Spaceship One - a partial sucess.

Brett Buck wrote:
snip
Just from my observation, the problem appeared to be a coupling from yaw
to roll - definitely had a significant yaw angle, and the effective dihedral
is extraordinarily high with this design - a lot like the lifting bodies
that had similar control issues. Maybe that resulted from a yaw thrust
vector misalignment, maybe just plain old roll/yaw coupling issues at high
speeds. But it seems very unlikely to be something that can easily be fixed.

Is it fly-by-wire, in which case it might?

On 29 Sep 2004 17:33:00 GMT, in a place far, far away, Ian Stirling
made the phosphor on my monitor glow in such
a way as to indicate that:

Brett Buck wrote:
snip
Just from my observation, the problem appeared to be a coupling from yaw
to roll - definitely had a significant yaw angle, and the effective dihedral
is extraordinarily high with this design - a lot like the lifting bodies
that had similar control issues. Maybe that resulted from a yaw thrust
vector misalignment, maybe just plain old roll/yaw coupling issues at high
speeds. But it seems very unlikely to be something that can easily be fixed.

Is it fly-by-wire, in which case it might?

No, stick and rudder.

wrote in message
oups.com...
Brett Buck wrote:
I don't have any direct professional experience with hybrids, but
I would
be amazed if it's worse in "swirl" than your average liquid engine.
And
that's always proven to be small to the point of negligible.

Is it true that this effect is negligable? The second Ariane flight
developed excessive roll from this problem. True, the torque is small
compared to the thrust, but just like today it exceeded the control
authority and forced a premature shutdown (though because the fuel was
spun to the outside of the tank, not deliberate shutdown.)

From: http://www.esa.int/export/esaCP/Pr_14_1998_p_EN.html

"The most significant anomaly, namely excessive roll torque during
cryogenic main stage flight, has now been explained. It will be
remembered that the possible causes of the 900 Nm (Newton metre) roll
torque registered in flight were narrowed down to two :

* roughness of the internal surface of the Vulcain engine divergent
causing the boundary layer of the main jet to spiral;

* snapping of one of the rods attaching the Vulcain engine turbine
exhaust lines to the aft end of the divergent.


Well, that's a point. But 900 n-m isn't very much torque at all for such
a large engine. Figure out how much you get in the other axes from say, 1/2
degree of thrust vector misalignment. I think (if I converted the absurd
unit of "tonnes" correctly) that the thrust of a vulcain is ~1.1 million
newtons. Assume the CG is 25 meters from the engine. 1.1 million*25*tan(.25
degrees) ~=123000 n-m.

Brett

On Wed, 29 Sep 2004 17:43:50 +0000, Rand Simberg wrote:

On 29 Sep 2004 17:33:00 GMT, in a place far, far away, Ian Stirling
made the phosphor on my monitor glow in such a
way as to indicate that:

Brett Buck wrote: snip
Just from my observation, the problem appeared to be a coupling
from yaw
to roll - definitely had a significant yaw angle, and the effective
dihedral is extraordinarily high with this design - a lot like the
lifting bodies that had similar control issues. Maybe that resulted
from a yaw thrust vector misalignment, maybe just plain old roll/yaw
coupling issues at high speeds. But it seems very unlikely to be
something that can easily be fixed.

Is it fly-by-wire, in which case it might?

No, stick and rudder.

I read somewhere that during ascent it's fly by wire with electrical trim
tabs, no computer. During a post flight interview, someone (?Rutan?) said
that the problem with the previous flight was caused by an actuator
overheating (overused) and the actuator shut down. It later reset itself
and worked correctly. They address this problem before this flight.
Although prior to the uncontrolled roll, it looked like the vehicle was
pitch, yaw and rolling quite a bit. Maybe the had larger upper level
winds, causing the same problem to occur.

Craig Fink

h (Rand Simberg) wrote in message .. .

Is it fly-by-wire, in which case it might?

No, stick and rudder.

(for those who watched this on NASA satellite)

Mike Mehlville is 'one of the best stick and rudder men' around! (only
heard that phrase 20 times this morning during the Webcast...) If he
can't control it, who can?... ;-)

A couple of things I noticed-

1. both the roll/yaw problems have happened very high in the
atmosphere. Melville probably figured the aerodynamic effects were
about to go away entirely anyway, so why not ride it out and then take
his time to correct any residual roll with RCS in a non-aerodynamic
environment (it's SS1's 'abort to space' option.)

2. Even if he had gotten spinning so fast that the RCS couldn't fully
stop the roll, he probably figured he still would have been able to
slow the roll enough to feather the tail, and then let aerodynamics do
the rest on the way down -- here's yet another shout out for passive
attitude control during reentry.

3. if there really was a yaw thrust vector misalignment of just a few
degrees, pardon my ignorance about SS1 design, but wouldn't some kind
of minor nozzle or engine shim allow a complete correction of the
misalignment? Or is it further complicated by the changing center of
mass due to fuel use?

