CIDCrazyJay
Well-Known Member
Moderator Note: This thread was forked off another about designing for stall behavior, as it's become a topic in its own right.
------------------------------------
The title was regarding stall tailoring. But it appears this is already pretty well understood.
I'll add; another way to ensure the root stalls first, is a change in airfoil section from root to the 75% b/2 where the highest lift airfoil is located. Keeping mean camber lines the same makes tailoring the lift distribution easier, E.g. NACA 66-218 root, with its sharper leading edge, and a slightly more blunt 63-215 located near 0.7 semi-span. You may have to rip one of the camber lines and apply the other thickness distribution to it for consistency; xFoil has this feature built in. For laminar airfoils, maximum lift and highest stall angle is found among the 15% t/c thickness foils. This is 13% for the classic NACA airfoils. (You have to scale them up from 12). If the section is to include flaps, then 18-21% will provide the greatest lift increment improvement.
Although the first part of this discussion was about stall speed as much as stall progression: I'm not sure what the design intent is, so my comment includes many generalizations. The following ramblings will be on stall speed:
Every textbook and aircraft design guide focuses entirely too much effort on stall speed.
And at least 100 otherwise perfectly good airplanes have been ruined by this obsession over this one requirement.
The FAA has been migrating away from the 61kt rule. Only LSA (someday maybe Mosaic) must adhere to an arbitrary stall speed requirement.
There is zero correlation between stall speed and flying-related fatalities. ~ The false assumption that Slow = Safe, is the single most ignorant idea which has lead to ten-thousand pilots and their passengers no longer being with us. For gods sake stop flying slow. If you want to go slow then WALK!
Airplanes with 7lb wing loading and 35 knot stall speeds are objectively and statistically more dangerous, and far less useful, than airplanes having 70lb wing loading.
What matters is; can the airplane take off or land in a certain distance which the operator finds useful? And; does it fulfill its mission requirements?
If grass strip, or off-airport landing is a requirement, either for bush flying or for potential emergency landing in a bean field, then a lower stall speed is certainly important. If the aircraft is intended for recreational flying in a small geographical area, then this consideration might be leas important. And what follows probably does not apply.
IMO, it is a waste of time for designers to agonize over this one detail, except for certain specific uses. If STOL ops, or emergency landing speed is important: Pick a general ballpark landing velocity, and see that wing size meets that need. Then see if the other requirements can be met at that wing area. ~ Generally, low stall speed makes every other part of the design 10x more useless for each knot you take off the low end.
If you get down to 24 knots (Part 103) you will have a nearly useless airplane. If you can live with a 124 knot stall, you can have a 747-800 with full fuel load, a hundred friends, and a pallet of Champaign. And If 210 knots over the numbers is acceptable. You could attain best L/D around 2,000 knots and 84,000' (SR-71).
Paradoxically: Low stall speed is the enemy of flying. Anyone designing a flying machine should fight tooth and nail to increase the minimum speed requirement by every knot which can be rationalized away.
Emergency landing speed could be the defining factor, but also might be assessed under the assumption of zero fuel remaining; because that is the leading cause of most unscheduled off-airport landings. Take gross weight, subtract fuel weight, or a certain fraction of it, and use that weight to spec the stall speed for cornfield ops.
A larger wing means a larger everything. Larger everything, assuming all load factors stay the same, means greater weight. More material means more building expense. More wing means more drag. Less of everything else that actually matters. Like max speed, cruise speed, maneuvering speed, range, low, or nonexistent comfort, even outright danger to life and property in turbulence. Larger wing means more roll damping, means larger, heavier ailerons, larger and heavier surface balances, more weight, Etc. The goal of low stall speed is actually counter-productive to itself, as weight gain is inevitable just to achieve it. (Unless the entire scope is changed to a new lower flight speed, lower in-flight loads, lighter-duty construction, and a worse airplane, etc).
Design the airplane to do what airplanes do: Fly.
If the end result can still be operated out of your intended fields, then great. If not, then worry about stall speed a little more. It is simple: Add wing area. Add flaps, or upgrade to higher performance high lift systems and suffer the added complexity and weight. Low speed is a compromise of everything else. Part 103 is exhibit-A of this fact. The farther you go away from a 1903 wright flyer, the better, more practical, and safer the airplane.
Here is an Example:
W: 1,000lbs
S: 100ft wing area.
Intent: 45kcas stall.
W/S = 10lb sqft
10 / CL 1.5 = 6.67 dynamic pressure. This is 44.4 knots indicated.
Now Try 10 / CL 1.4 = 7.15 ... This is 45.9 knots.
Only a 1.5 knot difference for a large 0.1 change in CLmax. And only over shooting the target speed by less than 1 knot. Can you even tell a difference in 0.9 knots? If you're gonna overrun the runway on landing for this small error, you are probably doing something wrong.
Barring any federal regulators squashing your $100,000,000 fortune into a very small one over a matter of 9/10ths of a knot; don't worry about it.
Make the airplane to do what its primary function is. That is flight at a given speed and altitude. Set the design lift coefficient for THAT condition.
