Center of gravity explanation?

[Motopreserve]

Hey folks. Done a ton of searching here and Internet at large, but can't seem to find any definitive answer. I'm looking for an explanation of the best way to set up CG in relation to multirotors specifically, including the Z axis. A link would be fine, I'd just like something I can understand.

Ive read about the Pendulum fallacy and found a few sites that speak about it for fixed wing. The basic concepts make sense. But with the multirotors there seems to be discrepancy between whether the CG should be below the prop, or in line with it.

For the record, my goal is to fly as stable as possible without flips or tricks of any kind.

Thanks in advance...

scott

 

[janoots2]

Keep x and y as close to center as possible. In my experience, keep z on the prop plane or below.

Not having z lined up with the prop plane does not affect battery consumption.

 

[Motopreserve]

Thanks janoots. Is there a preference, in line OR below? Or you're saying it doesn't matter as long as it's not above?

 

[janoots2]

My builds have always been with a traditional gimbal and a nex5, so by default my cog has been low. I don't know if this is the correct line of thinking.... But hold a long pole at the bottom with all the length at the top. You have to make all kinds of intelligent corrections to keep it balanced while moving your hand around. Now hold it with all the length at the bottom...you don't have to do anything to keep it straight up. I would assume that a lower cog would me more stable based on this simple observation.

 

[Motopreserve]

I could be off-base here, but while your example makes sense to me, it may be part of the pendulum fallacy. Also, if I read it correctly, the proponents for keeping the CG at the prop level helps individual motors not have to overwork to level the quad, which might be more difficult if having to compensate for inaccuracies in CG.

Without a a doubt, it seems that CG lower would be better than too high - but I just don't know how low - or whether for my flying style (REAL mellow and hopefully smooth someday) benefits from leaning one way or the other (pun intended).

Appreciate the reply!

 

[jbrumberg]

I sailed a "tippy" sailboat for +30 years, I look at my quads in some ways as flying boats. A different kind of vessel moving in a different kind of medium. The dynamics are very similar. In general the more balanced the better, and the more the COG keeps you "upright" the better too. We tend to look "down" and then "along" these builds. Just like a boat.

 

[Motopreserve]

Jay, that analogy also makes sense. But where I would possibly see a difference is that the boat has the water to help keep the weight "centered" underneath it. I read a very complicated paper on this and it just confused me more, hence the need for clarification.

What I was able to extrapolate is that it's possibly best to keep an even distribution of weight near the plane of the props, so that they do not have to work harder to adjust level/characteristics. It seemed perhaps lower weight may cause this overworking as the low weight acts in the rocket pendulum effect, in essence having to "swing weight to-and-fro."

Just not sure if I got any of this correct or not

 

[jbrumberg]

In keeping with this analogy there is a "plane" with all our builds. I "assume" my plane/waterline to be at the booms. In reality it is or should be somewhere parallel to above or below the booms. The better I "center" or balance weight in relationship to/in the chassis; the better I will "ride" in this "plane". This plane in theory in my overworked mind should be balanced physically/statically and actively.

 

[janoots2]

I've always kept the x and y cog centered because that just makes sense. Here is a link to an old thread when I experimented with z. The only variable measured was flight time, which did not fluctuate.

http://www.multirotorforums.com/showthread.php?t=9845

 

[Motopreserve]

Janoots, good stuff. How did this not come up when I searched?????

thanks so much for the link. I'm assuming that in addition to no flight time gains you noticed no problems with your motors? They didn't wear out any quicker due to increased strain?

the X and Y being centered seems sensible and obvious to me too. I think this becomes more difficult with the QAV/Disco style quads but most leave room for moving the battery back and forth to compensate.

 

[janoots2]

Can't say if they wear quicker, Avrotos are awesome, I've only had one sputter out in two seasons due to knocking a magnet loose. Didn't measure amps pulled or anything either so can't provide insight on that. But since flight time was the same, I can only assume everything faired out since, at least when I fly, a multirotor has "no front" most of the time.

 

[AirPix]

Good discussion. Balance and stability go hand in hand. I'm building my first quads ( a 350-size and a 500-size) both with DJI Naza-Lite FCs. One variable that will be introduced with the CoG settings in the FC setup would be weight differences in the flight battery-- I may want to use a 2200 3S or a 4200 3S or 4S Lipo, depending on flight conditions. The battery battery will go on the same place (centered below the main frame) and can be adjusted fore-and-aft to get the x-y balance. But a heavier battery will mean that the "Z" value will want to shift lower as the battery weight increases. Back-of-the-envelope calcs suggest that the shift will be small, assuming a 200 to 450 gram range in battery weights.

