Beagle Pup Flight Testing

[bomber]

There's a new beagle pup flight model in town...

viewtopic.php?f=15&t=816&start=10

My first observations are after a glide from 10000ft @ 50kts spawn.

I believe this plane should glide with engines off at about 80kts... I've read somewhere that this should be about 20kts above stall speed so as to give the pilot plenty of time aloft without having to worry about stalling or have a high workload to keep the plane flying... the plane should fly itself, and your job being to find a suitable landing place.

A quick test shows me she's coming in at 122kts, with an AoA of -0.948degs.
This gives her a vertical speed of -57fps, which might not be able to be classed as gliding, more plummeting with style.

So before any further testing I'd ask for a review on this by Richard.

Anyone else like to chip in with their findings ?

[Richard]

The only data I've current got is from "Comparison of flight test data with a computational fluid dynamics model of a Scottish Aviation Bulldog aircraft"[1] and it is as follows.

Code: Select all

Table 1 Flight test conditions and results for the straight and level flights
 
IAS   Altitude   OAT   TAS   Density   Re   AoA   Cl   Cd
kt   ft      ms-1   kg/m3   x10^6   
60   5000   8   33.4   1.045   2.25   11.4   1.349   0.218
70   4980   8   39.0   1.046   2.63   9.10   0.997   0.167
85   5020   8   47.4   1.044   3.19   6.60   0.683   0.108
95   5020   8   52.9   1.044   3.56   4.70   0.550   0.090
112   5020   8   62.4   1.044   4.20   3.40   0.398   0.067

Table 2 Flight test conditions and results for the power-off glides with data corrected for rotating propeller (ignition on)

IAS   Altitde   OAT   TAS   Re   GldAng   Cl   Cd   Propeller
60   5800   12   34.2   2.19   6.7   1.289   0.152   stopped
70   5800   12   39.9   2.55   6.4   0.944   0.106   stopped
80   5800   12   45.6   2.92   7.0   0.720   0.088   stopped
90   5800   12   51.3   3.28   7.5   0.565   0.075   stopped
100   5800   12   56.9   3.64   7.2   0.465   0.059   rotating
100   5800   12   56.9   3.64   10.2   0.462   0.083   corrected
110   5800   12   62.6   4.01   8.1   0.382   0.055   rotating
110   5800   12   62.6   4.01   10.8   0.379   0.073   corrected
120   5800   12   68.3   4.37   8.4   0.324   0.048   rotating
120   5800   12   68.3   4.37   14.3   0.317   0.081   corrected


In my experiences "hitting the numbers" is quite a broad approach to flight testing - it can be useful to get an idea if there is some underlying problem; but often just getting close to a published set of numbers is all that's ever going to be achieved - except when it is a consistent set of qualification test guide data that has been put together specifically for the purpose of validating an aerodynamics model.

However I'm obviously interested in all feedback, especially the handling qualities, takeoffs, landings, controllability, stalls, spins, flaps, control surfaces effectivity.

The propeller aero effects are something that I've considered allowing to be selectable - but for now I think we should leave them in as in my test flights this extra little bit of realism makes a difference, but that could be just my imagination.



----------------------------
[1] Comparison of flight test data with a computational fluid dynamics model of a Scottish Aviation Bulldog aircraft, N. J. Lawson, 1278 the Aeronautical Journal December 2013

[bomber]

My initial observation is that this plane as an inherent trim ie the decalage between the front and rear wings balance out at 122kts..

[KL-666]

Well, i gave it a short try. Installed beagle pup and added the stuff from the fdm file on Richards website.

Then tried to take off the vsp model. The plane is extremely unstable in the pitch. But worse, it only wants to fly in circles to the left, even with full counter rudder deflection. So i can not really try that one.

Then i tried the 160 model. That one flew decently as far as i could see with a take off and landing. Did not test anything yet, It was just a short first try.

Which one should actually be tested?

Kind regards, Vincent

[SHM]

I'll give it a try over the weekend.

[bomber]

Another observation is that the pilots aren't adding any weight to the plane....so she's a bit light.

[Richard]

Thanks for the testing and the feedback.

1. I'll add the pilot / co-pilot masses

2. Simon and Vincent, you may well be onto something with the pitch moment being wrong. I think this might be related to the wing incidence and/or CG - I'll do some work on this.

3. Vincent - it's the pupvsp that is mine; the rest are by Richard Senior. Can you confirm that the tanks are loaded equally; and that no trim has been applied. It could be that the propeller aerodynamics are be incorrectly affecting takeoff - flight reports that I've read, and videos I've watched seem to show that the aircraft may be a bit susceptible to yaw - one report said "requires a fair amount of footwork".

4. Maybe I should add a way to turn off the propeller aerodynamic effects ?

[KL-666]

Hi Richard,

I can not try to experiment with the yaw issue, because the pitch instability complicates controlling the plane too much. I'll wait for a pitch fix first. Btw. first flights i do with the plane as is. So i did not change the fuel, just used what you gave me

About the pitch: generally the plane gets instable when the center of weight is not far enough before the center of lift. But you may know that already.

Kind regards, Vincent

[bomber]

KL, Try spawning at 10kft and 50kts with no engine running, and touch nothing. You'll find the plane flys straight and true. This is a major clue that there's nothing inherently wrong with the airframe itself with regards any yawing or rotating.

A glide test also tells you the natural trim of the plane, the speed at which the pitching of the wing is equal to the counter pitching of the h-stab... At present this is at 122kts (140mph) which is past it's straight and level max speed of 110kts. Now I can't see why a designer would create a plane with this characteristic as what I expect them to be looking for is that if the plane goes into a shallow dive and picks up speed for the planes nose to gently rise. What we have here is that the plane at any speed that it's capable of achieving under powered flight for it to want to bunt into a dive, this makes it twitchy.

A plane gets twitchy when the AoA of the h-stab is not sufficiently negative in relationship to the AoA of the wing....by this I mean if the main wing is +6 degs the h-stab needs to be seeing +1deg...

for some insight

http://www.av8n.com/how/htm/aoastab.htm ... -stability

[Richard]

Helped by Simon's guidance I've spent some time today tuning the pitch moment; I settled on changing the incidence of the wings and htail to try to get to a balance, using the 2.5degress from volume8.pdf as a guide. I'm currently using wings at 4.0 degrees and 3 for the htail.

I've added an option to the Preferences dialog to dial in the propeller aerodynamics - on reflection I think these are quite wrong. Firstly I need to understand what JSBSim is doing, and secondly what VSPAero is doing to avoid duplication. The model will start with no propeller aerodynamics.

I've also removed the aero effects from the landing gear because after a good testing session I think the numbers are completely wrong; and probably need to be calculated differently.

Updated model files : http://chateau-logic.com/sites/default/ ... 6-10-30.7z