In practice, what is a good process for finding the best propeller for an electric model airplane? This is a lot harder than it looks.
I get a lot of emails from folks wanting to optimize the performance of their electric power systems. They want to know if they are going about it the right way. They also want to make sure they do not burn up their components in the process.
The following email does a good job of explaining the challenges. After the email, I will give some tips on how best to proceed.
Email That I Received
First I would like to thank you for the weekly editions of your tips. They have led me in the right direction numerous times. I am having a difficult time understanding exactly what to do in order to accomplish the correct combination of battery, motor, and propeller to make the plane perform.
Here is my tale. I purchased the cub from Hobby Zone with floats. It was having a difficult time getting off the water if there was no wind so I converted the 400 brushed motor using a 10×8 prop (similar to a slow flyer) to a 480 brushless 890Kv using a 9x6E prop and actually there was very little change for getting off the water and into the air. I read your article on wattmeters and purchased one. I found that the 480 with a 9x6E was developing 130 watts. I put on a 10x8E and it developed 190 watts and the thrust was much increased. The motor did not recommend that larger prop but rather the 9x6E. The 10x8E accomplished the added power I needed to get off the water easily. The motor was only warm.
Here is my question. Do I try to prop to get as close to the maximum watts (in this case it is 250) in order to have the best power? I would have assumed that the longer prop and more pitch would grab more air and is that why the wattage went up? It did not seem to affect the length of flying time but again I was using less throttle after takeoff. Is using the watts reading the proper way to measure to get close? Is that wattage correlated to thrust or is there another variable?
I am having the same issue with using a E-Flite 25 in place of a E-Flite 15 on a Apprentice utilizing the recommended props. In this case, I used a 4 cell on the 25 in order to increase the power over the 15. I haven’t tried my watt measurement yet on the 25, but I am thinking that increasing the diameter and pitch might accomplish what I did with the cub. Then I could continue to utilize the three cell if I can get enough thrust at 100% throttle.
I just want to know what I need to be measuring to accomplish (I think the thrust) and still be within the proper ranges for the motor. I always overcompensate with the ESC since I don’t like that burnt smell. It kind of spoils a good day when that happens.
The Goal is Efficient Thrust
Let me be very clear: the only goal for a power system is to provide thrust efficiently. Everything else is secondary.
Optimize the power system for the airspeed at which you are going to be flying most of the time. I call this the cruise airspeed. In this steady state situation, the thrust from the propeller has to match the drag on the airplane as a whole.
A practical model airplane also needs to provide thrust at other flying airspeeds. For example, at zero airspeed when at the start of the take off. But efficiency at this airspeed is not important, since you are only going to be in the process of taking off for a few seconds.
You also need to have excess thrust to climb and to accelerate. But again, efficiency in these situations is usually not too important.
Efficiency of a power system is output power divided by input power. You would need a series of sensors to accurately measure the efficiency of your power system during cruising flight. At a minimum measure the input power coming from the battery in watts (volts * amps). You also need an air pressure or pitot tube to measure the airspeed. Technically you also need to measure the battery temperature to gauge the loss of efficiency within the battery pack itself.
Something like an Eagle Tree eLogger can take and record these measurements. This is an off-the-shelf solution that is not terribly expensive.
Drawbacks to In-Flight Data Gathering
I understand why fully instrumenting your model airplane to collect in-flight data is not too popular. An eLogger with sensors will set you back about $150. Not too bad, but not free, either.
The bigger problem is time. It takes time to install the eLogger and sensors in your model airplane. It takes time to test fly all the different propellers that you want to measure. It takes time to analyze all the data that you would collect.
Measuring Thrust on the Bench
What most of us do instead is bench test the power system. Measuring the thrust from a propeller on the bench involves a bit of work but is very doable. The main challenge here is minimizing the thrust measurement errors.
You see, propellers do not behave the same way at zero airspeed than they do at cruising airspeed. With a thin electric propeller, the results will be unpredictable if the blade angle is more than ten degrees. The propeller in that case could be partially or fully stalled on the bench.
If the blade pitch angle is not too high, the bench results will be reasonably accurate.
Extrapolating from RPM to Thrust
It is a lot easier to measure the RPM of a propeller than to measure its thrust. Can you measure the RPM of a propeller on the bench and use that as a facsimile for the thrust it will produce at the cruising airspeed? Well, only as a poor substitute.
First, you need to be very careful with slow flyer propellers. These propellers distort at high RPMs. In fact, you could get to the point where increasing the RPM does not give you any more thrust. If you want to bench test these by measuring the RPM, stay well within their normal operating RPM range.
I have learned that low quality electric propellers distort even at moderate RPMs. They really behave more like slow flyer propellers.
Even if you avoid these pitfalls, there is no reliable way to extrapolate from a given RPM to the amount of thrust produced on the bench. Different brands of propellers behave differently. Even the diameter and pitch values printed on a propeller are often wrong.
Measuring Input Power
Using a wattmeter to measure the input power on the bench is pretty easy. But that is only the input side of the power system equation. If that is all you do, ignoring the amount of thrust produced, you are really just poking around in the dark. The efficiency of the motor and the propeller vary quite a bit. You would also be ignoring the loss of efficiency inside the battery pack, which in some cases can be quite large.
I hate to say it, but it is exactly these types of challenges that drove me to write my electric power system calculator in the first place.
The results from the calculator are only as good as the input data. I already wrote a series of articles that took you step by step through the process of measuring the motor constants. Someday I will write a similar series of articles on how to measure the propeller constants.
What did you say? That you have never heard of the propeller constants? You are wondering why propeller manufacturers never publish those, like they do the motor constants? These are all excellent questions! The answers will be in a future article.
Bottom line: relying on just a wattmeter, or a wattmeter and tachometer combination, is just not a good way to measure and increase the efficiency of your power system. I wish I had better news.
Test flying your model until the battery runs down and only using a stopwatch has value. The main challenge is matching the airspeed across test flights.
Avoiding Burnt Components
An entirely different issue is keeping the blue smoke inside your motor and speed control. That has really nothing to do with increasing its efficiency, but of course if you are trying out different propellers you are at risk of overloading your power system.
In practice, it is harder to burn up your components than it used to be. Batteries have gotten so good that staying within their C rating is usually not a problem. Most (all?) modern speed controls have thermal overload protection. If they are shutting themselves down, chances are that they will also protect the motor from damage in the process.
Regardless, bench testing the power system at 100% throttle with a wattmeter is still good advice.