap.

Right now I’m searching the web for the most OK set up. My only problem so far must be all the gadgets people have in their rotorcrafts (return to home, lights etc). want this sooooooooo bad :P

Share this post


Link to post
Share on other sites

I’ve been reading about quadrocopters day and night until my eyes became square.

There is one question that came up last night. I was watching a video where someone flew inside a building from room to room. Looks like he had rudder on his vehicle. Are the motors tilted to make motions around the yaw axis, or do you use torque from the propellers? I couldn’t see any “tailrotor” like a normal helicopter.

Share this post


Link to post
Share on other sites

Torque.

The front/rear props are turning one way, the left/right ones the other way, so that in normal flight the torque is nulled. To control the yaw axis, you simply speed up both props rotating in the one direction while slowing both others down. That way the total torque isn0t zero anymore, while the tilt/roll axes aren't affected.

Share this post


Link to post
Share on other sites

OK that explains a lot to me.

I'll have to read more.......vant a quadro......now....... :o

Have a nice weekend guys!

Edited by cfe7

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×