What's wrong with my scratch build design?

[ed1972]

Hi, I've just joined the forum so hello, I've been in the hobby about 1 year.

I must admit first off I'm a pretty lousy pilot, but I really got hooked on building. I've done a couple of Flite Test kits, but prefer working with insulation board, and own designs.

I recently tried to build a conventional, mid-wing, and it was (for my low skill level) a real handful, and I was wondering if anybody could give me some pointers. Of course "this plane is too much for your skill level" is probably completely fair, but I wanted to check if I made design faults as well.

It was a rectangular mid-wing, 80cm*14cm, which I called "the V1" due to it's very simple shape and lack of canopy. Weight was 295g (originally), and it had a 2200 motor on 3S battery. Wing loading would then be about 8.6 oz/sq.ft US.

It flew, but very twitchy, especially in pitch. It may be that I had not yet discovered "expo" on my radio, but the throws looked reasonable, like 10mm. What I noticed on the first flight was that I was all over the place the first minute, trying to keep it level without crashing, but once I somehow "found the envelope" I flew several patterns fairly reasonably, but still with a lot of pitching associated with turns, if I didn't modulate the thrust perfectly smoothly with the banking and yanking. I swooped in nicely for a landing at about 3ft off the deck, held it there to bleed off speed, and it stalled very abruptly straight down, breaking a prop. Still, I thought not bad for a first flight.

What I noticed thinking about the design, that I had a quite large tail area (about 1/3 of wing area), the fuselage was rather long. Also, the tail was more or less in line with the main wing vertically, so perhaps in the wake of the main? Also, the tail plane was angled up respective the fuselage, and the other wing, and not down as I think should be correct.

After that, I installed trike undercarriage and a nose wheel servo coupled to the rudder, and was pleased to be able to do a take off run and get airborne. Of course the plane was 20g-25g heavier. But now the plane was a real handful, it would (I think, my eyesight's not great at distance) suddenly tip-stall possibly and get into a dive before I could correct it. Wings had zero downwash angle so maybe that was it.

I thought the balance was a bit tail heavy when I first built it (but think the U/C would have made it more nose heavy), but still, I also tried a bit bigger battery in the nose. I flew a couple of more times, it was like a bucking bronco in pitch, really hard work to stop it diving or rolling. This was strange for me, because I thought that usually tail-heavy planes would behave in this way, but all the mods I made were making it nose heavier, and it was getting worse.

Anyway I gave up and worked on a delta for a while, which flies great but a bit too quick! But I wanted to try and work out what went wrong with my conventional plane for future designs.

Thanks to anyone who read this far, I attach a picture of the machine, and hope that some experts out there can give me a couple of pointers.

Ed.

 

[FDS]

Your pitch was probably partly due to rates and expo, I run at less than 50% total travel on ailerons with 30-50% expo for my trainer.
The mid wing with almost no dihedral that you have set up there is going to make it much more pitchy, it’s great for aerobatics but for learning a higher wing with a good amount of dihedral will help a lot, as dihedral will act like a “keel” to self level the aircraft a bit vs a flat wing. If the wing is in the middle of the fuselage it’s easier for the plan to rotate around it in pitch, vs a high wing design. Not bad in itself but you will notice nearly all the FT designs use dihedral to help stability. In an acrobatic design like the edge you want a flat centre wing but not so much in a general purpose plane.
Your prop is less efficient than it could be too, I would try a 6x4 or 6x3 twin bladed prop. You might find more thrust at lower rpm helps your flying.
More expo helped the smoothness of my flying, as did lower rates. If you can set it up a high and low rate setting on the TX is really useful.
Tricycle undercarriage looks really smart! It is better than tail dragging for take offs etc but is a lot heavier, especially if you go steerable.
Also watch out for all up weight, small to medium designs with greater wing loading are always going to want to have more airspeed to maintain lift, so won’t be floaty and easy to fly slower. Adding more weight like extra features or bigger battery to counter balance a heavy tail will cause these characteristics to be more pronounced. There is a good reason that the foam box fuselage is such a popular base for light models, you can get high levels of rigidity with low built weight, since the middle is mostly air.
When you make turns with ailerons you will usually need to add some elevator to keep the nose up, or some throttle, since you reduce the lift of the wing, if your wing loading is higher this will cause the nose to point down more sharply in turns, the more aileron authority you have the greater the impact on lift etc. Try setting the rates with the TT lift gauge, it’s fairly similar to what you have there and if you have low, med and full rates on a switch it’s easy to flick between them.
You have very nicely wrapped wings and tail on that build!

