Stability with fixed vs freely pivoting forward mounted canards

[Nytrunner]

Which canard behaved smoother on your test flights? If I was doing the experiment, I'd go with the one that wasn't wiggling all over.

 

[G_T]

Since this point does NOT change with angle of attack, what did we see in the onboard concerning the 2 differing pivot points?
"Source: www.theairlinepilots.com

For symmetric airfoils in subsonic flight the aerodynamic center is located approximately 25% of the chord from the leading edge of the airfoil. This point is described as the quarter-chord point.

Thus the aerodynamic center does not change with variation in angle of attack. Due to this, the aerodynamic center, rather than the center of pressure is used in the analysis of longitudinal stability."
"

In the real world, you have deadband issues with most any airfoil. That leads to a non-constant aerodynamic center for some portion of the angle of attack range. That is, aerodynamic center doesn't really exist in a simplified form for real wings. You get pretty close at large Reynolds numbers and incompressible flow conditions, close enough for real world applications for many cases. A little canard on a model rocket doesn't have large Reynolds numbers due to the rather short chord until considerable speed is acquired, then you get into the transonic region where that aerodynamic center is certainly not in a constant location. At slow speeds flow will tend towards fully laminar with probable separation issues around the trailing edge (leading to the non-linearity of Cm and Cl and a bit of pitch insensitivity). At higher speeds the flow may become substantially turbulent and therefore stay attached better.

A wing with substantial leading edge sweep gets substantial crossflow and when the lift is non-zero can change flow mode to a 3D vortex generated from detached flow near the leading edge at angles of attack that might traditionally be considered near to above the stall angle, which changes the characteristics once again.

Once you leave thin airfoil theory behind life gets messier. Going from a fin shape and assuming theoretical thin foil behavior of a symmetric foil to determine aerodynamic center and basing that as how to determine where to put the pivot isn't really safe for a little tiny delta wing that needs to cover a large speed range. And I didn't even mention interference effects from that big tube right next to it...

For an example of messy, just two moderately low camber foils though a variety of camber settings (variable camber foils via hingeline at 70% chord point giving a movable trailing edge). The middle two graphs, Cl vs Alpha and Cm vs Alpha, should readily demonstrate that there isn't going to be a fixed aerodynamic center. https://static.rcgroups.net/forums/attachments/2/6/2/6/5/0/a2525132-28-Zone52vsAG46c.jpg Type-II Reynolds number was held constant at 52K. I didn't find a quick example using Type-I Reynolds number being constant (which is what you will be used to; Type-II is Re*sqrt(Cl) which is more appropriate for a wing generating [sort of, under the false assumption that Cl is insensitive to Re] constant lift for a lot of analysis where people mistakenly use Type-I). I'm not where I can gen up or grab a more suitable example. The variable camber part won't matter for the discussion here, and doesn't matter for what I'm saying anyway.

So, that is the slightly longer form of why I said 25%MAC wasn't a safe choice. I skipped a lot and didn't stay technically accurate, to avoid typing in a lot. And, I wanted to avoid running actual analysis. I have too many other fish to fry and this isn't one of my projects. If someone really wants to play, use something like XFLR5, 2D viscous flow Type-I Reynolds numbers over a range suitable for pad through about Mach 0.8, with an nCrit of perhaps 9 or 10. After that you'll be wanting to look at Mach corrections. And only look at the results for low Alpha as the sweep and low aspect ratio requires more sophisticated 3D analysis for Alpha not relatively near zero.

Gerald - who has spent thousands of hours designing a crapload of foils, foil families, and wings, in use around the world. And I'm still handwaiving but hopefully got the point across in this post. One could write a book on this subject for just the canards. Most intro aero classes seem to teach way too many simplifying assumptions and many engineers never unlearn them... Those assumptions were good enough for the most part for designing planes for WW-II. Go too fast, too slow, too small, get into dynamic flow rather than static (which will be the case during fin flutter or any time Alpha changes relatively quickly compared to distance covered in chords), have swept leading edges, etc, and lots of the simplified theories break down. Badly.

Feel free to disagree, but as I said, I'm still handwaiving, and I'm not here to do everyone's homework. Sorry. I'm just trying to keep someone from doing something really dangerous without due consideration.

https://www.amazon.com/Flight-Vehic...262526441/ref=cm_cr_arp_d_product_top?ie=UTF8 - Should be a good reference. Dr. Mark Drela, MIT, who has also designed a crapload of foils, foil families, and wings, etc, including for the Dreamliner. He also wrote a lot of the code which is in use to do a pretty good job of aero modeling in the low Reynolds numbers range where darned near everything is a function of darned near everything.

 

[G_T]

I want to apologize for the tone of that last post of mine. It was not warranted.

Gerald

 

[Wallace]

No need to apologize. Seems well written to me. Thanks for the advice...