Thrust angle - is there an empirical method?

[MCarlton]

I don't want to get into the previous debate about datums etc, and I know enough to have a stab at working out a reasonable starting point.

(I'd usually plump for 2 degrees down/right)

But, is there a method by which a precise downthrust angle could be calculated?

I was thinking, if we know the pitching moment of the wing section, the incidence angle, rigging angles and everything else, we should be able to bash out an equation which would give a downthrust angle which would result in zero pitch change with throttle setting?

Shouldn't we?

[slipstick]

Yes absolutely. The main catch I can think of is that the "everything else" you need is going to include the current speed (because a lot of the other things like lift and drag and the ways the various centres move are speed dependent).

So calculating a perfect thrust angle at one given speed should be possible....even if not very useful. Well except for FF models which tend to fly at one speed ;).

BTW don't you mean "is there a theoretical method" ? The way we normally do it by experimentation and observation seems perfectly empirical to me :).

Steve

[MCarlton]

I did mean theoretical, I used the word empirical because I was being a doofus ;)

I see what you mean about the current speed thing. Which makes me think some more.

Adjustable downthrust, linked to a dial control on the tx. Would only be a minimal amount of movement I know, but it could be done. Then, when it was set up perfect, the thrust angles could be measured with a protractor and built in permanently and the adjustable system removed.

[vintage1]

First of all there is no correct downthrust angle, and sidethrust is itself fairly difficult to assess.

Dealing with sidethrust first, its there to largely counteract prop torque, so it depends on the prop diameter. It works by translating prop thrust intyo a ywa vector that couples via the dihedral into a roll vector opposing torque. As such it ain't perfect, and works surprisingly well, but it is dependent on how much dihedral you have and how much torque is related to a particular airspeed. If you are climbing, for example, sidethrust wont work as well as in level flight where airspeed is higher and the yaw-> roll coupling higher.

As far as downthrust goes, with a fairly over stable model..forward CG, lots of decalage, it can be totally variable, up to the point where I have noted on more than one model, slamming the throttle open pitches the nose sharply down, until the excess airspeed causes it to start to climb. Nasty in tight spots at low level. Better off lowering the decalage and pulling the CG back.

Really there isnt a better way than TLAR..with 2-3 degrees of decalage and the CG adjusted for some but not too much stability, 2 down 2 right on a plane with dihedral and the wing somewhere near the datum, works. You may need a bit more for a high wing, but that's enough for a 3 channel + set. It was different in the S/C days with no throttle or elevator. There we needed the stability to instantly get out of a dive and therefore more decalage was used, with a lot of downthrust them being needed to tame it.

[MCarlton]

Interesting Vint, thanks for that, I've always had a hazy notion of how sidethrust works, thanks for explaining.

Looking at the model I am building, the wing is low, and thus the centre line of the engine is going to be roughly 4" above the centre line of the wing. Incidence looks to be about 2.5 degrees, but difficult to judge. So 2 degrees will probably be about right, possibly a little too much, so I'll make a start with 1 degree down and maybe 3 degrees right thrust (low wing, low dihedral).

Thats why I'll be doing the first flights without a cowling, so that I can finalise the thrust angles and then build the cowl around that. I know the offset won't be perfect, but there's nowt much I can do about that without drilling the firewall umpteen times, or extending the bolt holes out into slots.

[bjr_93tz]

.
Dealing with sidethrust first, its there to largely counteract prop torque, so it depends on the prop diameter. It works by translating prop thrust intyo a ywa vector that couples via the dihedral into a roll vector opposing torque. As such it ain't perfect, and works surprisingly well, but it is dependent on how much dihedral you have and how much torque is related to a particular airspeed. If you are climbing, for example, sidethrust wont work as well as in level flight where airspeed is higher and the yaw-> roll coupling higher.
.

From my experience I'd tend to disagree with this but I come from a F3A not a scale background so why something is done on a scale ship MAY be different to why it's done on an F3A model.

Using sidethrust to generate a yaw, which in a positively coupled (roll-yaw) airframe could offset motor torque sounds like it's valid enough, however the resultant yaw would be a pain when trying to fly straight lines either horizontally or vertically. Also with my F3A eyes I've seen a lot of scale ships with not enough right thrust, they stick out like sore thumbs in the sky as they are the one which drop their noses on left hand turns and drop their tails in right hand turns, leading me to believe they are flying yawed to the left nearly all the time which again in a positively coupled airframe would result in a roll force ADDING to the motor torque, making it a real pig to fly.

The main benefit from sidethrust is not the little bit the motor will pull to the right but the aligning of the slipstream with the fin, this should only occur at specific combinations of power, speed, propellor size for any given airframe. You can also read into this that the wing root and elevators will "see" this spiraling slipstream as well and hopefully generate a roll force to help counteract the motor torque.