Tom Merkle

Bruce Hoult wrote:
In article ,
(Derek Lyons) wrote:


First returns; the 100km mark was broken, ascent phase control
problems remain.


And always will, with that design. Once it gets out of the atmosphere
it simply doesn't have enough authority on the attitide control jets to
compensate for the variations in thrust vector you're always going to
get from a rocket with solid fuel.

The higher a flight is aiming for, the worse the problem will be.

This is bunk. The problem on the first flight was straight
up Pilot Induced Oscillation. The cause for the second
problem is unknown but considering it was a roll problem
it's very, very unlikely to be due to thrust variations
(which would induce pitch or yaw variations).

"Christopher M. Jones" wrote:

Bruce Hoult wrote:

And always will, with that design. Once it gets out of the atmosphere
it simply doesn't have enough authority on the attitide control jets to
compensate for the variations in thrust vector you're always going to
get from a rocket with solid fuel.

The higher a flight is aiming for, the worse the problem will be.

This is bunk. The problem on the first flight was straight
up Pilot Induced Oscillation. The cause for the second
problem is unknown but considering it was a roll problem
it's very, very unlikely to be due to thrust variations
(which would induce pitch or yaw variations).

Yep. There was a good deal of discussion about that one at work today. I
haven't found any information on attitude control effectors. Does SS1 have
some kind of reaction jets? Something else? Today's problem looked like a
pure roll excursion. An aeroelastic effect? Roll trim?

Jon

In article ,
"Christopher M. Jones" wrote:

Bruce Hoult wrote:
In article ,
(Derek Lyons) wrote:


First returns; the 100km mark was broken, ascent phase control
problems remain.


And always will, with that design. Once it gets out of the atmosphere
it simply doesn't have enough authority on the attitide control jets to
compensate for the variations in thrust vector you're always going to
get from a rocket with solid fuel.

The higher a flight is aiming for, the worse the problem will be.

This is bunk. The problem on the first flight was straight
up Pilot Induced Oscillation. The cause for the second
problem is unknown but considering it was a roll problem
it's very, very unlikely to be due to thrust variations
(which would induce pitch or yaw variations).

Either way, Spaceship One is surely not yet in a flight-ready state. By
that I mean it is still basically a prototype. That is not to say it
might not win the X Prize, but if it does so despite its qualifying
flights encountering the sorts of problems which might have seen a
non-prototype grounded then what exactly will the X Prize have
accomplished, especially if better craft lost because they preferred to
wait until their craft had been more fully tested?

Which raises a issue: should prototypes have been allowed to qualify for
the X Prize?

In particular, instead of the X Prize rules stipulating that competing
craft merely be "able to carry three people", should they have specified
that these flights carry actual human passengers as well as the pilot.
(That way there would less temptation to use vehicles which have not yet
been fully tested to try to steal a march on competing craft.)

--
Stephen Souter

http://www-personal.edfac.usyd.edu.au/staff/souters/

In article ,
"Christopher M. Jones" wrote:

And always will, with that design. Once it gets out of the atmosphere
it simply doesn't have enough authority on the attitide control jets to
compensate for the variations in thrust vector you're always going to
get from a rocket with solid fuel.

The higher a flight is aiming for, the worse the problem will be.

This is bunk. The problem on the first flight was straight
up Pilot Induced Oscillation. The cause for the second
problem is unknown but considering it was a roll problem
it's very, very unlikely to be due to thrust variations
(which would induce pitch or yaw variations).

Then you didn't watch the June flight very carefully.

There was a roll each way just after engine ignition. This is
presumably what you are talking about.

But then there was another uncommanded roll in the last seconds before
engine cutout, much as with this flight, but less.

-- Bruce

"Stephen Souter" wrote in message
...

Either way, Spaceship One is surely not yet in a flight-ready state. By
that I mean it is still basically a prototype.

And no one has really claimed otherwise.

That is not to say it
might not win the X Prize, but if it does so despite its qualifying
flights encountering the sorts of problems which might have seen a
non-prototype grounded then what exactly will the X Prize have
accomplished, especially if better craft lost because they preferred to
wait until their craft had been more fully tested?

Define better. The goal of the X-Prize is to fly twice to 100KM in the same
craft in a two week period. It's not to develop a fully functional
commercial spacecraft.

As such, the definition of better is one that can accomplish the goal before
Rutan does. And that's extremely unlikely at this point.



Which raises a issue: should prototypes have been allowed to qualify for
the X Prize?

Of course. Or rather, why should it not?


In particular, instead of the X Prize rules stipulating that competing
craft merely be "able to carry three people", should they have specified
that these flights carry actual human passengers as well as the pilot.
(That way there would less temptation to use vehicles which have not yet
been fully tested to try to steal a march on competing craft.)

No.



--
Stephen Souter

http://www-personal.edfac.usyd.edu.au/staff/souters/