Space Shuttle: Difference between revisions

Jump to navigation Jump to search
Copyedit; Trying to get image captions to show; etc
(Copyedit; Trying to get image captions to show; etc)
Line 28: Line 28:
== Project Aim ==
== Project Aim ==


The aim of the Shuttle Project is to create a highly realistic simulation of the capabilities of the Space Shuttle in Flightgear. While most of the time the real Shuttle is under the control of automatic guidance systems, there are fallback modes to control the spacecraft manually, the so-called CSS (control stick steering) modes, and it is these modes we primarily try to implement.
The aim of the Shuttle Project is to create a highly realistic simulation of the capabilities of the Space Shuttle in FlightGear. While most of the time the real Shuttle is under the control of automatic guidance systems, there are fallback modes to control the spacecraft manually, the so-called CSS (control stick steering) modes, and it is these modes we primarily try to implement.


In addition to the real avionics and control modes, the idea is also to provide various 'educational' modes and instruments such as to explore and appreciate certain aspects of a Shuttle mission more.  
In addition to the real avionics and control modes, the idea is also to provide various 'educational' modes and instruments such as to explore and appreciate certain aspects of a Shuttle mission more.  


The [http://ntrs.nasa.gov  NASA technical reports server] supplies a large base of wind tunnel and in-situ performance data of both the mated launch vehicle and the orbiter, and the aerodynamics of the simulated shuttle is based on these documents. The authoritive source for procedures for trajectory management, instrumentation, limits and energency procedures is the [https://www.nasa.gov/centers/johnson/pdf/390651main_shuttle_crew_operations_manual.pdf Space Shuttle Crew Operations Manual] and currently a normal mission, i.e. ascent, orbital insertion, de-orbit, entry, terminal area energy management and landing can be flown largely 'by the book', i.e. following the real procedure for CSS. As of May 2015, this does not yet hold for emergency procedures.
The [http://ntrs.nasa.gov  NASA technical reports server] supplies a large base of wind tunnel and in-situ performance data of both the mated launch vehicle and the orbiter, and the aerodynamics of the simulated shuttle is based on these documents. The authoritative source for procedures for trajectory management, instrumentation, limits and emergency procedures is the [https://www.nasa.gov/centers/johnson/pdf/390651main_shuttle_crew_operations_manual.pdf Space Shuttle Crew Operations Manual] and currently a normal mission, i.e. ascent, orbital insertion, de-orbit, entry, terminal area energy management and landing can be flown largely 'by the book', i.e. following the real procedure for CSS. As of May 2015, this does not yet hold for emergency procedures.
 


=== Limit and failure modeling ===
=== Limit and failure modeling ===
Line 41: Line 40:
The general philosophy on limit modeling is that they can be treated dependent on a user setting as 'soft', 'hard' and 'realistic'. Where applicable, warnings when the state of the orbiter is getting dangerously close to a limit are called out in addition to a recommendation how to deal with the situation. Dependent on the trajectory of the orbiter, there may or may not be sufficient time to redeem the situation.
The general philosophy on limit modeling is that they can be treated dependent on a user setting as 'soft', 'hard' and 'realistic'. Where applicable, warnings when the state of the orbiter is getting dangerously close to a limit are called out in addition to a recommendation how to deal with the situation. Dependent on the trajectory of the orbiter, there may or may not be sufficient time to redeem the situation.


<b>soft:</b> Limit violations are called out, but their violation has no consequences for aerodynamics or component failures.
; soft
: Limit violations are called out, but their violation has no consequences for aerodynamics or component failures.


<b>hard:</b> Any limit violation immediately ends the simulation.
; hard
: Any limit violation immediately ends the simulation.


<b>realistic:</b> In reality, components do not necessarily fail immediately if used outside their design specs. This option applies a probabilistic failure model in which the chance for a component to fail grows with the degree of limit violation. The failure may or may not be immediately visible, e.g. too much qbar upon ascent may damage the heat shield, but this may not be apparent (unless specifically checked) until the heat shield fails upon atmospheric entry.
; realistic
: In reality, components do not necessarily fail immediately if used outside their design specs. This option applies a probabilistic failure model in which the chance for a component to fail grows with the degree of limit violation. The failure may or may not be immediately visible, e.g. too much qbar upon ascent may damage the heat shield, but this may not be apparent (unless specifically checked) until the heat shield fails upon atmospheric entry.


