Fullscreen HTML5Med Res (??MB)HTML4

We build and test an electric ELECTRIC FLI Electric powered model aircraft are becoming very common – but they do present pitfalls, not just for the beginner but for the experienced modeller as well! In this article we review the electric scene and build and test an electricpower Piper Cub. How did it go? Read on. M uch has changed in the world of electric-powered, radio-controlled aircraft since SILICON CHIP last visited the subject back in 1992. If readers may pardon the play on words, electric flight is undergoing a quiet revolution; a revolution so radical that the economic viability of internal combustion (IC) motor manufacturers must surely be under threat. This startling transformation has come about as a result of a number of electric flight technologies coming of age almost simultaneously. Of these developments by far the most important are: 12  Silicon Chip [1] Application of Rare Earth magnets to brush motors [2] High power, low-on-resistance FETs [3] Microprocessor-controlled smart speed controllers and smart chargers. [4] Brushless motors [5] Battery technology improvement. Prior to 1992 electric flight was in the hands of a small group of dedicated contest flyers. Today’s electric flyers owe this group a huge debt of gratitude, for without them electric flight would be nowhere near as advanced. It was this group and in particular, Peter Blomart of Belgium, that established the internationally recognised class of F3E competitions in 1986. Since those early days of primitive hand made soft start switches and analog electronic speed control (ESC), progress has been staggering, to the point where the modern microprocessor speed controller can now distinguish between brush and brushless motors and configure itself accordingly. International R/C aerobatic competitions have long held their place as the most prestigious R/C events. While traditionally dominated by IC motors siliconchip.com.au Piper Cub IGHT By BOB YOUNG of ever larger capacity and power, it is increasingly common for electric powered models to snatch places from the IC brigade in these showcase events and it is here that the real threat to IC motor manufacturers is developing most rapidly. The radio-controlled boat boys have also been hard at it. Currently the fastest R/C boat in the world is electric, with a speed of 120.7 mph. In Australia, Ray Cooper of Victoria set a world record for electric-powered models in the distance to goal and return class with a flight of 54.3km (108.6km total), lasting 1 hour 22 minutes. siliconchip.com.au Across the world, R/C flyers are scrambling to emulate their international heroes. Manufacturers of electric motors are springing up like mushrooms and battery manufacturers are continuing to confound, with batteries that are lighter, with more capacity and higher cell voltages. Manufacturers of the chargers for these batteries are hardpressed to meet demand and so the bandwagon has been set in motion. With all this going on, the time has come to review this wonderful world of ultra mobile electrons in the most practical way possible: building and flying an electric powered model suitable for park flying on those quiet, wind-free evenings that are an R/C modeller’s special delight. World Models Piper J-3 Cub EP The kit chosen was selected for several reasons. Small enough to qualify as a park flyer, it is simple to build and fly and is reasonably priced. The kit includes a geared, brush motor (Speed 400) and thus can be controlled by a simple and economic ESC. It also uses a genuine lightweight 4-channel R/C system providing four proportional channels. However the real reason for the choice of this kit was that I have had a soft spot for Piper Cubs for a very long time. It is a very pretty aircraft, easy to fly and is one of the nicest aircraft for take-offs, landings and especially “touch and goes”. If the reader loves to watch a graceful aircraft land and takeoff then there is no better model than the Cub. This emotional approach to the kit purchase was to have a dramatic effect on the ultimate outcome of this whole project but more of that later. The Cub does have one small vice and that is adverse yaw during an aileron-only turn. This is largely as a result of the flat bottom wing section (Clark Y) and poor aileron design. To turn an aircraft, the aileron on the inside wing must be raised at the trailing edge and the outside aileron depressed. This reduces the lift on the inside wing tip and the wing starts to fall due to the unbalanced lift distribution. However with the reduction in lift, the drag at the inside tip is also reduced. Conversely, when the aileron on the outside wing goes down to lift that wing tip, the lift shoots up dramatically, as does the drag. The result is that the aircraft rolls in the direction of the turn but the nose is pulled around in the opposite direction by the badly unbalanced drag forces at each wing tip. This gives rise to a very awkward situation known as “adverse yaw”. However, there are several tricks that will help improve the