Confusion over engine sizes, can someone enlighten me?

Im working on a UAV hybrid, and i have 2 engines with differing and confusing specs.

My aim is to gear down the output to around 5000 RPM to match the alternator.

The first engine, is a Kyosho GT 0.15Ci (2.5cc) rated at just under 0.5 Hp at 29,000 RPM so reduced to 5000 RPM should give me a theoretical 3 Hp or approx 2.2Kw..ish

The other is an elderly MAX OS 40, ive not found out much data on this engine, but even though from what i can find out, its 0.4Ci (about 6.2cc) but only rated at just under 0.7Hp at 12,000 RPM, geared down will give me less power.

Have i researched the wrong data, and are the facts ive found, wrong?

So from the above, which is the better engine to use?...I get the feeling it might be the Kyosho, can you please advise.


Views: 4469

Reply to This

Replies to This Discussion

@Jake: There are many reasons to favor an electric multicopter over a conventional helicopter, the most prominent ones being the far lower mechanical complexity and ability to incorporate redundancy for the latter.  Until now, the main drawback to an electric multicopter when compared to a conventional single-rotor gasser has been flight duration, and this is directly related to the energy density of the power source.

Gasoline has about 45 times the available energy for a given unit of mass than the best lithium-ion cells.  For a series hybrid there is a two-step conversion process, from chemical to mechanical, and then from mechanical to electrical.  The best fossil-fuel to power conversion comes in the form of a gas turbine engine, and the auxiliary power units (APUs) on jet aircraft use turbine M/G sets for this reason.  The bottom line here, again, is energy density.  Your car engine's peak efficiency is around 30%, and a small two-stroker might be only half of that.

Albeit some 2-stroke designs might get even better:

For the sake of the thought experiment, let's say that our power source cannot exceed 20 pounds of mass - or better yet, 10 Kg because using imperial units gets tiresome after a while.  At a total ship mass of 40 Kg, according to your recommended (and pretty close) 50W/lb formula, we'll need 4.41 KW. 

Let's calculate the energy density for a real-world LiPoly battery pack.  To deliver 4.4 KW, we're going to need some of these:

To stay within our 10Kg limit, we can afford to incorporate 7 of them.  At 37V and a total of 35 Ah, we can provide 4.41KW for 1,060 seconds (with a huge Peukert effect benefit of the doubt), giving us an energy density of .467 MJ/Kg or 4.67 MJ total.

Now let's compare that to a equivalent MG set.

So for a 4.4 KW delivery, we'll need a 6 HP motor and a suitable generator.  Here's one from HK which claims 8.2 HP, so let's assume it will do 6 for some sustained duration:

This little torquer weighs 1.7 Kg according to the listing, but that's without muffler or ignition.  So let's estimate that ready-to-run it's 2.5Kg.

That a motor can be used as a generator is beyond dispute, but I'll admit I don't have an exact analogy other than to suggest that they're going to be similar in power-to-weight ratio to this one:

Again, let's round up and say this motor used as a generator with the requisite accessories is another 2.5 Kg.  We have 5 Kg of our weight budget left for fuel.

At 47.2 MJ/Kg, there's a gross energy fuel capacity of 236 MJ. Let's assume this HK motor is dreadful at efficiency, giving us only 10% of that back in actual production (a reasonable assumption); we'd be left with a capacity of 23.6 MJ.  Let's say your estimate is correct - the generator should only be expected to operate at 60% efficiency.  That still yields a total energy capacity of 14.2 MJ.  Feel free to check my math, but that's still THREE TIMES the energy yield of the LiPoly 10Kg pack!

What if the two-stroke motor were optimized to 15% (very realistic) and the generator were really 85% efficient (again, quite obtainable)?  Now we're looking at a NET of 30 MJ - almost 6 times the real energy density of the LiPoly pack.

So, you see, if you really look at what's possible, the series-hybrid idea begins to look like a rather compelling solution.  If you back that up with a very high discharge rate electrical backup source (like a super-capacitor or lithium nano-phosphate cell pack) for emergency fail-over (call it the electric multicopter answer to autorotation), then suddenly the series-hybrid looks very, very attractive.

Certainly we should encourage research in this area.

Very nice analysis Brad.  I completely agree that there are very good reasons to favour a hybrid under certain circumstances.  Battery technology has a long way to go before it reaches the energy density of liquid hydrocarbon fuel, even allowing for the losses inherent in extracting it.

