How the fuck do you guys just slap drones together just like that without any calculations and still get it to work? I am building my first drone, an octocopter (I need it for work) and seeing as these things are expensive to put together, instead of trial and error I'm taking a mathematical approach and it's pretty deep stuff for something usually done as a hobby. You have to match the electronics with the physics like if the physics say you have to spin the prop at a certain RPM in order to get a certain thrust you need to supply a certain voltage for that. Any more or less and it won't work as good. I'm even having to use calculus to work some of it out. You guys are real lucky getting it right just by sticking parts together semi-randomly
>Accelerometer
The voltage thing is really mystifying me. For a 45 minute discharge each motor gets 13.3 A and it needs 375 W to fly. Therefore to provide that with 13.3 A I need 375/13.3 = 28.2 V. But the battery is 22.2 V? How does this work? The only way to reduce the voltage requirement is to reduce the power requirement and the only way to do that is make the props bigger. So is this calculation straight up saying that my drone can't fly with the propeller size I chose?
And the propeller RPM that is even more of a mystery, Thrust depends on RPM right? And RPM only depends on voltage right? I assume this because of the kv rating thing. If you want to increase thrust you have to spin the props faster and more thrust equals faster discharge. BUT increasing discharge rate only increases the amperage? Therefore for thrust to increase amps must be being converted to volts in the electronics?
Most people who build drones just use pre-made parts that do the calculations.
The flight controller takes the input from the receiver and then uses a PID loop to filter out stability failures, it sends the information on proper voltage to be supplied to the ESCs (one for each motor), and those put out the proper amount of power to each motor. If you're not going to use these parts, you're going to have to program it yourself, which is a lot of work.
If you buy a flight controller and ESCs, the only math you have to do is related to your battery, for the most part.
Volt * Amps = Watt: 384 Watt motors / 24 Volt battery = 16 Amps through the cables. Tells you what cables you need.
To run 30 minutes at 384 W you need a battery with 192 Wh capacity.
RPM depends on *average* voltage. Increased voltage leads to higher current and more RPM, more power and thrust.
>>967413
>>967453
I am using a stock flight controller and I know it varies instantaneously the voltage for stability, I just would like to know the average voltage output to each motor for the whole flight. With this information I can calculate the RPM from the kv and hence what propeller pitch I need. The problem I have is that if the voltage is capped at 22.2 V and that's the voltage you need just to get off the ground then how is it supposed to fly faster? The only explanation I that I can think of is that the flight controller converts amps to volts when more than 22.2 V is required. Is this true?
these controllers use PWM, i.e. they turn the fixed voltage off and on, varying from 0% to 100%. you control this with your behaviour. the motor takes its amps depending on load and rpm.
voltage drives amps, not the other way round. don't confuse this.
more pitch = less rpm for same speed = almost same power consumption, but at more amps.
>>967506
is 100% 22.2 V? if you are using a 6s battery? What I am trying to say is if you are spinning 300kv motors then does that mean the maximum possible RPM is 300*22.2 = 6660 RPM?
you input voltage to motor, motor takes whatever current it needs: R=U/I, but R depends on instantaneous load (blade pitch). i don't know about kv measure.
100% is almost battery voltage (small losses in controller), yes
>>967344
You don't do none of that shit.
You make sure your battery/controller/motor combination works together and that they will have more than enough power to lift the weight of the whole thing.
Half the point of such an aircraft is that it can be a complete mess and still fly. If it can produce enough thrust to maneuver, the accelerometer, gyro, and computer interpreting those things will take care of the rest. Barring some really egregious issues, anyway. You seem to be mostly concerned with developing thrust from a given battery configuration, which is a non-issue. The entire point of the controller is to modulate power to the motors and achieve variable amounts of thrust. There's no engineering requirement there other than "make sure 'maximum' is more than enough".
If you want it to be unusually efficient or build parts from scratch, that's a whole other issue not typically covered by the hobbyist crowd.
>>967554
So all I need to do is make sure enough watt-hours are available and the electronics will sort out the volts and amps?
>>967839
Yes. Battery Wh decide maximum flying time, battery voltage decides what cables/controller/motors you need. That's it.
