I'd like to try a quadcopter with motors facing down. This setup could possibly provide better stability in the wind (hypothetical), I'm only guessing here. Since nothing will obstruct the prop wash, this could help. Now I'm not sure if I can use normal quad APM firmware for this (with motors facing up).. What do you think ?
I am flying hexas withs props under the arm since the beginning. So I cannot provide any comparison. But it seems to be efficient and to produce less noise. I never had any problems.
Moving rotors beneath a copter is like moving engines back on a plane. Larger planes tend to be slightly forward of central so they pull the aircraft along, like pulling a trailer. Engines at the rear (think fighter) are like reversing with a trailer. Technically it's the same thing just rearranged, but it means a smaller adjustment results in an amplified turn. Good for things that want to change direction quickly, but it you're looking for more stability it's not the way to go. On the contrary, stability would come from making the rotors even higher, like a blimp, but any small quadcopter has such a large surface area to it's weight that I doubt you can make a significant improvement just by repositioning things.
Thanks, I will test both setups. Seems like I have to use rather 26" props for my 2kg light payload.. My general problem is to find an optimal trade between prop size, light payload, high wind resistance and long endurance. The position of the props (up or down) for X4 will be of minor concern, but still can play some role. Major issue is to choose X4 or X8. I cannot decide easily..
Nice ! Do you use normal APM quad firmware 3.3.rc4 ? When mounting the props on motors, do you have to shift props one turn to the other motor ? I mean they have to be mounted in the same way as if they were in upward position right, but change a position to either side of the motor setup. Am I correct ?
A few thoughts to consider.
1. X4 or X8 - An X8 gives you redundancy but adds weight, greater lifting capacity, and reduced flight time with same battery used on X4, and more things that can go wrong and added price. An X4 provides no redundancy if a motor goes out and APM code isn't there for using gryroscopic spinning to stabilize airframe for safe landing. See here. Obviously there are a few factors to consider.
2. Although I claim no fame to anything other than a drone enthusiast, I believe that the reason most copters have upward facing motors and propellers is due to clearance issues dealing with the ground for landing and take offs; also the fact the propellers sit fairly close to chassis which limits what you can suspend and that some folks use FPV which is slightly looking down and spinning props in view screen is distracting to some. I do not believe CG is a problem with motors facing up or down as long a the flight controller is generally centrally located. Keeping CG to center of airframe is very important in that all motors in a hover should work no more or no less than the others. Offsetting the FC to much or extending motors in a non-symmetrical configuration puts stress on certain motors to work harder to maintain stabilization thus reducing the life of that motor. But I digress.
3. Noise and Wind Resistance. Hmmmm.... I have to agree that downward facing props would in a hover produce less noise because there is no propeller downwash against the arm but I never measured the sound levels between upward and downward facing motors to compare noise levels.
4. Lastly, I do know that if you do go with downward motors do not use any motor that does not clearly allow air flow through the coils of the motor. I use Pancake motors for all my airframes (Y6B and OctoQuad (X8)) due to early testing of my original 3DR style motors. It was determined that the downward facing motors were overheating in flight because of the design of the motor casing. The upper motors ran fine and cool but again lower ones suffered. Use pancake motors especially if you plan to face them downward, you will be happy you did.
Photo below is from last year during test of different motors to mitigate heating issues. Lower motors are pancake style and upper one are original motors from 3DR. I eventually swapped all motors for Pancake style.
Good luck with whatever design you decide on and I hope this helps you along your journey.
Nothing changes for the firmware, but performing Auto Tune is a good idea.
Mounted the same why. The props are flipped (print always faces upward) but stay on the same motor, and the motor rotations are reversed. Very simple.
It can be a brain teaser, but with the motors facing normal up and spinning clockwise while viewed from above, when flipped over the props are spinning ccw when viewed from above. Therefore the motor rotation must be reversed from whatever direction they were spinning before flipping. I've done it at least 17 times so have retained in memory :)
Thanks Doug, inspiring notes.
My main enemy is the wind, especially unpredictable wind gusts. Noise level is not a concern for me. But since this copter has to operate in a windy environment and maintain longest possible endurance at the same time, I have to to a trade-off between X4 or X8. It seems that these two are the only options I have. Attached you may find calculated eCalc profiles for both options. The frame weight includes estimated 2kg payload so these should be the closest possible to real values. Now I don't have experience with such large props, low KV motors and high amp ESCs yet but I'm convinced they will fit nicely to the new GD 1400mm V2 frame well. The frame can be used as X4 or X8 and I can play with motors positioning easily. The problem is the wind again. I cannot estimate the behavior even if well balanced and tuned setup is complete. I expect that X8 will be more stable in the wind, but to what level, how much gain I can get over X4 ? These are questions I fear I will have to find myself during the testing of both setups if possible. Seriously is there any publication, blog or study concerning the comparison in a windy conditions ?
Another thing to consider is the motors temperature at max power reaching 80C. Even though KDE claims 240C resistance, I don't want to overheat them. And 80C seems to be overheating value for most other motors. Anyway, I can use 27" props in case overheating will provide motor failure..
To further complicate things, since the copter has to operate in a region with poor maps/elevation data coverage and it has to be able to do terrain following auto mission, I want to add a good quality analog laser range finder. It should help to maintain that 50m altitude over the mission. Now I was thinking how to mount it to the copter properly because copter tilts during the flight. This should be compensated by some sort of little gimbal that will provide a stable 90degrees position. But is it really necessary to have 2 gimbals on a single copter (second is the main one for the camera) ? Well I'm not convinced yet. I'd prefer some less complicated solution.
And at last another problem. How to train a another person without any drone experience to operate this huge copter safely ? I think that's going to be the biggest challenge :) So far I decided to use Iris for training and after some time try the big one.
Anyway thanks for great ideas guys !
What type of winds are you concerned with? My Octoquad is quite powerful and last winter I lifted it to 250 feet (76 meters) and hovered in loiter mode very well. What I didn't calculate was the winds at that height and when I switched to stabilized the wind started pushing it back and behind me at a high rate of speed. I later found out the winds were running around 35-40 mph (56-64km/h). It flew quite stable in that wind, but I had to fight the wind to get it back to me safely. I do understand your concern and your frame is massive to my (600-700) sized frame. Keep in mind you need to develop a vehicle that can out fly the common winds you have in your area. Most copters can only run on average about 40-45 mph all out. I don't know if you will be able to get more than that in that size of frame.
LiDAR recommendation. Try to keep it centered under frame. I do not think its that critical to have it on a gyro-stabilized platform just to keep it level with the earth. But I haven't given that much consideration to mine as I just installed two on my Y6 and X8 frames.
I think recall someone mentioning somewhere out in DIYDrones that LiDAR needs GCS map data for terrain following in auto mode. If you plan to use this feature, you will need maps on your mission planner for it to work correctly. You might wish to research this as I could be off on that.
Good luck with your development and keep my posted. I like seeing where people go and how they got there with their ideas.
Wind resistance at loiter required at least 12m/s (43km/h) or higher. From this point I see X8 has more advantages. So maybe I end up with it..
I will have to find out if maps data are available and to what details level. I will post something when it's ready to go.. But that will take several months from now..
Oh and I have to make some V/A senzor bigger then AttoPilot 180A.. something that works with Pixhawk and has 220A capability..
Anyway thanks !