Component failure is modeled gradually where applicable - while a tire can only blow or not blow, an airfoil or a thruster for instance may lose a certain percentage of its efficiency.
Component failure is modeled gradually where applicable - while a tire can only blow or not blow, an airfoil or a thruster for instance may lose a certain percentage of its efficiency.
Line 61: Line 63:
Each SRB weighs about 1,300,000 lb, out of which 1,100,000 is propellant weight. The propellant of the SRBs is shaped to provide a high liftoff thrust, followed by a thrust reduction during the phase of the highest dynamical pressure (max. qbar). The actual thrust as a function of time is fairly complicated:
Each SRB weighs about 1,300,000 lb, out of which 1,100,000 is propellant weight. The propellant of the SRBs is shaped to provide a high liftoff thrust, followed by a thrust reduction during the phase of the highest dynamical pressure (max. qbar). The actual thrust as a function of time is fairly complicated:


[[File:SRB thrust.png|400px|Thrust characteristics of the Space Shuttle Solid Rocket Boosters]]
[[File:SRB thrust.png|400px|thumb|none|Thrust characteristics of the Space Shuttle Solid Rocket Boosters]]


The distribution is faithfully modeled in FG and the definitions to match the real thrust characteristics is taken from the [http://jsbsim.sourceforge.net/download.html JSBSim code repository]
The distribution is faithfully modeled in FG and the definitions to match the real thrust characteristics is taken from the [http://jsbsim.sourceforge.net/download.html JSBSim code repository]
Line 83: Line 85:
The propellant for the SSMEs is carried in the ET. The tank has a liftoff weight of approximately 1,680,000 lb (760 tons) and a dry weight of about 66,000 lb. The ET is the only expendable component of the launch stack, it is dropped after MECO upon almost reaching orbit and then the shuttle uses the OMS to attain orbit while the tank re-enters the atmosphere half an orbit later and breaks up during entry.
The propellant for the SSMEs is carried in the ET. The tank has a liftoff weight of approximately 1,680,000 lb (760 tons) and a dry weight of about 66,000 lb. The ET is the only expendable component of the launch stack, it is dropped after MECO upon almost reaching orbit and then the shuttle uses the OMS to attain orbit while the tank re-enters the atmosphere half an orbit later and breaks up during entry.


[[File:Shuttle-ETsep01.jpg|600px| External tank separation]]
[[File:Shuttle-ETsep01.jpg|600px|thumbnail|none|External tank separation]]


In FG, the tank is normally separated using {{Key press|d}}. This is vetoed if the Shuttle has unsafe yaw, pitch or roll motion in which case the RCS should be used to stabilize the orbiter before ET separation. If an emergency separation needs to be performed, {{Key press|Control}} + {{Key press|d}} overrides the veto. At separation, a translational RCS burn will automatically push the shuttle away from the tank.
In FG, the tank is normally separated using {{Key press|d}}. This is vetoed if the Shuttle has unsafe yaw, pitch or roll motion in which case the RCS should be used to stabilize the orbiter before ET separation. If an emergency separation needs to be performed, {{Key press|Control|d}} overrides the veto. At separation, a translational RCS burn will automatically push the shuttle away from the tank.


After separation, the ET will approximately co-orbit with the OV, i.e. unless the Shuttle ignites the OMS engines, the tank will be visible for a long time, slowly drifting off, and it is quite possible to use the Shuttle's RCS engines to do a visual inspection of the tank.
After separation, the ET will approximately co-orbit with the OV, i.e. unless the Shuttle ignites the OMS engines, the tank will be visible for a long time, slowly drifting off, and it is quite possible to use the Shuttle's RCS engines to do a visual inspection of the tank.


[[File:Shuttle orbit02.jpg|600px| The ET seen from the Shuttle]]
[[File:Shuttle orbit02.jpg|600px|thumbnail|none|The ET seen from the Shuttle]]


=== A note on aerodynamics of the mated vehicle ===
=== A note on aerodynamics of the mated vehicle ===
Line 103: Line 105:
During ascent, the stick controls thrust vectoring for both SSMEs and SRBs. The following two DAP schemes are available:
During ascent, the stick controls thrust vectoring for both SSMEs and SRBs. The following two DAP schemes are available:


<b>Thrust vectoring:</b> This is the real CSS ascent mode for the shuttle in which stick motion controls rate, stick to neutral commands an attitude hold. Internally a PID controller vectors the thrusters and uses the stick input as a bias for the error. This is a very stable scheme and can be easily used to achieve high precision in controlling ascent speed or orbital inclination.
; Thrust vectoring
: This is the real CSS ascent mode for the shuttle in which stick motion controls rate, stick to neutral commands an attitude hold. Internally a PID controller vectors the thrusters and uses the stick input as a bias for the error. This is a very stable scheme and can be easily used to achieve high precision in controlling ascent speed or orbital inclination.


