Design considerations for an aerial photography platform.

[whoisthedaddy]

All,

What a great site, so many useful threads and posts. After reading many of these over the last few days I wanted to share my thoughts so far for your comments and advice. Before I go into the details and reasoning I think it worth mentioning that my personal requirement is for a craft that is capable of taking professional quality aerial video. (I am not interested in throwing it around the sky aggressively). Due to the commercial nature (I hope!) I need the craft to be as resilient and fail safe as possible and be capable of smooth and reliable flights - If something doesn't work on a customer site ideally it would be possible to repair without a complete strip down - I don't want too much!!.

Primarily due to the safety aspect I have decided on an Octo configuration (I like the Droidworx AD8HL frame) for three main reasons:

• A numbers thing, the effect of 1 motor failing out of 8 will be less than 1 failing out of 4 and hopefully the craft can be landed safely!
• A mass thing, 8 motors weigh more than 4 motors (not to mention the additional battery weight and electronics) and this mass should make the craft more stable (less responsive?).
• My thinking is that the additional thrust can carry additional weight (e.g. video camera, transmitter for FPV (future) / GPS modules (future too!)

I have been struggling with the choice of Flight Controller but seem to be settling on the Hoverfly Pro. I like this for a number of reasons over the MK equipment. I recognise that it probably isn't as configurable but I don't want to spend days or weeks tweaking electronics and setting parameters. Although I love most aspects of modelling and flying I am trying hard to remain business focused!

• It comes pre-assembled! - I am not a fan of soldering.
• It seems considerably easier to use - there is little programming as such.
• It looks like it has all the functions I require with GPS in the future (I've read the posts but this is not urgent for me)
  
• The external wiring connections look very neat, they are also plug in as opposed to solder in which means an easier and quicker change in the event of a problem).

(I think I read some posts concerning the ability of the Hoverfly Pro to hold a stable hover but can't refind it, it this was the case, maybe it isn't anymore but in any event it isn't such a big requirement for me and for the most part I intend to be flying it forward)

The Hoverfly Pro supports external and individual ESCs which although they need calibrating individually I still think I like the idea of for a number of reasons:

• Safety, if one fails the others still work!
  
• They can be located on/under the airframe (as opposed to in the dome) and therefore 'air-cooled' as opposed to some systems which need their own cooling fan.
• They are easily changeable - the MK ones are built into a frame and the motor wire soldered directly to them
  

I don't know how easy it is to calibrate the ESCs or whether there are other reasons why I shouldn't choose this system and any and all comments would be really appreciated.

Although I haven't really started yet on deciding on motors / prop combinations I have already read a lot of posts on these topics and I'll post again when I've read them properly.

Any advice or thoughts gratefully received and thanks in advance.
Lee.

 

[Droider]

Blimy.. you been busy. There are guys on this site with a lot more knowledge that I but I will give you my thoughts.

GPS is ESSENCIAL. Unless you are only flying very low altitudes. Max by law in the UK is 400ft I think. (121m) At that altitude even in a slight wind its hard to keep the MR stable and in a fixed position with out it. The HF GPS has been promised for months and talked about since the HF FC reincarnation from its previous life.

I went down the same road you are going down. I almost clicked the buy button on the Hoverfly for all the reasons you are talking about. But I went MK and am really glad I did.

Its actually not as complicated as it looks to get the MK FC set up. Yes it was absolutely mind blowing at first, poor information etc etc.

Would i go back NO chance. I am not saying the HF FC is bad cause loads of people rave about them and loads of people are frustrated with them.. much like the MK FC. The MK FC for me makes more sense and is more tuneable to your requirements.

You have started a very interesting thread and I am sure the members here will give you all there best advice mine.

I have found my ADX3 very stable and easy to transport. I am looking at building a heavy lift DW X8 so I will be also interested in all the comments that I am sure will come. One thing I know for sure that my build will be DroidWorx and the flight control Mikrocopter.

