6 ms·

> […] flying wing aircraft […] are aerodynamically unstable. They can't be flown by humans […] RC flying wings are regularly flown by humans without any softwa

> […] flying wing aircraft […] are aerodynamically unstable. They can't be flown by humans […] RC flying wings are regularly flown by humans without any software in between the input and control surfaces.

sithadmin 7y ago
Really now? As you might imagine, the scale and operating requirements of _actual_ aircraft introduces complex engineering challenges that aren't present in toys.
Zak 7y ago
There are very specific regulations relating to how transport category aircraft have to behave. One of these is the reason the 737 MAX has the specific MCAS behaviors that are probably responsible for two crashes: causing the aircraft to stall must require continually increasing the pitch-up control input (or something substantially similar), and that wasn't the case at low speed, high thrust, and high angle of attack. It may be difficult to achieve the required flight characteristics without software augmentation in a flying wing, even if RC hobbyists can usually fly them manually without crashing. The 737 MAX has raised the question of whether it's a good idea to approve transport category aircraft that require help from software to have acceptable flight characteristics.
- rtkwe 7y ago
  MCAS wasn't required for the MAX8 to fly it was required for it to maintain the common type rating with the existing 737 fleet. With training the MAX8 would have been fine to fly and pilots would have known to pitch down or decrease throttle when experiencing the engine induced pitching.
  - sokoloff 7y ago
    MCAS provides for the certification requirement of continually increasing stick forces with increasing pitch, which was otherwise not met at high angles of attack and high thrust settings. MCAS (or another means to accomplish the same goal) was required for the aircraft to meet certification requirements, regardless of whether it was to share a common type rating with the rest of the 737 fleet (so long as it shared the landing gear with the rest of the 73' line, it was going to have this adverse aerodynamic raw result). https://www.risingup.com/fars/info/part25-175-FAR.shtml https://www.risingup.com/fars/info/part25-175-FAR.shtml
    - rtkwe 7y ago
      That regulation seems to be at a static thrust level right? > 75 percent of maximum continuous power for reciprocating engines or the maximum power or thrust selected by the applicant as an operating limitation for use during climb for turbine engines; and Wouldn't that give a stable pitch response because the thrust relative to the center of mass is constant? Also if the MCAS is always pushing the plane out of the given AoA (which seems to be what it's logic is, if AoA > x trim down until AoA < x) would that be considered outside the flight envelope anyways? The stick force curve goes screwy for every aircraft in a stall.
      - Zak 7y ago
        It's an aerodynamic issue with the 737 MAX: the engine cowlings themselves produce lift at a high angle of attack, and the engines are mounted farther forward than they are on other 737s.
- mikeash 7y ago
  Airbus and Boeing have been flying fully fly-by-wire airliners for decades. It works just fine.
  - Zak 7y ago
    Those aircraft, as far as I know would meet certification requirements without the software adding control inputs that are opposite to what the pilots command.
briandear 7y ago
And helicopters are aerodynamically unstable as well.
- rocqua 7y ago
  They don't require active stabilization by software, which is the issue here. Specifically, does the plane become unfly-able if your computer crashes. It is fine if a fighter jet crashes when the electronics fail. Much less so for a passenger yet.
  - Rychard 7y ago
    > It is fine if a fighter jet crashes when the electronics fail. I certainly hope not.
    - tylerhou 7y ago
      It’s fine in the sense that a fighter jet can glide for quite a while and even land without engines. If that’s not possible, the pilot can steer it to crash in a remote area and the pilot can escape.
      - Max-q 7y ago
        A modern fighter jet is not a good glider. An F-16 can't fly without the computer. Have you tried to sit in the cockpit? The sticks don't even move, they are pressure sensitive only. You are right about the ejection, that's how the pilots survive a computer malfunction, with a complete hull loss as consequence.
    - blt 7y ago
      But it is. Militaries like to play the "protect our troops at all costs" card when it helps them, but ultimately they make decisions based on a nonzero amount of acceptable deaths.
      - azernik 7y ago
pedrocr 7y ago
> RC flying wings are regularly flown by humans without any software in between the input and control surfaces. Technically this is definitely not true. Any modern RC wing will use software to do elevon mixing and get elevator/aileron input into the actual mechanical surfaces. It's possible to do this mechanically but software is so much simpler it's a no brainer to just have a servo per surface and do everything else in software. It may be true in the spirit of this discussion in the sense that most RC wings just do fixed software mappings between inputs and outputs instead of having an on-board computer taking into account attitude/airspeed/etc and adjusting the inputs based on this. That's what's actually needed for aerodynamically unstable aircraft and RC wings generally don't need that, although similar kinds of hardware do exist in the RC world and are used sometimes.
- rocqua 7y ago
  I mean, you could just as easily do the mixing on the TX side rather than the RX side. In that case, there is not software between the RX-servos and the control surfaces. I think that counts as 'no software between the input and control surfaces'. The relevant question is whether there is an IMU providing feedback without which the plane would not fly.
  - T-A 7y ago
    It's my impression that "everybody" puts a flight controller between the servos and the receiver these days, either a sophisticated one running something like ArduPilot [1] or at least a cheap, dedicated flight stabilizer like (random pick) [2]. [1] http://ardupilot.org/copter/docs/common-choosing-a-flight-controller.html http://ardupilot.org/copter/docs/common-choosing-a-flight-co... [2] https://hobbyking.com/en_us/quanum-gyro.html https://hobbyking.com/en_us/quanum-gyro.html
    - dosshell 7y ago
      Maybe there is region differances here. Here in Sweden most RC-Wings, actually all, I have seen have been self made and do not use any active controller. What kind of observations do you have to support your statement? Ps. Most of them I have seen, and built, has been to play a game where you put a string on your tail and try to cut each others with the propeller. They were airodynamic stable.

sithadmin 7y ago

Really now? As you might imagine, the scale and operating requirements of _actual_ aircraft introduces complex engineering challenges that aren't present in toys.

