Response of FBW aircraft in updraft
Join Date: Jun 2013
Location: Or-E-Gun, USA
Posts: 326
Likes: 0
Received 0 Likes
on
0 Posts
@CaptainProp
No endless A vs. B debate from this reader! My only question is why so many (most, active pilots on one or the other types), so not instantly know the answer. The proverbial question, "What is it doing now???" is one that I do not like to hear. Regardless of the platform flown, IMHO the occupants of both seats 0A and 0B should already know those details.
Thanks and HNY.
Thanks and HNY.
Join Date: Jul 2013
Location: Medway towns
Age: 72
Posts: 40
Likes: 0
Received 0 Likes
on
0 Posts
what SHOULD FBW be doing...
Without FBW systems, gusts are dealt with by reducing alpha to recover from stall/overstress events. Talk to a glider pilot who has encountered extreme turbulence in stubble fires! Paradoxically, whether in up or down draught, the nose ends up pointing down a little, to maintain airspeed... and keep alpha sensible.
The FBW system does what the programmer tells it to, and there's the rub....
Up in the pointy end of the envelope, control deflections must be kept below any level that will cause structural pain, and this will come into direct conflict with the movement required to give an effective control of attitude in severe turbulence. Hence the need to go slower in rough air....
When the envelope is breached, priority should be given to avoiding stress by excessive control deflections, and hopefully the ups and downs may cancel themselves out in a reasonably short time scale.
Huge cu nim have descending air round the outside, and high speed jets in the middle... any speed can result in large stresses being imposed. Gliders have been spat out in pieces, and considering they enter at around 50 knots, it doesn't take too much imagination to figure out that 450 is too fast...
FBW is maybe not quite the right description, Fly By Computer should be employed. Maybe the data available needs to be researched further..
The FBW system does what the programmer tells it to, and there's the rub....
Up in the pointy end of the envelope, control deflections must be kept below any level that will cause structural pain, and this will come into direct conflict with the movement required to give an effective control of attitude in severe turbulence. Hence the need to go slower in rough air....
When the envelope is breached, priority should be given to avoiding stress by excessive control deflections, and hopefully the ups and downs may cancel themselves out in a reasonably short time scale.
Huge cu nim have descending air round the outside, and high speed jets in the middle... any speed can result in large stresses being imposed. Gliders have been spat out in pieces, and considering they enter at around 50 knots, it doesn't take too much imagination to figure out that 450 is too fast...
FBW is maybe not quite the right description, Fly By Computer should be employed. Maybe the data available needs to be researched further..
Join Date: May 2000
Location: Seattle
Posts: 3,196
Likes: 0
Received 0 Likes
on
0 Posts
The response of the airplane depends on the airplane and the modes of autopilot and autothrottle engaged when the updraft is encountered.
In normal high-altitude cruise, VANV will effectively be in an altitude-seeking mode, and the MCP cruise altitude will be the target. The FMS or MCP Mach will also be the target of the autopilot and autothrottles. I don't think it matters whether it's Boeing or Airbus at this point; manual flight is a different story, though.
So, in a severe updraft, the pitch will initially go down and the throttles retarded in an attempt to maintain target altitude and Mach. If the throttles get to idle, pitch will continue to seek target altitude until limit Mach is reached. At that point a climb at limit Mach will commence.
If the updraft abruptly ends, the throttles will advance (likely to the CLB or CRZ limit) and pitch increased to attempt to maintain altitude. Altitude will be sacrificed to keep airspeed above the low-speed limit.
Again, with autopilot and autothrottles engaged, it doesn't matter if it's FBW or conventional, Airbus or Boeing. Response will be essentially the same.
In normal high-altitude cruise, VANV will effectively be in an altitude-seeking mode, and the MCP cruise altitude will be the target. The FMS or MCP Mach will also be the target of the autopilot and autothrottles. I don't think it matters whether it's Boeing or Airbus at this point; manual flight is a different story, though.
So, in a severe updraft, the pitch will initially go down and the throttles retarded in an attempt to maintain target altitude and Mach. If the throttles get to idle, pitch will continue to seek target altitude until limit Mach is reached. At that point a climb at limit Mach will commence.
