An article by Phil Connolly:
Background the performance of an electric model is mostly dependant on the power to weight ratio. First, bear the following in mind. Any given model will need a certain amount of power for it to fly level in calm conditions. Suppose this is 100 watts. A 10% increase in power might not sound great, but the whole of the extra 10 watts is available for climbing, whereas no power was available for climbing previously. In other words, it might not take a major improvement to convert a model with marginal flight ability to one which performs quite satisfactorily.
The same principles apply regardless of the type of model, but here we will use an example of an electric glider using 7 cells. Lets assume that the current is about 20 amps for this discussion
Weight Depending on circumstances, you may or may not have full control over the weight eg if you are presented with, or already have a fully built model. If starting from scratch, then you should choose the lightest construction and equipment which will do the job. A joke amongst those electric fliers in the know is the false advice to build an electric model nice and strong so that it will survive the crash which the excess weight will inevitably cause. In some respects, electric models require less strength than other types they do not, for example have to withstand the strain of a bungee launch or the vibrations of an internal combustion motor
Servos should be of the mini or micro size. The weight difference between three 48 gm standard and three 18 gm micro servos is 90 gms. Choice of a modern lightweight receiver can easily save a further 20 gms. Altogether, in a 1100 gm model, these changes would improve the power to weight ratio by that 10% figure. You should make it an objective never to use lead ballast, so often required with gliders and this objective can usually be achieved by re-positioning the battery pack.
Other steps to save weight might be to carry a smaller receiver pack, since you can always recharge it whilst charging the motor battery pack. Receiver packs can be as small as 50 mAh Nicads, for short flights such as pylon races, but 150 mAh is a more common small size (Sanyo 150N). Compared with a 600 mAh AA size pencell, the 150 mAh pack would save another 56 gms or 5% of weight in our 1100 gm model. These days, there are numerous NiMH cells suitable for receiver packs, offering even better capacity and weight savings and, of course, powered gliders and other small electric models can use BEC speed controllers to avoid carrying a receiver battery altogether
Power and Efficiency
These go hand in hand, since a model will not fly without power, yet there is no point in consuming lots of power inefficiently. Lets take each component in turn.
The battery pack this should either have soldered copper strips to connect cells, or be end-to-end soldered. For all but the lowest powered models, the spot welded solder tags which are often fitted by the manufacturer are not suited to our use. For example a 3 milliohm resistance has been measured with solder tags 6 such tags on a 7 cell pack at 20 amps will lose 6 x 3 x 20 = 360 millivolts (over 1/3rd of a volt). This amounts to about 5% of the available voltage from the cells and would reduce the motor power by 15%, when compared with the ultimate result of end-to-end soldered cells
Of course, the choice of battery pack in the first place will have a major impact on the performance. If not limited to a maximum number of cells by competition rules, then the use of an extra cell will give a large performance improvement. For example, many a poor performance from a 6 cell model has been overcome by a change to 7 cells. The latter can still be charged from economy priced chargers.
Since the battery weight will, typically, be at least a third of the all-up-weight, there is also the question of the actual cell size. In recent years, battery technology has advanced to the point of doubling the energy to weight ratio of Nicads and NiMH cells. Thus a change to a newer type of cell, especially the latter type, could be expected to transform climb rates or motor run times
Finally on the battery pack, it is important to understand that Nicads and NiMH cells perform noticeably better when warm immediately after the end of charge than some time later. Always aim to give a top-up charge just before you fly, so ensuring that the flight gets off to the best possible start. A pack which was charged the previous day and is totally cold can be expected to give 20% less power.
Plugs and Wiring - Once again, the objective is to minimise resistance. Tests of a number of popular connectors showed the following resistance for each wire and the voltage loss at 20 amps for both the +ve and ve wires. Also shown is a suggested maximum current:
||Recommended Max Amps
|Tamiya style buggy
|2 mm gold bullet
|3.5 mm gold bullet
|4 mm gold bullet
Beware that variations will occur, depending on the actual manufacturer, the way in which the connectors are soldered and the amount of use. For example, there is a copy of the Deans ULT in either black or brown housing which has far less spring contact error and a resistance measured as 0.75 mohms. Even the genuine (red) Deans ULT plug can give poorer results if the red housing gets excessively hot during soldering since the plastic softens and the pins no longer line up accurately. The same has been found with gold bullet connectors in which the spring tension on the male plug can suffer if the plug becomes too hot whilst soldering the leads.
Carry out a periodic check with your multimeter on the 200 mV scale, measuring the voltage across each plug and socket and comparing the result with the above table scaled, of course, to match the current you actually consume. So, if you have a 50 amp current and you measure 40 millivolts across a 4mm gold bullet, then the resistance would be 50 divided by 40 = 1.25 milli-ohms and you should consider replacing the plug
Wiring two of the most common faults with electric flight installations are caused by a reluctance to shorten wires to the minimum length or the use of wiring of insufficient thickness. It should really only be necessary to have a total of 25 cm of wire to connect both the +ve and ve battery leads through a plug to the controller, out of the controller and to the motor. Indeed, some competition models can be found with only 10 cm. On the other hand, the novice builder often carries a total length nearer 50 cms and then uses a 16 swg wire thickness. Wire losses here would be approximately 200 mV or a considerable 3% of available volts (and 10% of available power)
For 20 amp operation it is acceptable to use 14 swg wire which has been measured as having a resistance of 8 milliohms per metre. A 40 cm length would then give a loss of 8 x 0.4 x 20 = 64 millivolts which is just about acceptable, but would be improved by halving the length of the wire. The importance of shorter and thicker wires is far greater at higher currents.
