Author Topic: Ariane 5 VA241-SES-14 (with NASA GOLD payload) Al Yah-3 Jan. 25, 2018-DISCUSSION  (Read 112430 times)

Offline Lars-J

  • Senior Member
  • *****
  • Posts: 6809
  • California
  • Liked: 8485
  • Likes Given: 5384
I remember, shortly after the incident, some people guessing that this had simply been an incorrect azimuth entry.  The main argument against this theory was that a 20 degree difference would surely have brought the vehicle outside of the range track quite early on.  For a 20 degree difference, every kilometer downrange is another 364 meters away from the expected position at that time, or another 342 meters away from the nominal flight line as the crow flies.

I'd still like to see a *real* map of what the allowed flight corridors out of that launch pad (showing what border would/should trigger a destruct), because like Craftyatom I remain unconvinced that there really was such a wide latitude of allowed azimuths.

What if it had been 40 degrees off rather than 20? Would they have let it keep going because it was "perfectly healthy"?
« Last Edit: 02/23/2018 07:17 pm by Lars-J »

Offline Chasm

  • Full Member
  • ****
  • Posts: 495
  • Liked: 230
  • Likes Given: 0
Maybe it is just my reading comprehension again. To me the report reads that the contracted launch trajectory required a 70° azimuth but that the specification stated 90°.

Space being the highly regulated industry that it is the various work tasks were performed and verified to specification. Oops.

Offline input~2

  • Moderator
  • Global Moderator
  • Senior Member
  • *****
  • Posts: 6810
  • Liked: 1540
  • Likes Given: 567
Quote
l’origine du problème est une demande inédite d’un des clients pour qu’en fin de mission son satellite soit séparé de l’étage supérieur à angle droit de la trajectoire de vol

translation:
"the origin of the problem is an unprecedented request from one of the customers so that at the end of mission his satellite be separated from the upper stage at a right angle to the flight path"

source: https://www.aerospatium.info/commission-enquete-dedouane-ariane-5/

Online LouScheffer

  • Senior Member
  • *****
  • Posts: 3383
  • Liked: 6111
  • Likes Given: 837
Quote
l’origine du problème est une demande inédite d’un des clients pour qu’en fin de mission son satellite soit séparé de l’étage supérieur à angle droit de la trajectoire de vol

translation:
"the origin of the problem is an unprecedented request from one of the customers so that at the end of mission his satellite be separated from the upper stage at a right angle to the flight path"

source: https://www.aerospatium.info/commission-enquete-dedouane-ariane-5/

Does the A5 guidance system have the possibility of gimbal lock?

This is the only way this makes any sense to me.  If it did have the problem of gimbal lock, then the obvious place to put the lock zone would be at right angles to the flight path.

Now if the customer requests right angles, they can't do that in the obvious way, since the gimbal would lock up.  So they would offset the bad axis by 20o degrees.  Now they need to shift everything else by 20 degrees to compensate, but they forgot the launch azimuth.

Presumably this would not happen with any modern INS.

Online LouScheffer

  • Senior Member
  • *****
  • Posts: 3383
  • Liked: 6111
  • Likes Given: 837

Does the A5 guidance system have the possibility of gimbal lock?
Anything with mechanical gyros has gimbal lock.
Only systems with only 2 gimbal axes can lock up, as the two axes can line up.  If instead they have 3 gimbals, the system can always keep the axes separate, and this allows arbitrary orientations with no lock.  However, the system is then heavier, more complex, and less accurate.

This was a problem for Apollo.   Gemini had the extra axis, and astronauts were free to maneuver as they pleased.   Apollo returned to the simpler system, and astronauts had to worry about gimbal lock.   Normally no big deal, but in some cases they had unexpected attitudes, and very nearly severe problems.

Online LouScheffer

  • Senior Member
  • *****
  • Posts: 3383
  • Liked: 6111
  • Likes Given: 837
The gimbal lock hypothesis explains everything we have seen, with just one mistake.
(a) Assume Ariane 5 has a 3 axis gimbal with the usual axes (roll, pitch, yaw).
(b) Such a system will always have an axis where the gimbals lock.  They pick this to be 90 degrees yaw, which is never used.
(c) Everything works great for many years.
(d) A customer asks for release at right angles to the path.   This is exactly what the INS cannot do.
(e) The solution is to rotate the INS 20 degrees clockwise, as seen from above.  Now the rocket can yaw 90 degrees to the right without gimbal lock.
(f) However, now to go due East, they need an azimuth of 70 degrees.  But all other GTO missions have used 90 degrees, and this is what is entered.  No one notices since it's the usual value.
(g) The rocket immediately goes to 90 degrees as measured by its INS.  Since the INS is rotated 20 degrees, that's 110 total.
(h) The rocket applies the correct delta-V for the intended orbit.  This results in a slightly lower apogee, since there is less rotational boost from the Earth at 110 azimuth.

