thank!
Let me try it. I have the Hobbywing V3 and the V4 ESC, all with separate speed sensor interfaces. I understand that the conversion needs to get the head speed through calculation.
Yes, you need the number of magnetic poles of the motor and the gearing information. Just devide the number of teeth of the gears.
Not impossible at all. Throttle opening (manifold pressure) and rpm are direct power readings for aircraft recips. For turbines it is N2 torque and rpm (shaft horsepower). All you need is the dyno chart for the engine. It is actually much simpler than trying to calculate losses in electrical systems to come up with actual shaft power. It has been the basis of aircraft power settings just about since the beginnings of aviation.
Mmh okay. Manifold pressure seems to be a good option. Can you measure that precise enough in small model engines? Too bad, we didn’t talk about that earlier… But my helicopter is flying now and it’s doing it’s job really well. And safe: Redundant power supply of the servos, by two separated battery packs, redundant supply of the Pixhawk by these packs and also the flight battery, Mauch current sensor, pitot tube, LIDAR, camera etc. I’ve got a 10s 10.1 Ah flight battery, so flight time is also not that bad.
Another thing related to the original topic: I’ve just noticed that my RPM sensor (the one with the hall probe on the main rotor shaft) isn’t working in flight. Just on the ground. I found out that this is the case because of the servorail being activated by the safety switch. I’m using a pullup resistor with 1 kOhm. That seems to be a too high resistance. The sensor pin of the Pixhawk pulls the high-signal down to just 2.6 V (when the safety switch was pressed before). I’ve calculated an internal resistance of the Pixhawk of just 498 Ohm. Mmh, seems a bit low, but I’m sure, replacing the 1k pullup resistor with 470 Ohms will solve the problem. Just for anyone who runs into the same problem.
No… that’s not working either. Strange.
It’s directly related to throttle opening and pressure altitude with normally aspirated engines. Altitude compensated engines (supercharged or turbocharged) are different but sort of the same, since 29.92" of manifold pressure is sea level at WOT. Altitude compensated engines are normally boosted to 52" at takeoff power, 40" climb power. Cruise power is normally 25 x 30 (2,500 rpm and 30") so they maintain sea level power no matter what the altitude is.
Few model engines in RC are altitude compensated. So if you have the dyno chart from the engine merely using throttle opening at rated rpm, and compensating for pressure altitude (barometer reading) provides a direct indication of crankshaft horsepower.
It is basically making standard dyno corrections based on the fact that recip engines produce maximum power at a higher rpm, and reach maximum efficiency at peak BMEP (Brake Mean Effective Pressure) which happens at WOT at peak torque.
The power output of recip aircraft engines is very well documented based on manifold pressure for over 75 years. And all it takes is a simple pressure gauge or sensor.
Now, when it comes to electric it is considerably more complicated. What you need to know is actual shaft power. Electric motors produce peak torque at zero rpm. But they don’t produce any power at peak torque because shaft power is torque times speed. So you’re taking voltage and current readings someplace, but this is not an indication of shaft power. We have I^2R losses in wiring, battery, ESC and motor efficiency to factor out. You lose some of your energy converted to heat in the battery, but how much? You lose some in the ESC because it is a square sine inverter, but how much because it changes with load? Same with the motor. Its heat dissipation vs mechanical energy produced varies with the square of the current. So this is all very difficult to calculate in real time without sensors at every loss stage.
In theory you could dyno an electric drive and try to apply the numbers but it’s not as accurate as recip power when doing power calculations in flight because of the fact that the voltage sags, current increases to maintain the same watts, and you have an efficiency curve that must be applied.
Sorry for the diversion from the main topic. But your research on power requirement and RBS is quite interesting. I’ve tried to put RC helicopters into RBS and found it is almost impossible because they don’t have enough continuous power. Electric can’t get them to the speeds required and maintain it because they fall flat in very short order when the batteries fail to deliver the current at huge voltage sag. Piston power can get them close and maintain the speed, but can’t produce enough power to push the machine to the point where you actually run out of lift on the retreating side. The worst I’ve ever gotten one to do is slightly nose up and rubber band a bit on the aileron, which naturally slows it and brings it back to normal flight. Actually getting one to completely stall and uncontrollably roll into the retreating side is very hard and I’ve pushed them to over 100mph trying to get it to happen.
