There is another piece of news in today's Mainichi newspaper, in Japanese.

I am now very pessimistic about 6 years from now on. This article says;

Akatsuki's failure is due to closure of reverse flow valve, JAXA said.

JAXA said on 27th Dec that the failure was due to the closure of the valve within the fuel supply piping. The valve cannot be operated from Earth. JAXA will do ground testing to see if firing in 6 years is possible. The valve is inside the piping and operates with a coil spring.

JAXA think that for some unknown reason this valve remains closed, leading to insufficient fuel supply and the wrong mixture ratio, leading to abnormal burning.

This view will explain all other abnormal records, it says.

Pandaneko

I do not think Akatsuki will make it..., sad

Thats odd, having a one time operation valve, not reversable by a command.

there are more pdfs released today by JAXA
http://www.jaxa.jp/press/2010/12/20101227_...akatsuki_j.html

Pandaneko, clarification re the valve: Do they mean that it is stuck in its current position & no longer responding to commands?

Thanks, Pandanenko, this is very useful information.

The valve in question is CV-F on the schematic. 'Check Valves' or 'non-return valves' are the fluidics
equivalent of diodes - let the fluid flow one way, not the other (thus you wouldnt expect to have to
command them). Usually a ball is held on an aperture by a spring - pressure one way just holds the ball
onto the gap more strongly and no flow occurs; pressure the other (desired) way causes the ball to push
against the spring and opens a gap allowing flow to occur (as long as the pressure difference is big
enough to overcome the spring).

It sounds like the ball somehow got stuck. Quick tap with a hammer might unstick it.

If the valve cannot be fixed (maybe it would just fix itself, or temperature cycling, or something
distressingly nondeterministic might make it happen - there are 6 years after all) then the only thing I
can think of is as follows, and this only works if there is an awful lot of fuel margin. (For those with a
copy of 'Space Systems Failures' to hand, this is reminiscent of the 2002 recovery of TDRS 9, which
also had a pressurant valve failure.)

Do lots of short burns to use up all but the minimum fuel needed for VOI and some minimal operations
(these burns will have to be arranged to not provide a nondesirable deltaV, and the burns will have to be
short enough to not drop the tank pressure too much, and separated by long enough in time to allow the
slow leak of helium past the check valve to restore the pressure to the regulator value between burns).

As the fuel gets used up, the space in the tank above the fuel ('ullage') increases, and so the pressure
drop per unit fuel use will decline. (i.e. if at the burn the tank is 90% full, then using 10% of the tank
capacity doubles the ullage volume and drops the pressure by 50% ; but if the tank is only 50% full,
using 10% of the tank capacity drops the pressure by only 20%.. a charitable interpretation of the
VOI telemetry is that things didnt go badly wrong until the fuel pressure fell by 30%... but that was with a
burn that only lasted a fraction of the time it needed to). Some combination of setting up
the main burn this way, and perhaps towards the end switching to the monoprop thrusters which are not
as sensitive to the fuel pressure, might squeak us into orbit (of course we still have to be spinning
in order to average out the asymmetric main nozzle torque).

The viability of this approach really depends on the margins in the fuel load.

That sounds like a rather lengthy process. Do you think they can get the Δv out fast enough to make it into orbit?

2.2 FTA

Here on this page there are 6 columns. There is only one box at the leftmost column. I will start with this box. This boc contains the tree top event.

Raw 1: Box 1 (Column 1): Brun stopped when attitude anomally detected. (This connects to Box 2 on the same raw)
Box 2 (Column 2): Anomally in propulsiion
Box 3 (Column 3): OME stops at 152 s and torque generated (and this box connects to Box 1 on Raw 2 in column 3)

Raw 2: Box 1 (Column 3): (coming from the torque box just above it) Attachment anomally at 152 s
>>> Box 2 (Column 6): negative becauselaunch environment was within expectation, also attitude histroy ndicates
no large enough force leading to attachment deformation

Raw 3: Box 1 (Column 3): (this box is just below raw 2 box 1, attachment anomally) Thrust gas direction anomally at 152 s
Box 2 (Column 4) (coming from thrust gas box) Thrust gas chamber deformation
Box 3 (Column 5) Thruster nozzle throat break (This particular box connects to a small box which contains A)

Raw 4: Box 1 (Column 5): Burn chamber break
>>> Box 2 (Column 6): Negative, because we obtained near constant deccerlation just before VOI-s stop, and propulsion coeeficient estimated from deccerlation is about 1.3. Thus, burn chamber is not damaged.

