Aggregate A9 (A4b)

Calm enough this morning to give it another flight. Field still very soft and muddy. The wind was from the SW so the driest part of the field was in the lee of a high cluster of trees which made things rather turbulent low down so its is poor video!.

Hard to see but it rolls nicely!
Flew without the stability on for the first time. It tends to lift the nose very readily and adding more down trim started to eat into its longitudinal stability. A touch more nose weight should improve things.
On the landing approach I noticed a dog (there was absolutely nobody on the field when I started!) making a bee line towards it so I had to adjust the actual touch down to full cross wind in order to land close by. Mud but no damage.

Roll on warmer weather. It is not exactly 'fun' doing development flying in these sort of conditions!

Calm yesterday so with a touch of nose weight (1/2 oz) tried for some low level shots for the camera
It was all going so well.........

Not really sure what happened. It definitely rolled to the left, removing the left wing. It was actually upside down when it stopped!
The electrics are fine, the EDF was still running.
This a 1/5th slomo of the actual crash.

I am not sure it helps any. You can hear the fan touching the casing on the impact.
Repairs well under way.

Eric
You could be right.
My guess at the nervousness is due to the fuselage lift. Its very stream lined shape will create little lift at normal 'wing' AoAs but above a certain figure it will start to build up and its effective centre of pressure will be well forward of the CofG. The big area tail surfaces should help to contain it but I would expect in this condition the relatively small elevators would have to work hard to have much effect.
I really should do some full stall tests.

So its fine at speed with low AoA but at slower speed and even modest alpha it becomes progressively more 'tail heavy' and the elevators weaker!

It may have to flown with an even more forward CofG and just put up with limited elevator authority.

Could the wing have just pulled away ? It looks like it did to me hence the roll and drop

Pulled away?
I don't think so particularly given the amount of mud on that wing tip and the fact that the trailing edge it cut back into the fuselage.
There is also a possible culprit, The lower fin servo has stripped a gear. I don't think it was a result of the crash as it was banked 90 degrees at impact and only the top fin had mud on it!
It is possible that the lower rudder simply stuck at an angle. All the surfaces move with aileron input.
In fact with the smaller EDF roll control at launch is not too much of a problem so the lower rudder will now be fixed and thus be more robust in belly landings.
As all four surfaces are only really necessary for a vertical lift off, which this cannot do, the 'duff' servo will be simply removed. This will also move the CofG forward a bit and save weight.
Win/win!

I reckon you had a positive G stall followed by a negative G stall Q. The aircraft then went into 'wing rock (classic swept wing characteristic), which needed more height than you had available to recover.

It looks very much like pitch-up (and/or pitch-down) due to classic swept wing aerodynamics to me. Swept wings get most of their lift from the vortices that attach to the leading edge of the wing. As incidence is increased, the vortices move and hence, the C of P moves.

The fix is to force the vortices to become fixed by adding a notch, saw-tooth and/or a fence to the wing.

Just a thought, but it might help.

The rebuild is nearing completion.

But the Orange RX3SM now refuses to bind!
Goes through the normal binding procedure except instead of the flashing LED going on solid it simply goes out. The Gyros still work.
Tried with another non gyro Orange and it binds perfectly.
The lower fin is now fixed and its servo removed.

The dodgy LH elevon servo has been replaced.

Quite a job as it and its long wire were completely built in.

Still it is doing what it was meant to - finding what worked on a rather unusual layout plane - so when I build another......

Now that did not go too well!

This morning I made at least two classic mistakes.
1. After the first flight which ended successfully - just - I did not pack up and go home to consider what was wrong but had another go!
2. Despite a breeze and rather marginal control I made a concious decision to "try to get it back" rather than "get down before it gets too far away"! Of course it went beyond the range of the 'park fly' radio and it does glide surprisingly well!
It ended up about 1/4 mile away in the middle of the golf course but fortunately it stands out rather well on a fairway!
So what was wrong?
Coupling the upper rudder with the aileron control was a mistake. It has the effect of inducing aileron adverse yaw which when coupled with the EDF torque meant it would not really turn to the right. Of course no problem to the left. I had hoped the adverse yaw would tend to counter its tendency to drop its nose in a turn but I probably would have done better to have simply used the rudder - as a rudder!
I also suspect some of its problems result from setting its true symmetrical section wing and tail at 0/0. This makes them behave as a single surface, like a delta, but without the delta's inherent stability. I fear it was only the 3 axis gyro that was giving the 'illusion' of longitudinal stability!

So this is the plan.
Build a brand new nose section!
Buy another gyro Rx.
Add a degree or two of wing incidence.
Add scale 'over hanging' extensions to the elevons.
Use the rudder as just a rudder!

First step - remove the broken nose.

It has been taken back to the original construction joint.
Usually for me I actually put bullet connectors on the motor wires which made the whole process a lot easier.

I'm not sure your analysis of a zero/zero wing tail set up being similar to a delta is, strictly speaking, correct Q, but adding decalage (longitudinal dihedral) will certainly increase the longitudinal stability. Remember that classic deltas need longitudinal dihedral too, normally in the form of a reflexed section or washed out elevons/elevators.

You'll need more nose weight, of course, but that is not a big issue as speed increase to carry the extra weight goes up as the square root of the relative weight increase.

Andy
I understood a true symmetrical wing section delta is actually longitudinally stable, within certain limits, but it is not the most efficient way of creating the lift. An asymmetric section with a degree of 'aerodynamic' reflex gives a better lift coefficient.

I suspect the A4b configuration of a low aspect ratio swept wing with a generous close coupled even lower aspect ratio fixed tail plane is likely to have rather complex stability and stall characteristics.

I am laying some driveway paviors at the moment so the new nose build is on temporary hold.

Q,

There's no way a true symmetrical wing section can be aerodynamically stable....where did you get that information from?

...and a reflex on the section would normally reduce the lift coefficient, so I'm interested where you got that one from, too.

Andy

There's no way a true symmetrical wing section can be aerodynamically stable

Agreed as a wing but a delta has specific characteristics.
I have seen reference to exactly this effect in papers on the work of Lippisch with his deltas and reference is also made in Wikipedia .
My own experiments would seem to confirm that a symmetrical section delta (above a particular sweep angle) is aerodynamically stable - just - without reflex.

Agreed reflex will not improve the lift coefficient of any wing section but it could nevertheless still be better than a symmetrical section which is not that good to start with.