flying characteristics of the Clark Y type wing. These include heavy differential aileron movement (more up than down) and coupled aileron/rudder mixing on either the transmitter or in the model (ie, turning using rudder and aileron together). Full size designers may resort to differential movement and designs such as the Frieze Aileron, which uses a complex hinge that allows the leading edge of the up-going aileron to protrude into the slipstream underneath the wing, thereby increasing the drag on the inside wing and balancing overall drag. However do not lose sight of the fact that these are only patches and they introduce other problems such The kit as she comes, straight out of the box. All of the difficult model work of past years – wing and body shaping, etc – is already done for you! February 2006  13 11.4V, 1800mah 3 cell Li-PO battery left rear. 3 cell ESC left front and two sizes of brushless motors. Two of the ultra mini servos fitted to the Cub. They’re significantly smaller than the servos you’re used to . . . as spoiling the rolling characteristics and reducing aerodynamic efficiency. As with all fixes, the real answer is in the initial design of the aircraft. The prime rule should be no flat bottom wings with simple centre-line aileron hinges on sport aircraft. Scale aircraft are a different matter. Why kit designers insist on using a flat bottom wing with simple centre-line aileron hinges on training and sports models is absolutely beyond me. They must subscribe to the theory that if you can fly this sort of trainer you can fly anything. They really are unpleasant to fly if the necessary precautions are not taken. Even a wing with a moderately curved underside (Semisymmetrical, eg, NACA 2415) will completely transform the flying characteristics of any model with a flat bottom wing section and may almost completely eliminate adverse yaw. While this is a diversion from the main topic it has been covered in detail because it is important for all tyro R/C modellers to understand the effects of flat bottom wings. It is the only factor that may spoil the delight of flying this really nice model. Nowhere in the Cub instructions does it warn of adverse yaw or mention the above precautions which is a pity as otherwise the Cub is a good value kit. The above applies to all R/C model aircraft, so do yourself a favour when purchasing your next trainer or sport model. Look for a model with a symmetrical or semi-symmetrical wing section. ings. No mention is made of the type of servos the kit was designed around and so the servo trays fitted in the wing were unsuitable for the servos used in this model. They had to be cut away and new ones fitted. Watch out for the cross-brace at the bottom of the servo well. This is actually a little pull out handle attached to the fine cotton pull-through used to pull the servo lead through the wing tunnel to the wing root. It is best to remove the servo connector and solder a 3-core ribbon cable extension lead long enough to reach the wing root plus an extra 75mm. When the wing halves are joined then splice the two 3-core cables together with a servo lead about 100mm long. This single servo lead then plugs into the receiver aileron socket. There is virtually no way that a servo connector can be threaded through the wing tunnel so conventional servo extensions and “Y” leads cannot be used. Remember here to set the aileron servos in such as manner that ailerons move up the recommended 10mm but contrary to the instructions only go down about 2mm, not 10mm. This can be achieved mechanically by offsetting the servo arm as shown in the relevant photo or by using the transmitter settings in computer radios. The rest of the wing assembly is straightforward enough with the exception of one final point. The wing uses a straight spar to join the two halves. It is a good idea when gluing the two halves together to place a moderate weight on the centre, giving the wing a very small amount of dihedral. A dead straight wing on a high wing model tends to Assembly The Cub is one of the new breed of kits called ARF (Almost Ready to Fly). This means that the manufacturer has already done all of the hard work. The model is almost completely built and fully covered in plastic shrink film when it comes out of the box. All that remains is some detail work and the installation of radio and motor. Thus for the enthusiastic tyro, building will typically take around 10-15 hours. The following is not meant as a detailed how-to of assembly, merely a guide to point out some of the difficulties and shortcomings in the kit and to help anyone building this model avoid the pitfalls. Assembly begins with the preparation of the two wing halves and it is here that the only real problem in assembly was encountered. The instructions are very poor and consist merely of photo sequences and a few odd draw14  Silicon Chip Engine Room with cowl removed showing the Speed 400 motor and gearbox as well as the folding prop. siliconchip.com.au look as if the wing tips are drooping and the slight dihedral eliminates the droopy