Raw 5: Box 1 (Column 4, just below thruster gas channell) Thrust gas seperation/peeling
Box 2 (Column 5) Inner nozzle surface anomally
>>> Box 3 (Column 6) negative, because test manouver was completed normally and there are no other factors which will change the status.

Raw 6: Box 1 (Column 5, just below inner nozzle surface box) Throat rear burn
>>> Box 2 (Column 6): possible, because burn was made in abnormal condition

Raw 7:

Here, I give up for this evening. Apologies. I jotted down box contents on 3 pieces of paper and I am no longer sure how each raw and cloumn are related. I will start all over again, perhaps with a different scheme. I do not have an easy access to a printer.

Please note that therefore, above information may not be correct...

Pandaneko

Yes, this is from my non-specialist knowledge. However, I can at least confirm the meaning of irreversible, or non-reversible, I mean and they mean that the valve in question is a one-way mechanical valve with no control.

I have no idea how much ajar the valve is at the moment. It should not be closed compeletely, I hope, and my only hope is that by slow and long burning at the calculated rate through this small gap they might be able to place the probe into the right orbit.

Pandaneko

There may be ways to free up the valve through thermal management.

Probably what they want to do is to try to force thermal contraction of the ball and/or spring, given that the spacecraft is already in a heat-positive environment. That could conceivably be accomplished by shading the valve & associated plumbing from the Sun, but the systemic consequences must be carefully considered before doing so. The main issues to consider would be whether the necessary attitude to shade & cool the valve would 1] still allow the arrays to generate enough power, 2] still allow communication with Earth (antenna geometry), and 3] be even possible within the dependent constraints of the spacecraft's thermal control system. which presumably is designed to disperse excess heat to the shaded side of the vehicle.

Such situations really provide the acid test for the 'robustness' of a given system's design. And, yeah, it's pretty difficult to devise a pre-flight test for these sorts of things...

Good luck, Akatsuki team; we're all rooting for you!!!

Paolo, you are driving me nuts! That is a joke, of course. I continue to be amazed at your ability to find new source files. I quickly looked at the first files, the first on is a Q&A document from SAC members in response to JAXA 17 December report. The second one is a much more detailed report by JAXA dated 27 December. All these are very informative.

For instance, they talk a lot about this one way valve. One SAC member in A&A even suggested that FTA tree top should be shifted to something else and I think they did that, too. What I will do from now on is to finish off with the rest of their 17 December report, without translating the bits I ignored this time and start translating these new files.

I was, before the demise of Akatsuuki, naively thinking that I will be watching Venus images over the Christmas period with fried potatoes and a glass of wine. That joy is now long gone.

About the only joy I can get now is to chew up on the demise of Akatsuki in close details with specialists' occasional comments. It is another way of spending my evening times, I think.

Pandaneko

Aw...

P, if it's any comfort, these things do happen. The drama of space exploration is not exaggerated; there are so very many critical things that have to happen for mission success (to say nothing of nominal performance) that it's nothing short of a miracle that they ever happen at all.

Akatsuki caught a bad break.

Although what you're doing right now is certainly not what you expected to be doing, it's arguably of even greater value. You're porting information that otherwise would be exceedingly difficult to obtain in a useable form in English to one of the few guys on the planet--rlorenz--who has world-class expertise on spaceships, and also (apparently) an association with the mission; that's a big deal, that helps, that's a net good thing.

