Servo Selection?

Futaba digital servoSIZE MATTERS
The 'standard' general purpose servo is of approximately thumb-size proportions and has an output torque of between 3 and 5kg/cm. Costing between £5 and £15 each, four of these are usually included when buying a new 'club fliers' radio set. Next in line is the ball raced servo (this may also be metal geared with a coreless motor), which will be of similar size and shape but will offer more accuracy, speed, torque and better quality compared to its 'cooking' equivalent. That said, it does roughly the same job of work.

Further up the ladder is the digital servo, which again does the same job as the standard unit but with vastly increased resolution and accuracy. Digitals are becoming more popular and far less expensive than they were just a couple of years back and will undoubtedly be the 'standard' servo of the future. If you do move up to digitals then be aware that they draw far more current from your flight battery, which will need upgrading pro-rata!

Moving down in size we come to mini servos, prime examples of which are the excellent Hitec HS 85 and 81 (amidst many others). Good quality examples often outperform 'bottom of the range' standard servos in both speed and torque. At half the size and weight, I now use minis where I once would have screwed a 'standard' in place. Mini servos are very useful for single aileron operation, with plenty of power to operate just one surface on quite large (relatively speaking) models.

Okay, now we come to the more exotic end of the servo spectrum, the micro and sub-micro jobs that enable us to fly models which, not so long ago, would have been deemed impossible to fly under R/C guidance due to their diminutive size. In my youth, I forced standard radio gear into the smallest free-flight model that would accommodate it, a pretty little Mercury Magna. This was powered by a .5cc Davies Charlton Dart diesel engine and it flew very well, despite a payload that trebled the weight of the model; four AA cells, a receiver of suitcase proportions and two monster-big servos forced into the tissue covered fuselage! All went as planned until I got a bit carried away one day and did a loop, at which point the wings clapped hands and that was that. Ah, happy days! How times change. Nowadays, of course, that model would be so easy to fit out with micro R/C gear.

Going to the other size extreme, take a look inside a really big model and you'll see servos the size of small buildings - very powerful, although invariably quite slow-moving. These are fine on the type of model they’re fitted to, where power and reliability are paramount and of more importance than speed of operation. These large 'brick' servos aren't unduly expensive, so if your model can carry the extra weight they represent a really economical alternative to an expensive digital device of similar power but five times the cost. It's worth thinking laterally when kitting out your model with servos, because less can sometimes be more. This is particularly relevant with some jet jockeys, who whizz around at super speeds with just a gnat's whisker of movement on their surfaces, using servos that are probably far more powerful than needed, rated right down on the Tx. A pointless exercise. More about that later.

Finally we come to the specialised servo, a device that's engineered with one purpose in mind. These range from helicopter tail servos that are super-fast and accurate (usually used in conjunction with the gyro stabiliser that drives them), to the ultra-slim wing servos demanded by today's high performance, thin-winged super sailplanes. Expect to pay a little more for these; a tail servo for a helicopter will run into three figures, with the dedicated glider servos about half that cost. Still pricey, but top performance never comes cheap!

So, the burning question: what servos to use where, and at what cost? Truth is, two people could build examples of the same model that would have exactly the same flying performance, yet one model could easily cost a lot more than the other due to the choice of servos. Take a typical 3D or pattern aerobatic ship by way of example, into which you may well put five standard-size servos (two for the ailerons and one each for elevators, rudder and throttle). Let's look at each application in turn.

At one time it was accepted practice to have just the one servo operating a pair of torque rods for the ailerons, but using a pair (one servo per aileron) theoretically requires only half the power. In fact, modern servos are so good that even the cheapest, bottom of the range standard unit will do the job of moving a single aileron. Granted, the model won't be the most accurate flying example when kitted out with basic servos, but rest assured a standard job will move one aileron on quite a big model without trauma, and for the average flier this performance will suffice.

If using just one servo to drive this (likely on all but the bigger models) you'll want something half reasonable for the job, but don't get carried away. We used to fly quite large models on a standard single servo that had very poor performance compared with today's devices, so on the average general sports hack, say something like a club flier's 3D fun-fly (Weston Cougar or the like), a standard ball raced servo with about 4 or 5kg/cm torque (costing around £10 - £20) will do the job very adequately.

