REFLECTOR: 65VT/ angle-of-attack system

[Jim Sower]

Brian Michalk wrote:

> Okay, Jim.
>  
> Your argument makes sense.  So does this mean for a standard 
> (non-canard) aircraft, that an AOA gauge will show different values 
> for stall speeds depending on flap extension? 

The short answer is yes.  Flaps change camber, different camber is a 
different airfoil and will have different attributes.  One attribute is 
stall AoA, and IIRC it is somewhat lower for most airfoils but I'm not 
positive.  I do know it's different.  My experience was on swept wing 
tactical jets, a LONG time ago..

> Since the main wing is fixed relative to the fuselage, and assuming no 
> flaps, I would see where the AOA gauge would perform.  When people 
> install those gauges, is it only for a single flap configuration?

Optimum (however you define that) AoA would be different with different 
flap settings.  It's sort of academic since AoA is most commonly (and 
necessarily) used on swept wing airplanes that have large (as a portion 
of empty weight) variations in weight and large speed envelope.  Like 
airliners and tactical jets with a payload that rivals their empty 
weight.  These guys never land with partial flaps (why would anyone want 
to?).  

> Okay, now back to canards.  Since the airfoil that stalls is the one 
> that the elevator is mounted, we no longer have a fixed chord.  My AOA 
> sensor that I am planning to grow myself uses a handfull of air 
> pressure sensors feeding into a very small microprocessor. 

Rutan had one he designed for (I believe) his Vari-Viggin that was just 
a vane attached to a scale inside the fuselage.  A vane attached to an 
angle sensor would work great and cost practically nothing, is easy to 
calibrate and etc.  How would you calibrate and maintain a "... handful 
of pressure sensors ..."?  I don't even know what that means.  Fighters 
and airliners use a vane on an angular transducer.

> I assume I would also need to feed it an aileron deflection reading.

Aileron deflection?  What for?

> However, this poses a problem.  I am a big fan of neural networks.  
> I'm not smart enough to figure out the non-linear equations (NASA said 
> so) that would return the AOA.  I could wash it through Excel, and 
> come up with something, but ... well, it's a lot of work, and I do 
> some of this _at_ work, and its no fun.  So, run the data through the 
> NN.  I was going to use a traditional vane as my reference, but now, 
> that's not even the true AOA.

The traditional vane is good enough for the big guys.  But I'm losing 
track of your objective.  This is starting to sound like the B-2 project 
- the thing has no mission ~ we're just doing it to demonstrate that 
we're *able* to do it.  What exactly is your purpose for all of this?

> I could run the plane through various speeds and loads at stall to 
> gather data, but I like feeding a NN with lots of data ... I'm talking 
> thousands of data points.

What's wrong with just stall AoA for froward, mid and aft CG?  You'd see 
how much they actually vary, and I don't know what else you'd need.  

> You made a remark that the canard does not stall at full elevator 
> deflection.  Okay, but wouldn't it always stall at the same 
> deflection?  If no, why not?

I don't recall saying anything like that.  I did say that at aft CG, the 
canard will stall at a modest elevator deflection, and at forward CG you 
might need full nose up elevator to stall.

Seems to me that the issue is landing slow.  That's the main thing I 
used it for ~ going aboard ship which you want to be as slow as 
possible.  FAA mandates landing speed 1.3 Vs.  If you want to land at 
1.05 or 1.1 Vs, you *definitely* need AoA.  Apart from that, I see no 
need for it at all.  If you want to land slower than 1.3, what I do is 
fly 80 kts in the turn and gradually decelerate to70 in the groove (or 
75 if it's gusty and choppy).  Depending on the situation coming across 
the overrun, I will cut the power sooner or later and touch down at 70 
or below (65 and below on a good day).  I get enough AoA input just from 
seeing/feeling how cocked up the airplane is.  It feels a LOT different 
below 70 kts than above, and that's good enough for me.

If I had good AoA, it might get me IN to a lot more trouble than it got 
me OUT of .... Jim S.


>     -----Original Message-----
>     *From:* reflector-bounces at tvbf.org
>     [mailto:reflector-bounces at tvbf.org]*On Behalf Of *Jim Sower
>     *Sent:* Monday, October 04, 2004 8:55 PM
>     *To:* Velocity Aircraft Owners and Builders list
>     *Subject:* Re: REFLECTOR: 65VT/ angle-of-attack system
>
>     Al,
>     <...The relative AOA of the two airfoils is fixed. ...>
>     Actually, they're not.  AoA / incidence are respectively the
>     angle(s) between the chord of the airfoil and the airstream /
>     fuselage centerline respectively.  The airfoil chord line is the
>     line between the "stagnation point" on the nose of the airfoil to
>     the tip of the trailing edge.  Wing flap or canard elevator
>     deflection increase the camber of the airfoil and increase both
>     the incidence angle and the AoA.  An AoA indicator measures
>     airflow relative to a *fixed* line (the fuselage centerline?) and
>     assumes *fixed *incidence angle (which is not the case with a
>     canard unless you never deflect the elevator).
>
>     <...could be useful for determining canard stall at one
>     configuration; full elevator deflection ...>
>     Actually, the canard stall does not always occur at full elevator
>     deflection either. 
>
>     With aft CG, relatively little elevator deflection rotates the
>     fuselage considerably, and the the canard will stall at an AoA
>     that consists of the largish fuselage AoA plus the relatively
>     small increase in incidence caused by relatively small elevator
>     deflection.
>
>     With forward CG, it takes much more (full?) elevator deflection to
>     rotate the fuselage less, for the canard to stall.  Here, the
>     fuselage AoA is much smaller, but the canard incidence increase is
>     much larger with full elevator deflection. 
>
>     The canard stalls in both instances, but the "AoA" that the
>     transducer sees is a good bit lower at forward CG than at aft CG.
>
>     You may recall from your Cessna days that the airplane had to be
>     pretty cocked up to stall clean, and much less so to stall with
>     the flaps down.  The AoA of the *airfoil* was not that much
>     different clean or dirty, but the incidence of the chord line
>     increased radically when you lowered the flaps.  By the same
>     token, cruising at, say 70 kts the airplane had relatively level
>     attitude, but drop the flaps and maintain 70 kts straight and
>     level and you were looking at the trees.  The AoA of the airfoil
>     didn't change radically, but the angle between the wing chord line
>     and the fuselage did.
>
>     So for a canard, AoA is a little like teats on a bull .... Jim S.
>
>
>     Al Gietzen wrote:
>
>>Angle of Attack is a very useful tool for conventional plan form 
>>airplanes.  It is of little value for canard type aircraft since the 
>>wing you're trying to observe is the canard, and the incidence and 
>>aerodynamic properties (like stall AoA) are constantly changing (with 
>>elevator deflection). 
>>
>>It just doesn't tell you anything particularly useful ... Jim S.
>>
>>Good point, Jim.  However, I do think that it could be useful for
>>determining canard stall at one configuration; full elevator deflection.
>>Using the pressure port system, I'd expect installation on the main wing
>>will work fine; you just need to calibrate it the canard stall point.  The
>>relative AOA of the two airfoils is fixed.
>>
>>Al
>>      
>>