I've been reading about Micro Drones claim of 1 hour flight and I have been pouring over some numbers trying to figure out if this is possible with LiPo batteries. My answer is simply no.
I've look at different batteries and different motors. Typically the motors are able to thrust 10g per W. This means that for every gram of thrust needed you are going to consume 0.1Wh (watts per hour). Even this number is slightly optimistic, some motors / props have half of that output others go as “high” as 11-12g per watt.
Let’s use 0.1Wh per gram of thrust.
Now, let’s look at batteries (I’ve compared thunder rc and hk batteries) and the best pack for the punch is about 650 grams for a 8400mah battery (3s). With LiPos you are supposed to drain about 80% which gives about 75Ws for this battery. If you divide the weight of the battery with the Whs you find that you get about 0.11Wh per gram of battery weight.
So, to simplify; just to carry the weight of batteries you will burn up all the charge in one hour. It doesn’t matter HOW many batteries you add. The equation is still the same. However, I have seen that typically LiPo batteries packs more punch the bigger they get, but I could not find any reasonably priced batteries larger than 8400mah.
My dream of building a 1 hour flight octo with 4lb carry capacity is just impossible.
The only possibility I see is to tweak and build a really large copter and gain a few % here and there or come up with another source of energy than LiPos.
Can someone correct my math if it’s off please, or direct me to more efficient motors and batteries?
be cautious with the 80% figure. motors are only 80% efficient across a narrow range of load, below that load, they can be ~65%, and above many go down to ~50%. Even though a motor/prop combo can deliver higher thrust, after that thrust curve passes above the bottom 1/3rd of its range, typically, you begin to drop below 75% for the motor, as a general rule. By the time you get to 50%, most motors are well into the 60's in terms of efficiency.
And the behavior for props efficiently delivering thrust has its own arc.
Murrau: You are of course correct that the g/W scale already takes into account motor efficiency loss.
But the g/W is calculated for a single propeller, so the figures when you have two propellers stacked and down wash will most likely be different.
But again I don't see having 16 or even 12 motors as something realistic for practical applications. Just 8 of them is already a chore. If I wanted 1 hours flights at all cost, I would just get a big fuel helicopter and learn to live with the mechanical complexity, motor noise and smell and vibration issues.
Is everyone is starting to get the idea that the optimal matching of motors to props isn't trivial? Then this is a great discourse.
In a BLDC motor, your throttle control directly controls RPM based on the KV rating, of course. Here's what a typical BLDC performance curve set looks like (courtesy of Johnson Motors):
The "X" axis is torque and the "Y" axis is, well, everything else. It's usually impractical "in the field" to measure either RPM or torque directly, but the two things you can measure are voltage (roughly equivalent to throttle position) and current. It is important to note that the maximum efficiency peaks at about 30% power, and (as Mike said) maximum power efficiency is only about 60%. This varies from motor to motor and among manufacturers, but the above curve set relationships are virtually axiomatic across all permanent magnet motors.
Most reputable motor manufacturers will supply data curves with real numbers. The real baseline measurement is current; if you know that, everything else is determinable from the chart, as current and torque are directly and linearly proportional to each other.
It stands to reason that if one wishes the most efficiency, using over-sized motors for the application would be the rule, provided the weight penalty isn't unreasonable. You can get more payback this way than careful propeller shopping, where almost all the figure of merit (FM) calculations from empirical data I've looked at place most model props in the 50-60% range in static thrust mode. The important point here is that small props up to about 18-inches or so actually gain FM the faster you twist 'em, due to the Reynolds Number (Re) paradox.
Empirical measurement always rules. But the basic rule-of-thumb for efficiency is buy a bigger motor than you need, and spin the prop as fast as you can, but not faster than the manufacturer recommends.
I recently ordered a laser tachometer. For $12.50 it might be a pretty handy tool for working out motor and prop efficiency. You just attach a little sticker reflector somewhere on a spinning part and point the laser tach at it.
As Brad said, getting a copter to hover with payload AUW at the ideal RPM where you have peak efficiency from both the motor and the propeller is a very large order indeed. Especially since performance curves from R/C brands can be somewhat lacking. But another rule-of-thumb that might crash a bit with the one Brad suggests? Is low kv motors and large (relatively) 14-16" propellers. But this is for heavy-lifter (6+ kg) applications. Not quads in the 1-2kg class.
Interesting to note the requirement for high RPM. That might explain why the standard purple Arducopter motors seem so efficient with 10x4.7 props. Hope to see their graphs sometime or do tests myself.
Also wanted to mention some links to micro generators. Don't know if they ever reached production stage, but seem promising.
http://www.sciencedaily.com/releases/2004/11/041124154548.htm and the same one 2 years later http://web.mit.edu/newsoffice/2006/microengines.html
Simpler design: MICE generators article
I have the RC Timer 4215 650kv motors and would like to use the carbon fiber 15"x5.5 propellers mentioned in this thread. But what would be the best way to mount the propellers to the motors? These motors have a three point prop adapter so I can't do direct mount. Is the best way to ream out the middle mounting hole on the propellers and put on the prop adapter?