Inverted flying rudder control

[John Mc]

I have a Seagull Extra ARTF which I balanced as per the manual with a mid C of G point.

It flies ok straight and level in light winds although I usually set a slight elevator downtrim to stop the model constantly gaining height when travelling fast or in windy conditions (most of the time it is the latter on our site)

I can fly inverted into wind without applying any down elevator at all. Sometimes I even have to apply a slight touch of up elevator. When I try to steer the model in a circle when inverted, if I apply anything more than minimal rudder the nose of the model pitches up and gains height rapidly as it turns.

None of my other planes do this and I cant understand how rudder can have this effect. Has anyone have any ideas on what causes this.

[Shaunie]

I'm no aerodynamics expert but I'm sure BEB will be along in a bit to correct me, however...

If the plane flies inverted with no down elevator to hold the nose up the CG may be a bit too rearward. The up elevator that you are applying shows the tail is making lift in this situation, any rudder you apply will cause yaw and that will tend to destroy the lift the taiplane is making causing the nose to rise and making the plane climb.

It's the same effect single channel fliers use this to make a plane loop by waggling the rudder.

Shaunie.

[Steve Hargreaves - Moderator]

Agree with Shaunie....it sounds like a slightly rearwards CoG to me. Have you performed a "Dive Test" at all to establish if the CoG is correct....?

[John Muir]

We had a 46 sized Seagull Extra at our field a year or two back which was set up as per the manual and was obviously tail heavy. We got it down in one piece but it was close. The owner fitted a bigger, heavier lump up front and it flew much better. Manuals aren't always correct on the subject of CG. Yours sounds too far back to me too.

A lot of planes pitch with rudder, mainly down when flying upright, so up when inverted. With the instability from a rearward CG the effect will probably be a lot worse. Do you know if it pitches down with rudder when it's right way up? My Wot 4, Cap 21 and Saphir all do it to some degree or another. Just a design 'feature' I'm afraid.

Incidentally I find inverted rudder turns really difficult - I keep going the wrong way. I have to stick to ailerons .

[Peter Jenkins]

Hi John, I've been writing a set of posts on how to improve your aerobatics and this covers trimming in some detail. If you are interested in this you follow the link here

From your description of the behaviour of your aircraft, your CG appears too far back. Rather than repeat what I've said in my posts take a look at the link.

You may already know this so my apologies if I'm telling you what you already know. All controls have a primary effect and a secondary effect. With rudder, as you know, the primary effect is yaw, with secondary effect being bank. Once the aircraft starts to bank then you need up elevator to maintain height as the lift is now acting at an angle with one component causing the aircraft to turn and the second component opposing the weight. Unless you increase the total lift being generated by the wing there is insufficient to counter the weight so the aircraft descends.

In your case of the flat inverted turn - at least I'm assuming it's flat as you are using rudder only - if you apply aileron to keep the wings level in the turn but you have accidentally banked the aircraft too much away from the direction of turn, the rudder now starts to have a slight nose up effect which your aft CG amplifies. Your aircraft sounds unstable as you say you need to apply some up elevator to fly level when inverted - that's instability. You should always aim for some down elevator to be held in when flying inverted - that's stable.

Hope that helps - but do read the far more detailed trimming for aerobatic aircraft in the above link.

[John Mc]

Thanks for the advice gents. I must have flown this plane 100 times over the last two years. It is powered by an OS 75AX two stroke. I always thought it was a bit lively to say the least.A 55AX is quite adequate for what is a 40 size model but I still enjoy flying it.

I have moved the RX battery forward as that seems to be the consensus. I will test it out after the Christmas break.

As a bye-product I have had a look at the link Peter sent me and there is loads of usefull aerobatics info to work through.

By the way - My inverted circles are more like egg shapes!

Merry Christmas all.

[Matt Jones]

Without wishing to start a riot by questioning your flying skills, can you be 100% sure you're not corrupting your rudder input with a small elevator deflection? This would be my first instinct. You could verify this is / is not the case by setting maximum expo on the elevator and trying again. Just a thought. Edited to add this depends on what mode you fly!

Edited By Matt Jones on 24/12/2013 17:52:47

[Peter Beeney]

When my models fly inverted they invariably tend to need a modicum of down trim to fly straight and level. I noticed that in John’s OP he tends to use a spot of down trim on his model when flying upright, so could it be that it’s this down trim that’s in effect holding the nose up? What happens if it rolls inverted with no down trim applied? Does it still need no down trim to hold S&L?