The fact that the speed is limited in helicopters compared to fixed-wings is not RBS. It is the fact that the advancing blade cannot exceed the speed of sound.
Thank you very much for the detailed explanation. That’s really interesting! Do you fly manned aircraft or why do you know that much about engines, pressure altitude and so on? I had to convert your numbers to bar, to get a feeling for it, but yes, that really makes sense.
I’ve got a supercharged 4 stroke OS MAX FS-120 Surpass in an aircraft which I once got on a flea market. Not sure if it has ever run. Beautiful thing. But yes, most of the model engines aren’t supercharged of course.
Okay, you convinced me that knowing true shaft hp is not easy with electrics either. But you can ignore battery losses and losses of the wiring to the ESC, when you measure the voltage directly at the input of the ESC, which I do. Current times voltage is power which goes into the ESC. The losses which heat the battery are just caused by the voltage drop due to the inner resistance of the battery. But you don’t have to care about it, when you measure the input voltage at the ESC. Of course you can’t ignore losses in the motor and ESC. I guess you just have to measure that. But you’d have to do that with a combustion engine aswell: A problem you have both in electrics and combustion models are the gearing losses. They’re also load and RPM dependant. I built a test stand for another helicopter to measure that. I mounted it on a rotatable plate, uninstalled the tail rotor and measured the force it would have to generate to hold the tail in position. With the distance from the pivot point you have the torque of the main rotor shaft. Multiplied with the angular velocity of the rotor you get the true output shp.
Concerning RBS: I think with modern electric helicopters you have at least the same amount (or more) of power than with piston engines. My motor has a max. continuous power of 5100 W (11 kW for 2 s), so it depends on the battery packs you’re using. There are 75C packs on the market. It’s just crazy. Anyway I’m not trying to get it as deep into RBS as possible. You can see signs for that a lot earlier. I’m looking at the cyclic pitch angle on the retreating blade side. I guess you’ll never see a model helicopter with a 2 bladed (teetering-like) rotor roll into RBS. That’s because of the gyroscopic effects. If you make a cyclic input that reduces lift on the retreating side, the helicopter won’t roll to the side either. It reacts 90 deg later and pitches up.
Of course it’s also the advancing side. The two things are connected. Otherwise you could slow down the rotor as much as you like. Speed of sound wouldn’t be a problem anymore. But then you get into RBS on the retreating side.
Yes, I am a Part 61 commercial pilot, both fixed and rotor wing.
You will have more power with an electric, but only for a limited amount of time. Even with 75C packs the voltage sags very quickly and things start getting very hot within seconds sending efficiency into a downward spiral. Which makes the power measurements skewed. A two-stroke engine OTOH can be run at wide open throttle continuous without hurting it so an aircraft can be flown on a test circuit at continuous high speed to gather data. Because of its duty cycle compared to electric, I’ve found higher continuous speeds are achievable with piston (or turbine) power. To use an analogy, in a drag race to 200 mph the electric will win. In the Daytona 500 at a continuous 200mph for 500 miles, the pistons are going to win.
This is where I think Sikorsky holds the edge over the EuroCopter x3 concept in developing a high-speed helicopter. Sikorsky went with a rigid coaxial rotor system which mitigates the problem with RBS and loss of control response as the main rotor is slowed in high speed flight, pushing speeds beyond 250kts. There is a limit to how slow you can run the rotor and still get acceptable control response with a single rotor helicopter.
So your research is quite interesting to me, would enjoy hearing more about it as you proceed with gathering data.
