Pull - Pull Control Cables

While at the stage of setting up the tail wheel and rudder pull-pull system on my FW190 it occurred to me that some may find this article of interest, with pull-pull cables on your own models. Something that is very common on today’s ARTF's.

The first rule is to remember that only the pulling cable is operating the moving flying surface. Ever tried to push a piece of string?

Second rule is that both cables should always be just taught at neutral. Never tightened like a guitar string. It won’t make the control surface move quicker, but will put unnecessary load on the servo, which at worse can cause servo failure and in the case of a hi-torque digital servo will burn them out resulting in seizing solid or the linkages will eventually fail……. hello terra firmer. If you can pluck the cables, they are probably too tight and is a common mistake by modellers who are in most cases trying to compensate (so they think) of taking up the slack on the idle (pushing) cable.

Some may have heard of the Ackerman principle when the control surface horn is not lined up with the hinge line and need for using offset servo arms. Hopefully the following will help you to understand if you should need one. I am not aware of a set formulae for determining servo arm offset that can be applied to all situations. From experience my YAK 54S, 33% Cap 232, 40% Ultimate and Vision 3D all required trial and error to get it right. One of the main factors is the size of the control surface horn, it’s position and max control throw, which invariably is different from model to model.

Question: Why does it matter if one cable is looser than the other, when in the neutral position? Well the flying surface will be able to move about and flap or correct term “induce flutter” the dreaded death of many a model. This will cause not only an unstable flight (more the movement – more erratic stability), but the flying surface has a habit of stripping itself off the aircraft ….. hello terra firmer. Also if the pushing cable is loose, when it comes it’s turn to be the pulling cable, there will be slack to take up and effect the handling characteristics. The looser the cable the more mushy the control response.

The following examples are from scale drawings for rudder pull-pull cables, but the same principles apply to elevators or tailwheel.

Example 1: The perfect set-up when running parallel cables. The mounting holes on the rudder horns are equally spaced and aligned with the holes on the servo arm. There is no offset on the servo and the rudder horn holes aligned with the hinge line. Both cables are identical length. (This is the set-up I’m using on my FW190 tailwheel and Rudder).

As the diagram illustrates, even at full rudder deflection the distances between the rudder and servo horns do not change, consequently the cable tension remains the same and neither go slack, always providing positive and instance response for the pilot.



Example 2: It is commonly found that ARTF’s have a set-up with crossed cables and some people quote that as long as the rudder horns are in line with the leading edge and the servo arm is the same length as the rudder horns, you can cross cables and the tension will remain the same ALL the way to full deflection. This isn’t true and can be of particular interest to those flying 3D with high rates and outrageous surface deflection where the slack increases.

As the Example 2 diagram shows the slack cable actually acquires 3mm slack at full deflection. Some pilots might find this acceptable but the bigger the model and deflection, the greater the slack.



Example 3: It is common to find that control horns are normally bolted onto the control surface, in a position where it is not possible to align with the hinge point. (the boys with big toys have an advantage to use “Rocket City” type fittings which allow adjustments to the linkages, to align up with the hinge line). It is said that that as long as any offset in the servo arm is the same as the offset in the rudder horns, then cable tension will stay the same throughout the throw and without any slack. This also isn’t true. In Example 3 the horns are 5mm back from the leading edge and to compensate there is a 5mm offset in the servo arm and the idle cable again acquired slack at full deflection. I remember that on my 46% Ultimate I had about 15mm of slack until I acquired the right servo offset arm.



What this means is that a servo offset should always be required when cables are crossed regardless of the position of the rudder (or elevator) horns.

Example 4: The rudder horn is offset by 5mm and by using a 10mm servo arm offset the idle cable has no slack at full deflection. This might lead you to believe that the formulae for servo offset is double that of offset on the rudder. Not so I’m afraid, it is dependant on the length of the cables and amount of maximum control surface deflection. The shorter the cables, the less slack and the longer cables produce greater the slack. Trial and error is the only way and I remember that I was pulling my hair out with my 46% Ultimate. My only tip is to keep the cables as short as possible to reduce the amount of slack and offset servo arm required, unless your flying 3D which throws in too many variables to work out.