If the updraft abruptly ends, the throttles will advance (likely to the CLB or CRZ limit) and pitch increased to attempt to maintain altitude. Altitude will be sacrificed to keep airspeed above the low-speed limit.
Again, with autopilot and autothrottles engaged, it doesn't matter if it's FBW or conventional, Airbus or Boeing. Response will be essentially the same.
Again, with autopilot and autothrottles engaged, it doesn't matter if it's FBW or conventional, Airbus or Boeing. Response will be essentially the same.
Join Date: Mar 2005
Location: Uh... Where was I?
Posts: 1,338
Likes: 0
Received 0 Likes
on
0 Posts
Airbus FBW is a path stable rather than a speed stable airplane, as conventional airplanes are. This is achieved by a C* pitch control law, which achieves and keeps a target value of g load (blended with pitch rate). Boeing FBW is very similar but it has added a speed maintaining element, and it is referred to as C*U law. So, Boeing are still speed stable airplanes.
It is important to understand that stability, and the tendency of an airplane to maintain angle of attack, or airspeed or whatever (flight path in the case of the bus) is just that: a tendency.
Therefore, as a result of a gust, updraft or downdraft, the airplane can end up in a slight climb or descent. It can¡t be predicted. It will depend on the gust, on how many, how quick and how intense they are, how long they last and in what sense they go.
The pitch normal law of an airbus is not like an ALT HOLD mode of an AP. It will resist to flight path change, but this does not mean that it will succeed in doing so. It doesn't have to. Flight path will change. All we need is that it resists to change. That's enough. That allows us to easily fly the airplane, just like "normal" longitudinal stability helps in conventional airplanes.
If you fly level and an updraft hits you, the airplane will pitch down (pretty conventional right?) But it is likely that some height is gained. Nothing in the control law makes it recover that, nor even to assume V/S zero. It tries to keep the flight path constant. But… Which flight path? As it changes, the reference to be maintained changes, too. Therefore, if after the upset the airplane it in a shallow climb, that is then the new flight path. If the pitch down reaction was a little "late" and "excessive" this may put the airplane descending, but not to recover an altitude.
If you understand this you will fly the FBW better in gusts. All you have to do is to let the airplane resist for a sec before you make inputs to put her the way you want. If you do it simultaneously you will start the "dance".
It is important to understand that stability, and the tendency of an airplane to maintain angle of attack, or airspeed or whatever (flight path in the case of the bus) is just that: a tendency.
Therefore, as a result of a gust, updraft or downdraft, the airplane can end up in a slight climb or descent. It can¡t be predicted. It will depend on the gust, on how many, how quick and how intense they are, how long they last and in what sense they go.
The pitch normal law of an airbus is not like an ALT HOLD mode of an AP. It will resist to flight path change, but this does not mean that it will succeed in doing so. It doesn't have to. Flight path will change. All we need is that it resists to change. That's enough. That allows us to easily fly the airplane, just like "normal" longitudinal stability helps in conventional airplanes.
If you fly level and an updraft hits you, the airplane will pitch down (pretty conventional right?) But it is likely that some height is gained. Nothing in the control law makes it recover that, nor even to assume V/S zero. It tries to keep the flight path constant. But… Which flight path? As it changes, the reference to be maintained changes, too. Therefore, if after the upset the airplane it in a shallow climb, that is then the new flight path. If the pitch down reaction was a little "late" and "excessive" this may put the airplane descending, but not to recover an altitude.
If you understand this you will fly the FBW better in gusts. All you have to do is to let the airplane resist for a sec before you make inputs to put her the way you want. If you do it simultaneously you will start the "dance".
