View Full Version : Virgin Galactic Tech
Despite extensive gargling I have not found a decent explanation of the ship's systems - if anyone has such I would appreciate a link. In particular, from looking at early test videos, it almost seems as if the wing/tail 'feathering' is drive by air loads at high alpha. If so, how is it unfeathered, and could an inadvertent higher alpha after feather 'unlock' cause movement or is movement regulated?
Peter H
3rd Nov 2014, 13:22
BOAC: The 2nd column of the 3rd page (numbered 56) of
http://www.boulder.swri.edu/suborbital/press-releases/Aviation-Week-Article-Suborbital-Spacecraft.pdf
states that:
Two main pneumatic 625-psi actuators with a 9.5-in bore and 31-in stroke, change the position of the feather ...
http://en.wikipedia.org/wiki/SpaceShipTwo
has a tiny image of the craft, but it's pretty unreadable even if you click to enlarge.
http://en.wikipedia.org/wiki/File:SpaceShipTwo_technical_diagram.jpg
Peter - thanks! I had found the mini-jpegs, but that pdf is excellent. I assume therefore they just use high-alpha to 'ease' the feathering load. I see from the 'sim ride' the tail was unfeathered well after apogee.
BOAC. Need to get away from the touchy-feely and other stuff on the other forum.
Apparently the NTSB has decoded several "tapes" early on.
- The mechanism for the feather mode seems straightforward and robust.
- I don't unnerstan why the mechanism is supposed to be unlocked before the vehicle is "coasting" up on its flight path and well above high "q". Passing 1.4M the vehicle is at a decent gee, so doesn't take much to move something too far or not far enuf.
Peter H
3rd Nov 2014, 18:49
gums: I'm a s/w guy trying to understand the pros and cons of when to unlock.
Not too early
At first sight unlocking just after rocket ignition seems premature and a needless hole in the cheese.
Not too late
I've seen something to the effect that the plane cannot re-enter without functioning feathers. Which gives a
reason for checking that it can unlock before passing the point of no return (wherever that is on the flight path).
To allow the feathers to be rapidly deployed for non-reentry reasons
To quote from another blog: http://www.orbiter-forum.com/showthread.php?p=486049
...because I wonder if SS2 lost attitude control (pitch/yaw/roll) after engine ignition? That might be one reason
why Alsbury pushed the feathering system to the unlock position. Remember the SS1 1st X-prize flight (16P)
in September 2004? Mike Melvill had to deploy the feathers early after it entered a violent roll late in the engine
burn. Maybe something was wrong early in the engine burn this time and Alsbury decided to arm the system
early in order to get it deploy ASAP after it gets above the atmosphere and engine shuts down?
... or course you might proactively unlock in anticipation of such an eventuality, especially on test flights?
sycamore
3rd Nov 2014, 19:56
It is not unknown for the Airspeed/Mach systems to fail,or under/overread in any aircraft,so it may be that it was an inadvertent early `unlocking`.However,I would have thought that a simple backup with `time vs expected acceleration` may have been useful as a cross-check,both on the ground and in the cockpit.
gums;
https://www.youtube.com/watch?v=0mCFxAsmnk0
This video below looks rearward at the moment of launch, through engine-start and the pitch-up/climb then the apogee and pitch-over. You can see the feathers working.
More interestingly however, the CVR is heard. I'm sure you'll be familiar with some of the sounds as the gee builds during the pitch-up.
Regarding the "unlock" question, from the video/CVR:
Release occurs @ 00:18 into the video. Engine light-off occurs at 00:22"
From crew comments heard on the CVR, unlocking the "feathers" is done @ 00:39, 17" shortly after initial engine light-off.
Actual release of the feathers, ("feather up" comment), near/at apogee for the descent occurs @ 01:18, or 49" after initial unlocking.
We don't have Mach information so don't know at what Mach the unlocking occurred.
The point is, unlocking shortly after engine light-off does not seem unusual according to this video.
In other words, the feathers aren't unlocked just before they're commanded to change position.
Perhaps something in the interlocks, (unlocked, not released) permitted the "feathers" to move; their relatively small structure seems (intuitively...not an engineer!), that they would not tolerate lateral loads well.
As we would know well, once such a failure mode started it would swiftly progress in the same manner as any airplane that had lost its rudder.
janrein
4th Nov 2014, 08:33
Thanks for that AWST article, a wealth of information!
By deduction and combination from the various sources:
Feather unlock (and actual feathering?) nominally at Mach 1.4 (says NTSB Hart)
Engine burn of 3rd powered flight had been 20 sec to Mach 1.4 71kft
Engine burn planned 4th (this) powered flight approx 20 sec to approx Mach 1.4
(my assumption, not to change fuel and durn-time/Mach in one step)
Operational goal 2 minute burn Mach 3.4, 370 kft
Feather unlock lever, I wonder, is this:
- A safety lever securing the actual movement of the feather lever?