Note:  Some steps must have accounted for the INS rotation.  The inertial alignment on the pad, before launch, should have shown rotation not aligned with the INS coordinates.  Since this did not raise any flags, this must have been anticipated.
« Last Edit: 02/24/2018 01:24 am by LouScheffer »

Offline MaxTeranous

  • Full Member
  • *
  • Posts: 154
  • Liked: 260
  • Likes Given: 55
I remember, shortly after the incident, some people guessing that this had simply been an incorrect azimuth entry.  The main argument against this theory was that a 20 degree difference would surely have brought the vehicle outside of the range track quite early on.  For a 20 degree difference, every kilometer downrange is another 364 meters away from the expected position at that time, or another 342 meters away from the nominal flight line as the crow flies.

I'd still like to see a *real* map of what the allowed flight corridors out of that launch pad (showing what border would/should trigger a destruct), because like Craftyatom I remain unconvinced that there really was such a wide latitude of allowed azimuths.

What if it had been 40 degrees off rather than 20? Would they have let it keep going because it was "perfectly healthy"?

Every launch vehicle that has ever failed was "perfectly healthy" right up until the point that it wasn't.

Offline Remes

  • Full Member
  • ****
  • Posts: 434
  • Germany
  • Liked: 246
  • Likes Given: 142
Ariane 5 always used ring laser gyros. No gimbal lock.

http://onboard.thalesgroup.com/ariane-5-sets-records-topaxyz-inertial-technology/


--

When the first Ariane 5 malfunctioned, the manufacturers where criticized for not doing hardware in the loop simulations. They still don't for each mission?

--

Quote
incorrect value in specifications for the implementation of the launcher’s two inertial reference systems. Given the special requirements of this mission, the azimuth required for the alignment of the inertial units was 70 degrees instead of 90 degrees, as is most often the case for missions to geostationary transfer orbit. This gap led to the 20-degree shift to the south in the launcher trajectory from the initial seconds of flight.

Can someone elaborate on this 70/90 degree?

From what I understand, the talk is about the initial alignment of the IMU. The IMU measures the gravitation vector and the earth rotation and concludes on its orientation. And now it should have taken 70 degree for azimuth, but it was programmed to overtake 90.

Is my understanding correct?

But what is the connection between azimuth and the rocket orientation? The rockets yaw-axis points to roughly -45 degree (NW, plus minus one/a few degree for tolerances). Azimuth is a reference system which is independent of the rockets orientation. Azimuth 90 degree is always east, regardless where your rocket is looking towards.

And if so, why would the azimuth change based on the mission? The rocket is always in the same position for every start. Are the 70 degree azimuth some internal coordinate system of the IMU? Did they mount the IMU differently?

The rocket was supposed to fly about 90° degree azimuth, plus minus a few degrees? (desired inclination was 3°) Or was it a mission, which would not fly on the equatorial plane but hit it from below or above? (I think the proton did this once).

I must be missing a lot, because I can't make anything plausible out of these numbers.

Online LouScheffer

  • Senior Member
  • *****
  • Posts: 3383
  • Liked: 6111
  • Likes Given: 837
Ariane 5 always used ring laser gyros. No gimbal lock.

If it's not the gyros, they why is flight at right angles to the flight path a problem?

Could it possibly be the software?  For example, if the attitude is computed with Euler angles, instead of quaternions, then some operations (such as inversion) don't work well at certain angles.  This underlying problem is the same as physical gimbal lock; when two axes become aligned, in one case there are physical problems, in the other case mathematical operations become ill-conditioned.

This seems unlikely to me - most navigation software switched to quaternions long ago, precisely to avoid problem like this.   But I'm running out of explanations for why a spacecraft requesting release at right angles to the flight path causes any problem at all.


