Thrust Line, ZLL, CG, tail moment

[HELModels]

O.K., I'm building a scratch built pusher and I mounted the motor on the back of the wing and used TLAR for the initial angle of the the motor. I hooked up the battery and attached the wing to the fuse with no tail section attached. I ran the motor up briefly, holding at the CG(.25 - .30 chord), and noticed a severe pitch down of the nose.

Should I see a pitch up tendency even without a tail attached?

[Sparky Paul]

Locate the c.g. vertically.
If the thrust line is above this, the nose will pitch down.
Machs nichts whether there's a tail attached.
Upthrust/downthrust and the tail are used to stop this.

[HELModels]

Locate the c.g. vertically.

Thanks Sparky.

So I should adjust the thrust so that there is no pitch up/down tendency with throttle? I probably want some pitch up under throttle, no?

[Sail 'n Soar]

So I should adjust the thrust so that there is no pitch up/down tendency with throttle? I probably want some pitch up under throttle, no?

No. Ideally you would like to avoid a trim change with throttle. Push the throttle up and the bird will accelerate. That will increase the lift at that trim. And THAT will cause your bird to climb at the angle where the vector sum of lift and thrust equals weight. No need for a throttle induced pitch up.

[HELModels]

.
If the thrust line is above this, the nose will pitch down.


Well, I adjusted the thrust line angle and it is better. To get this line colinear with the CG is tough. It is high wing with pod and boom fuselage, so most of the electronic guts are below and in front. I might end up with the prop pointing 45 degrees to chord line, arrrghh.

There must be a practical limit for this angle before it adversely affects performance.

[Sparky Paul]

Fly it.
Add the input of the tail to the airplane, don't just fiddle with the thrust angle, the whole plane contributes to stability.
This one needed the back end of the motor tilted up, to fly normally.

[HELModels]

Fly it.


I look forward to doing that, but will need a small sized mixer, hopefully under 10g or so.

Yours looks amazingly similar to what I'm building, minus the camera.
Form follows function.

I'll post some pictures when ready, thanks.

[peterangus]

A couple of degrees up or downthrust will have zero effect on the thrust moment about theCG. But it will have a very significant effect on the pitching moment produced by the tail, which is influenced by the propwash.

[HELModels]

But it will have a very significant effect on the pitching moment produced by the tail, which is influenced by the propwash.

Does the propwash influence the downwash off the back of the wing?
If so, is the prowash reenergizing the downwash and redirecting that flow?
Does the prop disc radius and span of the v-tail play a role?

Whatever the answer, most of the geometry has been decided, but
any helpful hints aprreciated.

[peterangus]

ElectroStorch

No need to involve the wing in this discussion.

If the tail is immersed in the prop slipstream, [or near it], then any change to the slipstream speed or direction will have some effect on the tail lift, and hence on the tail's pitching moment about the CG.

With your configuration, upthrust can be benificial. A thrust increase [which produces a nose-down pitch about the CG] will be accompanied by a change in the tail download [giving a balancing nose-up pitch ]

If the tail is not in the slipstream, a somewhat similar effect can be obtained by mixing elevator with throttle.

[John O'Sullivan]

This thread is particularly relevant to me as I just ran into a problem tonight.
I have been flying a high thrust line pusher photo plane (Eyespy) I designed last fall and I lost it in 1500 ft. cloud three weeks ago. I got a call from a gypsum mining company yesterday to photograph a large freighter at their wharf in Hansport Nova Scotia (site of the world's highest tides).
Well my flying buddy Chris Dean had built a similar model powered with a larger motor than I had and he offered it to me to use for the project. He has an AXI 2814-12 motor and I had an AXI 2808-24. My motor will haul the 41 ounce model up at 1150 ft/min. (from LoLo altimeter data), which is fairly respectable, and the larger motor is overkill.
Well, when I went to fly it, it just blew my mind. Opening up the throttle caused the model to veer violently to the right, and to keep it under control at higher throttle required significant left rudder input. I landed and checked the model out. Other than the larger motor than mine, with a larger, broader blade propeller, wing angles, washout and general setup were the same as mine. No discernable warps, twists or anything that would cause any rolling or yaw tendencies.
Well, after varying side thrust and downthrust angles, not much improved. I asked Chris about the right turn tendencies and he said that it did that since he first flew it and he had to apply left rudder all the time he had power on.
As my Eyespy displayed none of these nasty tendencies, I could not figure fhat was causing them in the larger motor version.
Then the penny (cent or peso) dropped. It must be "prop wash". The larger, broad blade propeller must be having a more pronounced wash effect on the vertical stabilizer, resulting in a right rolling effect on the model. Tomorrow it ill be fitted with a smaller narrow bladed prop to see if the rolling effect is reduced.
I will report on the findings.
The attached photo has Chris's model on the left and mine on the right. They are the high winged polyhedral ones.

[John O'Sullivan]

Looks like my photo attachment never got through. Hopefully here it is.