It might be possible to calculate things but there is no substitue for actual flight trimming. Changing COG, overall weight, even flying style can alter your final thrust settings. Also the more right thrust you use the higher the angle of attack the bottom propellor blade will be.

All this is why I fly my RES glider for fun, the only things I need to worry about are ballast, COG, elevator-spoiler mix, decalage, launch presets, washout....oh dear..... :confused:

[BMatthews]

Technically it should be possible to determine the downthrust needed for a particular setup and desired flying speed. But it'll require so much measuring and calculations of lift coefficients to determine it and it's so dependent on the decalage angle and resulting CG location in addtion to the airspeed that it's just not a practical thing to do. THis is why it's just easier to shove in an angle that is roughly in line with the model's design and expected trim. This is why hot dog mid wing aerobatic models won't have any downthrust and high wing, high decalage trainers with well forward CG's have up to perhaps 5 or more degrees. And this could well be for flying at the same speed and throttle setting.

[MCarlton]

Cheers folks.

Technically it should be possible to determine the downthrust needed for a particular setup and desired flying speed.

Is this done in the full size world, on the basis that most aircraft, especially of a GA nature (discounting aerobatic aircraft) spent most of their time at a particular cruise speed and RPM?

I'll stick with a degree of downthrust for starters and see where it goes from there.

[eflightray]

I have always felt that we only use down-thrust because we got the design of the model wrong ;)

The usual need for down-thrust is, -

we have chosen a lifting wing section instead of a symmetrical one,

and we want the model to have too big a speed range for the lifting section,

also we have probably put too much incidence on the wing.

I know there are plenty of models that need down-thrust because they were genuinely designed in a way that made it mandatory to get them to fly.

My comments come about because of the number of beginners who don't seem to understand that the faster a lifting section wing goes the more the nose will go up, the plane will probably slow and stall, and they will be repairing their model or dropping it in a bin.

To actually have a model with a lifting section wing that flies fairly level at all speeds, because the down-thrust and decalage has been set correct is a must for trainer type models. I wonder just how many of the current trainers are that good,......or is that why ARTF/RTF manufacturers usually under-power them. ;)

The wing section, decalage, balance, aerodynamics, etc, even how the model is to be flown etc, all come into play when down-thrust is needed.

[BMatthews]

MCarlton. In the full sized world yes they do set up the thrust line knowing what amount the need to put in and knowing how the design should respond. But keep in mind that the full sized stuff flies in about 4 to 5 very specific modes. Even less for an airliner where the vast majority of the time is spent only in climbing, cruising and descending.

Also keep in mind that in the full size world the craft are set up to let them do their thing naturally. There's none of this "fly flat from idle to full bore" such as modellers seem to expect and demand. For the full size stuff they EXPECT and use the tendency to climb when power is added and to descend when power is removed more as the elevator. In the full size training you'll often see instructions for a plane in cruising flight to use the throttle for altitude changes and the elevator or elevator trim to trim the speed. They are using the natural tendency for the airplane to climb from added power. Why this happens is that with normal trim setups as seen in full sized aircraft the CG is so far forward and the resulting decalage between the wing and stabililzer is so great that they really are a one speed wonder until elevator input is given. So what happens if you add power is the airplane slightly increases speed and then responds to that by nosing up and reducing speed until it's now climbing at the same, or VERY near to it, speed as it was previoius to the power addtion. Same when power is reduced. It'll slow slightly and then nose down into a glide slope that lets it accelerate and nose up gently until it's now descending at the original level flight speed.

We modellers, for the most part, try to fight this natural tendency so we can hotrod around the sky over a widely varying speed range. We do this by "cheating" and adding downthrust. But that's only half the picture. Speed gained by diving will still nose the model up strongly even if there is a lot of downthrust. This is where you'll find a lot of folks bad mouthing a design because it "balloons" too much. But it's just responding to the CG and trim setup they've set it to have. Move the CG back and retrim the stabilizer and you'll lose a lot of this tendency but at the expense of the "built in pitch autopilot" that is so desireable in a trainer. So what's the answer? Easy, we all set the model's trim to suit our own manner of flying. Hotdoggers that are setting up a "here, try it" model need to realize that what they are comfy with is not what a novice will be comfy with. But set it TOO far the other way and it'll confuse and befuddle the novice by seeming like it's possessed by pitch deamons. Like with the temperature of the porridge there the one setting that is "just right".

So where's this leave the downthrust angle? Up in limbo. Because the downthrust angle is there to fight the speed change response in pitch the best compromise angle is going to be dependent on the final CG location and the lift needed to trim that location from the stabilizer. But it'll also depend on the intended cruise speed of the model. And if that speed changes from one pilot to the next then the downthrust required to obtain the best response will change.