look. The remaining assembly is routine but with two points of concern. The hinges used are sheet Mylar. Glue only one side into the model using super glue. Do not allow superglue to get across the hinge line, as this will make the hinge stiff and brittle. Do not attempt to coat the hinges on the control surfaces with super glue and then slide them into place. The glue goes off immediately it touches balsa. Instead, fit the control surfaces and then drill vertical holes down through the control surface and hinge ready to accept a normal household pin. Wick super glue down the hole and then slide in the pin. Nip off the pin flush to the underside of the control surface. Hinges should always be pinned into place. Do not rely on the glue to hold them. Many a good model has been lost in this manner. Lastly, the elevator and rudder pushrods are blued steel and they slide into long Mylar tubes, giving a fit quite high in friction. To avoid this, before inserting the pushrods into the tubes, use a piece of sandpaper to remove all of the blueing to take the surface back to bright steel. Then coat the pushrods with graphite lubricant before finally inserting them into the tubes. Motor and radio installation Conference and Tradeshow Theme: “Meet the Pioneers” & Update your skills Queen Victoria Museum and Art Gallery Precinct - Launceston, Tasmania Australia 2-4th March 2006 Sponsorship opportunities available now GUESS WHO’S COMING TO THE SHOW! Live from Launceston, Tasmania nication satellite commu V DrDish<at>T rldwide TV the unique wo mote channel to pro Come and be a part of the audience and ask the hard questions The conference program will also feature seven of the industries top keynote speakers Ideal program for new comers as well as experienced industry professionals. Make your way to Launceston Download these documents from Conferenceplus.com.au/satellite2006/ As stated previously the Cub is supplied with a geared Speed 400 motor and a folding prop. Fitting the motor and associated electronics is very straightforward. The folding prop is a curiosity, as few people would attempt to soar this sort of model. The folding prop is designed to reduce drag during gliding flight when the motor stops. An excellent range of wooden e-props designed especially for electric motors is now available and this model would no doubt benefit from one of them. The folding prop does have one advantage though: resistance to breakage in bad landings! The motor is already fitted with suppression capacitors, 100nF (0.1uF) across the two motor terminals and 100nF from each terminal to the motor case. A simple, economical speed controller (ESC) was chosen and wired in to place. Be sure to follow the polarity instructions carefully or on the first take-off you will be run over by an aeroplane flying in reverse! The modern breed of speed controllers, designed for use in conjunction with lithium-polymer batteries, are very interesting. They feature a mandatory low voltage cutoff to prevent the Li-POs falling below 2.4V per cell and destroying themselves. They also feature a BEC (Battery Elimination Circuit) to provide receiver and servo power. Be sure when choosing a speed controller (the kit does not provide the ESC) to specify the number of servos to be operated from the BEC. The Cub uses four servos. The ESC uses a microprocessor to control all of these functions. One additional feature of micro-controlled ESCs is a degree of input pulse monitoring. This checks the in- Engine Room with cowl removed showing the Speed 400 motor and gearbox as well as the folding prop. Radio Room. Note servos mounted but push-rods not yet fitted. siliconchip.com.au Exhibition space on sale now - Hurry it’s selling fast Register your interest to attend • Exhibition Booth Application • Register to attend on-line Event Managers: Conference – Plus PO Box 1144 Legana Tasmania 7277 Phone +61 (3) 6330 1444 Fax + 61 (3) 6330 2190 email info<at>conferenceplus.com.au February 2006  15 coming signal for noise and should the input pulse count exceed the specified number or change pattern dramatically then the ESC shuts down the motor drive. Another very good safety feature is that the ESC cannot operate accidentally when turned on. To function correctly the ESC must first be set to LOW throttle and then the ESC is armed and ready for use. Be careful on the first powerup, as it may be that the throttle channel is reversed with low at the top of the transmitter (TX). In this case use the TX channel reversing function to set LOW throttle with the stick down towards the bottom of the TX. Now we come to the electronically juicy part of this saga. Range reduction Our experience with the analog speed controller presented in the 1992 articles indicated that there will be a loss of usable transmitter range with an electric motor running, as against motor stopped. Our testing in 1992 confirmed that this range reduction would be in the order of 10-12%, due to the noise produced by the commutator and brushes of the motor. So we decided that anything in excess 85% of the engineoff