STOL Performance Observations

[BradW1062]

Nev - That sounds good, thanks for the clarification. I see what you mean about regaining speed when you are low, not much time left to do that really. That's probably what was concerning me (I used the term "alarmed" above) because I was thinking why would you want an airplane that lost elevator authority on approach? The key is that I probably have lost sight that this post was originally about STOL performance, short and slow landings - not normal approaches. So it makes sense that keeping the airspeed up is the key. I mean, even in the Champ I am doing 1.3Vs. Stall is right at 45 mph, so I am usually flying these gliding power-off approaches right at 60 mph, even when slipping. So it's exactly as you say at #2. I don't ever really try to approach at minimum speed in the Champ. I guess until I start flying in the mountains and forests out west or something, STOL isn't really anything I care much about.
Brad

[BradW1062]

And the Bearhawk is already pretty much a STOL aircraft even with it's "normal" approach...

[Nev]

There's more written about BH 4B handling on the Bearhawk Blog website HERE.

[BradW1062]

It's an interesting question, for me. Does it actually matter?

Increasingly I am wondering which model (A or B) has superior STOL performance - or indeed, if there is any appreciable difference in the aerofoil alone. Other factors like weight, VGs, wingtips, build quality, and pilot inputs probably make a greater difference than the model does. I am increasingly convinced about this.

I would also note for completeness, I modified my aerofoil to increase camber, by lowering it at the trailing edge. My trailing edge sits about 1.5-inches lower than the template, from memory.
This isn't a big deal in practice, the ailerons are reflexed up slightly in flight due to wind loading, and the flaps are obviously deployed when landing. The wing roots and tips are the only parts not subject to changing effective camber.

Battson - Since I have the hyper-curious-analytical affliction, I just had to do a comparison of the airfoils in the computer. I know, I would rather be in New Zealand flying around with you guys any day - but I'm not, so here ya go. I use Xfoil, which is sort of an industry-standard 2D analysis package, very robust and allows for mods and such. As some will say, this is just a 2D wing-section flow analysis, not 3D computation fluid dynamics, and definitely not the real world. And they are right, but you just have to understand what this sort of comparison represents and act accordingly. In general, the real world coefficient numbers will be about 80% of these. This is mainly due to aspect ratio of the 3D wing (BH is about 6.5) versus the "infinite span" assumption in the 2D computer program, and the fact that the real world loading is not uniform across the span. It is basically elliptical for a rectangular planform wing, so at least it is subject to known approximations and estimation. What I am saying is: your mileage may vary...

I was intrigued by your mention that you dropped the trailing edge by 1.5". I didn't know exactly how you did it, so assumed that this resulted in a camber increase of 1.5/66=0.02273, and I put that in the camber line of the A-model 4412. Also, Rod Smith had mentioned above that Bob Barrows had drooped the nose of the BH A-model wing's leading edge - and I wasn't aware of that. I had never heard it mentioned, and hadn't looked at the plans close enough to see it - but there is clearly a 3/16" drop of the center of the leading edge radius on the plans. This has the effect of increasing the camber of the 4412 from 4% to 4.4% as far as I can tell. I assumed here that Bob took this droop back to 20% chord, which is typical but that is also just an assumption. Your mod on top of this, assuming I did it close to the way you actually did in reality, results in an increase from 4.4% to 6.66%. That's a whopping increase. This essentially makes the airfoil a 6412 instead of a 4412 as I had suspected before. Of course, all of this is idle (at about 3000rpm in my case) speculation.

Here are the results from Xfoil (if this works, I rarely attach anything):
Bearhawk - Barrows+Battson Mods - airfoil comparison plots.pdf
Bearhawk airfoil comparison - Xfoil polars.pdf

The airfoil plots show the standard 4412, the nose-droop mod that Bob did, and the tail-droop that you did. You can clearly see that your airfoil would be cambered a lot more than the others - I wonder if this is what it looks like? All of these lift/drag/moment polars were done with the same parameters in Xfoil. They are "constant-lift" polars, which vary the Reynolds number and Mach number with the square root of the lift coefficient. This simulates the fact that, in the real world, you fly with the need for constant lift (to support a constant weight) over a range of operating speeds, and low lift coefficients are required at high-speed, and high lift coefficients are required at low speed. So the Reynolds number and Mach number both increase/decrease with lift coefficient. Without this method, you would have a low-Re curve and a high-Re curve for each airfoil and it would get hard to see in about a minute. I am not a certified expert in this, just interested and know enough to be dangerous, I guess. So, like I said before, you may see different results.

I included the Riblett GA30-413.5 polar for comparison as well. You can see that although it has a lower pitching moment, as advertised, it also has less overall lift, and lower overall lift-to-drag than the 4412's. Also, the high camber of your modified airfoil (again, if this is even close) adds a good bit of lift in the high lift coefficient (1.2 to 1.7) region, just below stall. Exactly what you would want for STOL performance I would think - . But it appears, as usual with higher-camber airfoils, the trade-off is that there is also a good bit higher pitching moment. I wonder if this is what you have experienced? Lots of trim required in cruise? I dunno...

And - I can simulate flaps - but that's a whole 'nother level of analysis - and as we have all discussed, there's such a huge range of build variations and operating conditions that it probably is an academic exercise. The real world will be different, and the big flaps on the Bearhawk will negate most of the arguments about this-airfoil's-lift-versus-that-airfoil's-lift anyhow. So the question remains - does it really matter?

Brad

[Battson]

Wow that is really interesting - I wish I would have kept practicing this stuff, after 15-years of no fluid dynamics / aerodynamics, I am stretching by brain just following you through this thought process. Thank you!

[Nev]

BradW1062, nborer discusses the wing design in another thread HERE where there is also discussion of A vs B. Many of us enjoyed reading Nics observations about the design.

[BradW1062]

It is interesting that he (nborer) doesn't think there is any real difference, based on his "as supplied on the form block" Model B airfoil. That implies to me that Bob has once again modified the airfoil in some way. All of my analysis is based on data published by NACA and Riblett, because I wasn't aware of any modification. I used Riblett's coordinates in the lift-drag polars that I did to compare the airfoils.

When I found out that Bob had modified the 4412, I dug out my plans and verified it, and incorporated the mod in my analysis above. This still shows that there is a significant difference with the GA30-413.5 as provided by Riblett. So if someone could give me a copy of the new B airfoil as provided by Bob, I could do a more accurate (academic) comparison.

Outside of also drooping the nose on the GA30-413.5, I don't see how there is little to no difference. All of Riblett's airfoils show this tendency of having lower lift in exchange for lower moment - it was his "design" philosophy, to forward-load the airfoil for lower moment. Not to the extreme that NACA did with the 5-digit, which Riblett rails against at length (and I don't disagree), but his camber lines are clearly forward-loaded. I guess my ability to have a better understanding of this difference is on hold until I learn what mods Bob made to the Riblett airfoil on the B model to improve the lift situation, making it have "little difference" compared to the A model 4412-Mod.

[BradW1062]

Another builder graciously provided his measurements of the B-model airfoil, and they indicate that there is no obvious modification to the airfoil ordinates that Riblett published. So I guess for now, that means the polars I posted earlier are representative of the airfoils we have. For what that is worth.