When the real aircraft behaves differently .. what to change

[Bomber]

This is great news

[Darktrax]

Hi Simon
Re: Getting the JSBSim (T4T-Ask_13-set) to work.
The ASK13 download has ask13-set.xml by Patrice Poly, and a multiplayer passenger view ask13-passenger-set by Anders Gidenstam, and a maybe common ask13.xml. This works.

When I add the contents of your your ASK13(Simon) folder (except for the Docs), it then breaks.

I have a couple of little png images of the ASK13 content, but I don't know how to post images to this site.
I guess I have to try and figure out what the stumble is.

[Bomber]

Well you could pm me your email address and I'll zip up my working folder and send it you...

If that fails..... I'd guess forward vs backward slashes.

[Darktrax]

Hi Simon

It seems that to get it to work when LinuxMint uses /usr/share/games/flightgear, and one copies in your set, (requires to be root), one also then needs to set the access and read permissions to 644 for xml files, and 755 to have folders with contents executable flag set.

Now I at least get to the start point. It may now be working

On first tryout, with no attention to any config or help, the joystick and rudder work, but airbrakes lever not linked to throttle, no trim control, and suchlike. The log window from the fgrun launch control reports a very large number of properties "already defined".

Same log window mentions various "Failed to tie property fcs to object methods"

Also, Angle of Attack: Throttle: Pitch Trim: Roll Angle: all Failed.
Ailerons: and Rudder: Passed
Sat on the ground, it seems to wobble in pitch a bit by itself.

I don't know where to find a fgrun log file. I read all this from the log window.
Even though there is more to fix, I am quite pleased the first stumble turned out to be only a permissions thing.

[Bomber]

A quick real life ask13 flying question.. (RL)

at 45kts flying level with no trim.... to clarify is the stick aprox 30% forward away or towards you from neutral ?

[Darktrax]

Well you could pm me your email address and I'll zip up my working folder and send it you...
If that fails..... I'd guess forward vs backward slashes.

Hi Simon
Thanks much for your ASK13 file.
As it turns out, it was byte-for-byte exactly the same as the one I put together. and now that we have got past the file permissions thing, they even work the same. A big step forward!

This is not to say all is solved. Clearly the control mapping from my Logitech Extreme3D PRO stick is not the same as whatever you use. Brakes not tied to throttle, no trim, no yellow ball control. The view on the ground is a trembling wobble because Trim Results: Pitch Angle and Altitude AGL both failed.

There is "failed to tie property fcs/ to object methods" message in there.
Another message about not finding objects for animation: 'terminal_2' , (maybe because ground was at KFSO).

The most serious-looking was

Code: Select all

(process:4496): GLib-CRITICAL**:g_slice_set_config:assertion 'sys_page_size==0'failed

Not sure if fgrun log window output can ever be found as a log file.

Anyway - first try, 4000ft, 45kts held there with stick back a bit, descent rate on the variometer shows as 5m/S.
That is kind of fast!

Even so - we got started. From now on, much more is fixable.

[Bomber]

Don't try and fix that fdm as the new one looks nothing like it.

[Darktrax]

A quick real life ask13 flying question.. (RL)

at 45kts flying level with no trim.... to clarify is the stick approx 30% forward away or towards you from neutral ?

With trim in mid-position, as I recall, I have to keep the stick pulling towards me maybe to 25% (a guess!)

Allow that the instructor presence has almost no effect, because he is over the C-of-G
Me is 91kg+ 10kg for parachute in front seat.
To get the force off the stick, and maintain 45knots flight, needs trim set back something like 85%.

I have yet to get my scene fully working. Trim is still uncontrolled at start-up, so trying a flight won't work just yet.
Sink rate is 5m/sec. Actually quite impressive, like flying a dinner table, but I I am not concerned yet. Without controls mapping established, anything can happen.

[Darktrax]

Don't try and fix that fdm as the new one looks nothing like it.

Er.. you have an even newer one (FDM) ?
Until I can fix up some better stick & rudder hardware, I am OK, for now, to just report on the real thing.

I tried a cheapo dash cam in the car, and had considered using something like it in a glider.
There are LOTs of these things around. The pixels count is not any guide to the resolution.
Even a 1080p format is still rubbish if it started out with a tiny plastic lens and a 1/4" sensor.
Cams with a decent lens aperture, and a hi-res video obviously cost much more.

[Bomber]

That flight model is a year old I'd say... It's been in a re-write state since 1st Sept '15 when I started on spin which has required a complete rethink on AoA calculations.

Please do keep giving your reports on RL Ask-13 flying

[Bomber]

Can I ask you to investigate an issue I have...

On page 12 of the flight manual pdf I include in the docs you can see the H-stab has an incidence of -2 degs....

Now this will generate some positive moment, ie the nose will rise... however in previous posts you're saying to level fly you pull the stick back 25%... even more nose up moment.

What I'm struggling to find is the nose down moment from the wings....with the CoG being 6 inchs behind the datum I'm struggling to get enough if any positive static margin. To my mind the only way to acheive this would be to calculate the wing moments as if the Centre of Pressure is from 40% cord and not 25% cord..... maybe this is correct ?

Any ideas ?

Simon

[Darktrax]

...

... On page 12 of the flight manual pdf I include in the docs you can see the H-stab has an incidence of -2 degs....

Now this will generate some positive moment, ie the nose will rise... however in previous posts you're saying to level fly you pull the stick back 25%... even more nose up moment.