I’ve often observed the instant nose down pitch effect on many models when rudder is applied, but in general I’ve found that it always seems more pronounced when rolling into knife flight to put the model on it’s side and then some top rudder is applied to keep the nose up. I’ve always assumed this is because when upright or inverted the wings are a support against gravity, but in knife they’re not, however the lifting action still works relative to the model. I just take the coward’s way out and dial in a bit of elevator mix to compensate, that’s up elevator as the rudder is applied.

Slightly crosswind to the thread but still flyable maybe, there is an associated conundrum here that I never did fully get to grips with. When flying a standard 4 channel model upright, if it’s flying away and the pilot moves the rudder stick to the right the model yaws to it’s, and the pilot’s, right. When inverted, moving the rudder stick to the right still results in the rudder moving to the right, but the model yaws to the pilot’s left. Imagine you’re sitting in it. If the pilot now fly's a 3 channel rudder/elevator model the same thing happens. Except that it doesn’t… When inverted the right stick application will cause the model to still turn to the pilot’s right. So this must mean the rudder action reverses, when the rudder moves to the right the model yaws to it’s left!*? It’s now the same as flying with ailerons, or at least the rudder is acting in the same way.

A 3 channel model invariably has lots of dihedral and often something of a flat bottom wing, too, so when inverted it usually takes bags of down elevator to get the necessary angle of attack to maintain lift. Also the inverted dihedral, now definitely acute anhedral?, might tend to have a strange effect with the increased AofA. I’ve always vaguely thought this might be one reason for this odd phenomenon. The model doesn’t much care for flying like this, it’s constantly trying to get back upright; so it can be a bit tricky, but with plenty of practise it can be done, mostly a spot of careful balancing.
It wasn’t just me that noticed this, I’ve seen it mentioned elsewhere, written up in a mag of the time. But I guess these days 3 channel trainer types are not flown much anymore, so there are not too many opportunities to be able try it out..

Models sometimes do the most curious things. I one had a plane called, I think, a Stryker, this was a cottage industry ARTF aerobatic type. Fortunately it was very flimsy, so it didn’t last long, it wasn’t a very successful design anyway. 60 powered, about 5ft span low winger. One of it odder quirks was it’s ability, when you pulled to the vertical and then tried a flick roll off the top, that’s full up elevator, right rudder and ailerons it would simply just roll very slowly to the left!! A slower attempt made no difference. I never did bother to spend any time poggering* on this one, either…

* To pogger… Possibly an old Wessex word meaning a study of something Inordinately Incomprehensible by the Utterly Unknowledgeable…

PB

Edited By Peter Beeney on 25/12/2013 11:11:15

[Jamie sawyer]

I know its not related directly to the problem that you are describing but it will be helpful to you for setting up your transmitter settings. A video speaks better than words sometimes and much easier to soak up for some people.

**LINK**

The guy goes into depth how to do it and I feel the video is brilliant and easy to understand. I hope it helps a little bit.

Jamie

[Peter Jenkins]

Dear Peter

When you trim an aircraft to fly straight and level upright, if it is stable, then you will always need some down elevator when flying inverted straight and level. That’s because the relationship you have set up between the tailplane and wing gives the wing positive incidence when flying upright and negative incidence when flying inverted. One of the rules of aerodynamics, or flight physics as it is called today!

An unstable aircraft is one that wanders around at the slightest provocation such as flying into a gust of wind when its nose will pitch up and if you do nothing to correct this, the nose will pitch up even more.

A neutral aircraft will just maintain whatever attitude you or the environment has caused the aircraft to take up. If it pitches up, then speed will decay until the aircraft stalls. If it pitches down, speed will build up until either the ground intervenes or, ultimately, the structure complains by breaking.

A stable aircraft will always try and return to its trimmed straight and level state – that’s provided you did achieve such a hands off trimmed state.

As regards the aircraft rolling the wrong way on a snap roll, the easiest explanation is that you are flying too slowly for the manoeuvre so that as you initiate the snap the down going aileron stalls that part of the wing which, instead of rising, falls as the weight is no longer supported by the lift on that part of the wing. This leaves you with the situation you describe. So, you need to be flying a bit faster to achieve the desired effect whereupon the down going aileron produces the required upward force. That’s why when you flew slower, you got the same odd effect. If you had flown faster, you might have got the flick/snap roll going in the correct direction unless ….

