Setting Up for Electric Flight

I've often been asked to recommend the "best motor" for a particular electric model, or to replace an IC engine. Sometimes I can suggest a setup I've used previously, but more often I'm left trying to explain the basic rules I use to select a setup. Invariably these explanations end up with too much math and blank looks from the enquirer who then heads to Arkwrights Model Emporium to purchace the "easy option" recommended IC engine for their model.

So here we go, I'll try to explain everything step by step. Hopefully this will help. I will sometimes break my own rules, but if you want to convert an IC model to electric this is the best advice I can offer.

I've chosen to explain, step by step, how I'd go about choosing a setup for a Seagull Models Jumper 25. This is a model I'd recommend to anyone looking for an electric trainer. It may seem like cheating, since I've already set up a couple of these but it's interesting that, by strictly following my rules, I've come up with a different setup from my previous builds. I intend to build a Jumper with this setup, I think it will be the best yet.

The first thing I want to know is the models stall speed:

"A plane's stall speed in mph is approximately equal to 4 times the square root of the wing loading in ounces per square foot. This rule applies to both models and full scale aircraft." from stefanv.com

So the Seagull Jumper 25 trainer has a flying weight of 5-5.5lbs (80-88oz) and a wing area of 499 sq.ins (3.47sq.ft) (information from the box lid) Therefore the wing-loading will be:

weight/area -> 88/3.47 = 25.4oz/sq.ft

so the stall speed is about: sqrt(25.4)*4 = 20mph

Motor Power:

For reasonable performance, a plane's power system should draw about 60-70 Watts per pound (W/lb) of model. For reasonable aerobatic capabilities, 100 Watts per pound is more suitable. These figures are in terms of motor input power (Watts = Volts x Amps), and assume a motor efficiency of about 80%. (you might have noticed that Stefan's site uses lower figures, I find these numbers more reliable and leave a little power in reserve)

Back to the Jumper - for reasonable training capabilities we could select a power to weight ratio of 65W/lb so we need a motor that will cope with:

65*5.5 =357.5 Watts

In the interests of allowing a little extra power in reserve we could round this off to 400Watts, if you want to provide enough power to advance to aerobatics we'll aim for 550Watts (5.5*100).

For the Jumper I've had a look about and I've identified a Turnigy C35-48-A motor as it will handle the power I need with room to spare - it is rated to 55Amps and 700 Watts and has a Kv of 1100rpm/v.

Battery Selection

Battery voltage and capacity will dictate the flight time. Current draw at full power will be Power(W) divided by Volts(V). To provide reasonable flight times we probably want to aim for about 8 mins at full power (probably about 10-12 mins normal flight). When calculating this we should use a voltage of 3.3v per lipo cell, not the nominal voltage of 3.7v since the voltage will drop under load.

In the case of the Jumper using 3 cells:

Current Draw = Power/Volts = 550/9.9 = 55.6amps
Discharge Rate = 60mins/8mins = 7.5C
Required capacity = Current Draw/Discharge Rate = 55.6/7.5 = 7.4A/hr or 7400mAh

Using 4 cells:

Current Draw = Power/Volts = 550/13.2 = 41.7amps
Discharge Rate = 60mins/8mins = 7.5C
Required capacity = Current Draw/Discharge Rate = 41.7/7.5 = 5.6A/hr or 5600mAh

The 3 cell pack isn't really a practical option, for the simple fact that I can't find a pack with anything like that capacity, also it is borderline for the current handling of the motor I've chosen. Using 4 cells and a 5000mAh capacity is an option, the Turnigy 5000mAh 4s 25c Pack will have enough capacity and the high discharge rate (25c) will ensure that we have good voltage stability.

We could even step up to 5 cells. The advantage of increasing the voltage is that the current drops so we need a smaller capacity pack and we can use a smaller speed controller (more about speed controllers soon), however we need to make sure that our motor has a low enough Kv to allow us to use a suitable prop (more volts means more spin speed, so we need to spin a smaller prop to deliver the same power)

Prop Selection:

"The pitch speed should be about 2.5 to 3 times the aircraft's stall speed. Pitch speed in miles per hour is equal to rpm x pitch x 0.000947 (pitch in inches). For 3D models the value would change to 1.5 to 2 times the stall speed." from stefanv.com

We need to have started looking at the motor and batteries that will suit our needs to select the right prop.