 

[ed1972]

Thanks very much for replying !

You are saying no expo and rates too high was a likely cause, yes, I now discovered expo (while building the Mini Mustang which lasted 40 seconds). So I will try to set up.

There was some dihedral there, I sanded an angle then glued the roots together, but yes, maybe too little. Also to the wing loading, I'm building a low wing trainer now with ~6.5 oz/sqft and hope it will be better. Getting good wing loading, while not going tail heavy seems like an impossible task though !

When you make turns with ailerons you will usually need to add some elevator to keep the nose up, or some throttle, since you reduce the lift of the wing, if your wing loading is higher this will cause the nose to point down more sharply in turns, the more aileron authority you have the greater the impact on lift etc. Try setting the rates with the TT lift gauge

Yes, I'm not very good at that yet. I also had a lightweight sailplane foam kit (foldable front prop), and that flew much better, but I made similar "wobbles" in my pitch when trying to execute turns, just not so drastic.

My foam is a bit like a "foam box", because inside I "dig it out" like a canoe, so there's lots of air inside, not just solid box.

 

[clolsonus]

It's hard to say anything from just a picture. Pitchiness can certainly be due to pilotage (like moving the control stick too aggressively.) Expo could help with that for sure.
But I would also carefully check two other things:
1. Carefully check your CG location. For a plane like this I would expect it to be somewhere between a 1/4 and a 1/3 of the way back from the leading edge of the wing. As the CG moves forward the plane becomes more stable in pitch, but at some point you run out of elevator authority to hold the nose up and the plane turns into a lawn dart. As you move the CG aft the plane becomes more and more pitchy, and at some point will want to stall all the time. If you flew successfully, you are in the zone, but it might help to nudge it forward a bit?
2. The 2nd thing I'd carefully check is that your elevator, hinge, elevator servo, and linkage are all secure with no slop or play. I've seen a lot of things over the years and a few times my own intuition has failed me ... things I thought would be fine weren't. If you have a similar size FT design that you've flown successfully, maybe compare how solid the elevator/linkage system is between that plane and the one you just designed.
Those are my two thoughts for whatever they are worth. Good luck, it's great to see people designing and scratch building and making their ideas a reality! It's all about the fun of learning and doing!

 

[Merv]

I agree with @clolsonus, move CG forward slightly, use expo, & check for slop or binding in linkage.

 

[quorneng]

ed1972
My own view on your design is that aerodynamically the rear of the fuselage is too short and nose too long.
I suspect you have ended up like that for CofG reasons but it has left you with a plane with sensitive pitch and coupled with significant pitch inertia from the heavy tail and the motor well forward.
This means the plane reacts slowly to its natural stability forces so as a pilot you see it start to, say, dive and it it seems to just continue so you apply lots of 'up'. It does slowly respond but its inertia and limited pitch stability means it continues to pitch up even after you have centred the stick. In goes lots of down elevator to stop it and so the cycle continues!
Further more this sort of condition is made much worse with any structural flexibility or free play in the controls.

Looking at your plane it appears everything has been covered with a film to give a nice glossy finish. Fine for the wing and the nose of the fuselage but does the tail plane and fin actually need it? Remember every gram at the tail is likely to need four in the nose to compensate.

Your battery is probably the heaviest single component. Where is the battery in your plane? It needs to be as close to the motor as possible which will then allow the nose to be shortened and/or the tail extended to achieve a 'good' CofG.

Consider a typical powered foam glider. This is an almost ready to fly type so not my design.

Long tail, a relatively small tail plane and short(ish) nose. It more or less flies itself.
Unless you really want a good glide performance you don't need to have quite so big a wing span or a folding prop.