Here is a very good site built by Reece2. He documented his whole build very nicely here https://sites.google.com/site/mkokto2/home

Check out Digitech. Sandor will build you anything you want and on his site there is a drop down box selection to customise your build to your exact specification.. you can start here.... WOW WOW This is a bargin. If I had the cash I would not hesitate http://shop.jet-1a.com/product_info.php?cPath=468_480&products_id=1818&osCsid=e14d16ca9d259941abbc662d6e43f89c

 

[matwelli]

from my relatively recent experiance with multirotors, i have this little bit of input.

I think a X8 is the best solution

My reasoning is that, each corner has a motor pair, one top, one bottom, and you can handle one failing as long as you have made sure that 4 motors can carry your total load.
For professional work, thats important, esp as you are around people and property.

You can power the top 4 and bottom 4 seperately (more redundancy) and from the videos I have seen (on the hoverfly site) very very stable http://vimeo.com/9435221 http://www.hoverflytech.com/Videos.html

Frame - the droidworx ? http://www.droidworx.com.au/shop/products/airframes-heavy-lift/80-ad-4-heavy-lift-extended

 

[Droider]

This is THEDADDY whosyourdaddy!

http://www.youtube.com/watch?feature=player_embedded&v=7jRXK-JD-gM

I think this is the one on Digitech website.

 

[whoisthedaddy]

Some really interesting stuff here already, especially some of the links to video and websites.

@Droider, thanks for the link to the Reece2 build, it looks excellent and I'm going to review in detail later night.

@Mat, I have to say I didn't really consider the X8 configuration although the logic of redundant power by powering each 4 independently really does make sense to me. I wonder how efficient the prop arrangement is in an X8 configuration is as the ones underneath are working with disturbed air? I am no expert in aerodynamics but in Formula 1 you don't want to be in 'dirty' air?

I've never flown helicopters but with planes often the bigger ones are often easier to fly than the smaller ones and I figured the same logic applied to MR? (i.e. the Octo is bigger, therefore more stable and easier to fly).
I see there are a lot of references to motor to motor distances and I guess this is important for a reason but as of yet haven't been able to find out why?

Lee.

 

[Bartman]

Lee,
I'm chompin' at the bit to get an MK/DW Oktokopter going and my reasoning for the flat-8 vs. the coax-8 is that i would think the 4 lower props are going to be generating airframe vibration flying in the propwash of the upper 4. With a flat config the props are closer to each other and there is propwash going laterally out from the prop tips but I have a hard time visualizing the coax arrangement being smoother, especially for video.
just a guess at this point.
bart

 

[RTRyder]

Very quickly, a few thoughts before I head out to the shop to machine a few pieces to attach the Avertical mount to my octo X8.

If you want a reliable professional grade multi with proven capabilites, I cannot recommend the Hoverfly Pro at this time. I own one of the first H/F Pro boards sold to the public so I've been working with it for a while now, I also own two Mikrokopter Hexa's, one box stock and the other a Droidworx AD-6 heavylifter. The difference is the firmware is still not right on the Hoverfly and the last release actually made some things worse. The update releases are slow to come, so if there happens to be a bug that renders your setup unusable it could be a month or more before you see a fix and there's no assurance that the "fix" will actually fix anything.

Not slagging the Hoverfly folks here, just the truth of what I've been through in the course of the last 7 months trying to make one into a reliable A/P platform, still not there and I really have no faith that it will happen anytime soon. If you plan to purchase one on the hope that it will have a fully functional GPS with the same capabilities as the MK, I'm of the opinion that it will likely be a year or more before there is a Hoverfly GPS module that has even basic functionality if even that soon. If you want to know the track record of the folks doing the development go to RC Groups and search for threads on the Quadpower board which was their first attempt that came out several years ago, those folks are STILL waiting for a promised GPS module.