I've already chosen a motor and battery, we'll use the data from that to find a suitable prop.

speed = rpm*pitch*0.000947 -- so -- pitch= speed/(rpm*0.000947):

we are looking for a pitch speed of 50-60mph (2.5-3 times 20mph)

rpm is Kv*volts*0.9(for motor loss) = 1100*13.2*0.9 = 13000rpm

pitch = 55/(13000*0.000947) = 4.47

Looking through the available sizes of APC electric props, I can get a 4.5 pitch with a diameter of 5" or 9", if I go to 4" or 5" pitch I have a wider range, but the 9x4.5 seems like a reasonable option. I could look for a motor with a slighly lower Kv rating and use a coarser pitch

There are a couple of ways to verify the size :

The first option is to buy everything, stick it together and measure the power with a Watt Meter

or if you don't want to commit yourself to making a purchase just yet, download drivecalc and key in your selections to see the result. I've run my choices through and here are the results:

I had to make some substitutions as the motor and battery I want aren't in the database, I selected a 4500/30c battery and the axi 2826-08 motor (which the Turnigy claims to be equivilent to)

 

The numbers are looking good. 510 Watts is ideal, not quite the 550 we hoped for but still respectable. 78.4% efficiency isn't far off the 80% assumption we made to start with. Notice that Vopt (74Km/h - 46mph) is not the same thing as pitch speed.

It has also given us some additional information to consider:

• Total weight = 695g - we can use this to more accurately estimate the true flying weight by weighing the model and radio gear and adding this weight to it.
• Flying Time = 6:53 - close to my 8 minute target, and remember I used a 4500mAh pack in the simulation, I'll use 5000mAh in practice
• Static Thrust = 2087g (4.5lbs) - a figure of 1/2 the flying weight would be sufficient, this setup should provide plenty of thrust to take off from a short grass strip.
• Current = 35.9A - we can select a suitable speed controller from this info.
• Looking at the graph we can see that the setup is close to peak efficiency which can't be a bad thing.

Of course, I substituted components for the calculations. I could just buy the items I used in the calculator, but I'm reasonably confident that the figures will be close enough using my original choices. I also tried a 9x6 APC prop, this incrased the power to 607 Watts, but Flying Time dropped to 5:45, we also loss prop efficiency as our pich speed would be approaching 4x stall speed.

Looking at Alternatives:

Using the same methods I idendified a Turnigy 42-40B 900Kv motor which would work with the same battery pack and a 11x5.5 prop giving similar results but slightly lower thrust and a 68.1% efficiency. No doubt I could go on looking for days, but I don't think I'll get much better than my 78.4% efficiency.

Speed Controllers (esc)

We need to decide whether or not we intend to use a seperate power supply for the receiver. For models over 2lbs I would suggest that a seperate receiver battery is used, below 2lbs weight will probably be critical and we need to ensure we use a ESC which has a BEC (battery elimination circuit).

• BECs supply a constant 5volts to the receiver eliminating the need for a seperate supply.
• Opto ESCs have no BEC and isolate the motor circuit from the radio system by using an optical relay, this reduces the chances of electrical interference from the motor circuitry.

ESCs are sold according to their current rating, it is best to use a ESC with at least 50% higher curent rating than you expect the motor to draw. This 'overhead' prevents stress on the components and will provide more reliable operation. It is also important to check that the unit supports the voltage you will use.

For the Jumper my initial calculations suggested 41.7 Amp peak current, DriveCalc returned 35.9 Amps. I'll work on 40 Amps, with a 50% overhead this means I'm looking for a 60 Amp controller that can accept 4 lipo cells.

I found a 60Amp controller that will handle 4-8 lipo cells, however, for an extra few pounds I can get a 90Amp unit that will work with 2-7 cells and weighs 22g more. The weight penalty won't be an issue in a model the size of the Jumper, so I'll opt for the bigger unit and the extra safety margin. The flexibilty of being able to operate on less than 4 cells mean it will be more likely to be useful in future projects too.

Final validation:

We've done all we can to select the right components, if possible try to talk to someone who's used a similar setup. Once everything is installed the watt meter mentioned earlier should be hooked up and all the figures checked.

I built the model using the parts I selected, here are the results on the Watt Meter:
   

520 Watts, 35.4 Amps and a minimum voltage of 14.44 Volts. Thats pretty good for a computer simulation!

Of course, the real test is sitting the model on the runway and winding up the throttle. Good luck!

A quick update - After a season of flying it is agreed that, as predicted, this is the best electric Jumper conversion so far. A great trainer with a comfortable 15 mins training time per battery and also the fastest when the throttle is opened up