Or
- A lever disengaging physical locking pins allowing subsequent feathering by the pneumatic actuators
(or inadvertantly allowing uncommanded feathering by aerodynamic forces overpowering pneumatics-only)
Feather unlock c.q. actual feathering at Mach 1.4
- After motor burn-out around Mach 1.4 in third test
- Aerodynamic forces at that altitude/speed acceptable for coasting flight
- Aerodynamic forces at that altitude/speed also acceptable if still powered?
If yes on the latter, then it makes sence to unlock and/or feather after Mach 1.4 in the procedure for longer-burn flights, in order to discover anomaly in the essential feathering function for safe re-entry, and allow early motor-stop, in stead of continuing to push energy into the vehicle.
Hypothesis: Mach 1.4 and associated altitude and weight may be the limit for safe unfeathered re-entry?
Of course the unburned fuel mass after premature motor-stop is an additional weight/energy burden for reentry, the loaded SS2 is 30kLB (AWST), how much of that is the fuel mass? And can the oxidiser be dumped (through the extinguished engine) without re-lighting it?
A jettisonable rocket motor in case of feather failure comes to mind. Although that would also add complexity.
Any thoughts, additional information, corrections on that?
RetiredF4
4th Nov 2014, 11:45
Is it only I who thinks, that the lateral stability feathered and unfeathered does not look great in those videos?
Do the pilots on the controls oppose this wing rocking tendency or is some kind of stabilization system doing the job?
Here's my dumb question - why is 'feathering' required (?after apogee?)? Is it because at the high alpha of 're-entry' the nose-down moment of the unfeathered wings would be too great and require too much reaction control?
RetF4 - I expect with the high sweep there will be ample LE vortices which will almost certainly not be forming symmetrically and 'natural' roll damping looks minimal. I would propose 'leaving it alone' or otherwise the likelihood of 'out-of-phase' inputs would be high. I'm not sure (for many other reasons as well!) I would want to be in that cabin. You'd have terrible trouble with the champagne:)
"Is it only I who thinks, that the lateral stability feathered and unfeathered does not look great in those videos? "
Lateral, and longitudinal; I wondered too. The first thing that came to mind was "PIO".
It occurs on all flights I've viewed the videos for so it's not unique to one flight.
I would think you'd still want to dampen it before serving the bubbly...
In the aft-view video above, I looked for small movements of controls but couldn't discern any although they'd probably be tiny. And oscillations don't stop except briefly at two points: approach to the apogee, and later in the descent when the feathers retract to normal position.
BOAC, gums, Retired4 - if you can look past the soft stuff in this early commercial for Virgin Galactic, there's something of an explanation regarding the feathers @ 2:40 in this video, (which also shows the wing-rocking at the beginning of the video for a flight in October, 2004).
https://www.youtube.com/watch?v=MN-GE4D61Zk
Beginning at 00:45", another video, (first flight with "feathering"), shows what I would call a "significant" pitching & rolling pre-, during and post-feathering.
https://www.youtube.com/watch?v=Lqhzlq7UReA
janrein;
Re, "Of course the unburned fuel mass after premature motor-stop is an additional weight/energy burden for reentry, the loaded SS2 is 30kLB (AWST), how much of that is the fuel mass? And can the oxidiser be dumped (through the extinguished engine) without re-lighting it? "
At 1:20 in the video with the aft-looking camera linked in the post above there's a crew remark, "auto-dump" and a cloud of gas is seen leaving the ship; it occurs again at 02:30 into the video and I suspect that's unburnt fuel and/or oxidizer. It doesn't appear to be dumped from the engine from from behind the tail-camera.
janrein
4th Nov 2014, 14:03
RetiredF4
From an AWST article posted earlier (Sept 7, 2009)
For maneuvring outside the atmosphere, SS2 is fitted with a cold-gas reaction control system, with nozzles in the nose for pitch and yaw, and in te wing tips for roll thrust.
The SS2 reaction control system is remotely similar to the Space Shuttle Orbiter RCS. For the SSO re-entry was considered to start down from 400kft and a concept of "blended control" was applied, handing over roll control first to aileraons, then pitch control to the body flap, and rudder coming in last reaching the transsonic regime. This staged transition from RCS to aerodynamic control is said to have been the main requirement driving to Fly by Wire.
(see e.g. Free Online Courses From Top Universities | Academic Earth (http://academicearth.org/courses/) , then search ...)
The SS2 has to make the transition both ways and a learning curve has to be made again. One of the many challenges in the development program.
RetiredF4
4th Nov 2014, 17:52
@Janrein
My question was in regard to that rocking and rolling flightphase, where gas thrusters would not jet or later on during reentry no more be used for steering.
@PJ2
Indeed, it is present not only in roll, but looking at more scenes in yaw and pitch as well. It increases when feathering, winds a bit down when feathered, and increasing again when defeathering.