It'll also be dependent on how much natural climb with power you're willing to tolerate. If you don't mind a definite but not overpowering amount of climb with added power then you can get away with none to some angle. If you want the model to just hold itself to level flight and go fast then you're going to need some to more. The amount for each range is then dependent on the airfoil angle to the thrust line and the angle to the tailplane as well as the position vertically. It'll depend on the flight speed change induced by the power of the engine as well. A more powerful engine with a faster flight capable prop will require more downthrust to offset the extra pitching force from the extra airspeed than a light power "trainer" engine. This is another reason stuffing a .40 into a .25 size trainer often leads to the modifier scratching their heads when it tries to climb for the stars with the new engine but was just fine with the old. They just didn't understand how the speed to pitch couple all works.

So is it any wonder why we're all saying that there isn't any easy way to determine the angle before hand other than by experience with other similar designs? But even that will only get you into the ballpark even then.

And this fits in with eflightray's reply even though I'd say it's not so much a case of getting the design wrong but more a case of not understanding and ACCEPTING how differing styles of designs want to fly. It's also a case of trying to bypass the more correct method of dealing with this stuff by altering the CG and resulting tailplane trim by using some downthrust instead of getting to the root of the cause. But in the case of trainers where you WANT a strong'ish pitch recovery response then you need some downthrust to help keep the model from wanting to climb like a rocket.

And our tendency to overpower everything these days sure doesn't help at all. A typical .40 trainer these days pretty much as the power to weight of a front line WW2 fighter rather than the paltry power/weight of a Cessna. Keeping in mind what I wrote about too much power compounding the normal tendency to pitch up and you may see the problem. Power that same .40 trainer with a .25 and you'll suddenly find a lot of woes go away IF the new pilot is trained to not wrench the model off the ground but rather let it rise off naturally as flying speed is gained.... just like a low powered Cessna in fact. With that lowly .25 would come the lack of need for much, if any, downthrust because it would not be able to push the plane to speeds that require that much.

I've seen extreme cases where a "trainer" was SO over powered that when the pilot would chop the throttle the model would nose up into a climb and stall that he needed to jam in down to combat. He was totally puzzled by this and told me he kept adding downthrust but it was just getting worse. I gave him the same story as here and advised on taking out half the downthrust he added and moving the CG back in stages until it was flying the way he was trying to obtain. He looked at me like antlers had suddenly sprouted and didn't get it. I found out later that he told someone else that I was trying to get him to trim his model so it would get "unstable" so he would crash it. I had to laugh at that. And yes I told him the elvator would get more sensitive and he would need to lower the amount of travel as well as telling him about the "dive test" to set the CG close but not at the neutral point.

Full size designers have more time to devote to this stuff and run tests on engine stands and wind tunnels to figure this all out before hand. But even so I would not be surprised to find out that there's some prototype or early production changes made to cowls and engine mounts in a lot of cases.

[vintage1]

I have always felt that we only use down-thrust because we got the design of the model wrong ;)

The usual need for down-thrust is, -

we have chosen a lifting wing section instead of a symmetrical one,

A symmetrical section is a lifting one, or the plane wont fly :D

With respect, this is a red herring..

and we want the model to have too big a speed range for the lifting section,

also we have probably put too much incidence on the wing.

I know there are plenty of models that need down-thrust because they were genuinely designed in a way that made it mandatory to get them to fly.

Downthrust is there for really two reasons. To counteract overly stable models or to correct for an offset thrust or drag moment.


My comments come about because of the number of beginners who don't seem to understand that the faster a lifting section wing goes the more the nose will go up, the plane will probably slow and stall, and they will be repairing their model or dropping it in a bin.

That's nothing to do with the section and a lot to do with too much decalage and a forward CG to match it.

[bjr_93tz]

3 for 3 on the last post Vintage1.

I expect that those that use these so called lifting wing sections realise that these sections create an increasing nose down pitching moment as the airspeed rises.....cancelled out and mostly overidden by the decalage.

A lot goes on in an airplane, I'm sure we'd all agree, which brings things back to the original questions and it would be possible to calculate the precise engine thrustline settings for a particular set of conditions but the maths would be beyond me and the majority of us here........so put the calculator away and get out and do some flying :D

http://www.bom.gov.au/products/IDQ60901/IDQ60901.95551.shtml
Look what I've had to put up with for the last 72 hours :(

[vintage1]

Look, its fairly simple. Stability is ensured by having a model that wants to climb as its airspeed rises. So, in a dive, it pulls out.

The model can't tell whether its airspeed is increased because someone opened the throttle, or if its in a dive.

Irrespective of what wing you choose, you will, for stability, have it set up so that more speed=nose up. Unless you are happy flying a neutrally pitch stable pattern ship etc.

Downthrust is there to stop that getting out of hand, that's all, as power stalling is Bad.

[eflightray]

Look, its fairly simple. Stability is ensured by having a model that wants to climb as its airspeed rises. So, in a dive, it pulls out.