Controller make sure that your controller will handle the required current and, if possible, buy a unit with spare current capacity. This will not only allow you to go to higher power levels later, but will also have less voltage losses even at lower power levels. You should also take notice of the minimum and maximum number of cells specified by the manufacturer. Exceeding the maximum, could, clearly, cause the controller to fail, but use of less than the minimum number of cells may well cause it to be inefficient since the design may rely on a certain voltage to ensure that the switcher actually switches on fully.
By now, we have just about reached the motor, but first, here is a simple check which you can perform with brushed motors to test out many of the above components in one go. Switch your multimeter to the 2 volt or 2000 mV scale and clip one lead onto the -ve battery terminal (You may need to press a dressmaking pin through the battery lead and clip your lead to the pin)
Now connect the other multimeter lead to the -ve side of the motor brush gear. With the normal propeller fitted, connect up the radio and plug the battery into the speed controller. Carefully switch the motor up to full power and make a note of the voltage. Initially, it will be off the scale but, as the controller switches on fully, it should reduce to no more than 150 mV, most of which is across the controller.
Repeat the test with the meter between the +ve battery terminal and the +ve motor brush. This time, you should record under 40 mV. In theory half this is achievable, but your plugs and wiring will cause a small drop. If either figure is higher than expected, then pursue the reason perhaps your controller is not switching on fully or your wires are too long or thin?
Motor if you have a brushed motor, then a periodic check of the state of the brushes should be carried out, replacing them if badly worn. If the commutator is badly burnt or worn, then it may be possible to have it lightly skimmed (turned in a special micro lathe) to return it back close to its original condition. Your hobby shop may supply this service. If not, get in touch and we can arrange it for you. Look for sparks from the brushes/commutator while the motor is running as an indication that something is wrong
Motors have different operating points for maximum efficiency and maximum power and you should aim to operate between these two points. For example the popular 6v Graupner Speed 400 motor has a maximum efficiency at about 5 6 amps and a maximum power point at about 12 amps, depending also on the speed controller, wiring and type of battery. If you increase the propeller size and, therefore the current beyond this 12 amp point, you will actually get less power to drive the propeller.
If, on the other hand this same motor is consuming only 4 amps, then you will gain in motor efficiency and in propeller efficiency and in power by changing to a bigger propeller a triple benefit
Propellers and Gearboxes - As a rule of thumb, propeller efficiency is roughly proportional to the diameter, at least for electric gliders. Thus a change from a 7 to a 14 prop should, in theory give much better results from the propeller. The problem covered before is that you may well exceed the maximum power point of the motor.
The solution is to use a gearbox which allows the motor to run at better efficiency whilst also allowing the larger more efficient propeller In fact, the gearbox offers a further important advantage in that it allows a motor with a lower resistance to be used with the big propeller So, the 4.8 volt speed 400 could be used instead this has half the resistance of the 6 volt motor and maximum power occurs some 4 amps higher.
This whole question of motor efficiency is a very big one and, perhaps, the best advice would be to look elsewhere on the Web for one of a number of simulation models such as PCALC, ELECTRICALC or MOTORCALC. These allow you to input cell type and count, motor type, propeller size etc and will predict the results for you.
The final item to consider when looking for improved performance is right at the front of most planes. Whilst the propeller size is important for efficiency, even more so is the matching of the propeller to the task in hand. A Formula 1 Ferrari would not be the choice for towing a caravan up a steep hill, but would be excellent for top speed. A similar situation exists with propeller diameters and pitches. For top speed in a streamline model you need to have a high propeller pitch and high rpm, whilst a larger diameter, lower rpm propeller is far better for pulling a 2 meter glider into the air quickly
By way of example, the propeller diameter, pitch and rpm for one particular sports pylon model were measured as 6 x 6 (15 cms x 15 cms) and 15,000 rpm at launch. In theory, these figures should give a model speed of 0.15 x 15000/60 = 37.5 metres per sec. In practice, the model travelled 22 laps of a triangular course measuring 290 metres in 4 minutes. Allowing for overshoot of turns, this corresponds to an average actual speed of 29 metres per second which is a good match to the theoretical top speed.
Now suppose the same powerplant were fitted to a 2 metre glider, the actual speed during a climb would be no more than 10 metres per second and there would be a gross mismatch between the theoretical and actual speeds. This could be likened to trying to drive up the hill in 5th gear in a car. By changing down a couple of gears, much better results would be obtained. In the case of the 2 metre glider, use a gearbox and a larger propeller for the best climb rate, but choose the diameter and pitch ratio figures to match the expected speed. A ratio of 2:1 might be a good starting point with a 12 x 6 propeller, or a 3:1 with a 14 x 7 propeller. Once again, the motor simulation models on the Web can make this complex task much simpler.
As a final example, quite a number of 3D aerobatic models are appearing these days. This type of electric model is slow flying, but very aerobatic and can often hover like a helicopter. For this type of flight, diameter to pitch ratios of 3:1 are very successful.