range is an acceptable figure, with the ranges available from modern receivers. So imagine our surprise at the field one perfect spring morning for test flying, when we were confronted with 15% of the radio range on the throttle control. All of the flying Fuselage complete except for decals. Note the folding prop. 16  Silicon Chip No, it wasn’t a hard landing which tore the wings off . . . here the Cub is almost complete: motor, servos and radio installed and wings ready for fitting. controls worked well at the normal range but the throttle would run up to speed and cut off at about 15% of normal range. Obviously it was back to the drawing board. What had happened? Firstly in spite of my many years experience in R/C flying, the most fundamental rule of all had been ignored. That is, test everything at home before leaving for the field and that includes a retracted TX antenna range check with motor running and motor stopped, even though the motor running test is difficult to do on your own. This is especially true when preparing for any model’s first flight. So what had happened? Firstly, I’m an old power hound who likes models with only two speeds, stop and go very fast, and who believes that too much power is better than too little (you can always throttle back), so the Cub was fitted with an 11.4V 3 cell Li-PO battery. Thus the rating of the 8.4V Speed 400 motor was exceeded by the extra 3V. This is OK if one is prepared to accept abnormal motor wear. However the Speed 400 motor does not have good brush design and it arcs quite badly unless tuned properly. An old trick here was to time the motor in a darkened room for minimum arcing on the brushes. Clearly then the extra 3V was elevating the motor noise and this was getting into the speed controller and shutting it down via the pulse counting safety circuit. So we fitted a reverse-biased Schottky diode and a 40V varistor across the motor brushes as recommended in the 1992 articles but with little improvement. Where to from here? Change the receiver perhaps? The noise path was most likely coming down the receiver antenna and then through the receiver. Ultimately four brands of FM receivers were tried and located as far from the ESC and motor as possible; three imported brands and the original Silvertone. Three brands of TX were tried as well. These gave very little improvement and the results were still not acceptable for successful flying. It was then decided to change the operating voltage, reducing it to a 2-cell Li-PO battery delivering 7.2V. Unfortunately this introduced a complication in that the economical little speed controllers are set for the number of cells to be used. There is no built in facility for cut-off voltage adjustments. So a change in battery voltage called siliconchip.com.au Close-up of tailplane assembly showing control horns, pushrod connections and steerable tailwheel. Close-up of fin assembly showing the steerable tailwheel anchor/bearing plate. siliconchip.com.au www.elexol.com www.elexol.com www.elexol.com www.elexol.com www.elexol.com www.elexol.com and visibility problems, the last thing needed is a glitch or two to add to the misery. The result may very well be a smashed or lost model. I have four much flown and much cherished models approaching or exceeding 30 years old. One even has 30-year-old servos still fitted and functioning. The key to this sort of longevity is constant and effective servicing coupled with well-built airframes and a rigid pre-flight test procedure. This includes not accepting any shortfalls in performance. Models take too long to build and are too expensive to treat in an off-hand manner. Even the modern ARF still takes a lot of time to prepare ! W E ELEXOL Ether I/O 24 N www.elexol.com www.elexol.com www.elexol.com www.elexol.com UDP/IP-controlled digital input/output module featuring three 8-bit ports with 5V level signal lines. Each of the 24 lines can be independently programmed as either an input or an output. Connects to any TCP/IP protocol network. CT VIA CONNE TO TALK R ROUTE INTERNET TO ANYEVICE! D Supports ARP, BOOTP, DHCP, ICMP and UDP/IP protocols Standard 10BaseT Ethernet Interface; RJ45 connector 24 independently programmable signal lines with configurable CMOS, TTL or Schmitt Trigger thresholds and programmable pull-ups per line Easy connection by three 10-way box headers On-board 50MIPS flash micro-controller may be reprogrammed to suit specific applications Integrated switch-mode voltage regulator allows power supplies from 8-32VDC User 5V 500mA output to power external interface boards or sensors Compact module – 72 x 72 x 24mm Great range of I/O-24 peripherals, too ... CONNECTOR/ LED BOARD Provides screw terminal connections plus optional either pull up or pull down resistors and LED port status indication. OPTO INPUT RELAY OUTPUT SWITCH/PUSH BUTTON BOARD BOARD BOARD Provides 8 Opto Isolated inputs for each of the I/O-24 pins with LED indication. Provides 8 isolated relay contact outputs suitable for a variety of loads. Provides 8 switch/ push buttons with LED indicators showing the