The model can't tell whether its airspeed is increased because someone opened the throttle, or if its in a dive.

Irrespective of what wing you choose, you will, for stability, have it set up so that more speed=nose up. Unless you are happy flying a neutrally pitch stable pattern ship etc.

Downthrust is there to stop that getting out of hand, that's all, as power stalling is Bad.

Like one that uses a symmetrical section ?

For 'lifting section' perhaps I should have written 'undercambered section'. :)

[slipstick]

Nope you surely don't mean undercambered, you just mean cambered.

A symmetrical airfoil has zero camber, any other airfoil has some camber i.e. a line drawn midway between the bottom and top of the section is curved not flat. Cambered airfoils include those traditionally called undercambered, flat-bottomed and even semi-symmetrical. The crucial difference is that airfoils with camber produce lift even at zero angles of attack whereas airfoils with no camber require a positive AoA to produce lift.

Steve

[eflightray]

Thank you for putting me right Steve. :)

I won't mention 'over-camber' ;)

[HELModels]

The "pitch up" that requires downthrust isnt quite the same as "goes into a climb" when power is added. What I believe builders want to avoid is the as power is added, it wants to hang on the prop and fly like a pig until airspeed builds again. The only way I can do it, is to allow the angle to be adjusted before the design is finalized. When it flies right, take a picture and measure.

Here's how much this plane needed.

[BMatthews]

WOW! That's a BRUTAL angle. I'd never had guessed that it would require that much. In your case I gather the high thrust line was pushing the nose into a dive?

[vintage1]

The "pitch up" that requires downthrust isnt quite the same as "goes into a climb" when power is added. What I believe builders want to avoid is the as power is added, it wants to hang on the prop and fly like a pig until airspeed builds again.

They are the same, they only differ in degree. Whether you end up with a series of vicious power stalls or merely a very steep climb, or just a gentle one.. is a matter of degree. And that is set by the relation between the CG and the decalage, and the thrust to drag moment. And the downthrust.

[eflightray]

My word!, look at the und...., (sorry), look at the camber on that wing. ;)

[Montag DP]

WOW! That's a BRUTAL angle. I'd never had guessed that it would require that much. In your case I gather the high thrust line was pushing the nose into a dive?The angle has to be so big probably because the prop is so close to the CG.

[eflightray]

The angle has to be so big probably because the prop is so close to the CG.

So it's nothing to do with the wing section then ?, the uuuuuund! Ahhhhh!,......under..errr underside of the wing shape :D

[slipstick]

You're absolutely right Ray, it's got nothing to do with the high camber, low thickness aerofoil ;).

Steve

[Montag DP]

So it's nothing to do with the wing section then ?, the uuuuuund! Ahhhhh!,......under..errr underside of the wing shape :DThink about it like a see-saw. If you want the far end to go up, you could push down on the near end. If you pushed close to the fulcrum you'd have to push pretty hard, but if you pushed towards the end you wouldn't have to push nearly as hard.

It's the same thing with your pusher prop. Since the propeller is very close to the CG, you need a big downthrust angle, whereas if it were farther away you could get away with just a degree or two.

[rick121x]

Does anyone have a rule of thumb for setting the thrust angle on a flying wing?

I am putting two fanjets, the ones from a Phase 3 Fantom, onto a larger flying wing. This plane is an older Graupner "StarJet", similar in planform to a Fantom or Stryker but it has a 49 inch wingspan, and is a "lot" of airplane.. I just came by a new and unbuilt kit for this model. I imagine it is about seven years old.

My experience tells me that the two fanjets will supply enough power for my flying style, but the thrust line must be about 1 1/2 inches above that of the original design, and went approximately through the CG.

My instinct tells me that the thrust line should such that thrust should not cause up or down thrust moments, either due to inertia, or due to the flight forces ie, drag, lift, and weight.

Does anyone have experience in this area of design that might be of help? Or references I might use?

Thanks, Richard Ranney

[Chophop]

Can't help with the flying wing, and I have a question. I just converted a Herr Cherokee nose wheel to tail dragger because it just didn't handle an inch of grass at all. So now some wing off testing, it pulls to the right with power, and the plane will roll straight power off. I can give it a good kick of power and not much changes, it seems to want to go over about 1 1/2 feet right in about 8 feet roll under power. I alsa have a hard wood skid on a linoleum floor, no tail wheel.
It is a low wing, thrust line about 1 3/4 inch above the airfoil middle. It was designed for a 6 inch prop, I'm running an 8 inch and can now go as large as 10 inch prop.
As far as down thrust goes, again, wing off, if I hold it at CG and throttle up, the nose will pull down pretty hard.

So, it is designed for 3* right and down for a 6 inch prop. Any idea if I should reduce that for the 8 inch ?

CH