My question is how sensitive is the DJI NAZA FC to that "z" shift? Would there be a need to go into the Naza setup program and redo the set-up for different batteries? Or for different cameras (slung under the battery)?

--Bill

 

[Motopreserve]

That's a good question Bill. You might also want to post something like this in the DJI/Naza area, although I'd be curios to hear the answer even though I don't fly Naza.

Welcome me to the forum!

 

[AirPix]

I may post over in "DJI/Naza", too, but it's not purely a Naza issue-- any FC makes assumptions about the balance, but the Naza takes it a step further with the addition of GPS and it's USB configuration potential. I've always been curious and wanting to lift the hood to see what makes it tick...

--Bill

 

[Motopreserve]

I only suggested the other thread because it seems Naza does some things unique to those FCs (they all do). But of course all FCs will exhibit some type of compensation.

It seems hanging weight down seems to be the best bet vs building upward, which I was expecting to hear. But I thought maybe it would be more important to keep it balanced on the Z axis.

My big question is what happens on a QAV/Cinetank FPV style frame where the weight is distributed front to back???

 

[R_Lefebvre]

any FC makes assumptions about the balance, <snip> I've always been curious and wanting to lift the hood to see what makes it tick...

Arducopter makes absolutely no assumption about balance. It makes no difference whatsoever. We don't even have a parameter for CG location.

The basic fact is: multirotors are not more stable with weight hanging below them. They are most stable with the CG in the plane of the props. If it can't be in line, then having it above or below, doesn't make much difference.

 

[PeterLester]

Go to time 4min + 38 seconds in this video for a good idea of how IMUs and FCs can handle C of G above the prop plane......but much depends on the programming.
http://www.ted.com/talks/raffaello_d_andrea_the_astounding_athletic_power_of_quadcopters.html

Keep in mind that NASA used older FC IMU concepts so as to balance the Saturn V on the top of vectoring rocket motors for the moon launches..... and again in the LEM before and after moon landings.

Hey folks. Done a ton of searching here and Internet at large, but can't seem to find any definitive answer. I'm looking for an explanation of the best way to set up CG in relation to multirotors specifically, including the Z axis. A link would be fine, I'd just like something I can understand.

Ive read about the Pendulum fallacy and found a few sites that speak about it for fixed wing. The basic concepts make sense. But with the multirotors there seems to be discrepancy between whether the CG should be below the prop, or in line with it.

For the record, my goal is to fly as stable as possible without flips or tricks of any kind.

Thanks in advance...

scott

 

[Motopreserve]

I misworded that. My multiwii doesn't make any "assumptions" either. I meant "adjustments.' The FC is going to balance the quad using the onboard sensors.

The adjustments will be more or less severe depending on the CG. Work harder or less hard I would imagine if you have a well balanced craft.

 

[jbrumberg]

Arducopter makes absolutely no assumption about balance. It makes no difference whatsoever. We don't even have a parameter for CG location.

The basic fact is: multirotors are not more stable with weight hanging below them. They are most stable with the CG in the plane of the props. If it can't be in line, then having it above or below, doesn't make much difference.

I agree about weight above or below in air. I would think if you go one step further they would be most balanced when the COG is in the plane of the props and the COG and the Combined Center of Efforts of the 4 motor/props coincide. Which I guess is me rambling on and agreeing with the more everything is balanced the less the FC's work ie the adjustments that have to be made.

 

[Motopreserve]

Peter,

A bunch of the stuff I've read speaks directly to the benefit of having the CG higher - which is contrast to the typical multirotor build (especially with battery and gimbal underneath).

That's what sparked my question. The common build has a lower CG but I wondered if this is just most practical - or educated physics choice??

I don't claim to be able to wrap my head around this. I have watched that TED video a few times, and the spot you mention does show the water retained in the glass - but liquid is lighter than a lot of the solid crap we hang under our quads. So I wonder if this is not quite apples to apples.

The thing I tend to focus on is whether during flight (not hover) which choice of location for CG makes the motors work least.

I know that ideally we would have the weight sitting tight, not too high , not too low, right in the center of it all. But that becomes less and less realistic as we add more features to our crafts (FPV, cameras, gimbals etc).