A good starting point for any DIY design is to simply copy the proportions of something that you know works!

 

[Hai-Lee]

That your design has a few severe handling issues is easy to understand given your initial description of the first design/build/flight. For a conventional design it is important that the wing/tail incidence angles be such that the tail is slightly negative, or equal, in angle for stable flight. This angle requires to be larger as the weight of the plane increases, (for the same wing area).

If your model tries to obtain equilibrium or level flight then the tail and wing will need to find a balance where their individual incidence angles will provide sufficient lift to support and trim the plane. With a negative main wing incidence angle the model will need to fly nose up to obtain a positive angle of incidence but this attitude will give the tail a very large incidence angle and a greater lift than required for trim of attitude and effectively you will need a large amount of "Down" elevator trim just to fly.

This struggle because of wing incidence issues becomes more pronounced as the AUW, (All Up Weight), of the model is increased until you reach the point where high speed stall of the tail will occur and the model becomes uncontrollable at all but the highest of speeds. The statement of how the plane seemed to want to fly like it was more tail heavy after adding extra nose weight is a very strong clue!

Suggestions: Make the tail incidence angle equal to the wing incidence angle as a starting point. Do NOT worry about the fuselage length or tail surface areas at this time as a long tail moment actually has a stabilizing effect as does a large tail surface area. They can be adjusted later after you get it flying properly.

If you are able to fly it as it is do not waste effort in trying to add dihedral as obviously you are not mentioning any severe roll instabilities and so roll stability id not an issue at this time. Besides adding a pendulum weight, (the undercarriage), below the wings should have caused/added a small amount or increase in roll stability. In addition the vertical stabilizer should be fine as you make no claims in reference to yaw instability. Concentrate you efforts in overcoming what is obvious pitch instability. Nothing you adjust will make it fly properly until you fix the wing incidence issues.

Just what works for me!

Have fun!

 

[FDS]

Some great stuff in these posts, thanks all.

 

[ed1972]

Thanks again for more helpful replies !

I have to say I don't have the plane any more After the 3rd bad crash I didn't see any value in repairing again, and stripped it - but am building one at the moment where I want to include what I can learn from the last one, with help of the great feedback from experienced people here !

Clolsonus,

I think you have something about the linkage. The motion of the wire in the tube, and the fact that the elevator was attached to the tailplane with rather a lot of sellotape, not small glued hinges, did look kind of sloppy to me, and I could imagine under aero loads it wasn't responding smoothly to inputs. I will really focus on that job on the next one coming up.

Quorneng,

Yes, close-coupling of wing and tail with long nose, that was originally because of CG, this would decrease the pitch control, but I did at least have a biggish tail. Also, yes, the battery, ESC were all at the front, reciever just at the leading edge. Then I even increased battery size. The CG did appear to be fine by 1/4-1/3 chord rule (and even too nose heavy in later versions). Yes, at least my next plane I did not cover the tail fin, just light spray painting. Still foiled the wings though.

PS The sailplane you show is almost exactly same (but a bit bigger) than a training plane I use (children's glider with add-on RC kit) - I can still have pitch oscillations during turns with it due to my poor control, but nothing near so bad.

Hai-Lee,

I think I heard somewhere, you should have 1-2° relative tail incidence to main wing, but then did it the wrong way around. OK, you say go for zero incidence difference - your advice is timely because finalising and glueing wings on for new plane is coming up. I did think there was something potentially fishy about that angle.

Thanks again all for all the advice, this place is great!

 

[BS projects inc.]

Another thing that contributes to poor flight characteristics is where the controls are in respect to the cg. Your cg creates a single line that the ailerons, motor, elevator, and ruder all fall on. Look at the ft edge, the reason it has such sensitive controls (and it should because its a 3d plane) is because when you have your wing, elevator, rudder, and motor all line up with the cg on the same axis you have far less resistance when trying to do rolls which can cause stability issues on planes not meant for 3d.