Ah, I knew it! You can hear, that there is a loot of background…
I totally agree with that. Thought you were talking about short pitch spikes to accelerate. If you are trying to hold the speed over a longer period of time you’re going into the downward spiral with electrics, that’s true. I think, I’m going to set up waypoints to accelerate and then fly in a straight, horizontal line for something like 10 seconds to do my measurements, then turn around, keep speed and fly back to be sure, there is no significant wind influence in my data and that’s it. That should be doable with an electric helicopter.
I’m not sure about that. It’s definitely a good thing to have a coaxial rotor to counteract dissymmetry of lift. But on each rotor disc you’re getting very different loads. The advancing side of the lower disc is directly under the retreating side of the upper disc. So you have to build the blades and the whole rotor system incredibly stiff to prevent the blade tips touching each other. I’m not sure, how far you can push the boundaries there.
Of course X3 has the problem of low control response with the slowed single rotor, but I think they’re using aerodynamic control surfaces on the wings and tail to solve that. Besides the wings help unloading the rotor in general, so lift is produced much more efficient. Oh and a rotor in high speed flight is always source of a lot of drag. So having one is more than enough I guess 
By the way: Airbus Helicopters is already working on the next generation of the X3-concept, the RACER:
Pretty interesting time right now. I’m really looking forward to see, where that’s all going…
“So your research is quite interesting to me, would enjoy hearing more about it as you proceed with gathering data.”
Thank you very much! I’d be happy to keep you informed. But I guess, I should write you a PM, because there are some things which I can’t talk about in public yet.
To experiment with RBS I recommend starting at a high altitude and place the helicopter in a 15 degree nose down dive. You will have to fly it manually. The autopilot does not understand helicopter flight dynamics. The autopilot was designed for multicopter. It does not understand blowback or RBS. It does not understand helicopters fly like airplanes albeit the wing is rotary instead of fixed.
If you try it in level flight it is going to take a lot of continuous power to accelerate to Vne. Your batteries will go flat before you get there. In a dive it is achievable, but the autopilot being designed for multicopter does not do a dive to build airspeed. I’ve tried to program the autopilot to do ag turns with a helicopter and it refuses to do it - we need a TOTALLY different attitude controller to fly helicopters properly.
Russian engineers are the world’s foremost authority on coaxial helicopter design. It is interesting that coaxials are actually more power efficient at the same weight than a conventional single main with tail rotor. MiL Helicopters has been successfully designing, building, selling and flying coaxial helicopters for better than 50 years and they are some of the highest-performance helicopters on earth.
Flying it manually is just the thing which I don’t want to do. I could use every normal flybarless helicopter for that. But in manual flight I’ll never get repeatable conditions, which I definitely need for comparison of the measurements. It’s not necessary to actually reach RBS. The dissymmetry of lift is starting to build up beginning at zero speed. I’m trying to see that in the cyclic outputs of the flight controller in relation to flightspeed.
Yes, it is. They save power because they don’t have a tail rotor. That’s clear. But what they also do is saving power, because they deswirl the downwash. So they get a larger amount of thrust out of the same torque and RPM. But of course complexity is always high.
Felix,
Include me on that PM. I would also like to hear more about your research.
You’re very welcome!
I would think you would want to collect airspeed. Unless you conduct your flights on calm wind days. Are you going to put an airspeed sensor on the aircraft. I have miodified arducopter to allow use of an airspeed sensor. I could do a special build that supports the use of an airspeed sensor for you.
Yes, knowing the correct airspeed is essential in the project, so I definitely planned to put a sensor on the aircraft. It would be awesome, if you could make me a special build for that. Thanks so much for offering that!!
I just solved the problem with my RPM sensor:
I thought it’s a problem with my external pull-up resistors. But it’s not. First thing is: Pixhawk 2.1 seems to have internal pull-ups and I’m using it without external ones now and it’s working fine. But what solved the problem was using another input port. It seems that only AUX 5 and AUX 6 are working correctly as RPM-input pins. (I’m using them configured as Type 2.) AUX 1 – 4 are also working, but only until the safety switch is pressed. After that the RPM signal on the ground station shows just “-1”.