Servos - How Much Torque Required?

[Nigel Heather]

Hi,

 

Is there a rule of thumb for how much torque is required for a typical 40-size sports plane.

 

When I started out there were standard-sized servos and that was it.  Usually, you just bought the basic Futaba or Hitec servos.

 

Now, with the large range of budget brands, there is so much affordable choice, analog/digital, plain/ball bearing, nylon/metal gears, size.

 

Would you still go for a standard analog, nylon-geared, plain bearing, 40g servo with a torque of 3.5kg or would you consider a digital, metal-geared, ball raced bearing, 17g servo with a torque of 3.0kg?

 

The reason I ask is that I have a fun fly kit in my stash and to keep the weight down I have gone for some 17g 3.0kg mini servos - they are very nice, possibly the nicest servos that I have ever owned.

 

I am also about to add a Gangster 63 Lite to my stash and wondering what servos I should use on that - should I go for 40g 3.5kg standard or the the 17g 3.0kg mini.

 

Cheers,

 

Nigel

Edited September 26, 2023 by Nigel Heather

[Peter Jenkins]

Hi Nigel

 

Futaba S148s are rated at 3 Kg.cm at 4.8 v which used to the standard.  If you put it on 6 v you get 3 Kg.cm.

 

For a typical 40 size sports plane, a standard analogue, nylon geared, plain bearing servo with around 3-3.5 Kg.cm will do the job.  There is no need for a digital servo, or metal gears although nylon gears have been know to get damaged if you bash a control surface when moving the aircraft around.

 

For the Gangster lite, I would focus on getting a servo that centres accurately from either direction.  That is most important for an aerobatic aircraft like the Gangster.  Using cheap clone servos can be a waste of time if they have a wandering centre position.  For an aerobatic aircraft, not being able to have a set position for any of the main surfaces just makes it difficult if not impossible to trim out.

 

As I fly a 2 mtr competition aerobatic aircraft, I use digital servos as I can be sure that the controls will not blow back once I have demanded a set servo position so loop radius doesn't lose it's accuracy.  If you want to use a digital servo then that's fine but make sure that the rest of the control set up is as accurate as the servo will be. 

 

Hope that helps.

[Brian Cooper]

4 hours ago, Nigel Heather said:

The reason I ask is that I have a fun fly kit in my stash and to keep the weight down I have gone for some 17g 3.0kg mini servos - they are very nice, possibly the nicest servos that I have ever owned.

 

I am also about to add a Gangster 63 Lite to my stash and wondering what servos I should use on that - should I go for 40g 3.5kg standard or the the 17g 3.0kg mini.

 

It really depends on HOW the aircraft will be flown. 

If they are only going to be trundled around the sky, the mini servos will probably be okay. . But if the intention is to "get the best" from the models, the smaller servos can (will) break. 

 

The torque figures quoted on some mini servos can be optimistic to say the least, but the bigger problem is their gears... they simply cannot take the strain.  Fun Fly models in particular have large control surfaces and BIG throws.... trust me, mini servos WILL strip their gears in flight if the model is pushed hard.  

 

Beware of using servos without a ballraced output shaft (Futaba s148 for example). 

The shafts wear out very quickly if/when subjected to high loads. The shafts then become wobbly and vague.  Control authority and servo centring will suffer accordingly. 

They are okay in basic trainers but shouldn't really be used in any model which has a higher performance.   

 

 

[Jon H]

The original gangster probably used a single 2kg torque S100 or something for both ailerons and never had an issue. 

 

Personally i would probably still use the old school single servo/torque rod setup on the basis that the model is a simple sport job and it saves extra work to modify it. If however 2 servos are desired, standard size 3kg jobs would be my go to. 

 

3kg servos are more than adequate for a model of this size and even WWII fighters of 70-90 inch can be flown without any trouble using them. Personally i go up a bit on my warbirds (mostly on flaps and elevator), but an 88 inch Spitfire i bought had flown very successfully on 148's for over 20 years with the servos still in good condition. As the model was stored in a damp shed for about a decade before i got it i will be changing the servos, but only to 4kg savox jobs. Its a Spitfire after all and not a 3d monster so i just dont need that much power. 

 

On the wear thing, non ball raced servos do wear over time but i have never had an issue with it myself. Many of my 148's must be well over 20 years old now and none of them have problems with their output shafts. This could be partly due to my servos always being inside the model, so they are protected from dust etc getting into the bearing and chewing it up, but realistically its a non issue and i would not be concerned at all. Not that it matters much, 148's are discontinued and a ball raced savox 351 is cheaper than the plain bearing hitec 311 anyway. 