 

[Hai-Lee]

3D planes and most of those that use a 0/0 incidence setup through the CG are very aerobatic and have light wing loading. When you increase the wing loading the loading on the tail must increase in order to allow the wing to settle with a higher incidence angle in order to increase its lift or it may have difficulty in maintaining or even achieving flight. In other words a nose heavy setup can be almost fatal to its ability to fly. Sadly the increased tail loading reduces the planes low speed abilities and often crashes are the first sign of incidence/loading issues.

Sure a 0/0 plane can seem unstable but the distribution of mass can improve some inertial resistance to sudden attitude upsets or coarse control inputs. The normal requirement for a 0/0 incidence as stated is for high performance sports/3D style aircraft or those where high speed performance with low drag is required OR as in the case with 3D models that the control surfaces are all directly exposed to the high speed propeller driven airflow which allows for control functions below what would normally be considered as flying speed.

.The tail surface area and its distance from the CG/centre of lift is important but generally the larger the tail surface the greater its effect and the less critical it is to get the balance point exactly correct. The tail moment, (distance to the balance point), can determine how "twitchy" or how fast the tail is going to respond to control inputs. A short coupled tail applies its forces in pitch over a shorter distance and so gives less leverage BUT it has less distance to move to give greatest effect and hence in the extreme it can seem twitchy whereas a long moment can actually apply greater forces through leverage BUT it has to move a greater distance to effect large changes in pitch and so can seem slow and even sluggish.

The biggest issue any scratch design.build strikes is WEIGHT as it determines EVERYTHING! Thankfully I have only ever built one EMU. (For those from a different part of the world an EMU is a very large, heavy, bad tempered and unpredictable flightless bird). and that was many decades ago now. Sure I still make some heavy birds but they all fly following a simple rule of paying attention to the wing loading and wing incidence angles.

Have fun!

 

[BS projects inc.]

3D planes and most of those that use a 0/0 incidence setup through the CG are very aerobatic and have light wing loading. When you increase the wing loading the loading on the tail must increase in order to allow the wing to settle with a higher incidence angle in order to increase its lift or it may have difficulty in maintaining or even achieving flight. In other words a nose heavy setup can be almost fatal to its ability to fly. Sadly the increased tail loading reduces the planes low speed abilities and often crashes are the first sign of incidence/loading issues.

Sure a 0/0 plane can seem unstable but the distribution of mass can improve some inertial resistance to sudden attitude upsets or coarse control inputs. The normal requirement for a 0/0 incidence as stated is for high performance sports/3D style aircraft or those where high speed performance with low drag is required OR as in the case with 3D models that the control surfaces are all directly exposed to the high speed propeller driven airflow which allows for control functions below what would normally be considered as flying speed.

.The tail surface area and its distance from the CG/centre of lift is important but generally the larger the tail surface the greater its effect and the less critical it is to get the balance point exactly correct. The tail moment, (distance to the balance point), can determine how "twitchy" or how fast the tail is going to respond to control inputs. A short coupled tail applies its forces in pitch over a shorter distance and so gives less leverage BUT it has less distance to move to give greatest effect and hence in the extreme it can seem twitchy whereas a long moment can actually apply greater forces through leverage BUT it has to move a greater distance to effect large changes in pitch and so can seem slow and even sluggish.

The biggest issue any scratch design.build strikes is WEIGHT as it determines EVERYTHING! Thankfully I have only ever built one EMU. (For those from a different part of the world an EMU is a very large, heavy, bad tempered and unpredictable flightless bird). and that was many decades ago now. Sure I still make some heavy birds but they all fly following a simple rule of paying attention to the wing loading and wing incidence angles.

Have fun!

For those who still don't know what a emu is it's also known as the Australian military's biggest rival.

 

[Hai-Lee]

For those who still don't know what a emu is it's also known as the Australian military's biggest rival.

Nah! They're our military's most numerous and dangerous stealth weapon! They can run very fast, very quietly, are hard to see in the bush and if they attack their kick with razor sharp claws can disembowel in a single slash! That's why they are everywhere in the bush!

Did you ever wonder why our governments order culls for various species of local wildlife, from time to time, to control their numbers but NEVER Emus?

Invaders be warned!

Have fun!