status of the I/O pin. Visit our web shop <at> www.elexol.com Elexol Pty Ltd Ph: (07) 5574 3988 Fax: (07) 5574 3833 (PO Box 5972, Bundall, Qld 4217) www.elexol.com www.elexol.com www.elexol.com www.elexol.com www.elexol.com www.elexol.com www.elexol.com www.elexol.com www.elexol.com www.elexol.com for a new ESC with a low voltage cut-off exceeding 4.8V. After fitting the new ESC and battery the same tests were conducted with the FM receivers. The range was a lot better (50 –60% depending upon the FM receiver used) but still not up to the 85% figure adopted originally as the acceptable standard. Finally in desperation a Silvertone AM receiver was fitted and this gave the desired range. The model was at long last ready for flying. In conclusion, then, what had caused all of the problems and what had fixed them? Firstly, had I used my head instead of my heart and purchased a kit fitted with a brushless motor and a more advanced ESC perhaps things may have been different. That is another story of course and we may yet see the Cub fitted with such a system. However, the exercise was to learn about the modern electric flight systems and how to make them work in spite of whatever shortfalls there were in that equipment. People do buy with their heart and especially with their pocket in mind. The Speed 400 is a common motor in kits and the economical little ESCs are very attractive to beginners feeling their way into electric flight. The problems began when the recommended motor voltage was exceeded. Not clearly understanding the operation of the safety circuit in the ESC compounded the matter. It took time for us to realise that it was the safety circuit cutting off the motor. The FM receiver performing poorly against the AM receiver is easily understood. The powerful AGC on the AM receiver keeps the receiver in a less sensitive state for approximately 80% of its range. Thus noise-induced spikes may be much reduced in AM receivers, depending upon the nature of the noise. AM receivers in spark ignition and electric models can often give the best results. Again, this is a trial-and-error process. Perhaps re-timing the motor would have reduced the spark noise, but that is not a job for inexperienced modellers. It was dropping the voltage that reduced the interference most dramatically. Maybe 50-60% of the range is acceptable for some modellers. These are, after all, small models and cannot been seen clearly at long ranges. However, beginners often let their models get out of hand at times and they can very quickly be blown down wind a great distance. For a beginner struggling with wind February 2006  17 Aileron servo mounted in wing. Note the offset on the servo arm at neutral to provide differential movement of the ailerons (for non-computer radio control systems). for flight. Too much to just squander with a casual attitude. Beside this, there is the safety issue to consider. An out-ofcontrol aircraft may be a health hazard. So think carefully about your decisions to fly or not fly. You can always come back another day – if you have a model to fly that is! Flying the Cub As expected, the model flew just like a Piper Cub, looking as pretty as a picture. Also as expected, despite the differential aileron built in during assembly there was still an excessive amount of adverse yaw during aileron turns. In the course of trimming the model, Coupled Aileron/ Rudder (CAR) will be called up in the TX program with a switch to enable/disable the CAR in flight. CAR is a function in the TX program whereby a small amount of rudder control is mixed into the aileron control to help initiate the turn and hold out the adverse yaw. The switch is desirable for aerobatics, to switch off the CAR in flight during rolls etc. The radio worked perfectly with no sign of interference from the motor in flight. Take-off power with the 2-cell Li-PO battery was marginal but once airborne there was ample power for climb and cruise. There would be no hope of getting off even a smooth grass strip. We used a tarred road for take-off. Take-offs on grass and aerobatics would definitely call for a 3-cell Li-PO battery to be fitted. The real surprise was the lack of down thrust. The model climbed on full power and dived when the throttle was cut. At least another 3 degrees of down thrust will be required to correct this effect. This will be a real pain to retrofit as the motor slides snugly into holes in two bulkheads. To use a flat bottom wing-section is a tragic error but to provide a pre-built fuselage with the incorrect thrustline is unforgivable. Why do kit manufacturers do this sort of thing? How on earth are beginners expected to fight their way through a maze of annoying little problems? These things are not all that serious and relatively easily fixed, certainly in the kit building stage but they make potentially nice models unpleasant to fly. Why spoil what is essentially a really nice kit with lack of attention to some of the fine detail? There are several ways to handle the lack of down thrust. First you can carve out the bulkheads and set the