I would like to add that the efficiency of a brushless motor can exceed 90%, although motors that achieve this tend to be heavier.  I would also like to say that you are right on the money with your 85% for the AC to DC conversion.  At the risk of being accused of using DIYDrones as an advertising forum, I include the following efficiency curves for my generator power management unit.  Over 90% is possible with a sensibly chosen operating area.


People do tend to forget that the most efficient transportation devices on the planet convert fossil fuel to mechanical torque, which is converted to electrical power, and then back to mechanical again.  Admittedly, this is a very old application where power-to-weight ratio doesn't matter one whit (in fact, increased weight can be an advantage).

I give you...the diesel locomotive:

They've been around since the 1950's, and the technology is pretty well understood.  Note that the diesel engine is a V-12 2-stroker, but seriously...a series (really not a hybrid) chemical-mechanical-electrical-mechanical conversion process is not so far fetched as one would assume.  What is new are the more exotic rare earth magnet materials and cheap electronics hardware which afford far better power-to-weight ratios than a railroad locomotive.  Still, the undeniable efficiencies are there to be had with proper design.

   That's a good analysis Brad.  However, there's a lot of variables when you try to run all the way from energy density to actual output.  The theory might not translate as well as you hope.

   I took a quick look at some small gas generators and it doesn't look like there's currently anything in the realm of flying.  For example, a quick look on shows that the generators out there aren't doing well on a power to weight basis.

   Some numbers...

cc engine - KW Power / Weight (pounds) = KW/pounds

38 - 800/29 = 27.6

80 - 1400/46 = 30.4

99 - 1600/43.5 = 36.8

126 - 2000/49.6 = 40.3

   Those are numbers from Generac generators, since they had a nice table.  I also checked Yamaha and a couple of other more expensive brands and found they had similar power/weight ratios.

   Taking the best power/weight model shows that you'd have to strip about 10 pounds off of it to get 50W to the pound.  That's removing a fifth of it's weight just to get the generator alone to the power/weight we want for the entire aircraft.

   It probably is possible to strip that much weight from the larger ones, but I just don't see how you could really get it flying, let alone be efficient since you also have to add in the electric motors, frame, gas, etc..

   So there's a pretty big gap between theoretical performance and the off-the-shelf solutions available today.  It's always good to encourage research, but I don't see how you could do something similar to hybrid cars and actually improve efficiency over conventional designs.

  In theory, during stop and go driving, a car wastes 100% of it's energy input during braking.  So if you could capture all that energy with regenerative braking at 80% efficiency you could have a car that is 80% more efficient.  AFAIK this, combined with the fact that weight on the ground doesn't matter much, is the only reason a hybrid car design can work.

   Since there's no braking on a plane or copter, and weight does matter A LOT to efficiency, and aircraft are already A LOT more efficient than cars I just don't see how it could ever work.

   There's just no large efficiency waste that a hybrid design could eliminate or recapture on an aircraft.

I agree with you that there's no regenerative gain here.  But that was never the point.

Joules of energy versus mass is basic physics.  One joule equals one watt for one second.  The only "theory" involved is estimating the conversion efficiency, and I think I've been fairly conservative with the numbers.  As far as I know, there is no "off the shelf" solution for small aircraft - yet.  We certainly know there are such things for large airliners.

You're comparing generator solutions which are not designed for vehicle (and certainly not aviation) applications.  I bet those Generac units have cast iron blocks, etc.  Weight was certainly far down on the engineering priority list, favoring things like low acquisition cost, regulatory compliance, and durability.  For example, there are reasons the blower motor in your average residential furnace is only 1/3 horsepower (about 250 watts) and weighs 15 pounds.  Nobody is suggesting bolting one onto an airplane.

I think perhaps the "hybrid" aspect is confusing you.  The concept of replacing the main battery pack with an engine-generator combination is absolutely on solid theoretical and practical application footing.  That's exactly what an APU is doing in the tail of a 747, and high energy density is precisely why it's there.  It's turbine-powered for efficiency, and well, Jet-A fuel is widely available in a 747's normal operating environment.  So if we're talking about replacing a battery on an aircraft with something lighter and more powerful, and the larger airplane manufacturers are already doing it, then all it really boils down to is a matter of scale.  I provided links to just the type of scaled-down devices which would be required, and all that's left is a little integration work - something this original thread was ostensibly about.

