Blogs

New Wiki Advanced Multicopter Design Page

Posted by Gary McCray on February 18, 2014 at 5:28pm

Hi Everybody, I have added a new "Advanced Multicopter Design page to the Wiki.

3689575972?profile=original

It covers a lot of concepts in Multicopter design both for selecting components and frames and in designing your own if that's what you might really want to do.

http://copter.ardupilot.com/wiki/advanced-multicopter-design/

It doesn't by any means cover everything, it is primarily not oriented to "engineers" and it is a work in progress.

Please contribute any feedback, suggestions for additional material to include and criticisms (polite preferred) and I will update the Page to include worthwhile suggestions as best I can.

I am hoping this page can become a repository for serious undertaking of multicopter design and at least include descriptive links to any worthy individual projects.

Hope this meets with your approval.

Best Regards,

Gary

E-mail me when people leave their comments –
Follow

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • max mulhern March 7, 2014 at 6:00am

    hi

    are you willing to look at a design problem that i am trying to solve? Max

  • Glenn M February 20, 2014 at 5:16pm

    That's a comprehensive wiki entry. Good work!

  • mark February 19, 2014 at 12:39pm
    Gary, I am just a humble beginner amd as Oliver posted bow my head in great aprecciation of every ounce of experience that flys out from you. Great blog and profound information! Back to the issue of uninterrupted CF tubing. I have seen models with two tubes crossing each other perpendicularly with the whole lump of electronics in the center of the craft putting the gimble below this at the center of movement, or are spaced parallel with a central elongated brick housing of the control stuff and the gimble at the head of this contraption allowing for prop free viewing. In both outlays the CF tubes are continuous. Question is, is that really significantly helping flight manouverability and anti-wind-responsivness or is it just a crasher resistance issue? Are CF tubes better the the e.g. DJI flamewheel arms with regard to aerial behaviour? Mark
  • MR60
    Hugues February 19, 2014 at 11:42am

    Well done Gary! 

  • Oliver February 19, 2014 at 11:30am

    For those who don't know it already:  Gary McCray's work on the Wikis here goes far, far beyond this great new page. Without him many of us would be flopping around like fish out of water. He's that rarity,  an expert in a technical field who can write coherently (and is willing to) in a manner understandable to people who may not have a similar background. The value of that cannot be overstated. Thanks, Gary!

  • Gary McCray February 19, 2014 at 11:12am

    Hi again Mark only answered half your question,

    Clearly it would be nice to be able to have the CF tube be full lenght, but normally this is not practical and having individual pieces with two clamp type brackets at the central hub for each tube is the normal means of handling this.

    It is important to keep the clamps as far apart as is reasonable in the central hub so that they have the most leverage.

    I really wish that somebody made an internal + or X ferrule so that you could insert it into the four tubes of a quad because that could noticeably increase the crash resistance and rigidity of the frame arms.

    Slight update on above, there is at least one good and inexpensive silicone thermal glue available:

    http://www.ebay.com/itm/Fujik-Thermal-Glue-Silicone-USA-fcbusa-/360...

    Put this stuff on thin for good heat transfer and it will set up like regular silicone.

    A lot of people making High power LED light bulbs and flashlight bulbs are using it for retaining the high power LED pucks to their heat sink.

    Should be a lot more satisfactory to work with than regular silicone thermal grease.

  • Gary McCray February 19, 2014 at 10:25am

    Hi Mark,

    There a number of reasons to keep ESCs with as short DC input leads as possible, As Sgt Rick has said, power pulsing on the DC input to them can be problematic, but also the More DC lead length you have, the more induced EMF which is deleterious to your magnetometers performance. (The AC output EMF to the motors is much less problematic than the DC input leads (and they can be braided or twisted to reduce AC EMF even further)).

    However, for heavy lift copters in particular it is also necessary to insure that you have sufficient air going past the ESC to allow it to cool adequately.

    Generally you should be able to hover at no more than half of your Motors / ESCs rated current and that is normally the circumstance you should plan on.

    But if you need to carry more weight for whatever reason it is important that your ESC have sufficient airflow for cooling.

    I am going to be using a centrally located 30 AMP 4 in 1 ESC.

    It has a built in aluminum plate heat sink, but I may attach a finned aluminum heat sink to that using thermal grease between them.

    On the extreme end unless you are using too little ESC or trying to lift way too much weight you also don't want your ESCs placed where they interfere noticeably with the prop wash.

    Kind of a matter of balance and you need to figure out what will work best for you.

    I will expand a bit on this in the Wiki page.

    Best Regards,

    Gary

  • Daniel Allen February 19, 2014 at 10:00am

    Gray, great page. I think you really covered a lot of the areas beginners have trouble grasping. 

  • Moderator
    Sgt Ric February 19, 2014 at 8:48am

    ESCs should be located as close to the batteries as is possible to prevent feed-back pulsing which  can  cause them to fail.

    (I am not a multi-copter flyer so I can't comment on the actual design above, just on the ESCs themselves)

     

  • mark February 19, 2014 at 8:24am
    sorry: typos: with motors shown in the above graph
This reply was deleted.