E.g. same airplane, 200mph, 10,000ft. W/S is 10/ 75 slug-ft-sec. CL is 0.133...
You can never attain a decent lift/drag number at this mickey mouse lift coefficient. A cruise L/D of about 5-6 is all you can hope for. Therefore, your range is going to be very low. If you fly through an invisible thermal, and many exist at this altitude, you will exceed +11.25g almost instantaneously! (1.5/0.1333). If your wings remain intact, I doubt your canopy will after you head-butt it. But If you design your limit load at 6g, and desire a useful maneuvering speed of 150kias (200mph true at 10kMSL) then this requires a cruise lift coefficient of 1.5/6g = 0.25CL... x q75 = 18.75lbft wing loading. That is a requirement: It is an absolute.
This will readily allow a 10:1 or possibly a L/D of 12. Thereby Increasing range by at least 150%, possibly doubling range over a compromised design with a dumbo sized wing.
1000 / 18.75 = 53.33.. ft of area. But wait, by cutting the wing in half you also cut it's weight in half. Assuming 2lb per sqft, this is -93lbs saved. Gross weight 907, not 1,000. 907/18.75 = 48.4... of look we just cut the wing down even more. Try 900 gross, 48ft of area. Wing drag is now just 1/2 what it was before. You saved 100lbs. That weight was pure structure, at $30/lb for glass or $100/lb for carbon:$3000-$10,000 cheaper airplane! Probably a good 20-25% faster. Twice the range. Half the cost per mile.
18.75/1.5 = 12.5, this is 60.7 knots. Only 16.7 knots higher, or 60.7/44.4 = just 37% greater landing speed, which is a small price to pay for a better airplane, a cheaper airplane, a smaller airplane, that can do bigger airplane things.
Oh, let's add flaps. Electric, weight now 950lbs. CLmax now 1.8. Stall speed is 57 knots! Woohoo. We are 3.7 knots slower. It took an extra month of design effort, added 50 pounds, $1000, and reduced the payload by 50lb or fuel load by 8 gallons (1 hour at 200mph, -200mi range). For 4 knots less speed on the low end: Is that a good trade?
As to stall/departure, vortex lift is the undisputed champion. The problem is not a stall. It is not losing lift. The problem has always been and will remain uncommanded, exponential, divergent roll at the onset of a stall. Set up a strong vortex sheet between the 0.6-0.8 b/2 at all AoAs beyond +/-10deg, and go fly. This might take some outside the box thinking. (Simscale CFD could be your friend). This phenomenon is unbelievably easy to harness.
Any way, that is my philosophy.
YMMV.
------------------------------------
The title was regarding stall tailoring. But it appears this is already pretty well understood.
I'll add; another way to ensure the root stalls first, is a change in airfoil section from root to the 75% b/2 where the highest lift airfoil is located. Keeping mean camber lines the same makes tailoring the lift distribution easier, E.g. NACA 66-218 root, with its sharper leading edge, and a slightly more blunt 63-215 located near 0.7 semi-span. You may have to rip one of the camber lines and apply the other thickness distribution to it for consistency; xFoil has this feature built in. For laminar airfoils, maximum lift and highest stall angle is found among the 15% t/c thickness foils. This is 13% for the classic NACA airfoils. (You have to scale them up from 12). If the section is to include flaps, then 18-21% will provide the greatest lift increment improvement.
Although the first part of this discussion was about stall speed as much as stall progression: I'm not sure what the design intent is, so my comment includes many generalizations. The following ramblings will be on stall speed:
Every textbook and aircraft design guide focuses entirely too much effort on stall speed.
And at least 100 otherwise perfectly good airplanes have been ruined by this obsession over this one requirement.
The FAA has been migrating away from the 61kt rule. Only LSA (someday maybe Mosaic) must adhere to an arbitrary stall speed requirement.
There is zero correlation between stall speed and flying-related fatalities. ~ The false assumption that Slow = Safe, is the single most ignorant idea which has lead to ten-thousand pilots and their passengers no longer being with us. For gods sake stop flying slow. If you want to go slow then WALK!
Airplanes with 7lb wing loading and 35 knot stall speeds are objectively and statistically more dangerous, and far less useful, than airplanes having 70lb wing loading.
What matters is; can the airplane take off or land in a certain distance which the operator finds useful? And; does it fulfill its mission requirements?
If grass strip, or off-airport landing is a requirement, either for bush flying or for potential emergency landing in a bean field, then a lower stall speed is certainly important. If the aircraft is intended for recreational flying in a small geographical area, then this consideration might be leas important. And what follows probably does not apply.
IMO, it is a waste of time for designers to agonize over this one detail, except for certain specific uses. If STOL ops, or emergency landing speed is important: Pick a general ballpark landing velocity, and see that wing size meets that need. Then see if the other requirements can be met at that wing area. ~ Generally, low stall speed makes every other part of the design 10x more useless for each knot you take off the low end.