<b>Thrust vectoring (gimbal):</b> This is an educational scheme in which the stick motion directly controls the engine gimbal, i.e. the pilot needs to do the task of the PID controller himself. To make things somewhat easier, the engines are automatically vectored through the stack's CoG, i.e. outside the atmosphere stick neutral corresponds to zero moments acting on the stack. In the atmosphere, the control input hence needs to compensate for aerodynamical forces. Launch in this scheme is fairly rough and it is not possible to reach high precision, but it is possible to fly into orbit and gain a first-hand experience of the forces acting on the stack.
; Thrust vectoring (gimbal)
: This is an educational scheme in which the stick motion directly controls the engine gimbal, i.e. the pilot needs to do the task of the PID controller himself. To make things somewhat easier, the engines are automatically vectored through the stack's CoG, i.e. outside the atmosphere stick neutral corresponds to zero moments acting on the stack. In the atmosphere, the control input hence needs to compensate for aerodynamical forces. Launch in this scheme is fairly rough and it is not possible to reach high precision, but it is possible to fly into orbit and gain a first-hand experience of the forces acting on the stack.


{{Key press|m}} switches between the ascent DAPs. {{Key press|Control}} + {{Key press|m}} switches from the ascent to the orbital DAP modes (do not use an orbital DAP for ascent control unless you know very well what you're doing).
{{Key press|m}} switches between the ascent DAPs. {{Key press|Control|m}} switches from the ascent to the orbital DAP modes (do not use an orbital DAP for ascent control unless you know very well what you're doing).


=== Ascent structural and aerodynamical limits ===
=== Ascent structural and aerodynamical limits ===
Line 113: Line 117:
The following structural and aerodynamical limits need to be observed during ascent:
The following structural and aerodynamical limits need to be observed during ascent:


* dynamical pressure qbar < 819 lb/sqf (modeled)
* Dynamical pressure qbar < 819 lb/sqf (modeled)


This is a structural limit for the orbiter and mated stack, in actual operations the orbiter should be kept below 650 lb/sqf.
This is a structural limit for the orbiter and mated stack, in actual operations the orbiter should be kept below 650 lb/sqf.


* wing bending moment CBW between -0.019 and 0.019 (not modeled)
* Wing bending moment CBW between -0.019 and 0.019 (not modeled)


At max qbar, the wing bending moment is a function of Mach number and AoA. Since Mach number is close to 1.4 in this phase of the flight, this limit basically translates into alpha between -8 degrees and 2 degrees. This can only be achieved if the orbiter is in inverted flight.
At max qbar, the wing bending moment is a function of Mach number and AoA. Since Mach number is close to 1.4 in this phase of the flight, this limit basically translates into alpha between -8 degrees and 2 degrees. This can only be achieved if the orbiter is in inverted flight.


* translational accelerations Nx between 0 and 3.11 g (modeled), Ny between -0.18 and 0.18 g (not modeled) and Nz between -0.06 and 0.73 g (not modeled).
* Translational accelerations Nx between 0 and 3.11 g (modeled), Ny between -0.18 and 0.18 g (not modeled) and Nz between -0.06 and 0.73 g (not modeled).


These are structural limits of the mated stack to acceleration rather than aerodynamical forces. Especially the Nx (acceleration along the orbiter axis, i.e. main engine thrust) is important and requires to throttle down the SSMEs towards the end of the burn time.
These are structural limits of the mated stack to acceleration rather than aerodynamical forces. Especially the Nx (acceleration along the orbiter axis, i.e. main engine thrust) is important and requires to throttle down the SSMEs towards the end of the burn time.


* late ascent trajectory may not drop below 265.000 ft (modeled)
* Late ascent trajectory may not drop below 265.000 ft (modeled)


This is a heat load limit for the external tank insulation, if the thermal protection of the ET fails, it will explode.
This is a heat load limit for the external tank insulation, if the thermal protection of the ET fails, it will explode.
Line 141: Line 145:
Once in orbit, in FG throttle control is transferred to both OMS engines. They can be throttled from zero to 100% of nominal thrust and are automatically vectored by the flight controls through the CoG of the orbiter. The real shuttle has a DAP for thrust vectoring of the OMS engines as well as the option of using a single engine with partial thrust vectoring, neither option is currently modeled.
Once in orbit, in FG throttle control is transferred to both OMS engines. They can be throttled from zero to 100% of nominal thrust and are automatically vectored by the flight controls through the CoG of the orbiter. The real shuttle has a DAP for thrust vectoring of the OMS engines as well as the option of using a single engine with partial thrust vectoring, neither option is currently modeled.


[[File:Shuttle orbit04.jpg|600px| OMS burn for orbital insertion]]
[[File:Shuttle orbit04.jpg|600px|thumbnail|none|OMS burn for orbital insertion]]


=== The Reaction Control System ===
=== The Reaction Control System ===
Line 151: Line 155:
=== RCS DAP schemes ===
=== RCS DAP schemes ===


The real Space Shuttle has a multitude of (partially mission-specific) DAP schemes, each with different gains and deadbands, which control the thruster firing pattern in response to the controllers. A fair selection of these is implemented in FG. In the real Shuttle cockpit, there is both a rotational hand controller (RHC) and a translational hand controller (THC) to initiate either rotations of the shuttle or translational accelerations (e.g. for approach and docking). In FG, {{Key press|m}} corresponds to switching from THC to RHC and back, {{Key press|Shift}} + {{Key press|m}} switches between the different DAPs and {{Key press|Control}} + {{Key press|m}} is the override switch to aerodynamical controls. The HUD will display the currently selected mode for clarity.
The real Space Shuttle has a multitude of (partially mission-specific) DAP schemes, each with different gains and deadbands, which control the thruster firing pattern in response to the controllers. A fair selection of these is implemented in FG. In the real Shuttle cockpit, there is both a rotational hand controller (RHC) and a translational hand controller (THC) to initiate either rotations of the shuttle or translational accelerations (e.g. for approach and docking). In FG, {{Key press|m}} corresponds to switching from THC to RHC and back, {{Key press|Shift|m}} switches between the different DAPs and {{Key press|Control|m}} is the override switch to aerodynamical controls. The HUD will display the currently selected mode for clarity.


Due to the geometry of the thruster arrangement, there is significant mode mixing. For instance, a lateral translation firing nose and right pod thruster with the same thrust would also induce a yaw motion (since the modules do not have the same distance to the CoG) and a roll (since they are not in the CoG plane and in fact not even in the same plane). In most implemented modes, the FCS logic takes care of most of these effects by firing additional thruster to cancel the unwanted motion, however in some modes this is not easily possible and mode mixing has to be anticipated and accounted for manually. This is in fact the same as in the real Shuttle.
Due to the geometry of the thruster arrangement, there is significant mode mixing. For instance, a lateral translation firing nose and right pod thruster with the same thrust would also induce a yaw motion (since the modules do not have the same distance to the CoG) and a roll (since they are not in the CoG plane and in fact not even in the same plane). In most implemented modes, the FCS logic takes care of most of these effects by firing additional thruster to cancel the unwanted motion, however in some modes this is not easily possible and mode mixing has to be anticipated and accounted for manually. This is in fact the same as in the real Shuttle.


<b>RCS rotation:</b> This is a simple scheme in which the stick motion controls thrust, i.e. angular acceleration. Stick to neutral commands no thrust, i.e. the Shuttle will continue its current rotation.
; RCS rotation
: This is a simple scheme in which the stick motion controls thrust, i.e. angular acceleration. Stick to neutral commands no thrust, i.e. the Shuttle will continue its current rotation.


<b>RCS DAP-A:</b> A precision 'stick controlls rate' scheme in which stick to neutral commands an attitude hold. The mode has fairly strict deadbands and steep gains and hence uses comparatively much propellant to stabilize attitude.
; RCS DAP-A
: A precision 'stick controls rate' scheme in which stick to neutral commands an attitude hold. The mode has fairly strict deadbands and steep gains and hence uses comparatively much propellant to stabilize attitude.


<b>RCS DAP-B:</b> As DAP-A, but more permissive in terms of deadbands, trades less strictly stabilized attitude against reduced propellant consumption.
; RCS DAP-B
: As DAP-A, but more permissive in terms of deadbands, trades less strictly stabilized attitude against reduced propellant consumption.


<b>RCS ROT TAIL ONLY:</b> A 'stick controls thrust' scheme in which the nose module is not used. This causes significant mode mixing.
; RCS ROT TAIL ONLY
: A 'stick controls thrust' scheme in which the nose module is not used. This causes significant mode mixing.


<b>RCS ROT NOSE ONLY:</b> A 'stick controls thrust' scheme in which the OMS pod modules are not used. This causes significant mode mixing and has no roll control available (it would have some very limited roll control in reality).
; RCS ROT NOSE ONLY
: A 'stick controls thrust' scheme in which the OMS pod modules are not used. This causes significant mode mixing and has no roll control available (it would have some very limited roll control in reality).


<b>RS ROT ENTRY:</b> A 'stick controls rates' DAP designed for entering the atmosphere which enforces a 'no sideslip' attitude in which the nose module is not used. This has very strict deadbands and agressive gains to combat the yaw instability of the Shuttle upon entry, significant mode mixing and is very propellant-consuming. Do not use in orbit and only activate at the entry interface once the shuttle has the correct attitude!
; RS ROT ENTRY
: A 'stick controls rates' DAP designed for entering the atmosphere which enforces a 'no sideslip' attitude in which the nose module is not used. This has very strict deadbands and aggressive gains to combat the yaw instability of the Shuttle upon entry, significant mode mixing and is very propellant-consuming. Do not use in orbit and only activate at the entry interface once the shuttle has the correct attitude!


== Glossary of acronyms ==
== Glossary of acronyms ==

Navigation menu