( Firstly )  - please excuse what might be a really stupid  question - but, since I'm new to all of this... I cant help thinking....

Isn't there some way ( even on a very small scale ) to use a combination of a small 2 stroke powered motor ( or a typical RC gas powered engine ) to spin a small generator and thus create power to assist in "trickle charging "  - or even powering a quad?

I understand there is a weight issue, but I'm thinking the right combination of ESC and motors ( electric ) and prop size could possibly provide enough grunt to lift the thing.

If this were possible, then using a real liquid fuel would ONLY make the drone progressively lighter  compared to the total weight at the start  as it consumes its supply of fuel - in the meantime,

While I'm at it - what about using DC to AC inverter technology some how to an advantage in power output?

 - Just asking.. lol!

[ there are no stupid questions -  the only time a question is stupid, is when you should have asked - but you didn't.]


Views: 1791

Reply to This

Replies to This Discussion

Rather than trickle charging, which is likely to be impractical especially if multi cell lipos are involved, you'd likely be better off powering the unit with it or using the weight penalty to increase existing capacity.

Rick, this is not a stupid question at all.  In fact, you're discovering "the elephant in the room" with regards to vehicle power (flying or not).  No electrical storage medium can begin to approach the energy density (joules/mass) of fossil fuels.  It's not even close.

The reason electric flight in general, and multicopters specifically, are so popular, is the relative ease of precision control.  Well, that and less noise, no fumes, no volatile chemical handling, etc.  But if you want heavy payload capability and long flight durations, a "series hybrid" is, at least conceptually, an excellent solution for a power source.

Unfortunately, I am not aware of anyone attempting this on a model scale.  On a (much) larger scale, auxiliary power units, consisting of a turbine engine and an attached electric generator, are used in most large commercial aircraft instead of large batteries for electric power when the main engines are not running.

As a concept, I think this is a wonderful idea.  As a practical matter, unless you have expertise in power generation (or lots of time and money to experiment), adding a bit more battery to your craft and enhancing the efficiency of the system is the most direct route to a longer flight duration.

Firstly, thank you both for your input on this most perplexing,and yet, - fascinating subject.

About a month a go, I was given a 50 cc 2 stroke engine from an RC that was damaged beyond repair.

The motor works great, ( the REST of the aircraft was shattered pretty badly ).

I have a place near to me that creates custom generators and alternators for cars and trucks .

They ALSO have an old guy that's been working there for years there they refer to as " Old Man Tesla " - I think it may be worth my while spending an afternoon with a cold beer or  2 and a chat with him - I'll keep you posted.

For what its worth, I've seen a duel Dymo generator set up he made for his grand kids bicycle that runs lights, indicators, tail / brake lights, a  NAVMAN GPS and an I-Pod with a pre-amp an 2 speakers. (  all using peddle power. ) - Kid's these days! !! -....whatever happened to the SIMPLE things in life like a tennis ball and racket, and a brick wall for entertainment?


Take a look at this thread.  I started by softly admonishing a gentleman for apparently expecting the forums to design a multicopter for him, and then proceeded to cover most of the macro concepts. 


To my point: there are two inescapable conclusions once you "run the numbers" on electric multicopters for a larger scale (one disk per 10-20 lbs of load range):

1) Due to Re scaling issues, a fixed-pitch electric multicopter may provide superior hovering efficiencies over a classic single-rotor helicopter with swashplate control.  This is especially true if the former has the advantage of high-lift airfoil blade sections.  I recall that thread in these hallowed forums about a German company advertising an incredible 60 minute flight time for their quadcopter.  They obviously needed to pay attention to every design detail, but the most gains to be had were in propeller design.  In my opinion, the challenges involved in getting a cyclic-pitch single-rotor chopper in that weight range to approach that efficiency would be insurmountable.


2) The energy density (joules/mass) of the best lithium Ion batteries are incredulously uncompetitive with plain old gasoline.  I haven't seen any battery technology quoting a megajoule per kilogram ratio of better than 1.0 (if even that high).  Gasoline is 47.2.  Yes, a gasoline engine is only about 20% efficient in conversion, versus a LiPoly battery which is nearly 90% (depending on discharge rate of course).  That still makes the converted rates 0.9 versus 9.44, or TEN TIMES the difference.  Of course, you need an engine to convert the gasoline into power, and they do add mass.  Adding a generator to the equation increases the mass even further, of course.  However, even if you lose 40% of the energy converting a gasoline engine power into electricity (in other words, you're incredibly dreadful at generator design), you'll still wind up with a 5X advantage.  Five times.  If you run a two-stroke engine and an extremely lightweight generator, perhaps 4 X? 