Join Date: Jun 2007
Location: Wanderlust
Posts: 3,407
Likes: 0
Received 0 Likes
on
0 Posts
Tubby and capt prop
The moment you say airbus FBW cannot be stalled some people counter it by saying under certain circumstances it can. Let us understand what it exactly does. You cannot bank Airbus more than 67 degrees means exactly that you cannot. But that doesn't mean in violent storm it cannot be turned upside down. Similarly stall is an angle of attack phenomenon which you cannot achieve with side stick in normal law. If it happens due to severe atmospheric conditions the computer will try and reduce the angle of attack to Alpha Max which is not the case in conventional aeroplane. When you enter an updraft the angle of attack will increase the auto pilot will try to maintain the altitude by pitching down which will increase speed etc. How ever unlike normal aircraft if the angle of attack exceeded alpha prot it will latch and aircraft will maintain that angle of attack even after autopilot trips. Whether aircraft can pitch up in alpha prot yes it can. This protection is designed to pitch down to maintain it when you are increasing the angle of attack but when angle of attack is changing due to updrafts it will maintain the latched condition by doing whatever it takes even by climbing unless pilot pushes the stick forward and unlatches it. It has happened and caused Air prox incidents.
The moment you say airbus FBW cannot be stalled some people counter it by saying under certain circumstances it can. Let us understand what it exactly does. You cannot bank Airbus more than 67 degrees means exactly that you cannot. But that doesn't mean in violent storm it cannot be turned upside down. Similarly stall is an angle of attack phenomenon which you cannot achieve with side stick in normal law. If it happens due to severe atmospheric conditions the computer will try and reduce the angle of attack to Alpha Max which is not the case in conventional aeroplane. When you enter an updraft the angle of attack will increase the auto pilot will try to maintain the altitude by pitching down which will increase speed etc. How ever unlike normal aircraft if the angle of attack exceeded alpha prot it will latch and aircraft will maintain that angle of attack even after autopilot trips. Whether aircraft can pitch up in alpha prot yes it can. This protection is designed to pitch down to maintain it when you are increasing the angle of attack but when angle of attack is changing due to updrafts it will maintain the latched condition by doing whatever it takes even by climbing unless pilot pushes the stick forward and unlatches it. It has happened and caused Air prox incidents.
Vilas in your bank example I presume the aircraft will transition into abnormal attitude law (Bank>125deg).
This is an unusual law as I don't think that any sim is certified to operate to parameters where it could be shown.
This is an unusual law as I don't think that any sim is certified to operate to parameters where it could be shown.
Join Date: Jun 2007
Location: Wanderlust
Posts: 3,407
Likes: 0
Received 0 Likes
on
0 Posts
Tubby
In the sim you can not only bank 120 degrees but you can do a barrel roll also off course in alternate law. It will go in abnormal attitude law. You can recognise it by aircraft staying in alternate law instead of direct law after gear is put down. Landing becomes a bit tricky as there is no flare mode. If you reset ELACS you can get direct law.
In the sim you can not only bank 120 degrees but you can do a barrel roll also off course in alternate law. It will go in abnormal attitude law. You can recognise it by aircraft staying in alternate law instead of direct law after gear is put down. Landing becomes a bit tricky as there is no flare mode. If you reset ELACS you can get direct law.
autopilot mode on "updraft/downdraft"
Just asking, and the FCOM is no good. Gotta talk with folks that have been there and have the tee shirt.
Seems to me that the pitch correction to the basic one gee command could make for a rougher ride than a simple gee command ( not attitude command, and no correction for pitch attitude). So a severe updraft has the plane trying to reduce the gee and vice versa for a downdraft.
The "pure" gee command ( yeah, some rate and gains, but that's all) I flew with 35 years ago had the effect of dampening out the turbulence at altitude and same for the thermals we had flying real low over the desert. The tiny jet felt like it was much bigger and heavier. I had a joy ride in a 'vark one day, and that's a good comparison.
I would bet that the 'bus smooths out turbulence faster and better than we humans. Gotta get some feedback from real 'bus drivers.
PJ2???? You on here?
Seems to me that the pitch correction to the basic one gee command could make for a rougher ride than a simple gee command ( not attitude command, and no correction for pitch attitude). So a severe updraft has the plane trying to reduce the gee and vice versa for a downdraft.
The "pure" gee command ( yeah, some rate and gains, but that's all) I flew with 35 years ago had the effect of dampening out the turbulence at altitude and same for the thermals we had flying real low over the desert. The tiny jet felt like it was much bigger and heavier. I had a joy ride in a 'vark one day, and that's a good comparison.
