Elevator Trim Drift

[EvilC57]

I have a recently completed MPX Extra 330SC, which although a lovely model (for a foamie), suffers from the need to retrim the elevator for every flight.

One flight I'll find that it need 10 or so clicks of (say) up trim to sustain level flight hands-off, then the for the next flight (without having touched the trim), it'll need 10 clicks or so the other way - all of which wastes flying time while I retrim before doing anything 'clever'.

It has a large elevator driven from a single Hitec HS-85MG servo via a longish metal pushrod. There was some jerkiness in the movement caused by the metal rod sliding in the plastic outer, which I have eliminated by putting some silicone grease on the rod.

I also swapped the servo with the rudder servo a few days ago, to see whether the problem went away. It didn't .

One of my club mates has suggested various things from differential expansion of the metal control rod with temperature, to both ailerons drifting up & down, giving the impression of elevator drift .

If there's any mileage in the differential expansion theory, I guess I could investigate moving the elevator servo down the back end, to create a shorter linkage.

Any ideas anyone please?

Edited By EvilC57 on 18/08/2020 17:15:35

[Denis Watkins]

Battery moving, or something inside the model

Altering, ever so slightly, the C of G

Check the rig inside very carefully

Or Motor Mount loose

Edited By Denis Watkins on 18/08/2020 17:33:25

[Chris Walby]

Gluing things with grease my not be the best approach as the hysteresis (slop) remains.

View the model from the read and give it full up elevator, return to neutral stick (but no over travel), then full down and return to neutral stick. The elevator should return to the same place. If it does not then I would suggest removing the play/slop.

If the above is all ok then take another look at your control arm set up. Are you using the furthest out on the servo arm and elevator arm with full servo travel (to get the required throws for you)?

[EvilC57]

Posted by Denis Watkins on 18/08/2020 17:30:19:

Battery moving, or something inside the model

Altering, ever so slightly, the C of G

Check the rig inside very carefully

I doubt it. It’s secured onto Velcro, restrained with a Velcro strap, hard up against the firewall on a slot in carrier tray, which can’t move once latched in.

I should have said also; by ‘hands-off’, I mean with both sticks straight up. I.E. half throttle, and no control inputs.

[Denis Watkins]

In the olden days, Evil, Clicks were huge by comparison to the minute clicks we use now, so 10 clicks can be a mm.

Something is moving ever so slightly as your model needs trimming again and again.

A loose servo will do it. Thermal expansion is not a feature.

Check physically your 100% secure straps and bulkhead and check the tailplane is attached and not moving.

I can only suggest those areas that I have seen before. Have seen loose tailplane, loose straps, loose servos and loose battery tray. All adding up to be unable to leave the trims alone.

[EvilC57]

Posted by Chris Walby on 18/08/2020 17:37:38:

Gluing things with grease my not be the best approach as the hysteresis (slop) remains.

View the model from the read and give it full up elevator, return to neutral stick (but no over travel), then full down and return to neutral stick. The elevator should return to the same place. If it does not then I would suggest removing the play/slop.

If the above is all ok then take another look at your control arm set up. Are you using the furthest out on the servo arm and elevator arm with full servo travel (to get the required throws for you)?

I wasn’t sure about the grease, but it does seem to have eliminated the creakiness in the movement.

I’ve carefully checked on the ground for any sign of a ‘double neutral’ effect, when coming to zero deflection from each direction, and there is nothing measurable. Bear in mind that this is a large elevator, so the slightest deflection will have a noticeable effect in flight.

As for the control arm movement; it’s the usual Z-bend through a hole in the servo arm, and a turret with a grub screw the other end with little or no ‘slop’. I have actually moved the linkage in by one hole over where it was initially (as set iaw the instructions), as this will increase the mechanical advantage the servo has over the control surface, and reduce the torque load on the servo. It always seems to me that it’s better to have a servo wanging almost from one end of its movement to the other with little load on it, rather than using a hole out on the end of a long arm, with little servo movement, but a higher load - subject obviously, to having sufficient control surface movement.

[Martin Harris - Moderator]

If (and it's a big if!) my back of a fag packet calculations are correct, a 10 degree C change in temperature would give about 0.003 inches of expansion on a 2 foot steel push rod. Depending on the horn geometry, this could quite conceivably give around a degree of change in the neutral position.

[EvilC57]

Posted by Denis Watkins on 18/08/2020 17:50:16:

In the olden days, Evil, Clicks were huge by comparison to the minute clicks we use now, so 10 clicks can be a mm.

Something is moving ever so slightly as your model needs trimming again and again.

A loose servo will do it. Thermal expansion is not a feature.

Check physically your 100% secure straps and bulkhead and check the tailplane is attached and not moving.

I can only suggest those areas that I have seen before. Have seen loose tailplane, loose straps, loose servos and loose battery tray. All adding up to be unable to leave the trims alone.

Thanks Denis, I think I’ve just fixed it as a result of something you said. The tail is retained with a single screw through it from the bottom of the fuselage. Although this screw was fairly tight, there was some flex in the EPP foam, which allows the tail plane to rock from side to side somewhat. Rocking the tail (by pulling fore and aft on one side) caused the elevator to go up & down due to the changing geometry. Now fixed by application of some clear 3M tape top and bottom of the fus to elevator joint. Hopefully the next flight will prove this .

[EvilC57]

Posted by Martin Harris on 18/08/2020 18:19:34:

If (and it's a big if!) my back of a fag packet calculations are correct, a 10 degree C change in temperature would give about 0.003 inches of expansion on a 2 foot steel push rod. Depending on the horn geometry, this could quite conceivably give around a degree of change in the neutral position.

I can’t confirm or deny your calculations Martin, however I’m inclined to agree that despite what Denis W says above about thermal expansion not being a feature, I have anecdotal evidence from fellow flyers, that it can be. And can give a detrimental effect on trim on hot/cold days.

[Andy Stephenson]

It is true metals have very little expansion with temperature but the airframe material probably does so it only has to be different to the pushrod to have an affect. Plastic snakes are notorious for this and this is why precision fliers of patternships don't use them.

A.