It sounds like an electric multicopter with custom rotors and a gas-powered motor/generator set might be just the thing, huh?  At least it sounds plausible while we wait for the chemistry geniuses to invent the next generation of electrical storage cells (breath holding not recommended).  When that eventually happens, we just throw the M-G set away and swap the new cells right in.  It sounds like a plan to me.

Brad, have you actually run the numbers to calculate total system effectiveness in terms of energy/weight, including the mass of the engine, the generator, the fuel system, the cooling system, the control system, some sort of short-term electrical bank (capacitor bank or small battery) taking into account conversion efficiencies, etc....

I just don't see it.  Not to mention the incredible complexity you've just added.

I'm also a bit skeptical about the idea that a fixed pitch UNGEARED propeller design could approach the efficiency of a typical helicopter rotor system (which is geared) when considering the design RPM of available BLDC motors (high RPM=inefficient).

I've seen a 900-class electric helicopter lift 28lbs of payload, 42lbs all-up weight, drawing only 40Amps!  And that was with symetrical blades, flat-bottom blades would be even better.  It flew 4-5 minutes like this.  If that 28lbs was LIPOs instead of lead weight...

I have not run the numbers at this scale, but I have for a somewhat larger application.  You're right that what's missing from the analysis are the empirical numbers for engine mass, generator mass, etc. (I am estimating from previous design exercises).  When I have the time, I'll provide specific examples.

Certainly an MG set is more complex mechanically than a battery, even with an air-cooled two stroke engine (pretty darn simple).  However, how can that be compared to a cyclic pitch helicopter with a geared motor, mechanical servos attached to swashplates, and a variable pitch tail rotor for yaw control?  I don't wish to be confrontational, but surely your definition of mechanical complexity and mine are rather different.

I've demonstrated the formula for ideal power.  What was the rotor disk diameter of this helicopter?  Let's see what the FM was.  "Drawing 40 amps" is meaningless, of course, without the voltage under load.  I have tested several blade sets, ranging from 550mm to 800mm in my own (admittedly crude) lab and could never get a FM over about 35, even with one set that was labeled (and appeared to be) of an asymmetrical airfoil.

The bottom line is the FM and how it scales.  If you get rotor blades for a conventional helicopter that are long enough and with a wide enough chord to get the Re up (discounting other dynamic issues such as pitch moments and centers of pressure which add to the complexity), then yes, there will reach a point where the variable pitch long blade will become superior, as a similarly-sized fixed blade design approach will become untenable due to rotational inertia issues.  Even then, a variable-pitched direct-drive electric multicopter will still kick the daylights out of the efficiency of cyclic pitch helicopter (although the margin will be slimmer) because of the former's ability to incorporate high lift coefficient airfoils.

I refer you to Paul Pounds, who achieved a phenomenal FM of nearly 80 by crafting fixed-pitch rotor blades precisely for low Re applications:


The caveat here is that FM is a useful comparison only if the disk loadings are the same; the larger the rotor disk area for the same weight, the lower the ideal power, meaning a longer blade will have more thrust for the same power at the same FM. 

I know what I'm saying is aviation industry heresy, as most everyone is taught the Sikorsky way of doing things.  Examine the context of these forums...the ratio of electric multicopter enthusiasts versus the cyclic pitch crowd.  Something's happening here, and there are compelling reasons for it.  We are touting something highly disruptive in the Hayden Christensen sense of the word.

But we are all here to learn, as am I.  Notice I have nothing to say about the APM, the software running it, or the basic inertial sensor technology involved.  I thought I might "give back" to this community by sharing what I've learned over my 8 years of continual research and experimentation in electric VTOL flight.  Perhaps too, it is to dispel the notion that I'm just that crazy nut who crashed the manned multicopter in his driveway...:-P

Yes, of course we need the voltage or watts, really, and I know better.  He quotes 756W, 16.8A for a hover, so that probably means 12S.