Zak 7y ago

There are very specific regulations relating to how transport category aircraft have to behave. One of these is the reason the 737 MAX has the specific MCAS behaviors that are probably responsible for two crashes: causing the aircraft to stall must require continually increasing the pitch-up control input (or something substantially similar), and that wasn't the case at low speed, high thrust, and high angle of attack. It may be difficult to achieve the required flight characteristics without software augmentation in a flying wing, even if RC hobbyists can usually fly them manually without crashing. The 737 MAX has raised the question of whether it's a good idea to approve transport category aircraft that require help from software to have acceptable flight characteristics.

rtkwe 7y ago
MCAS wasn't required for the MAX8 to fly it was required for it to maintain the common type rating with the existing 737 fleet. With training the MAX8 would have been fine to fly and pilots would have known to pitch down or decrease throttle when experiencing the engine induced pitching.
- sokoloff 7y ago
  MCAS provides for the certification requirement of continually increasing stick forces with increasing pitch, which was otherwise not met at high angles of attack and high thrust settings. MCAS (or another means to accomplish the same goal) was required for the aircraft to meet certification requirements, regardless of whether it was to share a common type rating with the rest of the 737 fleet (so long as it shared the landing gear with the rest of the 73' line, it was going to have this adverse aerodynamic raw result). https://www.risingup.com/fars/info/part25-175-FAR.shtml https://www.risingup.com/fars/info/part25-175-FAR.shtml
  - rtkwe 7y ago
    That regulation seems to be at a static thrust level right? > 75 percent of maximum continuous power for reciprocating engines or the maximum power or thrust selected by the applicant as an operating limitation for use during climb for turbine engines; and Wouldn't that give a stable pitch response because the thrust relative to the center of mass is constant? Also if the MCAS is always pushing the plane out of the given AoA (which seems to be what it's logic is, if AoA > x trim down until AoA < x) would that be considered outside the flight envelope anyways? The stick force curve goes screwy for every aircraft in a stall.
    - Zak 7y ago
      It's an aerodynamic issue with the 737 MAX: the engine cowlings themselves produce lift at a high angle of attack, and the engines are mounted farther forward than they are on other 737s.
mikeash 7y ago
Airbus and Boeing have been flying fully fly-by-wire airliners for decades. It works just fine.
- Zak 7y ago
  Those aircraft, as far as I know would meet certification requirements without the software adding control inputs that are opposite to what the pilots command.

briandear 7y ago

And helicopters are aerodynamically unstable as well.

rocqua 7y ago
They don't require active stabilization by software, which is the issue here. Specifically, does the plane become unfly-able if your computer crashes. It is fine if a fighter jet crashes when the electronics fail. Much less so for a passenger yet.
- Rychard 7y ago
  > It is fine if a fighter jet crashes when the electronics fail. I certainly hope not.
  - tylerhou 7y ago
    It’s fine in the sense that a fighter jet can glide for quite a while and even land without engines. If that’s not possible, the pilot can steer it to crash in a remote area and the pilot can escape.
    - Max-q 7y ago
      A modern fighter jet is not a good glider. An F-16 can't fly without the computer. Have you tried to sit in the cockpit? The sticks don't even move, they are pressure sensitive only. You are right about the ejection, that's how the pilots survive a computer malfunction, with a complete hull loss as consequence.
  - blt 7y ago
    But it is. Militaries like to play the "protect our troops at all costs" card when it helps them, but ultimately they make decisions based on a nonzero amount of acceptable deaths.
    - azernik 7y ago

pedrocr 7y ago

> RC flying wings are regularly flown by humans without any software in between the input and control surfaces. Technically this is definitely not true. Any modern RC wing will use software to do elevon mixing and get elevator/aileron input into the actual mechanical surfaces. It's possible to do this mechanically but software is so much simpler it's a no brainer to just have a servo per surface and do everything else in software. It may be true in the spirit of this discussion in the sense that most RC wings just do fixed software mappings between inputs and outputs instead of having an on-board computer taking into account attitude/airspeed/etc and adjusting the inputs based on this. That's what's actually needed for aerodynamically unstable aircraft and RC wings generally don't need that, although similar kinds of hardware do exist in the RC world and are used sometimes.

rocqua 7y ago
I mean, you could just as easily do the mixing on the TX side rather than the RX side. In that case, there is not software between the RX-servos and the control surfaces. I think that counts as 'no software between the input and control surfaces'. The relevant question is whether there is an IMU providing feedback without which the plane would not fly.
- T-A 7y ago
  It's my impression that "everybody" puts a flight controller between the servos and the receiver these days, either a sophisticated one running something like ArduPilot [1] or at least a cheap, dedicated flight stabilizer like (random pick) [2]. [1] http://ardupilot.org/copter/docs/common-choosing-a-flight-controller.html http://ardupilot.org/copter/docs/common-choosing-a-flight-co... [2] https://hobbyking.com/en_us/quanum-gyro.html https://hobbyking.com/en_us/quanum-gyro.html
  - dosshell 7y ago
    Maybe there is region differances here. Here in Sweden most RC-Wings, actually all, I have seen have been self made and do not use any active controller. What kind of observations do you have to support your statement? Ps. Most of them I have seen, and built, has been to play a game where you put a string on your tail and try to cut each others with the propeller. They were airodynamic stable.