Offline Remes

  • Full Member
  • ****
  • Posts: 434
  • Germany
  • Liked: 246
  • Likes Given: 142
most navigation software switched to quaternions long ago, precisely to avoid problem like this.   But I'm running out of explanations for why a spacecraft requesting release at right angles to the flight path causes any problem at all.
Seems like they use quaternions and euler angles:
AIAA 2011-6340
https://www.researchgate.net/publication/269128453_Intermediate_eXperimental_Vehicle_IXV_the_ESA_Re-entry_Demonstrator

Page 6:
Quote
The IMU is the QUASAR 3000, also used in Ariane 5 and Vega launchers. It provides the GNC algorithms with
the IXV attitude quaternion, Euler angles, cumulated angular increments and velocity increments delivered at 100
Hz via 1553 MIL Bus .

Of course, not knowing whether they use the same sw version or if the A5 sw is frozen to a older state. At least an interresting hypothesis. Specific mission requirements lead to a reparametrization of the flight software (like an internal reference frame) and cause some not foreseen or not tested side effects.

Offline the_other_Doug

  • Senior Member
  • *****
  • Posts: 3010
  • Minneapolis, MN
  • Liked: 2191
  • Likes Given: 4620
Independent Enquiry Commission announces conclusions concerning the launcher trajectory deviation during Flight VA241

http://www.arianespace.com/press-release/independent-enquiry-commission-announces-conclusions-concerning-the-launcher-trajectory-deviation-during-flight-va241/

The key para:

Quote
Investigations by the Independent Enquiry Commission showed that the trajectory anomaly resulted from an incorrect value in specifications for the implementation of the launcher’s two inertial reference systems. Given the special requirements of this mission, the azimuth required for the alignment of the inertial units was 70 degrees instead of 90 degrees, as is most often the case for missions to geostationary transfer orbit. This gap led to the 20-degree shift to the south in the launcher trajectory from the initial seconds of flight. The cause of the trajectory deviation, therefore, was due to a bad specification of one of the launcher mission parameters that was not detected during the standard quality checks carried out during the Ariane 5 launches’ preparation chain.

Yep -- exactly as I thought.  The launcher never deviated from its planned trajectory, flew it perfectly.  It was simply programmed with inertial space references that were 20 degrees off in azimuth.   :)
-Doug  (With my shield, not yet upon it)

Offline Lar

  • Fan boy at large
  • Global Moderator
  • Senior Member
  • *****
  • Posts: 13463
  • Saw Gemini live on TV
  • A large LEGO storage facility ... in Michigan
  • Liked: 11864
  • Likes Given: 11086
Quote
How did a 20 degree offset from get go not trigger range safety and FTS?

https://twitter.com/toruonu/status/967648161404739584

Quote
It's a question that will be asked. Industry officials say a judgment was made that the rocket was performing well (except for trajectory) and that the danger to local population of debris from flight termination outweighed the dangers of continuing flight.

https://twitter.com/pbdes/status/967649386699411456

Interesting case study for AFTS?
Wouldn't that depend on how it was programmed? An AFTS that merely evaluated how well the rocket was doing compared to planned trajectory would have been perfectly content. Only if "are we nevertheless flying into keep out zones" logic were added would there be some possible termination.

I think?
"I think it would be great to be born on Earth and to die on Mars. Just hopefully not at the point of impact." -Elon Musk
"We're a little bit like the dog who caught the bus" - Musk after CRS-8 S1 successfully landed on ASDS OCISLY

Offline CyndyC

Ariane 5 always used ring laser gyros. No gimbal lock.

If it's not the gyros, they why is flight at right angles to the flight path a problem?

FWIW, WikiP says that ring laser gyros can undergo laser-lock:

Quote
Ring Laser Gyros (RLG)
Main article: Ring laser gyro
A ring laser gyro splits a beam of laser light into two beams in opposite directions through narrow tunnels in a closed circular optical path around the perimeter of a triangular block of temperature-stable Cervit glass with reflecting mirrors placed in each corner. When the gyro is rotating at some angular rate, the distance traveled by each beam becomes different—the shorter path being opposite to the rotation. The phase shift between the two beams can be measured by an interferometer and is proportional to the rate of rotation (Sagnac effect).
In practice, at low rotation rates the output frequency can drop to zero as the result of backscattering causing the beams to synchronise and lock together. This is known as a lock-in, or laser-lock. The result is that there is no change in the interference pattern and therefore no measurement change.
To unlock the counter-rotating light beams, laser gyros either have independent light paths for the two directions (usually in fiber optic gyros), or the laser gyro is mounted on a piezo-electric dither motor that rapidly vibrates the laser ring back and forth about its input axis through the lock-in region to decouple the light waves.
The shaker is the most accurate, because both light beams use exactly the same path. Thus laser gyros retain moving parts, but they do not move as far.
https://en.wikipedia.org/wiki/Inertial_navigation_system#Ring_Laser_Gyros_(RLG)
« Last Edit: 02/25/2018 04:53 pm by CyndyC »
"Either lead, follow, or get out of the way." -- quote of debatable origin tweeted by Ted Turner and previously seen on his desk