On the other hand, a Mikrokopter in box stock form with essentially no tweaking of the software flys many times better with a lot more stability than anything I've been able to get from the Hoverfly, no BS, just fact. My MK's sit on the shelf and when needed they get the job done without problems. Yes I tinker with them a lot simply because I like to see if I can improve the way they work, really haven't succeeded in making them any better yet. On the box stock one I keep winding up with it back to the way it came from the factory and when I need to get some aerial shots I know I can rely on it to do exactly what I need, no worries.

That said, I'm a glutton for punishment and am in the final stages of building an octo X8 using the Hoverfly board as the flight controller, you can find my build thread here... http://www.multirotorforums.com/showthread.php?247-X8-build ...the main reason I'm doing this is more a proof of concept project than one that I expect to be a flawless A/P platform simply because I know the controller isn't capable of that at the moment. My backup plan, should the platform turn out to be something I really like, is to install another Mikrokopter flight controller board with an I2C to PWM convertor so I can use the standard ESC's already on the frame. If there should happen to be a Hoverfly firmware release that addresses at least 50% of the really annoying problems the board has then maybe I'll leave it as-is and hope the remainder of the less important problems get resolved soon, but I'm not holding my breath waiting for it.

Anyway, time to go make some parts for the project...

Ken

 

[RTRyder]

Lee,
I'm chompin' at the bit to get an MK/DW Oktokopter going and my reasoning for the flat-8 vs. the coax-8 is that i would think the 4 lower props are going to be generating airframe vibration flying in the propwash of the upper 4. With a flat config the props are closer to each other and there is propwash going laterally out from the prop tips but I have a hard time visualizing the coax arrangement being smoother, especially for video.
just a guess at this point.
bart

Bart,

Just did the first proper test flight of the X8 about an hour ago, I think you'd be surprised at how smooth the configuration is. Because the top and bottom motors are spinning in opposite directions there is zero torque on each arm and any vibration tends to be cancelled out by the opposite motor. I was truly amazed at how smooth it was in the limited flying I was able to get in between rain showers, now I'm working to finish off the last details of the Avertical camera mount so I can do the final assembly and see if this mount works as well as it as I think it can.

Ken

 

[Bartman]

that's an interesting point ken, about the torque on the arms.
let's see what kind of results you get with the camera mount and we'll go from there. i'm still undecided on the mount as well. there are a lot of great claims out there regarding two axis camera mounts but they each have their problems.
bart

 

[whoisthedaddy]

Yes Ken, I agree the point about the torque on the arms is neutral and I hadn't really thought of it that way.
On a flat 8 the torque of each of the arms should make the entire frame neutral (obviously) but each arm of the frame has both a lateral and vertical force, on the x8 the only force is a vertical one. Whether this alone leads to less (or reduced) vibration should be considered.

@Bart, I read another thread somewhere where you talked about tip propwash but don't know how far this extends from the tip, guess it will vary based on rpm and other variables. i.e. would a bigger octo where the props are further apart make it less prone to propwash? Has anybody any views on the size of a MR and whether bigger ones are more stable (and less reactive)?
@RTRyder, I am going to take another look at the MK equipment as all the experienced people here are telling me the same thing! - thanks for taking the time to let me know.

Thanks
Lee.

 

[Bartman]

The thing about MK is that it just works. You don't have to wait for anything to be developed, you don't have to learn how to hit a curveball, it's just there and it works. Mine is still in beginner mode and I can't think of a reason to change it. It's a pack mule with a camera strapped to it and it's perfect at its job. Get too creative with the software though and it'll bite you. Worthwhile add ons are the Spektrum diversity board and I2c isolator board. Plan on having at least two satellites if you go with the JR radio.
Bart

 

[jes1111]

...would a bigger octo where the props are further apart make it less prone to propwash? Has anybody any views on the size of a MR and whether bigger ones are more stable (and less reactive)?

The props on a flat-8 are not further apart than those on a coax-8 of the same overall diameter with the same props - they're definitely closer and that to my mind, makes the coax-8 the better option. The "sidewash" from props can create all sorts of hard-to-identify problems - the general advice for sizing a frame is to have one propeller diameter of space between each propeller, i.e. if you have 13" props there should be about 13" between them - therefore bigger props need longer arms. Having the arms much longer than this would be redundant, simply slowing the responsiveness of the craft for no benefit. Much shorter and you're likely to encounter issues.

