Strangely enough we are doing the exact same thing right now. We are finishing up our testing tomorrow of all of our tests. Long story short, we are sweeping through the throttle range of our esc and recording the thrust given off by our motor and the power draw by our esc. Here is our data in the same format that you presented your data with. I am curious as to how you calibrated your thrust readings.
Here is our calibration data from two calibration runs as well as the raw data from my above post.
My goal was to find the
1) optimal number of motors
2) best props for a certain thrust
3) most efficient motor… of the ones I have
Attached is my xls sheet… It’s a mess for someone who never saw it…
Once I found the prop I like, I fit a curve to it and used the curve to predict flight times with more or less motors and batteries..
Luckily… The props curves do not cross but I did not always use the best prop.. 14 x 8.5 were very efficient but did not fly the copter well..
So I use 12 x 3.8 SF on everything. Its not the best but its only off by 5-8%.. best I can tell
I looked at your spreed sheet... That is very nice.. much more data per run than I have...
I made some changed to it on sheet3... I think you will like it..
If the 14x8.5 did not fly well but a smaller propeller did, it probably means the copter was to lightweight for the propeller size. When that happens you get liftoff at a very low rpm range, making stabilization difficult. Bigger propeller = bigger batteries = more flight time. :)
I agree %100 percent... Everytime a copter does not tune well, we just add another 5500 4S..
and everything gets far better...
Last night we were flying HobbyKing NX4006 530kv... 13x6.5E....on a quad that weighted nothing.
Only when we added the second 5500 4S was it a rock... also flew for > 30 min... mostly in loiter at night..... got tired of flying it.
I wanted to submit some of our data on this topic as well. One of the things that we wanted to test was the impact of ground effects on the overall magnitude of the thrust as well as the stability of the output thrust. We put a plywood board under the motor as it was running at approximately the same height that it would be off the ground when installed on our vehicle during takeoff and landing. As expected, it improved the performance of our motor and rotor but it also had a lot more noise in the higher throttle range. I suppose the primary takeaway from this is that if your UAV has to throttle up into the unstable region to take off then you are probably going to have some serious stability issues.
Thanks for reporting this practical data. It seems so much better than reading all the speculation and eCalc. I have to ask if the "black" props you used are the rather flexy thin props like I got from 3DRobotics? Mine were way out of balance and seemed too flexible. I decided to buy a batch of APC 12 x 3.8 and I like them. I felt like the lower pitch would give my 880kv motors a "lower gear" to haul a modest GoPro payload on my 3DR quad. And throttle response has been too sensitive for me. Even the smallest throttle adjustments yield too great an altitude change. I want very stable altitude to match the roll and pitch stability.
Your data supports my 12 x 3.8 choice. Yeah!
These are the "Black props"... I have a bunch if anyone wants them... LOL...
I to thought that 12 x 3.8 would be by far the best... but, test show APC 12 x 6 E... is just as good or a
few % better... 13 x 6.5 E... is just a tad better again..
These test are with the same motor... A good watt meter and really good Grams scale... not many places for errors to skew the data... But the 13 x 6.5E are them selves heaver... May even everything out...
Summary of results...
Black props suck...
Keep the load on the motor LOW..
Get as low KV as possiable and use a big prop...
In many cases... going from 3 motors to 8.. the copter would fly far longer...
mostly because the g/w was much higher at lower power settings... more than enough to make up for the extra motor weight...
If your copter has large motors and big props, it may not fly well... so there is a limit !!
So just put another battery..... That often fixes many problems.. !
Good info guys.
I just ordered 9 NTM 3536-900 motors. Hopefully that wasn't a mistake, I see you guys seem to be testing the pancake motors, but when I was looking around, seems like everybody else is using standard profile motors? I almost went with 2836-750 motors, but a huge goal of this machine is to have enough reserve power to have a good chance of surviving a motor failure while carrying a payload, and I wasn't sure I'd get that with the smaller motors.
Anyway, I've been trying to figure out however, which props to use and thus how far to space the arms on the Octo.
I'm actually thinking of doing overlapping props, but flipping alternate motors up/down. I've got a design for the Octo with 12" props and the ideal 20% overlap (as detailed by Brad Hughey). The idea here, if you aren't aware, is that in theory, having 20-30% overlap can actually increase the thrust vs. a single motor due to some beneficial aerodynamics.
The diameter with this design works out to about 630mm, so it's a compact Octo, but with lots of power.
If I wanted to go with 14" props with the same overlap, it would require a diameter of 750mm.
And then a "standard" Octo with 12" props would be about 850-950 with 12 or 14" props.
So the potential is there to benefit with more efficient lift. And also the compact size is easier to transport. But I do wonder about stability, if the shorter arms will make it less stable? There seems to be some trend there, but I don't have the experience to say.
Anybody have any input? Wanna guess which props will be better for these motors, 12" or 14"?