A further explanation might be that your wing is not as torsionally rigid as you would like. The down going aileron can therefore twist the wing and produce a roll in opposition to the aileron movement. This effect was used for relieving control forces on manual flying controls in full size aircraft and it was called a servo tab effect – in other words, it helped the pilot to move the controls with reduced effort – not what you want for a flick roll!

The effect of dihedral is to increase stability whereas anhedral decreases stability. That’s why when you are inverted with a 3 channel model you have your work cut out to keep it flying straight and level – it’s like trying to balance a cone on its tip as opposed to its base. One is naturally stable and the other is unstable.

If you are a devil for punishment, you might like to read the thread that I’ve written on how to set up an aerobatic aircraft which is here.

[Peter Beeney]

Hi Peter,

Thank you very much for your concise explanation about the behaviour of models under differing flight conditions.

However, I feel that we might be at ever so slightly cross purposes here. In John Mc’s OP, he mentions that he usually applies a modicum of down trim to keep the model flying S&L upright; he then goes on to say that when inverted the model requires no down trim to maintain height. Subsequent posts tend to suggest this is due to the models instability, such as an incorrect CG and so on; I was only merely thinking that as the model already had some down trim wound in anyway, then it might perhaps not require any more to fly S&L inverted. But that could be entirely wrong. From experiments I’ve done in the past, such as hanging lumps of lead on models in various regions such as the tailplane, ‘just to see what happened’, I’m of the opinion that moving the CG around doesn’t always greatly affect the planes normal flying ability; however, it does become a bit obvious when you start tinkering with the control inputs!

Relating to the ‘Stryker” flick roll of long ago, again apologies for not making myself quite clear. This particular situation occurred when the model was going vertically up.The speed would start to decay, due to insufficient power to sustain unlimited vertical, and just as it was coming to a standstill I would try and flick roll. As I see it at this point, because of the very low speed the wing and ailerons would now have little effect anyway; but the tailplane, fin, rudder and elevator are squarely in the prop-wash of a 60 still turning at full chat. So I’d consider that any surface deflexion here would result in an opposite reaction from the model; I’ve always found this to be so, but in this case it didn’t happen. Unfortunately, (or maybe not!), it’s very unlikely that I’ll ever have a model with the same characteristics so I’ll never be able to replicate it; and, as I said, I remember it as a bit of a dog overall anyway. The same principle can be seen in the stall turn for instance, on something like a Stampe or a Tiger Moth. One way of doing it is to pull to the vertical and close the throttle, as the model virtually reaches standstill open the throttle and at the same time a quick full defection and return on the rudder, at the same time closing the throttle again. Get it right, and the model does a nice pivot around it’s CG and starts to descend.

It would now tend to get very confusing if the model swung the opposite way to which the rudder was applied.

This is utilising the effects of the prop-wash and rudder together; but as always, timing is everything, if it goes wrong then some anomalies soon creep in, such as the pendulum effect.

With regard to the rudder/elevator model, eventually, after much practise, I found I could fly the three channel model inverted reasonable well. The odd situation that I was actually trying to point out, though, was that I found that when inverted the rudder effect on the model was reversed. Very convenient for flying but not quite so obvious as to why it it happens. It certainly can be a bit precarious flying sustained inverted, with the higher angle of attack only adding to this, I feel.

Happy Landings, and All The Best for the New Year.

PB

[John Muir]

Hi Peter,

I think what's going on is that the reversed rudder effect is down to adverse roll to yaw coupling which is fairly common on low wing models with little or no dihedral (an upside down trainer is effectively a low wing model with anhedral). The inverted trainer's anhedral increases this effect to the extent you can steer around as normal, as right rudder (which you would expect to yaw the plane left when inverted) will now give a very substantial secondary roll to the right. Of course the plane will want to fall out of the turn and roll upright because the of the destabilising effect of anhedral and high CoG.

When the plane is right way up, the high wing and dihedral do what they are supposed to do and increase stability while also exhibiting proverse roll to yaw coupling which means the plane banks into a turn produced with the rudder only.

[Peter Beeney]

Thanks very much for that, John, that does sound exactly right. Back in those days we didn’t give anything much thought really, the problems soon found us, so we just looked for the practical solutions. The trick was to see who could keep going inverted the longest, and as you say, if the model leaned a bit too far one way it just snapped back upright. We had a lot of fun with the three channel models.