[leccyflyer]

In the size of models that I typically fly I think Brian's point about the servo gears is most apposite - on micro and smaller servos the teeth on the nylon gears are really, really small -little more than flash in some cases, and can strip if you look at them sideways. I haven't had many failures in flight, the most common cause of stripped gears being catching a control surface when putting the model in the car, which will do them in nicely.

[Cuban8]

Somewhere tucked away in my cashe of useful information documents I have an article that gives the method for calculating the loads on flying surfaces at various angles of deflection, airspeeds etc. I think it might be from a BMFA News article when that publication was produced as a newspaper, so quite a few years ago. I'll see if I can find it.

[Cuban8]

Professor Google just came up with this..............

 

The torque on the control surface is roughly proportional to the square of the airspeed, the area of the control surface and, 1/2 the average chord of the control surface times the sine of the angle of deflection. The constant of proportionality depends on the mass density of the atmosphere and all the units of measurement.

[Peter Jenkins]

17 minutes ago, Cuban8 said:

Professor Google just came up with this..............

 

The torque on the control surface is roughly proportional to the square of the airspeed, the area of the control surface and, 1/2 the average chord of the control surface times the sine of the angle of deflection. The constant of proportionality depends on the mass density of the atmosphere and all the units of measurement.

Does it say what units to use and does it give an equation to give a precise figure?  Otherwise, Prof Google is merely telling us what most of us know.  The OP asked for a way of working out control forces.

 

I suppose one way to check, a bit rough and ready though, is to apply full power and check if there is any noticeable difference in speed of elevator movement from full up to full down compared with no power.  Assuming this is a tractor engine and the elevator is in the full slipstream.

[Ariel]

I asked this question back in April and David Hayward was kind enough to post this link.

https://www.radiocontrolinfo.com/information/rc-calculators/rc-airplane-calculator/

 

[Cuban8]

I was being a bit mischievous posting Prof Google's gobbledegook...........

Edited September 27, 2023 by Cuban8

[EvilC57]

There was an interesting article on this very subject by Mike Freeman in the July 2023 issue of RCM&E. I scanned and kept it, attached here...

 

Servo Torque Calculation.pdf

[Simon Chaddock]

There is this question:-

Do you actually need to be able to apply full deflection on any control surface at the planes maximum airspeed speed?

 

I don't claim to know the answer bur as an RC system has no feed back I prefer to work on the principle that the servo will move the control surface to the servo's maximum torque and then stall rather than continue so possibly exceeding the structural limits of the plane.

Even when stalled a servo does continue to apply a force and will resume the motion when the torque load reduces.

Just an observation.

[GaryW]

Unless the models instruction manual specifies a particular servo size I pretty much stick with using bog standard servos ie: Futaba 148, 3003, 3004 3001.... JR 507, 517 & Hitec HS 300 in pretty much all my models including my current models with the exception of current Hirobo shuttle which has JR DS 8417 on pitch, DS8411 on cyclic DS811 on Tail & 3001 on throttle,, which i has in the past used all 3001's on helis as for my style of flying whether its planes or helis  I don't see the need for high powered high speed high torque servos,, however my ben buckle scorpion has prehistoric futaba 128's but thats only what was in it when i bought it and I can't get my hand in the narrow fuselage to change them

 

 

 

 

Edited September 27, 2023 by GaryWebb

[Nigel R]

On 26/09/2023 at 22:16, Nigel Heather said:

I have gone for some 17g 3.0kg mini servos

 

Two questions;

 

Which mini servos?

 

How big is the gear train?

 

Fun fly models with massive surfaces, lots of power, and servos with tiny gear trains do not make for a long lived mix.

 

Standard servos have reasonably sized gear trains which can work for many years without issue, I have heaps of JR 517 (standard, single ballrace) that are still perfectly ok for most of my purposes (sport models, 40 or 60 size, like yours).

 

(as an aside, 17g is micro size (my opinion), whereas ~25 - 30g would be mini)

 

[Nigel Heather]

7 hours ago, Nigel R said:

 

Two questions;

 

Which mini servos?

 

How big is the gear train?

 

Fun fly models with massive surfaces, lots of power, and servos with tiny gear trains do not make for a long lived mix.

 

Standard servos have reasonably sized gear trains which can work for many years without issue, I have heaps of JR 517 (standard, single ballrace) that are still perfectly ok for most of my purposes (sport models, 40 or 60 size, like yours).

 

(as an aside, 17g is micro size (my opinion), whereas ~25 - 30g would be mini)

 

 

Thanks, may have to re-purpose them by the sound of it.

 

They are EMAX ES3054 - I bought them for a fun fly to keep the weight down following advice - but sounds like I should put something bigger in their.