>>967978
Great!
do your research, expect to weigh more than you think, and then buy accordingly
>Lots of people saying to voltage supplied to motors is used to control power
No, only small quadcopters (like 50mm class) use brushed motors.
Most use brushless motors, which are AC. They have 3 or more phases, and 2 of them will be shorted through windings in the stator to generate a magnetic field that moves the rotor, because the rotor has permanent magnets so it also has a magnetic field. And as you probably know magnets repel/attract magneteach other. Gif related.
You can control how fast the rotor will turn by varying how long to wait until you short the next 2 phases (called switching frequency).
If you *really* want to know how it works,
http://www.st.com/web/en/resource/technical/document/user_manual/CD00236524.pdf
(Pages 14-26).
>>969042
So why do all brushless motors come with a kv rating?
>>969046
Because if you are at the maximum switching frequency (full power requested), the only variable that remains is "how strong is the magnetic field?", which is determined by the voltage.
Now you might say, "but then you assume all ESCs have the same maximum switching frequency, if they didn't the kV rating wouldn't be accurate!"
And indeed they do not and it isn't.
For example this guy does ESC benchmarks:
http://www.miniquadtestbench.com/escs
For one benchmark it's the same motor with one kV rating, same prop, same voltage, only the ESC is different; yet you get different RPMs and by extension thrust. Because the processors in the ESCs are different, some are faster, allowing the ESC to switch phases faster.
>>969049
How do you know the switching frequency of your ESC?
>>969055
By looking at benchmarks like the one I used.
As you can see in some of them they can even vary when you use different software on the ESC.
All this switching frequency and kV rating stuff is like the mile per gallon on cars. It's a good estimate, but the Chinese manufacturers would rather die than tell you the truth.
If you have a motor with 2000 kV and 1000 kV from the same manufacturers, you can assume the first one is going to turn faster than the second one with the same voltage at maximum thrust.
With such a large difference it would also work for different manufacturers.
But a shitty white label 2000 kV motor from Banggood might be worse than a RC Tiger 1800 kV one.
(Assuming same motor size for all of the above)
The switching frequency of ESCs doesn't differ so much that you quad won't fly, it's not a dealbraker. I just used them as an example to show how kV is more an estimate instead of a mathematical truth. But if you build racing quads you're going to notice the difference.
Also keep in mind that kV is measured without load (no prop). So the maximum RPM is going to be lower with a heavier prop than with a lighter one.
Really finding benchmarks with thrust in grams for a motor/prop combination is the best thing.
>>969055
Oh I forgot. There is a firmware for ESCs, BLHeli. The developer also tests ESCs.
You don't have to use BLHeli, but look at these 2 PDF files and try to find the ESC you're looking at:
https://github.com/bitdump/BLHeli/raw/master/Atmel/BLHeli%20supported%20Atmel%20ESCs.pdf
If it says "Switching speed is quite fast" it's a good one. If it says something about slow, like "Switching speed is quite fast, although high side is slow to turn on." it's not so good.
https://github.com/bitdump/BLHeli/raw/master/SiLabs/BLHeli%20supported%20SiLabs%20ESCs.pdf
Same here, it says "switching speed is slow" or "not fast" it's not a good one. If it says "The MCU runs at 48MHz" it also has a fast CPU which further boosts performance. For example the X-Rotor ESCs which are at the top of all benchmarks in my link say both.
Not mentioned are Flyduino's KISS ESCs, which are extremely fast but also fragile.
>>969069
>shorted
Yep, page 14 of the PDF I linked
>In BLDC motor control, the electrical cycle is subdivided into six commutation steps. For each step, the bus voltage is applied to one of the three phase windings of the motor while the ground is applied to a second winding.
What is that if not a short?
>>967554
Pretty much this.
Quads are a brute force approach to flying.
As such, they can be fairly ridiculously shaped, yet sill stable.
This comes with the price of absurdly high power consumption and generally poor efficiency.
If you want a quad with a flight time in the 15 minute range or better, then you have to do some designing.
good for you OP, doing your own engineering instead of just playing with expensive legos