[peterangus]

John O'S

Did you preview your post? I don't know if it still happens, but preview used to wipe-out attached pictures.

[peterangus]

John O'S

You say "no discernable....anything....that would cause rolling..."

I discern a high-torque motor.

[John O'Sullivan]

Hi peterangus
I agree that the motor is a high-torque one and is much more powerful thai I would recommend for this size model (46" x 10" wing). There would be a strong torque effect on throttling up and could well be causing the problem, but at all constant throttle settings, from low to high the problem persists.
I was using 3 cell 2 parallell Lipoly packs. Chris Dean says that the effect is much less on 2cell packs and would tend to support your torque explanation. Three cell packs on my original did not give any problem.
I'll experiment today with different props and battery packs to try and tame the beast.

As the powers that be are sripping my attachments, the following thread has some photos of the model: http://www.rcgroups.com/forums/showthread.php?t=223258

[John O'Sullivan]

Been thinking on the torque effect. With the propeller rotating to the right, torque effects would tend to rotate the model to the left. So torque can't be the problem as the model veers to the right.

[peterangus]

John O'S

I notice your very slender tail boom. Structural deflections can have unpredictable effects, which can be acutely dependent on airspeed.

[John O'Sullivan]

The tailbooms are carbon golf club shafts and are quite stiff and are more than overkill in wall thickness.
I flew my friend's model yesterday for more than 40 minutes using a variety of props. The model glides well with all controls at neutral. It turns right on throttlling up. As power increases more left rudder is required to track straight, At full throttle it will climb vertically but needs almost 2/3 left rudder.
At 1/3 throttle and a little left rudder it tracks well.
Realistically this motor is way too much for this model which was designed to be a sedate, conservative, slow flying photo platform.
It could be tamed by mixing left rudder to the throttle stick, but this would be only a quick fix for this particular model and does not get to the real problem of the turning.
I have a suspision that the problem is related to prop wash effects on the tail.
It still does not explain why I never experienced the problem on my original model, which since last November, must have had more than 50 hours of airtime.

[Sparky Paul]

John, "accidentally" knock off the vertical, and put twin fins on... :cool:

[John O'Sullivan]

Yes Sparky Paul, this should get the vertical stab out of harms way, but would require a fixed stab and complex rudder linkages (it now has a flying stab).
A bit of good news, however. I set it up with left rudder slaved to the throttle and on my initial setup I had a bit too much compensation and had reverse response from the model. When I reduced the mixing, things have now worked out fine. However, this is only a stop-gap measure until I get to the bottom of the problem.
Thanks for your input.
John O'S

[HELModels]

To get back to the original topic of pusher V-Tails..

I built a sheet balsa horizontal tail and slapped it on the fiberglass boom - In a V configuration - Lawn Dart. I salvaged an old free flight tail which used the same airfoil for the horizontal as was used on the wing - Lawn Dart. Then it dawned on me that if I wedged a piece of scrap under the main wing, I could adjust the decalage - Bingo. The amount of incidence needed to get a glide with the FF tail was ridiculous, 10 degrees with CG at 25% chord, about 15% ahead of the Aerodynamic Center of 40%.

If I go back to a flat plate/sheet for a horizontal stab, should I expect that the needed decalage will go down to about 7 degrees? Incidentally, the area for the 2 stabs was almost identical. When I eventually try the V, I will increase the tail semi-span by 18% to maintain total effective area.

Suggestions welcome.

[HELModels]

Tested with motor mounted - glide tested, that is! Lawn Dart again with the CG and incidence unchanged from successful glides. I guess increased wing loading with motor, amplified a very subtle forced descent which I thought was a good glide without motor attached, so I boosted the incidence just a hair. Bingo, she glides with motor attached.

It is ironic but I thought the whole point of "modeling science" was to avoid destructive testing in order to get things in the ball park - "machs nicht"

Next test, hopefully nondestructive, will be to glide her and give her a shot of juice to see what Nasty tendency power introduces. Another brief static test while holding at AC and again at CG showed no nose down tendency as mentioned in post#1.

Happy Modeling!

[HELModels]

O.K., conducted another semi destructive test on her and she is nasty under power - turns Right! Put the wing on crooked so wing and motor is pointed for left turn and flies straight under power, cut the power and turns left.

This is all with the conventional tail. If using 2 smaller verts mounted at the horizontal tips helps, would a V also reduce this tendency? Since with a V, the vertical stab component is projected and not really in the prop wash?

Sparky? Anyone?

[peterangus]

At launch we see high thrust [and accelleration] and low airspeed.
With the thrust line above the CG , there will be a nose-down couple [thrust and inertia] which requires to be countered by up elevator. And at low airspeed the elevator may not be very effective.

Possible remedies:
1....Try a faster launch
2....Try less power at launch
3....Hold up-elevator prior to launch [be ready to neutralise quickly]
4....Relocate tailplane higher, into the propwash [this will increase elevator effectiveness at launch].