 

Question is what to do with the 'mini' servos because they are very nice - shame to throw them away or stuff them in a cupboard unused.

 

I have three kits on the stash

 

Fusion II Fun Fly

Cambria Fun Fighter

Gangster 63 Lite

 

I wonder whether to put them in the Fun Fighter - before you say no way - my intent with it is quite different - I'm using electric power, want to make it as light as possible, and I want to fly it much more gently, much more slowly than the mad all-out displays they do at airshows.  So maybe I could use them in that.

 

 

[john stones 1 - Moderator]

I would fit them in the Fusion 11 myself, bumf on the model recommends Corona micros, those Emax you have are metal geared.

[Nigel Heather]

1 hour ago, john stones 1 - Moderator said:

I would fit them in the Fusion 11 myself, bumf on the model recommends Corona micros, those Emax you have are metal geared.

Yes, I’ve noticed that - though those coronas are a bit bigger/stronger.

 

Cheers,

 

Nigel

[Nigel R]

Fusion II Fun Fly

Cambria Fun Fighter

Gangster 63 Lite

 

I would happily use the Emax servos in the wing on the fun fighter, one per aileron. They're probably just fine for the G63 Lite ailerons. I'd look for slightly larger on the elevator/rudder, something HS225 sort of size  (~25 - 30g).

Fun fly, I'd go for HS225 size all round.

[Nigel Heather]

The makers of the Fusion II fun fly manufacturer recommends Corona DS-238MG (which are 22g 4.0kg) so I was going to buy those.

 

When looking at the Corona site I was think of going with the Corona DS-339MG (which are 32g 4.4kg) for the Gangster.

 

For the funfighter I’ll probably use the EMAX in the wing as you suggest and probably Corona DS-238MG on elevator and rudder.

 

That leaves me with two EMAX spare.

Edited October 2, 2023 by Nigel Heather

[Martin Harris - Moderator]

I’ve found Corona servos to be good value for money in terms of performance but with serious shortcomings in the plastic used for the casings.  Most of those that I’ve used  have ended up with cracked mounting lugs. Heavy handed?  It’s never been a problem with any other make I’ve used. 

[Nigel R]

45 minutes ago, Nigel Heather said:

makers of the Fusion II fun fly manufacturer recommends Corona DS-238MG

 

Don't forget that's a competition fun fly airframe, all out light weight is top priority, day in day out sport model type durability isn't... pilots in the winners circle will have "a few spares" on hand to account for attrition. The Corona is £10 or thereabouts a throw, so in terms of overall competition spend it's, well, probably considered expendable.

 

Going up a size to the DS-339MG at 32g, for four servos is only adding 40g or 1.5oz overall and whilst us average mortals will not notice the difference in flight performance, we might notice e.g. a lack of elevator, if the gears strip.

 

[Nigel Heather]

6 hours ago, Nigel R said:

 

Going up a size to the DS-339MG at 32g, for four servos is only adding 40g or 1.5oz overall and whilst us average mortals will not notice the difference in flight performance, we might notice e.g. a lack of elevator, if the gears strip.

 

 

Not quite sure what you are saying her - I wasn't suggesting putting the 339s in the Fusion II, that would be for the Gangster 63 Lite as an alternative to standard size.  But I want my Gangster to fly quite precise so wondering whether I should put something better in it.

[Nigel R]

Sorry - didn't make that very clear - I was suggesting 339s for the fun fly.

 

If you want precise flying I would take recommendation from the aerobatics or high speed glider / pylon racer guys.

[Erfolg]

In the past, Biggles, Dr Dave Burton, provided a methodology how to calculate the probable torque required for a specific model.

 

I t comprised Cl=0.5*Roe*V^2*A then taking moments of the clevis and its position.

 

In reality the toque required was pretty low. Just think back to the days of SLM, Westwood(or was it Winwood), Waltron servos. Where a central wing servo would drive via snakes, or push rods/bell cranks. Many quite large, reasonable fast models managed to be controlled.

 

I suspect that many models have servos that are many times over specified relative to duty. Yet this approach is reassuring, if not expensive.

 

Just consider how many ARTF models manage quite well with essentially 9g servos. 

 

Then again, what price is safety?

 

Must order a RSJ for the spar for my next model. Or could I undertake a simple calc, or see what others manage with?

[Nigel R]

12 minutes ago, Erfolg said:

Just consider how many ARTF models manage quite well with essentially 9g servos. 

 

Some time ago my <1lb foam 3d hack chewed through 9g servos, in flight. They were exactly as recommended in the instructions. Flight speed, next to nothing, almost certainly <15mph. Prop hang, yes. Fast, no. 9g up to the job? No.

 

Up a class, to ~13g branded metal gear servos, no further problem.