They can also be a bit of an unknown quantity. A chap came along once, with his young son, who was wanting to learn to fly. It transpired later that part of this gentleman’s day job, and night as well, I should think, was flying in very fast two seat jets; it seemed he’d wanted to be the pilot, but because he was a bit good at maths he’d had to play the navigator’s part. They’d brought along a small three channel model and when I saw it I said “Oops, this might not be quite the right starter!” But I was never more wrong, and I soon did have to admit this. It was a near ARTF, designed by Mark Tilbury, the heli champ that’s recently done the same with the Riot. From the moment it took off it was a little cracker! There was another super dinky, too, by Bowman’s Models, for about a 0.15 glow as I remember, I think this might be a classic with today’s radio kit in control.

So now, with luck, someone will turn up with a rudder/ele model, and I can have another little whizz; with maybe now at least some understanding at what’s occurring…

PB

Edited By Peter Beeney on 30/12/2013 22:47:40

[Peter Jenkins]

Hi Peter B,

CG position is one of the most important as regards the behaviour of any aircraft - full size or model. Too far forward and the aircraft is very nose heavy, very stable and difficult to manoeuvre - the opposite is true and if you take the CG back too far the aircraft becomes unstable and unflyable - at least for a human, Many of today's front line jet fighters are unstable and require the flight computer to fly the aircraft depending on what the pilot commands - he is not flying the aircraft as he would not have sufficiently fast reflexes to prevent the aircraft from breaking apart in less than a second! You do not say how much lead added to the tail of your aircraft to see what effect it had on the flying characteristics. I can assure you that you can transform the way a model handles by moving the CG by quite small percentages. Keeping track of how close you are to getting unstable behaviour is what flying inverted and seeing how much down elevator you need to keep the aircraft flying S&L. When a properly trimmed aircraft needs no down trim when inverted it is in the unstable area and you will need to fly the aircraft all the time - you will not be able to rely on any help from the aircraft in recovering from unusual attitudes.

The other point I omitted to address in your first post was the fact that the rudder appeared to have an opposite effect when the aircraft is inverted. This is to be expected. With the aircraft upright, if you stand behind it, right rudder is right rudder. If you invert the aircraft and view it again from behind, right rudder moves the rudder to the left - the servo still moves in the same direction as does the rudder but you are viewing it from a different position. Applying right rudder with the Tx stick when the aircraft is inverted will, therefore, cause a turn to the left. Nothing strange there. Remember that the elevator action is also reversed when you are inverted. The only controls that are not are the ailerons. Left stick still produces a roll to the left and vice versa.

Peter J

[John Muir]

I used to have great fun with my three channel Kamco Kadet back in the eighties. It was quite possible to do barrel rolls with it but inverted flight wasn't easy. Half loop up, push in loads of down and avoid the rudder control like the plague. If you touched it the plane righted itself almost instantly - which is why it was such a good trainer of course.

PJ, I think you're maybe missing the point. PB's model would turn right with right rudder while inverted, as did my Kadet, due to, I reckon, adverse roll coupling. Turning left (as seen from the ground) with right rudder command is, as you say, what we would have expected with most models.

John.

[John Mc]

I moved the C of G forward by re-siting the Rx battery and then took out the slight down elevator trim. I flew the Extra 300 on the 28^th which was a perfect day in our neck of the woods with a wind speed of about ten mph. Normal flying was much smoother than before. Loops ,rolls and stall turns were much easier to execute well. I now found that both normal and inverted flying required only the smallest amount of elevator movement to maintain level flight. I could turn inverted without the nose rearing up unless I tried to tighten the turn too much on the cross wind leg of the circle when there was an easily correctable nod upwards. Even the landing was much smoother. Thanks for all the advice.

Happy new year

John Mc

[Peter Beeney]

Peter, I have to say then, rightly or wrongly, I’m afraid I’ve perhaps not really been paying enough attention to the CG position. As I’ve said before, I’m invariably more interested in the wing loading, and occasionally in the wing and tailplane incidence in relation to the model’s centre line. I have to admit that often I don’t even check the centre of gravity position, or if I do, it’s only a cursory glance. When it comes to something like ARTF, if made as per the instructions, then I’d consider it going to be more or less in the right spot anyway.