Aerobatic models can have extremely large rudders that impose equally large flight loads on servos, so don't skimp here if you want the rudder to stay deflected during those spirited knife-edge passes! Peering at the innards of a big TOC type CAP, you might see as many as four large servos driving the rudder. You may think that this is over the top but believe me it isn't, such is the performance demanded from these models. So, bear in mind that the rudder servo often needs to be the most powerful of all those on board.

At last, this is where we can get away with almost anything, right? Er, no... very wrong actually! Although the throttle servo isn't doing much work it still needs to be reliable and accurate. Just imagine for a minute: there you are with your model idling away nicely in the pits when, all of a sudden, that ropey old servo you threw into the airframe (because it was ‘good enough for the throttle’) gives a twitch and a glitch, whereupon the engine leaps into life and the prop eats its way into your arm. Not very satisfactory! The most basic standard or mini servo will indeed suffice for throttle use, but a brand new one costs so little. Do please fit a fresh one, if only for safety's sake!

Not fitted on our aerobatic example, but for models that are equipped with flaps the servos that drive them tend to be bigger and of better quality. Imagine the flight loads imposed by a fully-deployed set of warbird flaps, hanging out like sails in the breeze, overlooked until the model goes in for no apparent reason! Incidentally, the linkages on flaps can, with a bit of forethought, easily be configured so that no flight loads whatsoever are transmitted through the servo on full throw.

There you have it then, your model's all sorted out with a nice set of servos of the right type and quality and all is well with the world. Or is it? So-called 'computer' radios are the blessing and the bane of today's modeller, if only because they'll electronically facilitate the most awful bodges that previously would have doubtless been engineered properly. You can now effect an apparent fix with a prod or two of a button on the transmitter when your first port of call should always be a mechanical solution.

Take a typical servo of, say, 4kg/cm torque output and affix it to one end of the elevator control rod in the normal way. Assuming that the linkages and clevises are as slop-free as you can make them and don't bind at any point over their travel, the servo's full power will be transferred to the control surface, which is exactly what we're after. You then adjust the link until the surface is centred and adjust the throws by choosing the appropriate hole in the servo disc or control horn to give the required movement. However, many fliers invariably just hook up the surfaces and immediately dig into the programming of their computer transmitter to alter the servo centre points, sub trims, ATV's and the like, even resorting to rate switches to achieve the throws desired on any one control surface; this is just bad aeromodelling practice and not the way to do it.

Back to that super duper, state-of-the-art, five grand's worth of jet model that I mentioned earlier where the ailerons just move a smidgen, yet a brace of dirty great big and powerful digital servos are hooked up to the surfaces in question. Given that a fairly run-of-the-mill standard servo will lift around 4kg a distance of 1cm from end point to end point of its travel, then it will lift twice that weight (8kg) for half that travel distance (5mm). So, if you only need 5mm servo travel, then by mechanically altering the linkages you can double the power of your cheap and cheerful servo to perform like one of twice the power (and four times the cost no doubt), which is really good news (unless you sell servos, that is!) It's even better than that, because although servo motion appears to be one continuous analogue movement as the output disc swings through its arc of travel, it actually comprises of many discrete points of reference. As a result, when using just half the available travel to operate the surface, for a given arc of travel at the servo, the resolution is doubled. How good is that? Just by setting up the linkages properly you can have better resolution and more power than a really expensive servo that’s installed poorly.

Those who fly super-expensive models bedecked in top quality digital servos yet rate their ailerons right down because they don't need much movement could probably get away with £10 worth of servo, if only they had half a clue how to build a model properly. Learning to mechanically engineer a radio installation, as we were forced to 30 years ago, is the single most important part of any radio set-up. And as the old saying goes: "less can sometimes be more!"

There's an awful lot more to the humble servo than my ramblings here have covered, and there are more servo types on the market than anyone really needs, but it's certainly nice to have the choice. Years ago we would take servos to bits, clean the pots and grease the gears and press them back into service, but those days are long gone (although I do know a few skinflints who continue along this path!).

Replacing damaged gears is certainly worthwhile as these are cheap and simple to fix, and if you have one that's a bit iffy around the centre then a bit of switch cleaner squirted into the (worn) pot can sort it, but on the electronic side, forget it. Buy yourself a new servo, lest for the sake of a few quid it costs you a model!