What I'm struggling to find is the nose down moment from the wings....with the CoG being 6 inchs behind the datum I'm struggling to get enough if any positive static margin. To my mind the only way to achieve this would be to calculate the wing moments as if the Centre of Pressure is from 40% cord and not 25% cord..... maybe this is correct ?
Any ideas ?
Simon

Firstly - I have temporary problem in relating the flight manual reference to those you sent.
There is ASK13_Flight_Manual.pdf. It's natural page 12 is nearly empty, containing only the line "7) Approved Versions".
The PDF document scanned page 12 is a 3-View drawing 1:50 scale, scanned page 19, with no direct reference to "h-stab".
On that drawing, left side view, is a "Flosse Stabilizer" dimension showing a physical -2° angle setting. I am guessing that is the value you mean.

About the built-in positive moment.
With my 91kg + 10kg parachute (all approximate) weight in front, I do not think the C-of-G is at the nominal place any more.
I (speculate) that any positive moment is used up, and more, by moving the C-of-G forwards. This could modify the flying to the extent that I have to use some elevator (stick back) to fly normally, replaced by pulling the trim lever back nearly all the way to take the force off the stick.

ASK13 is extraordinarily sensitive to C-of-G loading. Natural page 11 of the flight manual mentions at the bottom of the page that if no parachute is carried, it gets replaced by a cushion of compressed thickness of 4 inches.

Just above, we can see that for occupancy single or two, regardless, the weight range for the front seat is 143 - 220 lbs (64.8 - 99.8 kg)
Clearly the C -of- G is under the instructor seat, so it makes no difference if instructor is present.

We can tell from the page 11a that 17lb weight at the forward foot board will compensate for 22lb on the pilot.
The scaling to shift the C-of-G is then (17 / 22).

The FDM features.
The performance vis. stalling and spinning and the tendency to roll over, caused by lightweight pilot uncompensated by steel weights, and inability to fly at maximum lift or flare out properly because of heavyweight pilot, need not be modelled unless you want to include them as featured inputs. Simply assume the guy in front is the right weight to leave the trim tab neutral as a default.

The (old) model.
The only flyable model I have is YASim one by Patrice Poly, because that one connects to my control stick, and uses the "throttle" to control the air-brakes, and uses thumb switch inputs to move the trim forward and back . The T4T-Ask_13-set controls the main stick only, and various other things go uncontrolled. It nods and wobbles about when on the ground!

[Bomber]

...

On that drawing, left side view, is a "Flosse Stabilizer" dimension showing a physical -2° angle setting. I am guessing that is the value you mean.
[b]

yes this is the reference..

The problem I'm having with the CoG is that we know exactly where it is.. on page 6/18 of the pdf ..left hand side it tells us the empty weight CoG positions max and min at various weights... all aft of the datum and roughly above the centre wheel...
on the right hand side we see the known CoG flight positions that have been tested... 2.76 inches to 9.7 inches aft of datum...

so I'm left with a small -ve static margin that wants to push the nose up.... yet this is opposite to what you describe. (I believe what you're saying)...

I can't change the physical positions of the plane, so I'm looking to see what I can change.

regards

Simon

[Darktrax]

The problem I'm having with the CoG is that we know exactly where it is.. on page 6/18 of the pdf ..left hand side it tells us the empty weight CoG positions max and min at various weights... all aft of the datum and roughly above the centre wheel...
on the right hand side we see the known CoG flight positions that have been tested... 2.76 inches to 9.7 inches aft of datum...

so I'm left with a small -ve static margin that wants to push the nose up.... yet this is opposite to what you describe. (I believe what you're saying)...
I can't change the physical positions of the plane, so I'm looking to see what I can change.

OK - allowing that my aeronautical knowledge is somewhat behind yours..
Conditions:
1. The loading includes the weight of a ideal forward seat pilot (is that P1 or P2?).
2. The C-of-G is at the place the designer hoped for.
3. The centre where all the summed lift may be considered to act at is slightly forward of C-of-C, where the designer intended.

Then the need for the -2° stabilising is to cause positive rotation (climbing) to counter the rotation coming from elevator, body, lift centre on wing chord, etc.
Do we possibly have a sign convention confusion here?
It may be just semantics, or ignorance on my part, but what exactly is that h-stab?
Is it a physical angle of some part or surface?
Does a negative h-stab angle cause positive moment?
Is it the convention that positive moment means climb?

I expected that with ideal weight pilot, the -2° h-stab thing is necessary to get the glider to fly along it's 1:28 slope at 45kts, because the -2° h-stab provides the moment to counter the rotational pitch coming from everything else. It is just not completely clear to me whether a (say) 0° h-stab would leave it climbing and slowing down, or diving and speeding up.

I am toying with the notion of somehow instrumenting the stick to record position at (say) 10Hz, to make a captured data set of (say) a 5 minute test fly sequence. Then we also have a dash-cam or similar placed to see instruments and horizon. The graph might be similar to the stuff they use for F1 racing, where such can reveal if driver braked too late, or messed up in various ways. If I ever do this, it would to be from proper controls position instrumentation. It cannot really be done with a converted Wii device, or from an Android app in a phone strapped to one's wrist! The artificial horizon app I found thinks a steep coordinated turn is still horizontal level flight!

Maybe a fast and cheap way to do this is to run two dash cams, one set to record the controls movements, with another looking out.
Maybe a single camera can be set to show wide angle, and rely on audio to record what one is attempting at the time, and mentioning the settings.

[Bomber]

yes a positive moment means climb

a force acting upward is negative... so lift is negative. The H-stab at -2 degs is creating a downward force ie +ve.
The armature is +ve towards the tail and -ve towards the nose from the CoG or datum...

so h-stab moment is a +ve force multipled by a +ve armature.

the wing however has a -ve force (lift) and would need a +ve armature to create a down down moment,..

I've another question that might shed some light on my poor thinking...

When a plane is at best glide and decending at 28:1.... is it flying at 0 degrees AoA or somewhat closer to 2 degree ?

open the the floor this question.

Simon