   I'm sure the off-the-shelf generators could be reduced in weight considerably.  However, size and weight was certainly a design consideration to some extent for those small, throw-in-your-car mini generators.

   I just don't think the idea is going to fly.  It's going to take a lot of engine wattage to get the motor-generator off the ground.  That much wattage is going to take a heavy generator to produce.  I think there's too many factors unaccounted for.

   With a straight forward design of gas engine + generator + electric engine(s) you're talking about flying essentially at least 3 engines.  I think it would be even harder to get a multicopter design flying.

   I challenge anyone to come up with a design that can even theoretically fly using any combination of gas engine, generator, and electric motors.  50W of electric engine output per pound would be what I think is reasonable to fly, so it should at least come close to this.


I don't see any efficiencies in running a nitro engine in place of a battery for any reason.

I'd just like to point out, (siding with Brad and OP), that the hybrid concept is still a valid one for aircraft in general, and is actually under study ATM by Boeing and others for use in civil aviation. What a lot of folks are missing in this thread is a simple point: gasoline engines have a very wide range of RPMs at which they function, over which their power band varies quite a bit. For EVERY gasoline engine, there is a speed at which maximum output power occurs, and a speed at which maximum energy conversion efficiency occurs. The two are very rarely the same. 

It's also an agreed upon fact that fossil fuels store more energy per unit weight (have higher energy density) than LiPo or any other form of battery. Ultimately, the idea is simple: run a gas engine at it's peak efficiency constantly, and extract as much energy from it as possible. Then convert the energy to usable form through electric motors. While the hardware on the hobby level may not be capable of doing this without more losses than either electric or gas hardware now, it is possible on a much larger scale. So why not do this now, do the research, and push the market towards higher efficiency products while giving yourself a huge plus on flight time?

EDIT: Sorry, I forgot my key point. In flight, we use our motors under varying conditions: you add throttle to climb, and you take it away to descend, etc. This need for on demand power is very inconvenient in the design of efficient power delivery systems, but mostly with gas engines. What if you could take away this nuisance from the gas engine, but still garner the benefits of longer range/flight time it delivers? That's the point of hybrid.

The problem is, we're not talking about large, thermodyamically efficient diesel engines.  We're talking about lightweight, 2 stroke, really poorly carbureted, lightweight model aircraft engines.  If they achieved 10% thermodynamic efficiency, I'd be surprised.  That compared to 40+% in a decent diesel engine?  Generally speaking, a 2 stroke model aircraft engine converts fuel into noise and heat, with only a little thrust as a fortunate byproduct.

A 4 stoke engine is a bit more efficient, but not much, and then weight is greater.  At the end of the day, I can say with experience, that a Saito 4 stroke has about twice the runtime as a 2 stroke in a given airframe.  But at the expense of performance.  You don't see many 4 strokes doing prop-hangs, which is what a quad basically is.

To Dr. Lannigan, I've seen somebody marketing model-aircraft sized fuel cells.  I think you might be better off looking at them.  Neat stuff, but out of my budget.

O.S is always a more reliable and lighter motor, but if your goal is only to generate then I suggest the O.S 50 which I believe idles at about 5000 rpms

And keep in mind, the higher performance two-strokes are way less reliable than the lower performance ones.  The incidence of flame-out on the high performance ones is fairly high.  My first engine was an OS 40FP, low performance, but it just worked.  Started easy, even in freezing weather.

On my Saito 4 strokes, I use remote glow driver, I've never once had a flame-out.  Very reliable.

Regardless, for a multi-copter that can't autogyro, and can't glide like an airplane, you'll *HAVE* to have a battery backup to at least get down to the ground.

I believe the theory I proffered is solid.  That does not mean I also believe that the casual experimenter could cobble something useful together with off-the-shelf parts.

It seems that many people are starting to confuse my admitted pontifications about what's possible with the notion that I'm suggesting immediate plausibility.  The fact is, nobody, today, makes a hobby-class motor-generator set you can waltz down to your local hobby store and buy, regardless of price.

However, I do predict that day is not far away.  The applications (read market) for extended duration electric flight are exploding, and the demand is not going to wait for the next electrochemical breakthrough in battery design.  I see M/G sets of "drone" scale in a couple years, not the 10-20 years for the next generation of chemical cells to emerge (even with the untold billions of dollars being thrown into the latter).  Hey, I'll be the first to step up to invest when new battery technology gets commercialized, but for now, I'll keep assuming that such an event is a long way off.

Reply to Discussion



Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2020   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service