[peterangus]

Both models turn right under power when trimmed straight on glide.

If torque were the dominant effect they would turn left. So we must seek some other reason. I suspect propwash on the tail plus structure flexing.

Try these ground tests: Hold the model at the nose.
1....At full power, with rudder & elevator neutral, is there any yaw tendency?
2....At full power, move elevator. Is there any yaw tendency?
3....At full power, move rudder. Is the yaw tendency in the expected direction?
4....At various power levels, move rudder. Is the yaw tendency proportional to the power?

[HELModels]

peterangus,
those are worthy suggestions, assuming 3 channel control. what I gots is 1 channel - motor. I figure if she wont glide, no need for a motor or radio :)
If wont fly straight with power, then hooking up control surfaces could help gain control, but I want something that wants to fly and controlling that flight is easy,i.e. trimmed and stable is what I want. I've got trimmed glide finally, just not trimmed powered flight.

so, I took a real hard look at how things are aligned and noticed the horizontal stab was not perpindicular to the vertical axis. Fixed it and will see if it improves powered flight. This does jive with the flight where the wing was intentionally pointed left and flew straight under power and then glided left with no power.

http://www.modelaircraft.org/templates/ama/300_11.asp

This might be the root of the problem and is one of my original suspicions:
Excerpt from the above link:

"Third, there is the propeller slipstream swirl effect. This swirl tends to change the flow direction over the rest of the plane, especially the tail surfaces. The effect is strongest when the airplane is moving slowly with high power, since this causes the greatest swirl angles. As the airplane speeds up, the swirl is diluted by the faster flow, and the angles are reduced. Thus slipstream effect is reduced as speed increases. Direction of the slipstream swirl moment is nose left."

This excerpt written for a tractor config and would be more pronounced on a pusher since the motor is closer to tail surfaces. would it be opposite on a pusher?

Maybe the swirl angle is a function of prop pitch? Prop Disc radius would probably be a factor too? So, Sparky's recommendation that the vertical stab be out of the prop wash agrees. Whether a V helps remains to be answered.

[HELModels]

Still only have 1 channel operational, but the flight tests continue.
Took the suggestion of trying 2 outboard vert stabs, since this didnt require rebuilding the horizontal - Bingo, what hoot, works like a champ! Hunk of Junk no more!! Tracks straight with power. Glides straight and gentle with no power.

Now, to add control surfaces by decidng how much area to use - Full flying or 25%, and how to accomplish that?

[Sparky Paul]

Full flying on one side. Easy to do.
With two verticals, making both rudder work... use a snake to one, instead of monkey-motion bellcranks, then a direct link across the horizontal to the other.

[HELModels]

Youza, easy? I tried just simply attaching some strips to see what would happen if the horizontal were used for turning. It didnt seem to work right.
If I only have one side fully flying, then wouldnt I only have turning at best? I like how the twin fins confine things on the horizontal, but wouldnt full flying both sides, allow turning And attitude adjustments? The V-tail is calling again.

[Sparky Paul]

There's no law that says both rudders on a twin vertical must be used for control.
One side can be fixed, the other provide the steering.
In the full scale world, such a scheme would result in a misbalance side-to-side due to the need to have a movable surface on one side only; weight and area.
We don't have that problem, the difference in weights is minimal in our sizes.
Make the moveable part large, or full-flying. The other side can be a simple slab.
If this offends :), then make both sides move.

[HELModels]

It is hard to be offended by whatever solution works. The single rudder control surface occurred to me. It's just another design decision that has to be made. If the testbed survives, I'll post the picture.

[HELModels]

Well, I cant post pics because no camera batts. The design decision that I made was to go with the V-tail. It glides and seems to have alot of lateral stability as each semi span was increased by .24, not .18. I say seems because I gave it toss cocked left and it oscillated left right until it settled straight ahead - that seems to hint release stick to recover. The .24 was chosen because I used salvaged FF parts from 2 identical planes - The spacing on the ribs dictated where the break/join would go.

The interesting part of this is that the original FF had a CG located almost 50 percent chord with same tail moment as this new design.

I've been fighting tail heavy situation and wonder if 50 percent CG is risky for R/C? I can manage .35 CG without adding weight. It worked for the FF.
The V-tail does not use flat plate style foil, but is a flat bottom 10% thick foil. Anyone? Sail n Soar?

[Sparky Paul]

An aft c.g., 50% is aft, can use an lifting tail. A forward c.g., 35% is forward, doesn't need it, but it won't hurt. A flat or symmetrical tail would be better though.
With the forward c.g., the tail lifts down. so a lifting tail is not working properly at that position.

[HELModels]

Right you are Sparky. I ditched the lifting tail, tried the flat plate Vee.
Worked much better.

Now development process requires rebuilding the pod so that a carbon sheath can be used as a cover over the balsa/foam. slip the sheath over to fly and unslip to access the gear.