motor at the correct angle. This is the right way to do it aerodynamically. Or you can mix some elevator trim in with throttle so that when the throttle is opened the elevator moves down to compensate for the climb. Finally and for the experienced flyer only, try moving the CG back until the full aerobatic position is established. (Not for beginners.) However the nicest part of this whole story is being able to drive for five minutes to a large clear local area and fly without annoying people nearby. The model was as quiet as a church-mouse in flight. This is another priceless legacy of technological progress. So there you have it, a true, warts-and-all introduction to electric flight. Ready for take-off! The Piper Cub complete with decals (they’re all supplied in the kit) and ready to fly. The “bendy” propellor, so disconcerting to some, is clearly visible in this shot – it straightens up once it starts pushing air! 18  Silicon Chip siliconchip.com.au PRECAUTIONS WHEN USING Li-PO BATTERIES Li-PO batteries contain volatile and toxic chemicals. For your safety please read the following carefully. • NEVER leave batteries on charge unattended! It is also a good idea to place the battery in a steel or ceramic dish while charging and keep well away from inflammable materials. • Never leave the battery connected to the speed control as these units have a small leakage current. Do not allow batteries to fall below 2.4V either in use or by self-discharge. Always recharge batteries at least once every 3 months. Batteries that fall below 2.4V are ruined and will never work again. • IMMEDIATELY remove a Li-PO battery from a model if it is involved in a crash. Carefully inspect the battery for even the smallest of dents, cracks, splits, punctures or damage to the wiring and connectors. CAUTION! Cells may be hot! DO NOT allow the battery’s internal electrolyte to get in the eyes or on skin – wash affected areas immediately if they come in contact with the electrolyte. A Li-PO battery might not appear to be damaged after a crash but it could smoulder over a short amount of time and suddenly catch fire unexpectedly. If in doubt, place the battery in a fireproof location indefinitely. • Disconnect the battery IMMEDIATELY from the charger if it begins to swell, emits smoke or is warm to the touch! Place warm or hot batteries in a fire-safe location, such as a container made of metal (such as an empty ammunition box) or ceramic. Always monitor the area with a smoke or fire alarm, and have an “ABC type” fire extinguisher available at all times. • DO NOT set the battery charge rate to a value greater than the battery’s 1C value as permanent damage could result. Do not exceed a 9C discharge rate. • DO NOT allow LiPO cells to overheat at any time! Cells which reach greater than 140°‑F (60°C) can and USUALLY WILL become damaged physically and could possibly catch fire! Always inspect a battery which has previously overheated for potential damage and do not re-use if you suspect it has been damaged in any way. Do not leave a battery near a heat source above 80 °C (Stove, heater etc). siliconchip.com.au Leaking batteries must be kept away from naked flames. Keep the battery as cool as possible at all times, particularly when charging. • Always provide adequate ventilation around Li-PO batteries during charge, discharge, while in use and during storage. If a battery becomes overheated, remove it from the charger immediately and place it in a fireproof location until it cools. • Use a charge lead that is directly compatible with the “charge” connector on the Li-Po battery. It is strongly recommended to use pre-assembled charge leads. These can be found at most hobby retailers. • Do not use automotive chargers to power Li-PO chargers. • It is preferable to charge individual cells for best results (parallel charging). If series charging is used, do not attempt to charge more cells in series than the charger is designed for. • Always disconnect chargers from the input power source when not in use. • Keep out of reach of children. • Dispose of discarded batteries responsibly! WARNING!! You MUST NOT care for lithium-polymer (Li-PO) cells in the same way as other battery types!! It is very important to have a good understanding of the operating characteristics of Li-Po batteries – especially their exact rated voltage and maximum acceptable charge current. Always read the specifications printed on the label of your Li-Po battery prior to use. Failure to follow the care and handling instructions can quickly result in permanent damage to the batteries and its surroundings and even start a FIRE! Do not mistake lithium-polymer cells for other lithiumbased cell types (such as lithium-metal, lithium-phosphate, etc.), as other lithium hybrids have different care and handling characteristics . It is strongly recommended to use packs that have been assembled with built-in charge protection circuits. Such circuits help to regulate the maximum voltage per cell in the pack to ensure that that they do not accidentally become overcharged. SC February 2006  19