Current draw of high voltage digital servos

[David Sargent 1]

Hi,

I've just bought myself one of these - **LINK** and I'm planning on running it on 2s 7.4V Lipos with 5 Hitec HS5565MH for the control surfaces and a HS5665MH for the throttle.

Having read this article however I'm a bit concerned about potential current draw. Basically what it says is that digital servos can pull almost their full stall current when they start moving from one position to another. Assuming that is correct, you could have a worst case where the servos suddenly draw over 12 amps. (I checked some of the test results on servomances and it does seem plausible at least)

The main problem would seem to be the wiring and connections from the battery to the RX, 12 amps through those slim wires and regular servo connector feels like a lot and potentially dangerous.

So the questions really are:

Is the system really likely to draw that much current?

Even if it does is it so transitory that it won't cause a problem?

Anybody with a similar setup who can share their experiences?

Thanks in advance

David

[Peter Beeney]

David, I’m afraid I’ve not had any experience with those servos and your particular type of installation, but this might be something in the way of a possible explanation.

I would suggest it’s all just to do with the way electric motors work. At the moment in time when the motor is stationary and the voltage is first applied maximum current will flow; but because of this the armature is now subjected to maximum torque and therefore it very quickly begins to turn; and as the revolutions accelerate the current flow will decrease proportionally to the speed as a result. This happens fairly swiftly, to the observer this appears to be instantaneous. I’ve always thought that servo motors actually do have a pretty easy life, which is perhaps one reason why they last so long. Because we don’t required the servo arm to move particularly fast, relatively speaking, the gear ratio, motor to arm, can be designed to be quite large. As a consequence they spend most of their working life running at very close to their full speed, so the least current is flowing, this is their ideal situation.

So I’d expect there might be many brief current surges in a flight, as will happens to any model. But they will be extremely short in duration. As far as any heating effect is concerned I think I’d consider this to be negligible. However, if you have lots of long thin wires in the loom any volts drop spikes across the total circuit might be significant. But as you are using the high voltage configuration this is also likely to make little difference, in my opinion anyway. If there is any apparent discrepancies that give you concerns then I’d tend to think that something is well and truly way out of line in the first instance!

So a general rule of thumb might be, whenever possible, use short fat hairy wires, this can only ever be beneficial. Looking at the servo from the other side of it’s performance capabilities, if it’s heavily loaded and the motor is appreciably slowed then significantly more current will flow, and in the worse case, with the servo locked in some way so the motor can’t run then we are back at the start, with the current at a maximum. Here, though, this will now continue. The motor’s resistance will be low, the lower the better, and soon something might just cry enough is enough. A situation to be avoided if possible.

The heavy current predicament did crop up many years back, a mechanical problem with the receivers PCB failing. A friend who was heavily into helicopters at the time said this was related to the then brand new digitals which were constantly stopping and starting. The short answer at that time was two switch harnesses, one to each end of the receiver. The neg and pos circuit board tracks on the servo output pins can be a bit of a vulnerable area; although I very much suspected that nowadays these have been suitable beefed up. Might be worth taking a look, though…

This is just my take on it, only just for what it’s worth, really.

PB

[Steve Hargreaves - Moderator]

It is an interesting point David....I myself suffer from a similar paranoia....all that servo current passing through the Rx bus....how much is too much...etc etc....its enough to keep a modeller awake at night.....

Maybe try one of these & see just how much current your servos are pulling

[ben goodfellow 1]

i have too thought about this , i have 1 plane flying with powerful servos. and another under way with even more powerful servos , ive gone the route of an ar12120 in one and an ar 9200 in the other both can handle approx 50 amps max each . ive have seen some 50 cc + plane going with a little 6 channel orange rx hooked up to 5 + hi torque servos which imho dangerous............ fly low

Edited By ben goodfellow 1 on 14/12/2013 18:08:48

[Peter Beeney]

I’ve now had a glance at the links, and the short answer to Ron’s Heliproz South forum post is that a BEC is the very least suitable device used to power a model radio. As I’ve remarked once or three times before, aeromodellers sometimes seem to be past masters at trying to make something do that which it’s not intended to do; and in general it might be a bit tricky to adjust the general principles of electricity; they are well established. The fact that many BECs do operate successfully might be, in my opinion anyway, often simply down to good luck rather than good design. Here the model’s radio power demand is simply low enough to be within the specification of the regulator, and in this case it (the BEC) will function without fail. Unfortunately too, generally the BEC on the ESC answers a law of diminishing returns, as the input to output voltage ratio increases so the output power lessens. Which is exactly what you don’t want, as the power might be increased to a model by adding a cell say, so all the other power demands might also go up; such as more servos required etc.

With regard to the current demand, if the servos are not constantly stopping and starting, such as in a fixed wing, then surely the average current will be relatively quite low. All the evidence points to the fact that control surface loads do not require vast amounts of power, I reckon the engine/motor is doing all the work. But again, back at the suspect applications as above, why run all the current though the rx PCB tracks anyway? I’d perhaps make a power bus, and run everything from that. If the servo power supply wires are considered capable enough, just make any extension wires heavy duty. Alternatively, if there is enough clearance in the rx case, lay a length of 0.5 mm sq. solid along the tracks and flow a good coating of solder on. This if they don’t appear to be solid enough.

Like the rest of the model, the power supply needs to be designed for the job, and certainly with a built in sufficient safety factor……

PB

[David Sargent 1]

Thanks for all the replies so far.

I've just looked up the manual on the AR12120 and it's got some good info in it. **LINK** Page 6-7 has got some traces of current for a couple of different models. One of them for a 40% Yak has a peak of 17.8 Amps! However, the average is only 2.62 amps. The second example is a 33% Sukhoi which is a bit closer to my setup which has a peak of 6.92 amps and an average of 0.82 amps.

Having seen those figures I'm a bit more convinced that everything isn't going to melt the first time I twiddle the sticks! But once I've got all the bits bought (or hopefully Santa will be kind to me) I am going to do some bench testing, just to satisfy my curiosity if nothing else.

Unfortunately I have a Futaba transmitter so I can't use the Spektrum stuff. It does look like a good solution to this problem though.

[Mark Hancock]

We use a Powersave (common supply) with a twin voltage regulator from S M services on models up to 50cc and a powerbox on bigger models . Saves pulling the current through the receiver . We use hitec 7955 on all our models .Never had any problems