So we need to extrapolate a bit, say 12x4.0Vx40A=1920W.

When I mentioned mechanical complexity, I did not intend to compare a MG-multi to a TradHeli.  I meant to compare an E-multi to an MG-multi.  IMO, the only advantage, and the only reason that multis are used at all, are because of their mechanical simplicity.  4 moving parts. Simple straight arms.  The fact that quad guys talk about crashing 3 times in a day is funny, because I'd be lucky if I could crash my heli 3 times per month. ;)

I am open minded about the possibility that a multi can be more efficient than a trad-heli, but I'll have to see the data.

The primary harbinger of hovering efficiency is disk loading.  I assume "900 class" means a blade length of 900mm (we'll call it a 3 ft. radius)?  That's a disk loading of 42 lbs/28.3 sq. ft = 1.48 lbs/sq ft., which is not very heavy (Pounds' quad for example is more than twice that). 

So to calculate induced power at the shaft of your heavy lifting cyclic machine (it seems like deja vu from another thread): induced velocity sqrt (42/2 * 0.00238 * 28.3).  The answer is 17.66 feet/second ^-1.  42 * IV is 741.7 pound feet/sec ^-1.  Dividing by 550 yields an ideal power of 1.35 horsepower or * 746 = 1.007 KW.

Induced power at the disk is probably about 80% of 1920, and that's an FM = 65.  I have to admit, that sounds pretty darn good at 36.6 watts/pound, but imagine the same disk loading with high lift/drag ratio airfoil blades incorporating taper and twist?  

The crashing part is all about who's doing the flying and the response times.  A small quad can flip upside down faster than you can react, ergo the heavy reliance on computer augmentation.  With a large cyclic ship, you have an underslung CG which slows things down considerably.  Get something wrong with those PID values in a small quad and they can auger in quickly.

I completely agree that commercially available helicopter rotor blades are not efficient profiles.  If I could find scaled up blades like what they use on the little coaxials (which have taper and twist) we'd really have something.  But unfortunately the market is saturated with symmetrical profiles due to the popularity of "3D" flying.

My point about mechanical complexity and quad crashing is not about flying skills, or anything like that.  What I'm trying to say is that the current popularity of multis is not because they are more efficient, or fly better, or anything like that.  It's just that they are mechanically simple.  People are more willing to work with one instead of a helicopter, because helicopter mechanics is somewhat daunting, and crashes are an order of magnitude (or two!) more expensive.  It's far cheaper and easier to learn to fly a multi.

I also suspect part of the reason multis are so popular is because it is *required* to have a stabilization system to fly them, so it is socially acceptable to use one. The heli guys still seem to be stuck in this world where stabilization systems are seen as a "crutch", and their use is implicitly discouraged.  This makes it unnecessarily hard for people to learn.  

I'd bet that a lot of people successfully flying multis with a stabilization systems, also have a "rekitted" helicopter in a box somewhere that did not have a stabilization system.

This deserves its own thread.  I shall make it so.

The Case for Large Scale Electric Multicopters

Funny enough, I brought up a very similiar idea of using a stripped out Honda inverter generator in this other thread.


Being that it's a Honda, already built reasonably light once you remove all the housing, inverter and controls to just a bare engine and the alternator, one could easily build a decent power source. No, it's not as powerful as a 2 stroke, but at least it could be a start to a rather large drone with Honda reliability. Some quick searching for cheap inverters led me to this gem, a 2 stroke $299 version that surely is lighter than the 4 stroke when stripped down to just the bare minimum. 



One interesting aspect is the ratings for power. Since we know this is likely a 3 phase alternator, rectified to DC, then converted to AC, I'm thinking the DC portion is more than the 900watt rating (has to be for losses alone). I'm not sure how voltage control is implemented or if they just let it free runs say 90-180 volts DC, and expect the inverter section to handle the final output.

I take that back on the voltage, most of these have the 12 volt charging option and surely that means they are attempting to produce high current around 14 Volts output (maybe slightly more) feeding into a pretty bog standard inverter. This sounds a little more pratical if that's true.

This diagram seems to indicate how simple they are. The alternator is the flywheel.



© 2014   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service