Offline CyndyC

Those are some very educational theories & details about how the "unprecedented request" would have added complexity. There are also the questions of why & how much was the request necessary, and knowing who made the request would be part of the answer.

Would Ariane Group have wholeheartedly agreed the deviation from the norm was necessary, or was another company or organization being a little quirky? I can see SES making a special request, because I've seen them ask for and/or receive special treatment from SpaceX. I can see the Saudis not trusting that Westerners would otherwise employ the maximum precautions necessary to protect their satellite from collision, interference, or interception. And then there's NASA with GOLD riding on SES-14. Maybe GOLD needed special treatment for unspecified reasons.
« Last Edit: 02/25/2018 11:57 pm by CyndyC »
"Either lead, follow, or get out of the way." -- quote of debatable origin tweeted by Ted Turner and previously seen on his desk

Offline mn

  • Full Member
  • ****
  • Posts: 1033
  • United States
  • Liked: 872
  • Likes Given: 333
If you adjust the INS by 20 deg in the yaw angle at release, wouldn't that mean 20 deg off vertical during launch until vehicle goes horizontal, so they had to compensate for that in software otherwise my guess is the vehicle should have tried to correct an imaginary 20 deg pitch immediately after liftoff?

Does this make any sense or am I totally not understanding?

What I'm getting at is that if it's clear that they did adjust the software to handle the change to the INS, it makes it all the more difficult to understand how they still ended up wrong.

Offline sewebster

  • Full Member
  • ***
  • Posts: 307
  • British Columbia
  • Liked: 190
  • Likes Given: 155

Wouldn't that depend on how it was programmed? An AFTS that merely evaluated how well the rocket was doing compared to planned trajectory would have been perfectly content. Only if "are we nevertheless flying into keep out zones" logic were added would there be some possible termination.

I think?

This was discussed to some degree previously in this thread. Presumably a prudent AFTS design would have limits programmed in an independent way from the planned trajectory. However, there will always be some chance of failure in both manual and automated systems.

Online LouScheffer

  • Senior Member
  • *****
  • Posts: 3383
  • Liked: 6111
  • Likes Given: 837
Ariane 5 always used ring laser gyros. No gimbal lock.
If it's not the gyros, they why is flight at right angles to the flight path a problem?
FWIW, WikiP says that ring laser gyros can undergo laser-lock:

Though they both use the word "lock", the two effects are quite different.   Gimbal lock happens at two very specific orientations.   Laser-lock happens when the angular velocity on any axis is very low.    It's usually well guarded against since one of the worst cases is the INS simply sitting still on the test bench.  There is no reason I can see that laser-lock would happen when the rocket is pointed at a right angle to the path, but not otherwise.

Offline tobi453

  • Full Member
  • **
  • Posts: 250
  • Liked: 81
  • Likes Given: 15
French press report:
https://www.latribune.fr/entreprises-finance/industrie/aeronautique-defense/trois-questions-sans-reponse-autour-du-vol-va241-d-ariane-5-769778.html

An investigation by CNES regarding flight safety is ongoing. Yannick d'Escatha is in charge.

Quote
Ils ont donné la chance au lanceur", estime un bon connaisseur du dossier.
Flight safety en francais!

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 37441
  • Cape Canaveral Spaceport
  • Liked: 21451
  • Likes Given: 428
Quote
How did a 20 degree offset from get go not trigger range safety and FTS?


Because the vehicle did not go outside the range safety boundary limits

Offline guckyfan

  • Senior Member
  • *****
  • Posts: 7438
  • Germany
  • Liked: 2332
  • Likes Given: 2891
Quote
How did a 20 degree offset from get go not trigger range safety and FTS?


Because the vehicle did not go outside the range safety boundary limits

The latest info said it could not be exploded over Kourou because of debris risk. How is this inside safety boundaries?

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0