I would bet that the 'bus smooths out turbulence faster and better than we humans. Gotta get some feedback from real 'bus drivers.
PJ2???? You on here?
Join Date: Jun 2007
Location: Wanderlust
Posts: 3,407
Likes: 0
Received 0 Likes
on
0 Posts
Tubby and safe life
Nobody teaches aerobatics in the SIM. But if you want to experience the implications of abnormal attitude law I showed you how to do it. Take it or leave it. SIM will do what is programmed and cannot trusted for unknown contingencies. Abnormal laws are known parameters.
Nobody teaches aerobatics in the SIM. But if you want to experience the implications of abnormal attitude law I showed you how to do it. Take it or leave it. SIM will do what is programmed and cannot trusted for unknown contingencies. Abnormal laws are known parameters.
Join Date: Jun 2011
Location: france
Posts: 760
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by gums
I would bet that the 'bus smooths out turbulence faster and better than we humans. Gotta get some feedback from real 'bus drivers.
Last edited by roulishollandais; 4th Jan 2015 at 16:50. Reason: spelling,gramar...:-( apologies
Join Date: Jun 2011
Location: france
Posts: 760
Likes: 0
Received 0 Likes
on
0 Posts
A case of missing observability -which is the other system design worst fault- can be pointed in the missing AoA under 60 KTs in AF447.
Remember : it is MANDATORY -state of art by serious guys- that any system be always
1. Observable
2. Controllable
Remember : it is MANDATORY -state of art by serious guys- that any system be always
1. Observable
2. Controllable
Last edited by roulishollandais; 4th Jan 2015 at 16:52. Reason: fault instead failure
Join Date: Jun 2009
Location: somewhere
Posts: 451
Likes: 0
Received 0 Likes
on
0 Posts
@Roulishollandais:
So buses don't have such additional actuators to control the effective system in that case who only creates climb until physic laws get master again and drop the plane.
A case of missing observability -which is the other system design worst fault- can be pointed in the missing AoA under 60 KTs in AF447.
Join Date: Jun 2011
Location: france
Posts: 760
Likes: 0
Received 0 Likes
on
0 Posts
First quote : concerns the loss of controllability reaching the limitation/protection. Boeing are not hard limited. Airbuses and F-16 are hard limited.
Second quote : concerns the certification of the AF-447 AoA sensor from the report and our threads i.e.
Second quote : concerns the certification of the AF-447 AoA sensor from the report and our threads i.e.
Join Date: Jun 2009
Location: somewhere
Posts: 451
Likes: 0
Received 0 Likes
on
0 Posts
First quote : concerns the loss of controllability reaching the limitation/protection. Boeing are not hard limited. Airbuses and F-16 are hard limited.
In case of sensed overspeed both Airbii as Boeing FBW raise their nose to arrest the exceeding airspeed.
Both needs pilot intervention to restrict the induced climb.
Second quote : concerns the certification of the AF-447 AoA sensor from the report and our threads i.e.
Join Date: Jun 2011
Location: france
Posts: 760
Likes: 0
Received 0 Likes
on
0 Posts
Sensors, actuators and algorithms, observability and controllability
Hi A33Zab,
You are pointing other systems' faults... If sensors are missing for any reason and observability cannot be reached in the closed loop you must modify or add sensors, perhaps threads of wool on the frame! Engineers will be able to find something if the certification guys get able to verify observability and controllability.
And the same with controllability and missing actuators. Pilot intervention is possible only so long you have an actuator to jump in the loop again, creating a system interruption and doing the correction quickly enough...and adequatly, if it is possible (the 30 seconds don't fit, too slow)
We are in the time to reject every fault in systems, from A, B or C.
You are pointing other systems' faults... If sensors are missing for any reason and observability cannot be reached in the closed loop you must modify or add sensors, perhaps threads of wool on the frame! Engineers will be able to find something if the certification guys get able to verify observability and controllability.
And the same with controllability and missing actuators. Pilot intervention is possible only so long you have an actuator to jump in the loop again, creating a system interruption and doing the correction quickly enough...and adequatly, if it is possible (the 30 seconds don't fit, too slow)
We are in the time to reject every fault in systems, from A, B or C.