Would some dehidral configuration of the main wing have a better outcome?
janrein
4th Nov 2014, 18:13
PJ2
At 1:20 in the video with the aft-looking camera linked in the post above there's a crew remark, "auto-dump" and a cloud of gas is seen leaving the ship; it occurs again at 02:30 into the video and I suspect that's unburnt fuel and/or oxidizer. It doesn't appear to be dumped from the engine from from behind the tail-camera.
Thanks, triggered me to looking further into that.
( video https://www.youtube.com/watch?v=0mCFxAsmnk0 )
This is what I see and what I think I hear.
0:44 End of motor burn observed
0:53 "There comes ullage"
Fumes observed on centre line, appear to be coming from rotor nozzle
(flushing motor fuel grain clean? dumping of remaining oxidiser?)
1:15 "Feather up"
1:18 Feathers observed coming up
1:20 "Feather dump"
A cloud observed escaping on port side for next 3 seconds
(venting of air from pneumatic feather actuator?)
(with water vapour condensation?)
!:47 "OK feathers down"
2:11 Feathers observed coming into view
2:17 Feathers observed down
2:26 "Autodump are on"
Cloud observed escaping on port side until end of video
Fumes appearing to come from nozzle may actually be excess oxidiser being dumped through the motor (without burning the fuel, flow started well after end of motor burn).
The oxidiser is nitrous oxide, liquid or gaseous.
The fuel is a solid.
BTW the plume from the motor burn can be observed around 0:50 just after end of motor burn, with a slight cork-screw twist due to roll oscillation in last stage of the burn.
At 1:26 the plume can be seen again, twist seems to have further expanded.
Around 2:00 the plume top is seen rising through the horizon, that 1:15 of coasting parabolic flight at that time.
Amazing footage.
janrein
4th Nov 2014, 18:54
@ RetiredF4
My question was in regard to that rocking and rolling flightphase, where gas thrusters would not jet or later on during reentry no more be used for steering.
I find it hard to estimate at what stage the jets will be applied in ascent and until when in descent, for the SS Orbiter the "blended use" was over a wide regime of altitudes and airspeeds.
Equally hard to say if the aerodynamic controls will be frozen during any portion of the trajectory, if they are, the regimes may again be different for each of the axes.
The feathers - going by Rutan´s own statements - are indeed meant to provide most of the stabilisation function when the air becomes progressively denser, I am not sure if the RCS system would be totally inactive in feathered configuration.
Other than that the aerodynamic active or passive stabilisation functions see transitions from subsonic through transsonic, to supersonic (low supersonic so far for SS2). There is hardly a constant regime, it´s all transients, and maybe that´s also a factor in the observed oscillations. I can imagine that the oscillations increase with decreasing q while the jets - if applied to complement the aerodynamic controls - are known to behave less predictably when there is still an appreciable effect of the thinning atmosphere (shockwave patterns).
That said, I have only touched on these complex matters and in a more distant past, so I welcome any other views and corrections.
RetiredF4
4th Nov 2014, 19:22
@Janrain
I have no knowledge concerning SS2 and how it is designed to behave.
I've done a few handful of flights up to M2.3 in the phantom and up to 50.000 feet. This jet was getting the more stable the faster it flew. If i would have encountered such a rock'n roll motion in any flightphase I most probably would have aborted the flight. The problem with such non linear oscilations is, you never know how big the next one will be. And they must create some loads on those feather fins too.
I'm curious to see some footage of the fllight prior to the breakup, wether those oscilations had been present too.
RetiredF4;
Re, "Would some dehidral configuration of the main wing have a better outcome? "
Would that come with slightly increased drag?
And given that, I wonder how "critical" or unstable the design is in order to perform its layered tasks? I haven't flown anything but commercial transports and would never (ordinarily) see such motion but I certainly agree with you regarding the rock 'n roll and aborting the flight.
janrein, thanks for the interpretations - makes sense that "venting" oxidizer from anywhere but the rear wouldnt' be done.
If the controls aren't frozen, the RCS would be very busy and perhaps that's why it remains within certain limits? It's a lot of gyrating and given the air density, without knowing but only guessing, would be difficult for a pilot to control using conventional controls.
janrein
4th Nov 2014, 21:18
@RetF4 and @PJ2
The SS2 has only started to open the lower part of its intended flight envelope and has not even come close to the achieved performance of SS1. It´s a major scale-up, and clearly it needs time.
Here is footage of one of the SS1´s above 100 km flights, have a look at the rolling motions.
SpaceShip One X-Prize Flight #1 Launch Mike Melvill
https://www.youtube.com/watch?v=LXNkUNP75-Q
I appreciate that in increasingly straight-up motions the notion of bank angle vanishes, however there must be limits to roll rate both for crew and vehicle. Not sure what those would be though.