Thanks, man.

er, no. (blush)
I think any competent spacecraft systems engineer would be able to work this stuff out. I just happen to
have a vested interest in Akatsuki as a NASA Participating Scientist, and have researched (informally, btw) many
spacecraft failures for my book with Dave Harland, so I posted an analysis before anyone else did. I just wish I
could nip up there with a hammer and fix it...

btw I totally endorse your gratitude to Pandanenko for exposing this material to a wider audience.

Pandaneko,
I think we are all as grateful as one can be for your translations. Even in the era of machine translation, there is no replacement for the man-in-the-loop for making the whole comprehensible to everyone.
I wish I could be able to make such in-depth analyses as Ralph, but even though I have a degree in aerospace engineering and a specialization in space engineering, it's been 10 years since I last went into these subjects

I have given some thought to these FTA pages translation. Rawwise translation scheme is just too complicated.

What I will do from tommorrow on is to go columnwise. With each of 6 columns I will be translating boxes downward, ignoring their vertical positions in the column.

They will be given unique box code names and at that initial stage I will not even explain how each of these boxes are connected.

There is also this branching out issue, between boxes from connecting lines, and these branching out points will be also uniquely coded. I have not yet decided how.

Having translated all these box contents first, I will then indicate connections. That will be a lot easier and perhaps you will not even need to refer to the original Japanese image page, I think, because I will be saying XXX connects to YYY, etc etc. And, these XXX, YYY will be unique, so there will not be mistake made in connection.

Also, if I need to make my own comments with translated boxes in between I will be calling myslf P for short. I hope it will work!

P

2.2 FTA

(Here is my renewed attempt. Each box is now uniquely coded and there will not be mistakes. Each box has been freshly translated. I will later indicate how these boxes are actually connected)

C1B1: (Tree top) Burn stop on detecting attitude anomally

C2B1: prop anomally
C2B2: attitude/orbit control anomally (ACOS)
C2B3: external force due to large meteorite

C3B1: OME anomally at 152 s and torque generated
C3B2: attachment base anomally at 152 s
C3B3: thrust gas direction anomally at 152 s
C3B4: RCS anomally at 152 s
C3B5: liquid flow out at 152 s
C3B6: attitude seonsor anomally at 152 s
C3B7: attitude control harware anomally at 152 s
C3B8: control computing anomally at 152 s

C4B1: thrust gas flow channel deform
C4B2: thrust gas peeling off (??? by P)
C4B3: burn status aanomally (unsymmetrical burn)

C5B1: thruster nozzle throat break
C5B2: burn chamber break
C5B3: nozzle inner surface anomally
C5B4: rear part of throat burn (this box is shaded, i.e., suspect)
C5B5: unstable burn (shaded)
C5B6: injector thrust anomally (shaded)
C5B7: burn chamber inner surface anomally (shaded)

VB1: negative: launch environment stayed within expected values, attitude history telling us not enough force as to cause deform in attachment protion

VB2: negative: we had roughly constant deccerlation just before VOI termination and prop coeeficient estimated from deccerlation corresponds to 1.3. Therefore, there is no possibility of the upstream of throat breakage

VB3: negative: test manouvour was completed normally and there is no other factor affecting this after the test

VB4: possible: cannot exclude this out because burn was made with unkown conditions

VB5: Ditto

VB6: Ditto

VB7: negative: test manouvour was completed normally and there is no other factor affecting this after the test

VB8: negative: RCS functions were normal immediately before and after VOI

VB9: negative: each section's pressure change is in line with delta V, estimated from observed delta V. Therefore, P must have remained unaffected. Outflow impossible.

VB10: negative: threefold redundancy was in place

VB11: negative: hardware is functioning properly at present and telemetry data suggests no fatal anomally

VB12: negative: both before and after the anomally these were functioning as designed

VB13: negative: possibility of collision at this particular moment (152 s) is very small and above all no anomally has been detected with the probe (This one fails me, meteorite collision can happen any time, no?, P)

Anyway, all that remains now is to show how these boxes are connected. My wife is perhaps about give me a shout for dinner, so I send this out right now and upload box connections later on this evening, I think.

Pandaneko