To summarise: To get the pull-pull cable set-up you want, you have to determine how much slack you can live with, some say a couple of millimetres is OK, others argue there shouldn’t be any. Never allow flutter to develop due to your set-up. Do not over tighten the cables. Use an “offset servo arm” when cables are crossed. The only perfect set-up is when cables are not crossed, both ends are the same distance apart and in-line with pivot points.

Happy landings.

good write up Richard thanks

Mark

Don't forget that the choice of material for the pull pull cables will have a big impact on how susceptible to flutter, especially on larger control surfaces. Something with extremely low stretch is was what you should be looking for, so materials like dyneema, steel wire, etc are good. Braided nylon isn't so good

Cheers
Andy

Personally so long as I have no slack in either pull pull cable at neutral I've never bothered overmuch about "some" slack on the none pulling cable bacause as you say Richard you can't push with a cable. I've never heard of rudder flutter and most of the major ARTF manufacturers don't bother sealing the rudder hinge line whereas to a man they all seal or recommend sealing aileron and elevator hinge lines to prevent flutter.The only place you would be likely to notice an issue maybe when flicking or rolling from KE to KE but I think todays efficient digital sevos render it unnoticeable and many guys fly with loads of expo on rudder anyway which would theoretically induce the same effect as slack in the none pull cable. So as I say I'll try and adhere to Mr Ackerman's principles as much as possible but won't give myself grey hair (further) finessing over it

Whenever I utilize pull pull / closed loop systems, I always try and arrange a slave system of main "tiller" to which I attach my lines quite tightly, and then use a short stout pushrod to connect this tiller arm to the servo - this eliminates the cabling from imparting any direct force to the servo bearings etc. Thought this was the "standard" way of doing this

All three of my large scale aerobats run closed loop straight to a single high spec digi servo. Richie's got a floating pull pull system on his Yak with two servos in tandem but arranged in such a way that if one servo completely locks up you still have some rudder to continue flying

I have one of these on my super stick (closed loop) but on my tutor 40 it just has a steel pushrod, what is the advantage of the closed loop?

In linkage terms there's no advantage Sam the only advantage really is in being able to mount the rudder servo/s nearer to the nose of the model and thus "sometimes" helping with the balance. Since my last post I realised I told a porky the rudder on my Sukhoi is not P/P but a single servo on one side of the fus with a direct push pull rod Another possible advantage in scale terms is you don't have a servo/s hanging out the side of your fuselage and spoiling the aesthetics

How unless you have stretch due to inadequate cabling can you have slop on a cable that is only pulling

Ohhhh I think im out of my depth here!

Sam in the perfect ideal world it is desireable to have the servo next to the control surface with a short rod to operate and not to use pull-pull at all. However generally pull-pull's are used where it is not possible to mount a servo close to the moving surface due to the lack of space at the tail end within the airframe. As Brian commented, some models don't balance well with too much weight at the tail end, so the servo is too far away for a rod. Some people use a snake rod, but they can flex if incorrectly mounted.

As I said you get slack at maximum servo deflection with an incorrect set-up which can cause mushy control when you have to take up the slack with the opposite control throw. Neither cable is taught?

Anyway I thought it was worth posting.

it was worth posting and some good comments from others to, I hate with a pasion doing closed loop cables but i am getting better

M

who now even uses closed loop on the ELE of his shokkies

Some years back I acquired scale drawings for offset servo arms, which show the most common offsets. You will note that there are four drill holes around the centre and these are intended so that you can bolt them onto a servo wheel directing on top of the servo. If you want to use them as a bellcrank configuration, just put a bolt in the centre. The drawings allow you to make your own out of plywood so that you can experiment without the cost of purchasing varied servo arms, then you can make one out of G10 composite or aircraft grade alloy. You can of course buy off the shelf from within the R/C industry, mainly from large R/C aerobatics stores.

now that is handy to have

now that is handy to have

twice