I’ve always had a sneaking suspicion that the designer of a model is always going to err slightly on the forward side with the CG, just to make sure. So if I put together something like a Boomerang say, as the club trainer, I always put the battery behind the servos, back in the fuselage. I personally think this makes it more pleasant to fly. With a nice powerful lump too, it has a bit of sparkle and bounce, I’ve never had any complaints from any beginners. Just my opinion, but I’d consider it all to be a cumulative anyway, if you have a model that’s heavy and it needs lead on the nose to balance it, that’s saying that the back end’s too heavy. So it may need a lot of weight to counterbalance, which is now only adding to the overall weight. Much better if you can reduce the rearward weight, but sometimes it’s that’s not easy, if not downright difficult. If the model is very light to start with, however, the amount of nose weight required will most likely be considerably less, and in fact I’d try it first without doing anything. Lobbing a bigger engine in doesn’t always add weight to the front, either; as I remember the Super Tigre 90 two stroke all up was 3 grams lighter than the 75!

To tell the truth we never weighed the lead we fixed on the models, although we certainly tried this on two or three, but it certainly must have amounted to a few ounces. One I particularly remember, a Jetta, an aerobatic plan from the 1980’s, I think, always took off and flew S&L ok, just as you’d expect. But eventually a fairly positive application of elevator saw it instantly flicking out all over the place. So we removed the weight, of course. However, what we did glean from all this, I think, was the fact that a rearward CG was never going to hold any particular concerns or perhaps even terrors. Some folks can get very nervous about their C of G position. At around that time I used to watch some aerobatic lads practising for their competitions, and I do remember the inordinate lengths they sometimes seem to go to to get their models right. Usually trying to lose weight.

One day, when the weather gets better, I might well have another go at some experiments with the centre of gravity, using a hack model, but this time I’ll be a bit more observant; and make some notes, too.

John, Your mention of the Kamco Kadet brought back some memories, although perhaps of the man rather than the machine. A really good friend, he first brought his son along, the lad soon gave up though, but he carried on instead. His Kadet, 4 channel, was covered with nylon, he didn’t find flying at all easy, but his tenacity was legendary. He progressed slowly to a Kavalier, and from there to the quarter scale Precedent Stampe, which I had much pleasure in flying for him; although on the odd occasion he would pluck up enough courage for a little pole around; OS 90FS powered, one of the first. A tad marginal on tugging power, but perhaps that’s just scale. It was here I first got into the intricacies of engine detonation, model four stokes tend to suffer from this. And then one day he just popped off…

PB

Edited By Peter Beeney on 31/12/2013 15:11:02

[John Muir]

John Mc, That's good news. Nice when a thread like this produces a positive result.

On the subject of CoG, I've found that a little experimentation can transform a model. My old Wot 4 was a fine flying sport model from the off, but it always needed a fair bit of elevator to keep the nose up inverted and I found it hard to get a nice axial roll. On the advice of a clubmate I started adding weight to the tail. The difference was amazing! Rolls became instantly easier with a lot less elevator input needed and the plane just felt more responsive in general. On the other hand I currently have a little foam 3D model which has the CoG too far back. If I trim it for level flight while it is upright, it climbs when inverted. If trimmed inverted, it climbs when upright. It is very annoying. I can't get it to lock into a flat spin as it flicks out all the time, on a sequence of rolls, if you don't touch the elevator, it tends to gain height and a lot of other irritating side effects. Now I know for a fact this is because of the CoG, as it used to be almost 100% neutral with a different battery and the CoG forward a little. I have now been inspired to go and relocate the battery position to get it back the way it was. I find all this type of experimentation a large part of the fun of flying I have to say and will spend a lot of time tinkering with my models until I get them just the way I want them.

One more thing, for me a well balanced but heavy model will always be much better than a tail heavy but light one. As I said earlier on I had to land a Seagull Extra with the 'out of the box' rearward CoG and it was awful. Elevator control was terribly twitchy even though the owner had toned it right down and added loads of expo and we were lucky it survived the roller coaster approach. With the extra weight up front, yes it was heavier, but it was much better to fly.

John.

[PatMc]

Posted by John Muir on 31/12/2013 11:36:38:

I used to have great fun with my three channel Kamco Kadet back in the eighties. It was quite possible to do barrel rolls with it but inverted flight wasn't easy. Half loop up, push in loads of down and avoid the rudder control like the plague. If you touched it the plane righted itself almost instantly - which is why it was such a good trainer of course.

John.

John, if a 3 ch R/E model is inverted with too much down held the rudder becomes pretty ineffective. I discovered this way back when I used to roll the Tyro that I learned with. A roll would hesitate half way through leaving the model inverted but could be completed by leaving the rudder full over & releasing some of the down elevator.

OTOH some aileron equiped models that have dihedral will roll faster with rudder held in the same direction as aileron all the way around depending on how much down is applied during the inverted section. The MPX Easyglider being an example.