If you get down to 24 knots (Part 103) you will have a nearly useless airplane. If you can live with a 124 knot stall, you can have a 747-800 with full fuel load, a hundred friends, and a pallet of Champaign. And If 210 knots over the numbers is acceptable. You could attain best L/D around 2,000 knots and 84,000' (SR-71).
Paradoxically: Low stall speed is the enemy of flying. Anyone designing a flying machine should fight tooth and nail to increase the minimum speed requirement by every knot which can be rationalized away.
Emergency landing speed could be the defining factor, but also might be assessed under the assumption of zero fuel remaining; because that is the leading cause of most unscheduled off-airport landings. Take gross weight, subtract fuel weight, or a certain fraction of it, and use that weight to spec the stall speed for cornfield ops.
A larger wing means a larger everything. Larger everything, assuming all load factors stay the same, means greater weight. More material means more building expense. More wing means more drag. Less of everything else that actually matters. Like max speed, cruise speed, maneuvering speed, range, low, or nonexistent comfort, even outright danger to life and property in turbulence. Larger wing means more roll damping, means larger, heavier ailerons, larger and heavier surface balances, more weight, Etc. The goal of low stall speed is actually counter-productive to itself, as weight gain is inevitable just to achieve it. (Unless the entire scope is changed to a new lower flight speed, lower in-flight loads, lighter-duty construction, and a worse airplane, etc).
Design the airplane to do what airplanes do: Fly.
If the end result can still be operated out of your intended fields, then great. If not, then worry about stall speed a little more. It is simple: Add wing area. Add flaps, or upgrade to higher performance high lift systems and suffer the added complexity and weight. Low speed is a compromise of everything else. Part 103 is exhibit-A of this fact. The farther you go away from a 1903 wright flyer, the better, more practical, and safer the airplane.
Here is an Example:
W: 1,000lbs
S: 100ft wing area.
Intent: 45kcas stall.
W/S = 10lb sqft
10 / CL 1.5 = 6.67 dynamic pressure. This is 44.4 knots indicated.
Now Try 10 / CL 1.4 = 7.15 ... This is 45.9 knots.
Only a 1.5 knot difference for a large 0.1 change in CLmax. And only over shooting the target speed by less than 1 knot. Can you even tell a difference in 0.9 knots? If you're gonna overrun the runway on landing for this small error, you are probably doing something wrong.
Barring any federal regulators squashing your $100,000,000 fortune into a very small one over a matter of 9/10ths of a knot; don't worry about it.
Make the airplane to do what its primary function is. That is flight at a given speed and altitude. Set the design lift coefficient for THAT condition.
E.g. same airplane, 200mph, 10,000ft. W/S is 10/ 75 slug-ft-sec. CL is 0.133...
You can never attain a decent lift/drag number at this mickey mouse lift coefficient. A cruise L/D of about 5-6 is all you can hope for. Therefore, your range is going to be very low. If you fly through an invisible thermal, and many exist at this altitude, you will exceed +11.25g almost instantaneously! (1.5/0.1333). If your wings remain intact, I doubt your canopy will after you head-butt it. But If you design your limit load at 6g, and desire a useful maneuvering speed of 150kias (200mph true at 10kMSL) then this requires a cruise lift coefficient of 1.5/6g = 0.25CL... x q75 = 18.75lbft wing loading. That is a requirement: It is an absolute.
This will readily allow a 10:1 or possibly a L/D of 12. Thereby Increasing range by at least 150%, possibly doubling range over a compromised design with a dumbo sized wing.
1000 / 18.75 = 53.33.. ft of area. But wait, by cutting the wing in half you also cut it's weight in half. Assuming 2lb per sqft, this is -93lbs saved. Gross weight 907, not 1,000. 907/18.75 = 48.4... of look we just cut the wing down even more. Try 900 gross, 48ft of area. Wing drag is now just 1/2 what it was before. You saved 100lbs. That weight was pure structure, at $30/lb for glass or $100/lb for carbon:$3000-$10,000 cheaper airplane! Probably a good 20-25% faster. Twice the range. Half the cost per mile.
18.75/1.5 = 12.5, this is 60.7 knots. Only 16.7 knots higher, or 60.7/44.4 = just 37% greater landing speed, which is a small price to pay for a better airplane, a cheaper airplane, a smaller airplane, that can do bigger airplane things.
Oh, let's add flaps. Electric, weight now 950lbs. CLmax now 1.8. Stall speed is 57 knots! Woohoo. We are 3.7 knots slower. It took an extra month of design effort, added 50 pounds, $1000, and reduced the payload by 50lb or fuel load by 8 gallons (1 hour at 200mph, -200mi range). For 4 knots less speed on the low end: Is that a good trade?
As to stall/departure, vortex lift is the undisputed champion. The problem is not a stall. It is not losing lift. The problem has always been and will remain uncommanded, exponential, divergent roll at the onset of a stall. Set up a strong vortex sheet between the 0.6-0.8 b/2 at all AoAs beyond +/-10deg, and go fly. This might take some outside the box thinking. (Simscale CFD could be your friend). This phenomenon is unbelievably easy to harness.
Any way, that is my philosophy.
YMMV.
Last edited by a moderator:
