Help - Search - Members - Calendar
Full Version: Akatsuki Venus Climate Orbiter
Unmanned Spaceflight.com > Inner Solar System and the Sun > Venus
Pages: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
tasp
Asbestos snowballs (which would be worse than snow snowballs). smile.gif

Perihelion is 1/2 the distance of Mercury.

Stuff coming off this rock is going to be slag.

Which makes it all the more interesting!

What angle does the sun subtend to that close? 10, 20 degrees? Ouch . . .
siravan
The encounter speed with Phaethon will be around 43 km/s. As far as I know, it would be the second fastest encounter after the Halley armada.
tasp
Erf.

43 clicks per second!

Well, makes sense, Phaethon wouldn't be that far from perihelion either inbound or outbound, and it has a 'year' almost 600 days, so it would be screaming around the sun.

One tiny saving grace, that close to the sun, Phaethon is pretty much going to be lit up as much as it ever gets, so the camera will get the advantage of quick exposure time(s).

(Yeah, (s), I am thinking even one pic is going to be a trick)
tasp
Went and looked up some specs on this interesting rock:

day= ~3.6 hrs

IRAS first spotted it back in '83

Apollo asteroid

Mercury, Venus, Earth, and Mars orbit crosser(!)

peri =.14 au

ap= 2.4 AU

yr.= 524 days

inc.= 22 degrees

dia= 5 km

highest temp= 750 to 1000C


Fine ride around the inner solar system, but take your sun block . .






Paolo
QUOTE (tasp @ Jan 16 2011, 10:15 PM) *
It is apparently the parent body for a meteor shower so would that be the first spacecraft encounter with a body like that?


Halley was the first. It produces two different meteor streams.
And Itokawa is suspect of producing one
pandaneko
Page 36: Modified FTA 1st part box connection is as follows

1. C1B1>C2B1>C3B1>C3B2>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB1

2. C1B1>C2B1>C3B1>C3B3>C4B1>C5B1>>> box A

3. C1B1>C2B1>C3B1>C3B3>C4B1>C5B2>>>>>>>>>>>>>>>>>>>>>>>> VB2

4. C1B1>C2B1>C3B1>C3B3>C4B2>C5B3>>>>>>>>>>>>>>>>>>>>>>>> VB3

5. C1B1>C2B1>C3B1>C3B3>C4B2>C5B4>>>>>>>>>>>>>>>>>>>>>>>> VB4

6. C1B1>C2B1>C3B1>C3B3>C4B3>C5B5>>>>>>>>>>>>>>>>>>>>>>>> VB5

7. C1B1>C2B1>C3B1>C3B3>C4B3>C5B6>>>>>>>>>>>>>>>>>>>>>>>> VB6

8. C1B1>C2B1>C3B1>C3B3>C4B3>C5B7>>>>>>>>>>>>>>>>>>>>>>>> VB7

9. C1B1>C2B1>C3B4>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB8

10. C1B1>C2B1>C3B5>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB9

11. C1B1>C2B2>C3B6 >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB10

12. C1B1>C2B2>C3B7>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB11

13. C1B1>C2B2>C3B8>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB12

14. C1B1>C2B3>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB13

end of box connection, FTA continuation to follow, followed by its own box connection after that

P
pandaneko
We are not exactly sure about the future of Akatsuki right now. About the only other thing of interest to me is BepiColombo MMO due to go out in the summer of 2014. Surely, that will have to get it right as it is a joint with ESA.

I do not know much about it (or them, two probes in one go).

P
pandaneko
Page 37: FTA modified (from that presented at the first investigation meeting on 17 December 2010) (continuation)

I am sure some of you may be fed up with FTAs, but bear with me as I guess that this may be the last one.

(this FTA continues from box A on the eralier part and I am not giving it any column/row number, P)

C1B1: thruster nozzle/throat destruction

C2B1: insufficient mechanical strength
C2B2: strength decline due to excessive external force
C2B3: excessive thermal stress

C3B1: bad design
C3B2: bad manufacturing
C3B3: excessive mechanical burden at launch
C3B4: collision with a meteorite
C3B5: excessive thermal flow
C3B6: bad film cooling
C3B7: thermal energy input from outside

C4B1: excessive fuel ingredient supply

(Here, I am alarmed. There is a small probability of my earlier translation of "fuel", meaning "fuel ingredient", but that probability must be small gievn the fact that I noticed this here, P)

C4B2: burn conditions shifting to higher temp side (towards larger F/O ratio, I think, P)
C4B3: insufficient fuel supply
C4B4: film cooling theurst direction anomally (shaded)

C5B1: insufficient fuel supply
C5B2: excessive oxidant supply

C6B1: box B
C6B2: box B

C7B1: insufficient fuel pussher gas pressure
C7B2: excessive pressure damage to fuel liquid system
C7B3: mixture of fuel ingredient from fuel liquid system

C8B1: bad regulation
C8B2: excessive pressure damage to gas system
C8B3: gas mixture from gas system (there is something else right after this, but it is not clearly seen, P)

C9B1: closure of piping system
C9B2: closure of CV-F (shaded)
C9B3: cosure of fuel tank outlet (BWI,P)

VB1: negative: design confirmation test already doen with QT

VB2: negative: AT already done with real flight model with equivalent flight loading

VB3: negative: launch environment was normal

VB4: negative: probability calculation of meteorite collision during flight to Venus is extremely low

VB5: negative: no excessive propulsion was generated given the obsrved acceralation record

VB6: negative: P2(regulated pressure) and P4 (oxidant tank pressure) are all normal and they receive gas supply from the same regulator valve

VB7: negative: pressurising system's capacity was checked to be normal by the pre-launch water flow test. Cleanliness test afterward was normal. Also, we can confirm that there did not exist such a low temp environment leading to fuel vapour freeze

VB8: possible: we cannot rule out this possibility as far as the specs checking, test contents, tested items are concerend (WBI, P)

VB9: negative: leak to outside cannot be possible because each pressure variation around delta V are consistent with delta V as calculated from the observed acceralation and they cannot possibly affect the value of P3, leading to leak to outside the probe

VB10: negative: siginificant pressure loss with the liquid system is unthinkable given the history of P3 valve

*: given that it took about one hour for P3 to rise to the value of P2 and this is confirmed by the P3 pressure history after VOI there is no possibility of excessive pressure damage to sections between the tank and P3 (WBI, P)

++: the possibility of excessive pressure damage to sections between P3 and injector. This is confirmed by the fact that P3 pressure history is consistent with the telemetry data of acceralation during VOI-1 (WBI, P)

VB11: negative: given the remaining amount of fuel on board it is not possible that the diagphram moved to the position to block fuel tank outlet (WBI, P)

VB12: negative: Ditto

VB13: no excessive propulsion than anticipated was produced, confirmed by observed acceralation

VB14: possible: cannot rule this out as the burn was made in conditions not checked before

VB15: negative: there was no temp anomally that might have led to nozzle strength decline, confirmed by injector temp and fuel ingredient valve temp records

(there are some more comments on the inverted (WB inverted) boxes, P)

1. closure of CV-F
2. closure of fuel tank outlet
3. excessive pressure damage to fuel liquid system

reasons for these modifications are:

1: this was considered to be an inapproapriate reason for judgement

2: this factor was placed downstream of gas system excessive pressure damage in Investigation 1-2, but since this is a factor below fuel system excessive pressure damage. Position was shifted accordingly

3: this factor, in Investigation 1-2, was judged by the reason that" P3 pressure history indicates no significant excessive pressure damage to liquid system". However, this was a statment referring to two different closure positions. Thus, it has been divided into two separate parts

(I will upload the box connection tommorrow, P)

P

pandaneko
Page 37: A3 (continuation's box connection, P)

1. A>C1B1>C2B1>C3B1>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB1

2. A>C1B1>C2B1>C3B2>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB2

3. A>C1B1>C2B2>C3B3>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB3

4. A>C1B1>C2B2>C3B4>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB4

5. A>C1B1>C2B3>C3B5>C4B1>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB5

6. A>C1B1>C2B3>C3B5>C4B2>C5B1>B>C6B1>C7B1>>>>>>>>>>>>>>>>>> VB6

7. A>C1B1>C2B3>C3B5>C4B2>C5B1>B>C6B1>C7B2>C8B1>>>>>>>>>>>>>VB7

8. A>C1B1>C2B3>C3B5>C4B2>C5B1>B>C6B1>C7B2>C8B2>>>>>>>>>>>>> VB8

9. A>C1B1>C2B3>C3B5>C4B2>C5B1>B>C6B1>C7B3>>>>>>>>>>>>>>>>> VB9

10. A>C1B1>C2B3>C3B5>C4B2>C5B1>B>C6B2>>>>>>>>>>>>>>>>>>>>> VB10

11. A>C1B1>C2B3>C3B5>C4B2>C5B1>B>C6B2>C8B3>>>>>>>>>>>>>>>>> VB11

12. A>C1B1>C2B3>C3B5>C4B2>C5B1>B>C6B3>>>>>>>>>>>>>>>>>>>>> VB12

13. A>C1B1>C2B3>C3B5>C4B2>C5B2>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB13

14. A>C1B1>C2B3>C3B6>C4B3>B>B

15. A>C1B1>C2B3>C3B6>C4B4>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB14

16. A>C1B1>C2B3>C3B7>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> VB15

(Comments at the very bottom of this page are the same as those found on the earlier page of A3, P)

P
pandaneko
This, just below is for your general information.

There was an article in today's Asahi newspaper about a new launch vehicle planned, H3.

I have been worried about the launching capability of JAXA for some time now, because the only other launcher I could find was this Epsilon, a solid fuel thing. Apparently, they use one PC (actually, 2 PCs) to launch it where previously they had to man 300 to 500 people to launch these solid fuel rockets.

That is where my ocncern was. JAXA talked just recently about sending larger probes for interplanetary science missions with enough redundancies. This Epsilon thing seemed to me just going backward on their words, as obviously, their capacicity is so much limited as in the case of earlier M-V.

Now, this H3 thing might be OK, as it uses the second stage of H2, at least 3 of them clustered as the first, with another one atop as the second, due to be launched in 2020, earliest. So, JAXA may well be sending out larger probes for science missions, after all.

P
pandaneko
This news just in.

Apparently, Hayabusa's journey will be made into a cinema film, with a lot of VFX scenes, views only seen by Hayabusa itslef on large screens, due out next year.

I want to see it!

Actually, I was at the National Astronomical Observatory (NAOJ) some years agow. There, I was shown a 3D universe silumation video, and that was fantastic, and I presume there are similar things around the world, but seeing deep space in a cinema, what it might look like, is exciting to me!

Pandaneko
centsworth_II
QUOTE (pandaneko @ Jan 19 2011, 05:12 AM) *
This news just in.

Apparently, Hayabusa's journey will be made into a cinema film...

Yay!
Maybe you should post this in the Hayabusa thread.
pandaneko
QUOTE (centsworth_II @ Jan 19 2011, 08:18 PM) *
Yay!
Maybe you should post this in the Hayabusa thread.



Apologies...

Pandaneko
pandaneko
Now, I am coming back to Investigation 1-2 published on 17 December 2010 by JAXA. I have translated all pages after and including section 3 onward. What I want to do is to translate sections before 3, but not all of them. I will concentrate on technical pages.

Page 8: (of Investigation 1-2) 2.2 Attitude and orbit control system outline

(this page is a schematic. There are 3 columns, left column has 5 boxes, centre column 2 boxex, right column 6 boxes, and I will translate them from the top left box, by the way a short chracter string just below the long solid horizontal line at the top says "data bus", P)

C1B1: coarse sun sensor
C1B2: attitude detection gyro
C1B3: accerometer
C1B4: solar pannell paddle drive
C1B5: medium gain antenna (MGA) drive

C2B1: attitude and orbit control system
C2B2: valve driver

C3B1: sun direction sensor
C3B2: accurate sun sensor
C3B3: star tracker
C3B4: reaction wheels (RW)
C3B5: attitude control thrusters (RCS) valve drive
C3B6: orbit manoever engine (OME) valve drive

NOTE at the bottm left:

valve drive commands are sent to RCS thrusters and OME via valve driver from the attitude control system

P
pandaneko
Page 9: 2.3 Propulsion piping system outline

In the left schematic, top circle is the helium pussher gas tank, down below on left fuel tank and on right oxidant and the characters on thei right are; thruster structure with the propulsion system

There is a large glossary box; they read on the left

LV: latching valve
FDV: pour and drain valve
TP: test port
CV-*: check valve
FLT: filter

character on right, I return to my usual designation

C1B1: RCS thruster (including fuel ingredient valve) (1 liquid system with 23 Newton class propulsion)
C1B2: RCS thruster (including fuel ingredient valve) (1 liquid system with 3 Newton class propulsion)
C1B3: OME (including fuel ingredient valve)(2 liquid system with 500 Newton class propulsion)

P
pandaneko
Apologies, I left out a large chunk of characters in the 2.3 Propulsion piping system outline. This chunk is found top right of this page 9 and they are:

Proplusion system's thruster structure:

One OME (500 N), 8 RCS (23 N), 4 RCS (3 N)

OME is used for orbital insertion to which is supplied pressurised fuel and oxidant. To RCS pressurised fuel is supplied.

In order to prevent fuel and oxidant vapours mixing and reacting inside the pressurised gas system we added check valves

P
pandaneko
Page 10: 2.4 Attitude control policy during OME burn

There are three schematics on left (thruster arrangement schematics), they are, from top down:

seen from above Z+
seen from +X surface
seen from -Z surface

box lower bottom in middle must be self-explanetary, so mus be the lower bottom schematics, but with this I translate these character strings just in case.

up top from left to right:

fuel ingredient valve>burn chamber>throat>nozzle skirt

below, injector and these are the major components of OME thruster

Now, texts, uop top on right, say:

1. a little bit of attitude change is expected during OME burn due toinaccuracy of centre of gravity etc

2. for that reason RCS will be used to control attitude during OME burn

3. those which will be used are T1 to T4 RCS whose propulsion direction are at right angle to OME thrust direction. Also, AB1 to AB4 RCS whose directions are paralell to OME burn direction.

4. if, during OME burn, attitude maitain motion was judged to be impossible (NOTE) thrust is discontinued and operation is then shifted to attitude maintain mode.

NOTE: Probe's thruster arrangement:

This is where the control torque , calculated from attitude angle error, angular velocity error,exceeds for 5 seconds or longer continously, exceeds pre-determined values (5 N around X and Y axes, and 2.5 N around Z axis, in terms of angular acceralation, these correspond to 1.5 degrees/m*2 around X axis, 1.7 degrees/m*2 around Y axis, and 0.9 degrees/m*2 around Z axis)


P
pandaneko
This completes my translations of the two major JAXA reports Investigation 1-2 of 17 December 2010 and Investigation 2-2 of 27 December 2010). I did not bother to translate non-technical pages of Investigation 1-2, they are preliminaries, mostly.

However, there was one page talking about international cooperation. Obvisouly NASA and Australia, but there is Swedish involvement.

On this subject, I was very upset about recent press reports that only a limited number of Japanese institutions are given grains for initial analysis. I thought that it was selfish on the part of JAXA.

However, I am now a little happier to know that NASA and Australian scientists, plus presumably some from Sweden will be invlolved right at these initilal analys stage.

Apparently, there was an MOU beforehand, dealing with the possibility of the samples returned (when they could not have been sure of the return!). I am in a way also amazed that a budget had been allocated to construction of the super clean room for handling the returned samples, again when they could not have been sure of the return!

If there is an MOU all those concerend must be content with what they agreed on. I only hope that they will be able to tell us something interesting in the near future.

P
MahFL
It would have been more embarassing to have samples and a non built clean room back on Earth......
pandaneko
about the failed check valve (fuel) with Akatsuki, there is a similar case reported with Nozomi that also failed, as follows.

http://www.stp.isas.jaxa.jp/nozomi/index-e.html

You may have read this before, but just in case...

P
Paolo
There were a few updates on Akatsuki this morning on Twitter.
On the 17th, the probe will pass perihelion and temperature are being monitored to see how well the probe behaves. It is still in the vicinity of Venus, so the first images of its dayside were taken, although the planet spans a mere 4 pixels
there is also an image of the current position of Akatsuki and Venus http://twitpic.com/4k8691
elakdawalla
The updates seem to coincide with the release of a new SAC report. The diagrams look interesting -- they seem to have something to do with the heating of the spacecraft, with the suspect valve, and with the failed thruster, but I didn't have time to run text through Google Translate so I'm not sure what it concludes:
http://www.jaxa.jp/press/2011/04/20110413_sac_akatsuki.pdf
Paolo
note on page 5 one of the unremarkable infrared images of the dayside of Venus, taken on 9 March
Paolo
a good summary (in French) of the latest status report
I am about to board a plane so I have no time to translate it. It looks like they are having higher temperatures on fuel valves than designed, so they are reorienting the probe, and they fear degradaition of the thermal insulation
Paolo
If I understand correctly yesterday's tweet, Akatsuki is well past the perihelion, temperatures are under control, and they are reorienting it to point the HGA to Earth
Paolo
JAXA has released a picture of Venus taken last 9 December by the 1-micrometer IR camera showing a heavily bloomed dayside as well as what looks like surface features in Aphrodite Terra. Unfortunately, the release is only available at the moment in Japanese (translation welcome!)
see:
http://www.stp.isas.jaxa.jp/venus/top_news.html#20110518
http://www.isas.jaxa.jp/j/column/akatsuki/10.shtml
and a Japanese captioned version
http://twitpic.com/55wf14
also, if I understand correctly recent twits, it should go through solar conjunction in late-June
Paolo
Akatsuki probing the solar corona http://twitpic.com/5bqva7
pandaneko
QUOTE (Paolo @ Jun 3 2011, 12:01 AM) *
JAXA has released a picture of Venus taken last 9 December by the 1-micrometer IR camera showing a heavily bloomed dayside as well as what looks like surface features in Aphrodite Terra. Unfortunately, the release is only available at the moment in Japanese (translation welcome!)


Yes, Paolo, you are right about the Aphrodite Terra. With these few photos the area within a circle is Venus, the dayside to the left and the nightside on the right. On-photo captions refer to the wavelength at 1.01 micron from a distance of 600,000 km, stray lights vertical and sideways, all this looking back at Venus over the shoulder, as it were.

Main text says all kinds of technical things, and reference to Aphrodite says "it is called Aphrodite continent, but it really is a higher altitude area, colder because of that, and therefore less emittance of IR, hence darker than surrounding areas"

I am not going to translate that text because it is not terribly interesting, and instead I will be translating the Akatsuki failure 3rd report just out yesterday (and reported to SAC). I got this from local newspapers and they said that Akatsuki will fire its OME in early part of September.

However, the 3rd report says JAXA will also try another engine on board in addition to OME. I do not remember off hand what it is called, but I will mention it right at the start of my translation. The report is 48 pages and my guess is that the end of my translation will probably coincide with the burning of the OME, minus one or two weeks, I hope.

This 3rd report must surely be their last one and I have been forgetting that they have been working on this report. My translation may turn out be sluggish as I am now more earth bound (evenig Internet timewise) due to the recent incidents in the north of this country.

Pandaneko



Paolo
thanks you in advance for the translation! and welcome back! I hope things are better where you live, now
pandaneko
(Page 1)

Investigation 1-1

Akatsuki's orbital insertion failure - causes and measures to be taken (Part 3)

Following report was submitted to Space Activities Commission (SAC) today (30 June 2011)

JAXA

(Page 2)

Summary

It was estimated through the arguements up until the second investigation meeting that Akatsuki's failure was due to;

1. Fuel side reverse valve remaining closed
2. and this valve closure leading to OME burn conditions status exceeding designed parameters during VOI-1 phase
with resulting OME burn anomally

Given this we have summarised our current and latest thoughts into this third report. The report is based on, after the second investigation meeting, further investigatory considerations, analyses and a host of experiments conducted on ground in order to;

1. explain the status of our investigation into the causes of valve failure
2. our current estimate of the conditions in which OME lives
3. our preparedness for a renewd attempt at orbital insertion

(Page 3)

Outline of the report (for tommorrow, I think)

Now I know what the second engine to be burnt in September is. It is called RCS, and I do not know what it stands for, or I have forgotten by now what it is...

P



pandaneko
(Page 3)

Contents list

1. Past events and the current status of Akatsuki

1.1 Failure estimates and scenarios up until 2nd meeting and summary of this report
1.2 Outline of reverse flow stopping valve and OME

2. Investigating the causes of the closure of the non-reversible valve

2.1 About the candidates for the causes
2.2 Discussing "operational mulfunctions with the valve due to the formation of fuel/oxidant reaction products
2.2.1 Estimating the flow speed of the propulsive agent between up and down streams of the valve
2.2.2 Estimating the amount of migration of the propulsive agent within Akatsuki's propulsion system
2.2.3 Evaluating the possibility of salt crystal formation in the vicinity of the valve
2.2.4 Evaluating the possibility of salt crystal formation leading to the closure of the non-reversible valve
2.2.5 Views at design stage of Akatsuki about oxidant movement and our current understanding based on increased knowledge
2.2.6 Other spacecraft and how they fared
2.2.7 Plans and provisions for our future missions

2.3 Looking at the possibility of contaminants wading through the gaps
2.4 Verification results of the failed non-reversible valve (summary)

3. About the influences and effects suffered by OME due to the failed valve

3.1 About the influence candidates of OME suffering
3.2 Results of the injector thrust motion confirmation tests
3 3 Ground test burn recreating VOI 1 phase
3.3 VOI-1
3.3.1 Test burn (part 1)
3.3.2 Test burn (Part 2)
3.4 Results of investigating into the influences suffered by OME (summary)

4. Examining renewed attempt at orbital insertion

4.1 Orbit plans for insertion
4 2 Trade-offs expected of re-insertion operations
4.3 Examining post re-insertion operations
4.4 About contingencies during closest approach to the Sun
4.5 Schedules from now on up to the next nearest Sun approach manouvre

5. Summary of this report and appendices

(There may be a few ambigous expressions in above list and I hope they will become clearer as main bodies are translated, P)
pandaneko
(Page 4) - not translated as there is nothing new here that we do not know of

(Page 5)

1. History and the current status of Akatsuki (continuation)

3. At the two investigation meetings of 17th and 27th of December 2010 we carried out facts confirmation, and through FTA, estimates, washing out and narrowing down of possible causes as follows

3.1 Via FTA we estimated that the mulfunction was fundamentally due to the closure of the non-reversible CV-F in the gas piping system
3.2 We estimated that as a result the burning condition of OME was affected
3.3 We estimated that as a result OME gave an external torque to the craft and this activated the automatic burn termination signal

4. We therefore conducted tests, experiments and verifications in order to arrive at more detailed estimation of the mechanisms which led to these mulfunctions. However, since the number of items we could use for these tests was limited we had to proceed very carefully in order to obtain the most useful/valuable results.

There were some delays in the schedule for these tests/experiments compared with the suggestions made by the second investigation meeting. This is because extra provisions had to be made to the testing installations in order to recreate an abnormal situation well beyond the scope of initial design conditions and we were forced to adopt a trial and error approach.

Also, there were safety issues to consider with these tests/experiments.

5. On 13th April 2011 we advised SAC with the latest information we had as a result of these tests/experiments. We also explained our operational policy with respect to Akatsuki's nearest approach to the Sun expected on 17th April

6. Right now Akatsuki is flying normally in solar orbit with all of its devices on board functioning normally. (end of page 5)

(I realise that item 2 of this page 5 is somehow missing. I wanted to check by going back to the original PDF, but in case I might lose this in so doing (as I do not know how to hold a temporary translation) I am uploading this as it is.

If there is 2. as I suspect there is I will have to deal with it separately, perhaps at the start of tommorrow's translation, P)
nprev
Thanks, and welcome back, Pandaneko! smile.gif

Re "RCS": Normally, that stands for "Reaction Control System", which would be the attitude thrusters. Makes sense; if the main engine is permanently unusable, then those are the only available motors left (if I remember Akatsuki's design correctly.)
pandaneko
QUOTE (nprev @ Jul 5 2011, 11:57 PM) *
Re "RCS": Normally, that stands for "Reaction Control System", which would be the attitude thrusters. Makes sense; if the main engine is permanently unusable, then those are the only available motors left (if I remember Akatsuki's design correctly.)


Thank you. RCS appears on page 6 and the direct translation would be attitude control system, but, yes, same thing, same engine. Re missing item 2 on page 5 this is OK. There is no item 2 on page 5 as it is a continuation from page 4 which does have the item 2.

(Page 6)

1.1 Scenarios for estimating the causes of mulfunctions up until the 2nd investigation meeting and the summary of the report for the 3rd investigation meeting

1.1.1 Closure of fuel side non-reversible valve

At earlier two meetings we estimated that due to the fuel supply pressure drop because of CV-F fault the amount of fuel to the combustion chamber decreased and the mixture ratio of oxidant and fuel for OME exceeded the design value

1.1.2 OME was therefore affected and craft attitude changed and this led to automatic OME burn termination, destruction of thruster nozzle and/or abnormal burning, all resulting from the abnormal burns which exceeded the design values

Given above estimated scenarios at the 3rd investigation meeting here we will try and report on;

A. Causes for the closure of CV-F
B. Influences and effects experienced by OME due to the closure of CV-F
C. our current plans for orbital re-insertion

Schematic on the right: Akatsuki propulsion system flow chart

There are 3 character strings, topmost reads CV-F, bottom left is RCS (reaction (attitude) control system), and bottom right OME

(end of page 6, P)
pandaneko
(Page 7)

(translated and lost! while searching for the following URL, I will re-translate tommorrow)


Anyway, for some reason I am unable to copy schematics with the texts and I would ask you to refer to the following URL to find relevant pages for those schematics. Surely, at least page numbers must be legible.

http://www.jaxa.jp/press/2011/06/20110630_sac_akatsuki.pdf

P
rlorenz
QUOTE (pandaneko @ Jul 7 2011, 09:07 AM) *
(Page 7)

(translated and lost! while searching for the following URL, I will re-translate tommorrow)
http://www.jaxa.jp/press/2011/06/20110630_sac_akatsuki.pdf


Thanks for the link and welcome back. The upstream migration of oxidizer to form crystals in the fuel-side
check valve is a very interesting explanation for the mishap - I am looking forward to learning more about it from
your translations.
Ralph
pandaneko
(Page 7)

1.2 Outline of CV-F and OME

Outline of CV-F is shown with schematics below

If the differential pressure (upstream pressure - downstream pressure) exceeds the prescribed pressure (cracking pressure) then the valve moves downstream and the passage is clear for flows in both directions

If the differential pressure goes down below the prescribed pressure (re-seat pressure) then the valve moves upstream by the force of the coil spring and the passage is completely closed

A. The same valve mechanism for the oxidant (CV-O) functioned normally during VOI-1 phase. However, CV-F failed to function normally.

B. Both fuel and oxidant systems use the identical valve mechanism

Outline of OME is shown with a schematic below

- One OME installed on board
- Propulsion is of 500N class
- 2 liquid propulsion system using hydrazine as fuel and NTO as oxidant
- He gas is regulated for pressurising fuel and oxidant
- Combustion chamber is made of monocock monolithic ceramics
- Injector is made of Titanium alloy
- Inner surface of the combustion chamber is film cooled by the fuel
- Pushed out (from injector) fuel and oxidant are made to collide and mix before burn

C. OME burn was autonomoucly aborted during VOI-1 phase when disruption in attitude was detected

(end of page 7)

This page is actually devided into 2 areas, left and right areas. Left hand area carries CV-F schematics and right hand area that of OME. I am not supposed to carry pictures here and I refer you to the original schematics in JAXA pdf file. They are very simple to understand if you take a look at them. However, there are more pictures later on in this report and some of them are not that simple to describe and I am worried. Perhaps, I should say something like "picture 1 top left, picture 3 lower bottom right", or something like that. We will see.

On a separate issue, I am hoping that somebody may be able to help me with the approapriate wording in English. My electronic portable dictionary does not have entries for this phenomenon I am going to describe below. Perhaps, mech eng people would know readily, but I do not...

This report, most of it, is about valve closure and they use rather unusual wording for the mechanism. Direct translation may be something like "chewing or eating in of grains into gaps".

Grains and particles, for whatever reasons, migrate into tight spacing between hard surfaces and get sutck in there due to friction (I think). What do you call this movement in English? I would like both noun and verb forms for my translations. I then will be able to be much less descriptive.

P
Syrinx
The noun is "grit".

The verb is "settle" or "sit".

You can say "grit settled in to the tight gaps."

I'm not a mechanical engineer. There may be better technical terms available.
Gsnorgathon
Unless the settling occurred on Earth or during thrusting, "settle" might be misleading.
pandaneko
(page 8)

2.Investigating the causes of CV-F closure
2.1 About the candidates for closing the CV-F

At the two investigation meetings we suspected the closure of CV-F as the cause of Akatsuki's mulfunction. Also at these meetings we filtered out 16 candidates via FTA (section A, 2d (in earlier report somewhere, I think, P) ) for further investigation. We subsequently assigned following investigation policies to each of these candidates.

(Actual layout of this page is divided into two areas, candidates (E-#) on left and policies on right and they are linked by arrows. However, pasting of these made them into a single entity, with coherence still preserved. So, translation is based on this structure, P)


E-1) Bad materials matching at the sealing areas as a result of using different materials
E 4) Bad materials matching at the sliding areas as a result of using different materials
E-5) Deterioration of clearance due to bad fixing methods
E-6) Bad design and bad manufacturing of clearance at the sliding areas
E-7) Bad alignment of the valve and the supporting body
E 10) Bad manufacturing of the sliding areas

Investigation policy for above 6 candidates: Look further into design and manufacturing relating to CV-F

E-12) Grit produced as a result of sliding sitting in gap due to the unexpected number of activations
E-14) Damage to and/or drop out and /or gritting of mechanical parts due to the unexpected number of activations
E-15) Dislodging of coil spring system

Investigation policy for above 3 candidates: Look further into other candidates which may relate to the dynamic behaviour of the valve

E-2) Excessive valve insertion due to plastic (I may be wrong here, P, such as clay deformation meant) deformation of the sealing areas
E-3) Excessive valve insertion due to long term reverse impression

Investigation policy for above 2 candidates: Look further into other candidates relating to excessive valve insertion

E-8) Surface corruption and wear due to sliding motion
E-9) Surface corrosion due to bad materials matching at the sliding areas
E-11) Grit forming in fuel environment and subsequent sitting in gap

Investigation policy for above 3 candidates: Look further into candidates relating to wear and corrosion

E-13) Stoppage of valving motion due to formation of products (salts) in fuel/oxidant reaction

Investigation policy for above 1 candidate: Look further into other candidates relating to formatoon of reaction products in fuel/oxidant reaction

E-16) Contaminant grits sitting in gap

Investigation policy for above 1 candidate: Look further into other candidates relating to contamination

(end of page 8, with thanks for wordings used here, P)
pandaneko
(page 9)


2.1 About the candidates for the closure of CV-F (continuation)

-summary of CV-F failure candidates investigation-


1. About the candidates relating to design and manufacturing of the CV-F

We ruled them out based on analysis of design and manufacturing information provided by the valve manufacturer (section B.1)

2. About the candidates relating to dynamical behaviour of CV-F

We ruled them out based on analysis of dynamical behaviour and experiments using the CV-F (section B.2)