On the footage that has appeared of the accidented SS2 flight in those 9 seconds or so I have seen no indication of any significant oscillation, maybe more close-up and detailed footage will come out lateron. The NTSB has so far been quite open sharing factual information on the accident.
Not sure if that video is actually first hop for the prize. I watched both on live TV ten years ago. Guess I'll look at the long videos tonight. The clip is the first hop, sorry.
Seemed to me that flight two was the hairy one with the uncontrolled roll. An interview with Mike revealed he just let her roll as he was so high that aero forces were not a player and the RCS had some limits like time and authority. So he just let her roll and took control with the RCS at apogee or maybe lower.
The best part of that video is how effective the feather system worked.
Burt has favored the twin tail design forever. I would have liked a large vertical fin way back on the fuselage. Keep the outboard stuff, but increase roll and yaw stability.
I'm also fascinated by the way the guy survived - being forcibly ejected from the wreckage at 45k+ and to be able to deploy a chute.............
sycamore
5th Nov 2014, 22:11
BOAC,possibly a static line to a stabilising drogue,then an auto barostat at 10-15k.....
Yo BOAC!
I have many friends that had the experience of being in free fall after being hit by AAA or a SAM. One guy told us that first thing he saw was no "floor" on his jet.
It's the luck of the draw.
Most of us had a barometric doofer that deployed the chute at 14,000 feet or so without any action on our part.
The biggie is being ejected from the wreckage via a system or blind ass luck. I would take either, heh heh.
A static line needs to be attached to something...
Gums - how many of your mates did it from 45k+ and supersonic? I know of only a few actual 'controlled' ejections from that environment, an X-15, a Lightning and probably a few others, but disintegration?
Bill Weaver SR-71 Blackbird Breakup (http://roadrunnersinternationale.com/weaver_sr71_bailout.html)
Yes - with pressure suit and emergency oxy. We do not know what the clothing was for these guys.
thcrozier
7th Nov 2014, 04:20
PJ2
At about 0:55, after "unlock" but before "feather", I think I see significant random independent buffeting of the 2 feathers, the worst of the entire 3 minutes or so. Suggests to me that "unlocked" allows quite a bit of uncontrolled movement of those surfaces.
Peter H
7th Nov 2014, 15:12
Cross reference of pchapman & Redredrobin's efforts on the rumours thread.
From New spaceship restoring hope after Virgin Galactic crash | Reuters (http://www.reuters.com/article/2014/11/06/us-space-crash-virgin-new-idUSKBN0IQ02920141106)
As the ship is rocketing upward, the tail is held fast by a large hook that is supposed to remain engaged until the craft reaches supersonic speed, Mike Moses, Virgin Galactic vice president of operations, explained in an interview with Reuters.
At that point, the pilots release the hook, though the tail remains pinned back by aerodynamic pressures. The command to actually move the tail into descent position comes after the rocket motor burns out, near the apex of the ship's altitude.
Unlocking the tail is done well before then so that if the mechanism fails, the pilots can abort the flight.
"It's a great safety feature, but if you use your safety feature in a regime that it's not designed to handle, bad things are going to happen," Moses said. "It's like your car airbag going off at 65 miles per hour."
@BOAC tell if you don't want this posting here and I'll delete it
thcrozier;
Yes, I wondered about that too - lots of movement.
For me anyway, the issue isn't the "early" unlocking as attributed to the PNF "without orders from the PF" but the fact that a mere unlocking resulted in unintended deployment of the feathers.
By another description, it's the thrust reverser deployment in mid-air problem - it just shouldn't happen. However, simplicity of design and weight considerations may play a role in a more elementary mechanical or even electro-mechanical design which required stricter SOP requirements, (lots in that statement to unpack, I know).
I'm sure someone knows but I'm curious as to what Mach No. the feathers were unlocked in the third launch, (the one with the CVR where "unlocked" is heard about 17" after light-off). Probably it's above the required M1.4 but it would be interesting if it weren't...
DozyWannabe
7th Nov 2014, 18:44
@PJ2 - My first thought was actually "sounds similar to the TR locking problem that brought down Lauda Air 004". Still, we wait for more concrete info...
I looked up my Viper chart for movement of the c.p. as we went from 0.9 M to 1.4 M.
We were not static stable until 0.95M, but went from a c.p. of 43% MAC at 1.1M or so to almost 60% MAC at 1.4M, and then it leveled off.
So looks like 1.4M is a good rule of thumb for less concerns about changes in center of pressure. Have to look at Concorde and Blackbird stuff to confirm.
My guess is the pneumatic system allowed the feathers to move, unlike a hydraulic system ( compressible versus not-so-compressible acting fluids). So the shock waves and movement of the c.p. seems to be an area of concern.
How and when are the pneumatic actuators that move the feathers pressurized ?
RetiredF4
9th Nov 2014, 08:45
I'm still mulling over the reason to unlock the feathering system in the early stages of the flight. The plausible argument, that the system is esential for a safe re-entry and that therefore the functioning of the system has to be tested before leaving to space makes sense, but does that apply to a testflight where neither achieved Mach nor achieved altitude would be high enough that usage of said system is critical?
In search for a more detailed description of the system i found the following patent, which was filed 2004 for the spaceship One re-entry system, and as Spaceship Two is known as being a scale up, it might still apply to the system used in Spaceship Two as well.
The unlock system and the feathering system are pneumatically operated.
Patent US20060108479 - Winged spacecraft - Google Patents (http://www.google.com/patents/US20060108479)
What is the reason behind to unlock a safety feature that early in flight? Is it to have it available already during ascent if stability issues arrive to slow the ascent or to stabilize the ascent by other means when flight control inputs or gas thrusters are not effective or failed?
Something does not sum up, there is no urgency at all in a test flight like that to arm the system that early in a flight.
janrein
9th Nov 2014, 09:13
@RetiredF4
What is the reason behind to unlock a safety feature that early in flight?
Good find that patent!
About the unlocking, to my understanding:
The earlier in flight that you can allow unlocking - provided already having reached safe margin to uncommanded feathering through the transsonic regime - the earlier you can detect an unlocking failure, the earlier you can cut-off the rocket motor.
The earlier you stop pushing energy into the vehicle - which has been found to be unable to perform the nominal high drag reentry - the better the chances for surviving a non-nominal low drag re-entry.
In addition to that I would expect after engine cut-off the oxidiser to be dumped overboard thus reducing weight, to further improve re-entry margins.
but does that apply to a testflight where neither achieved Mach nor achieved altitude would be high enough that usage of said system is critical?
My post perhaps superfluous, reading the later portion of yours.
Hmmm, ...
I guess that also in a lower energy testflight an earlier than engine cut-off verification would add to re-entry safety margin as well.
Other than that, to test the procedure as such as closely as possible to the procedure for nominal energy flight.
Or, ... perhaps to more precisely quantify the first safety margin, i.e. to step-wise determine in subsequent tests how soon you can safely permit unlocking.
Admittedly more speculatively all that ...
RetiredF4
9th Nov 2014, 09:44
@Janrein
I agree with that view, if the flight profile itself could lead to a situation where that reconfiguration would be essential for a safe return profile. The more speed and altitude is planned for the special test profile, the earlier you would like to know if the system is working. In the end this only ensures that the unlocking is working, but not the feathering itself.
But as unlocking removes also an important safety feature against premature activation of the system, there has to be a tradeoff between removing that safety feature ( and thus the danger of deployment of the feathering system at an unwanted phase of the flight) and risking to be not able to unlock it in a later part of the flight. As a pilot i would not bother about a system, I most probably would not need anyway (except for testing purposes) and I would acttivate it at the latest (safest) time.
Next thing is, there must be thoughts about the reliability of this pneumatic feathering system concerning possible failure modes (to the feathered position without activation), otherwise a guarded switch would do the trick to prevent the pilots of inadvertent activation of the system.
I really would like to understand the operation of the system, which parts of the actuators are pressurized by what action, wether the locking and actuating system uses the same accumulator, and what the fail safe default modes are in case of loss of pneumatic pressure. Still searching though for that information.
Peter H
9th Nov 2014, 10:14
@RetiredF4 Nice find
@ janrein
I tried to start a list of the pros & cons of the unlock timing in post #5. Perhaps we can start to expand on this.
Obviously it needs something like:
- After any potential sources of [high?] aerodynamic [upward?] tail forces:
--- shock waves?
--- clean engine burn?
--- uneven engine burn?
--- flight-path induced?
- Exactly where the point-of-no-return (without feathers) occurs in the flight.
BTW what is the likely split of responsibilities in a two-person rocket-powered test crew?
- PF/PM seems inconsistent with the PNF apparently making an unrequested & unannounced change (inconsistent with the flight plan as far as we know it).
- If the PNF was handling the engine (and perhaps stability) the unlock seems more within his remit.
The fact that the unlocking seems to have been very early in some other Virgin test flights suggests that the high & M1.4 'rule'
was not strictly adhered to. The apparent need to deploy the feathers early in some flights [in response to instability?] suggests
that might have been unrealistic (at least for test flights examining the planes stability).
janrein
9th Nov 2014, 10:23
@RetiredF4
PeterH upthread had found some details about the pneumatic system.
http://www.pprune.org/tech-log/550547-virgin-galactic-tech.html#post8726231
More questions remain.
- Pneumatic cilinders single or double working?
- If double working, what are the numbers for the reverse stroke?
- What redundancies in the pneumatic system?
- Any redundancies for the pneumatic system? (e.g. aerodynamic and reaction controls "working" the configuration)
- ...
I would expect numerous failure mode effect and criticality analyses (FMECAs) have been performed prior to any flight including test flights, the details of much of that we cannot expect to ever see.
OTOH it´s amazing how much information over time can be gleaned from te public domain, I am sure we will see more answers coming.
RetiredF4
9th Nov 2014, 11:43
Patent with better pictures
http://www.spacepatents.com/patented_inventions/pat7195207.pdf
A PiRep about a simulator expierience in SpaceShipOne
SpaceShipOne Pilot Report (http://www.airbum.com/pireps/PirepSS1.html)
The project SpaceShipOne
http://robustdesignconcepts.com/files/papers/files/SpaceShipOnePaper.pdf
janrein
9th Nov 2014, 12:15
@PeterH
My hunches for the main drivers:
Before unlocking:
- Assure sufficient Mach number (M 1.4 plausible number) to stay clear of transsonic regime´s agressive and complex dynamics;
- Other factors assumed of lesser or no relevance.
Margins to respect after unlocking:
- Pressure distributions in the supersonic and hypersonic regimes (no complex transients) within prescribed limits, as determined by a.o.:
- - Trimmable elevons within prescribed trim angles
- - Other aerodynamic control inputs within prescribed limits
- - No reaction control inputs or within prescribed limits
- Sense and magnitude of actuation forces from pneumatic feathering system within prescribed limits
- Thrust vector misalignments within prescribed limits (determined from design criteria in combination with static motor tests and powered flight test data recordings), or else early motor cut-off.
Margin to latest permittable unlocking, determined mainly by:
- Aerothermodynamics of unfeathered re-entry, driven mainly by vehicle energy (altitude, speed, weight, i.e. by duration of motor burn)
Other than that I can say very little about the re-entry.
The aerothermodynamics of the re-entry is probably the least known of all factors involved, even for the vehicle designers.
The same can be said for Apollo lunar return re-entries at the time and for the first Space Shuttle Orbiter re-entries, the actual margins were only established after the first flights and there have been narrow escapes.
In comparison SS2 has no or limited thermal protection and the design concept is to not need it, hence the feathers and the unlocking as soon as safely possible, that´s my best guess.
janrein
9th Nov 2014, 12:50
@RetiredF4
What a wealth of information, amazing reading that PiRep.
Must re-read and digest.
I found the patent and other document descriptions of the unlock system disturbing.
Pneumatic unlock actuator?
I can understand the actual feathering being pneumatic, but the unlock? With all the simplicity of the design, why not a physical cable to rotate the "hook"?
Hell, at apogee, I could even see a cable system to move the feathers. Zero aero loads and simplicity. Hardest thing would be moving the feathers once back at 60,000 feet or so.
My prediction is that there was a physical problem with the deployment actuators that allowed the feathers to move once they were unlocked. As I mentioned, our Viper c.p. moved a significant amount in the transonic regime, but then essentially stopped moving aft at M 1.4 Who knows, waiting until M 1.4 may have helped keep the feathers "streamlined".
The NTSB statement "of fact" that the feathers deployed needs to be clarified, ya think? Did they simply move on their own or did the pneumatic actuators move them?
RetiredF4
11th Nov 2014, 13:54
SpaceShipTwo: How it's Supposed to Work : Discovery News (http://news.discovery.com/space/private-spaceflight/spaceshiptwo-how-its-supposed-to-work-141110.htm)
Granted that this source is credible, it gets more confusing.
Initial analysis ruled out an engine problem. Instead, onboard video and data relayed during the flight quickly led investigators to conclude that for some unknown reason co-pilot Mike Alsbury, whose body was found in the wreckage, moved a lever that unlocked the spaceplane’s pivoting tail section early, before conditions were right for aerodynamic forces to hold it in place.
Once unlocked the tail is not held in place by the pneumatic system, but by aerodynamic pressure on the tail?? Or does it mean that the pneumatic system is not powerfull enough to hold it in place when arrodynamic loads exceed a given limit?
I tried to follow the logic in the patent, while the unlocking is explained in great detail, there is not that much info on the working of the tilting system itself, or I was not able to grasp it.
After SpaceShipTwo is dropped by its carrier jet some 50,000 feet above ground, its tail section, unlike the tail of a regular aircraft, is generating up-lift on the vehicle, rather than a downward force (which on airplanes counters the wings’ lift to generate stable, horizontal flight.)
Once its rocket motor burns out and its fuel is spent, SpaceShipTwo’s center of gravity shifts forward, so that the tail generates downward force, like an airplane. In between boost and re-entry, aerodynamics forces dictate when structural loads shift for safe reconfiguration of the vehicle.
Aerodynamic loads on the tail are not only influenced by speed (subsonic, transonic, supersonic) and altitude (less loads higher up), but also by shift of CG due to burnt fuel. According to the source that leads to a reversal of lifting force on the tail. after burn out. If I follow that logic, then the unlocking of the system would not only depend on the airspeed, but also on the remaining rocket fuel.
That must be a hell of a ride for the pilots to keep that ship pointed on the intended flightpath, now i understand these flight path oscilations some more.
Does that also mean, that unlocking the feathering system at 1.4M gives the tail some freedom to align with aerodynamic loads and thus helps in dampening the flightpath?
Any thouhts on that?
Once unlocked the tail is not held in place by the pneumatic system, but by aerodynamic pressure on the tail?? - I am totally confused by what we know so far of the tail 'feathering' system, but I think we must assume the tail movement is 'powered' somehow, and I can see significant aerodynamic forces moving the tail to 'feather', but I cannot see how the reverse will happen - unless the 're-entry' is inverted.
Peter H
11th Nov 2014, 19:52
BOAC
- I am totally confused by what we know so far of the tail 'feathering' system, but I think we must assume the tail
movement is 'powered' somehow, and I can see significant aerodynamic forces moving the tail to 'feather', but I
cannot see how the reverse will happen - unless the 're-entry' is inverted. I too am confused, but some items in the patent seem to clarify some points and raise others.
From para [0018] Pilot-controlled pressurization of actuators 33 drive the aft wing sections and tail booms upwardly and downwardly.
From para [0019] When the wing sections are retracted after reentry, actuators 41 are again pressurized to extend pistons 43 to reengage the locking system.So it seems that:
- Intentional feathering and un-feathering is by pneumatic upward and downward actuation.
- The feathering lock is re-applied as soon as the downward actuation is completed.
Leaving the following questions open:
- What keeps the booms in the feathered position, as there is no locking mechanism for this.
--- Pneumatic pressure?
--- Aerodynamic forces?
- What keeps the booms in the unfeathered position after unlocking, but prior to feathering?
--- Aerodynamic forces? But only in appropriate flight regimes?
--- Pneumatic pressure? But have the downward actuators been pressurized yet?
thcrozier
11th Nov 2014, 21:30
It looks to me like Cylinder 33 controls the tilt. In the diagram it appears much larger the Cyl 41. From the description, I infer that 33 and 41 have the ability the ability to both push and pull rather than rely on some outside force (aerodynamic in the case of 33 and maybe a spring like force - like a leaf sping - in the locking hook in the case of 41) for one direction. If I'm right, then the pneumatic systems have the ability to apply differencial or equal pressure to both sides of the piston.
Applying equal pressure to both sides of the piston would "lock" the piston in place, but only up to a certain force because the air inside is compressible.
Under this scenario, 33 might provide stability above Mach 1.4, but be overcome (depending upon altitude) by aerodynamic forces in the transonic zone.
Another possibility is that a pneumatic failure in 41 could have caused a premature unlock and the pilot was moving the unlock lever to try to fix it.
I'll be reading the NTSB final report with interest.
janrein
12th Nov 2014, 12:27
With the additional insight from the Discovery News ref. (RetiredF4 post)
Once its rocket motor burns out and its fuel is spent, SpaceShipTwo’s center of gravity shifts forward, so that the tail generates downward force, like an airplane. In between boost and re-entry, aerodynamics forces dictate when structural loads shift for safe reconfiguration of the vehicle.
So not only must a sufficient Mach number have been reached (to be clear of the complex dynamics of the transsonic regime), also a sufficient CoG forward shift must have been made.
The first transition occurs around Mach 1.4, or about 1.4 times the speed of sound. It’s not an exact time, but dictated by a combination of several technical factors including vehicle speed, altitude and motor nozzle angle. Around Mach 1.4, the lock (which actually is a chunk of machined aluminum) is moved so that the feather is now mechanically free. It won’t move though because at Mach 1.4, aerodynamic forces keep it nailed back.
The Mach number 1.4 in the applicable flight regimes appears to imply both proper "clean" supersonic conditions and the required CoG for tail surfaces down-lift in that regime.
“In addition to the lock, the feather itself has a big actuator that drives it up and down. So just because it’s unlocked doesn’t mean it’s just flopping in the wind and it can go where it wants. In the (initial) boost phase, the aerodynamic forces can overcome that, which is why we lock it in place,” Moses said.
Which suggests that after "the (initial) boost phase" the aerodynamic forces cannot overcome the pneumatic downforces on the feather. Which would imply that positive lift on the tail surfaces up to a certain magnitude can be tolerated. Such positive lift occurances (assumed limited in time and magnitude) may result from the trim of the horizontal tail surfaces, elevon deflections and thrust vector variations (e.g. due to uneven burn and outflow).
Two main pneumatic 625-psi actuators with a 9.5-in bore and 31-in stroke, change the position of the feather ...
Some numbers related to force and stroke had been found by PeterH, with an estimate of tail boom dimensions (someone may have these?) these numbers may then be related to maximum adverse (positive) lift on the tails that can be tolerated.
Which may serve to answer part of the recent questions from PeterH.
And leaves the following questions:
- What keeps the booms in the feathered position, as there is no locking mechanism for this.
--- Pneumatic pressure?
--- Aerodynamic forces?
In line with the previous I would expect Yes and Yes. In addition the proper re-entry attitude relative to flight path - when not controllable aerodynamically - may be set up by the reaction control gas jets.
@thcrozier
Applying equal pressure to both sides of the piston would "lock" the piston in place,
A hard locking in any halfway position is not possible with pneumatics, and I would not expect the feathers to be in any other position than their extremes except when transitioning. Maintaining some pressure on the retreating side of the piston may serve though to smoothen the transition. In the feathered position I can imagine - rather than bi-lateral pressurisation - a uni-lateral pressurisation pressing against a hard mechanical stop.
@BOAC
- unless the 're-entry' is inverted. ,
I understand the the unfeathering motion to be driven by the pneumatics, possibly assisted with forward trimming of the horizontal stabilisers and up-deflection of the elevons to reduce required forces.
Upon inadvertant failure of the pneumatics during the unfeathering phase perhaps a roll to (partial) inverted attitude may be an emergency option (inside or outside of designed procedures). I recall this trick had been pulled by an aeroplane with structural wing failure (distant past, must search deep ...).
Other than that: bail out at safe altitude and speed.
It would be interesting to know if parachutes are forseeen for space flight passengers (?)
wiggy
12th Nov 2014, 16:13
janrein
I recall this trick had been pulled by an aeroplane with structural wing failure (distant past, must search deep ...).
At the risk of thread drift I guess you mean this:
Zlin wing Structural Failure Report - Neil Williams (http://www.aerobatics.org.uk/repeats/zlin_wing_failure.htm)
Oh yeah, the aerobatic pilot's famous inverted approach and crash.
As with me on my Viper leading edge flap failure ride, you have to remain as calm as possible and see what works and what doesn't. But you don't have a minute, or even a few seconds in some cases. As with the aerobatic dude, I was flying fast enough to have a small degree of roll control and plenty of yaw, even with the right LEF folded up at 50 or 60 degrees. So don't do something stupid, just sit there!!! Don't slow down, don't go faster, don't pull up, don't roll, just stay there while the gal is happy.
http://sluf.org/misc_pages/rightwing.jpg
The SS2 guys didn't have a chance. And I seem to recall Mel unlocking the feather mechanism on his severe rolling ride for first leg of the X-prize. Alleged reason was for "dampening" or such.
I do not like allowing or depending on my control surfaces to move on their own. Just my preference, but I have reasons. Even at M 2, the SS2 should be able to shut down the motor and come back safely. Problem is the sucker goes ballistic at "x" altitude/"x" speed, and you are along for the ride to apogee. I would guess that the RCS might be used at or slightly above 100K with motor still running at a low setting, but once that motor shuts down for good you're screwed. Not sure about SS2 nozzle gimbals such as shuttle had. Dick Covey and John Blaha told me that the "return to launch site" procedure was a pipe dream, but they practiced it anyway because the main motors could be gimballed and throttled way back. Imagine rotating a zillion pounds of vehicle using the remaining fuel, main motors and OMS system and then gliding back home, heh heh.
Lastly, I do not understand the procedure to unlock the feathers early except it appears that the mechanism is pneumatic, not pure mechanical. Why not a simple cable and long handle to rotate those big hooks? Leverage. Ditto for deploying the feathers coming back. But maybe loads on the feather suckers require a huge mechanical advantage as they are exerting pressure against the locking hook.
I am sure we will see some design and procedure changes, ya think?
janrein
12th Nov 2014, 17:24
@wiggy
Quick find!
And I would say relevant (admitted remotely) to the SS2 mission.
RetiredF4
10th Mar 2015, 16:21
Virgin Galactic to drop Scaled Composites from SpaceShipTwo rocket testing - Aerospace Technology (http://www.aerospace-technology.com/news/newsvirgin-galactic-to-drop-scaled-composites-from-spaceshiptwo-rocket-testing-4496427)
Scaled Composites is reportedly being dropped from testing Virgin Galactic's new SpaceShipTwo (SS2) rocket.
Scaled Composites president Kevin Mickey told Los Angeles Times that the company will no longer be engaged in SS2 testing and would work as a consultant to Virgin Galactic.
Following the SS2 crash, Virgin modified the spacecraft based on the investigation findings.
Virgin Galactic has appointed Mark 'Forger' Stucky as pilot to its commercial flight team responsible for flying WhiteKnightTwo and SpaceShipTwo.
Consequences of the crash and / or disagreement how or if to proceed further?
http://www.virgingalactic.com/introduction-second-spaceship/