There's a lot of misunderstanding around about "stability". The most common misconception you'll hear is that hanging a big camera under the frame will "make it more stable" (with the word "pendulum" usually involved in the explanation). In short, it won't! What it will do is a) catch more wind (which will disturb the craft) and then b)require more energy input to correct the craft's attitude (meaning the correction will take longer). The higher you can get the camera, the better - but that gives you problems to keep the arms/props out of the shot. The flat-8 doesn't have lower props but the arms have to be longer, which means they'll be in the shot... and so it goes on, round and round. Enter the Y6! 120 degrees between the front arms means the camera problems are reduced. Six engines gives you "some" redundancy and the whole ot should weigh less than an octo, so you've got more chance to carry battery weight instead of dead weight. Hence the popularity of the Y6 - it's probably the best all-round compromise.

For a video platform you want mass (harder to disturb), absolute rigidity (one of the plus points of the Droidworx), lots of power (quicker reactions) and, of course, an electronics package that's up to the job (which, I agree, excludes Hoverfly).

You need to spend well on the camera mount (kind of like needing a good tripod when on the ground). Consider dedicated stablisation, but remember that the camera mount can only stabilise rotational disturbances - if the wind moves the whole craft a foot to the right there's nothing the mount can do about that!

 

[matwelli]

@ RTRyder - good stuff, happy to see the X8 flies !

@ jes1111 - love the Y6 design, they seem to handle the wind well (another consideration) but not sure on redundancy, if you lost one motor you would be in for a world of hurt.

@whoisthedaddy - i have that on a tee-shirt "whos your daddy!" the flat 8 design, there was a discussion about stability in the wind, with the flat 8 having a large prop surface that would be affected by wind gusts.
how much of a hurry are you in to build ? last nite a customer crossed my palms with gold (well he used his CC and paypal) and comissioned a X8 build. for my own interests, I am going to whip up a flat 8 design as well, and do a back to back comparison, same motors/props/etc, and will let you guys know what differences I see.

 

[jes1111]

Just my quick thought on redundancy... No doubt it is good to have and 8 motors gives you more of it than 6, but... by all means dismiss 4-engined craft but there are so many differences between 6 and 8 motors - it is much more complex than just saying "8 has better redundancy than 6, so I'll have 8".

Another aspect is that we can lengthen the odds of failure by building more reliable craft. Better engines, better ESCs, better frame, better batteries, better soldering, better maintenance, etc. A well-built, well-maintained craft with quality components is, IMO, better insurance than two more engines.

 

[matwelli]

@ jess111 - sorry , mis understanding my friend - the x8 with motor pairs rotating in the SAME direction has redundancy, in that if one fails, you double the speed of the good one to get the same thrust and tourque, so the system remains stable.

In a x8 with counter rotating motors, and the Y6 with the same, i dont see how the redundancy would work, as when you loose one motor, and speed the other up to compensate, you have a massive tourque issue ?, or do you not ? i dont know as havent tested it, will be testing a x8 with a dead motor this week

 

[Bartman]

........................ last nite a customer crossed my palms with gold (well he used his CC and paypal) and comissioned a X8 build. for my own interests, I am going to whip up a flat 8 design as well, and do a back to back comparison, same motors/props/etc, and will let you guys know what differences I see.

i knew we let you in here for a reason!!!! how fast can you have some answers for us? is monday too soon?

 

[matwelli]

haha, more like thursday, waiting on 16 motors to turn up brought 100 props last week !

 

[Droider]

Got to say.. I LOVE my Y6 DroidWorx ADX3.. In the coming few weeks I will be posting more and more pic and video from me beauty.

 

[matwelli]

@ droider - that one sure handles the wind nice, i guess i have to add Y6 to the build for a tru comparison

 

[matwelli]

Doh ! thats my weeks holiday gone, just got a second order for an X8