3. About the candidates relating to excessive CV-F insertion

We ruled them out based on the results of our long term reverse imprint experiments (section B.3)

4. About the candidates relating to wear and tear

We ruled them out based on the result of sliding characteristics experiments in fuel environment (section B.4)

5. About the candidates relating to formation of fuel/oxidant reaction products (salts)

We confiremd that they can be the candidates based on the measurements of oxidant flow across the valve in both up and down stream directions (section 2.2)

6. About the candidates relating to contaminants

We cannot rule them out although the possibility is low. Our judgement is based on the record of inspection and work procedure during assembly and testing stages (section 2.3)

(end of page 9)
pandaneko
(page 10)

2.2 Examination of CV-F mulfunction by fuel/oxidant reaction products

Corresponding to item 5 in section 2.1 we will take a renewed look at the amount of fuel and oxidant flow within the gas (He) supply piping system by estimating the amounts existing within the system and consider the influence produced by their reaction products and how this might have led to CV-F mulfunction.


Assumptions:

1. Oxidant vapour moved upstream
2. Fuel/oxidant reaction produced salts
3. Salts led to CV-F mulfunction

(With this page 10 there is one schematic and captions are as follows)

(Topmost character string: Gas supply pipings)

(One character string below it on the right hand side: Liquid propulsive substance supply pipings)

(Between the two tanks there is an arrow in green: Oxidant vapour)

(Actions to be taken to check on the validity of these assumptions above are as follows, P)

1. Estimate the flow speed of fuel vapour through the valve using real fuel (section 2.2.1)

2. Estimate the flow volume of fuel within the fuel piping system using the fuel flow speeds obtained by experiments (section 2.2.2)

3. Confirm production of salts through fuel/oxidant reaction (section 2.2.3)

4. Conduct experiments to see if CV-F is actually closed by the salts (section 2.2.4)

(end of page 10)

(Details of 1 to 4 above are going to be described in following pages of this report, P)
pandaneko
(page 11)

(this report is getting more and more involved from around this page. Also, I now realise that I have been making a mistake in my translation of the valve(s). Apparently, there are two types. I can only translate them as "closing valve, or closure valve, or stop valve" and what can only be translated as "reverse flow stop valve" Up until now I have translated the only one type that I believed existed as "non-reversible close valve", or sometimes simply CV-F. My apologies are due and I will be careful from now on.)

(Also, there is a schematic of the valve here on this page and the crucial part is separately described, still on the same page with its blow up enclosed in a circle with an arrow pointing from it (the circle) to where it comes from on the valve. Since they are images I could not copy, let alone translate those captions which seem important. Therefore, I think I will have to create a ficticious page 11+ immediately after this translation and deal with the schematic in there)

2.2.1 Estimating the flow speed of propulsive substance (fuel and oxidant) across the valve (method employed at time of design)

At the time of designing the valve we assumed that the flow of the vapour of the propulsive substance is the volume of vapour moving through the clearance of the valving mechanism (leak model) and that we should be able to control it by monitoring the flow volume of helium gas and converting it into the (amount of, I think, P) propulsive substance vapour.

More concretely;

1. We decided that we should measure the flow speed of the propulsive substance (fuel and oxidant) vapour across the valve using the more easily measurable flow speed of the standard gas (helium) under normal flight temperature environment

2. To this end we actually set up pressure differences across the valve and measued the actual flow speed of the standard gas (helium). In so doing we assumed an imaginary hole through which the gas flows and from the actually measured flow speed we calculated an equivalent orifice diameter to be used as a model. (Leak model, see section B.5 for more details)

3. We then calculated, using this equivalent orifice diameter, the leak speed of the propulsive substance and used it as the propulsive substance flow speed at the time of our design.

TABLE of flow speed (mg/s) of propulsive substance at design stage (table got broken in C/P, so I will describe the contents below, P)

Close valve: Oxidant is 1×10‐08 and fuel is 1×10‐10

Reverse flow stop valve: Oxidant is 2×10‐08 and fuel is 2×10‐10

※ for more details refer to section B.5

(end of page 11)
pandaneko
QUOTE (pandaneko @ Jul 13 2011, 09:48 PM) *
(page 11)

I now realise that I have been making a mistake in my translation of the valve(s). Apparently, there are two types. I can only translate them as "closing valve, or closure valve, or stop valve" and what can only be translated as "reverse flow stop valve" Up until now I have translated the only one type that I believed existed as "non-reversible close valve", or sometimes simply CV-F. My apologies are due and I will be careful from now on.)


I now know that there has not been any translation mistake with the valves. The first type is a simple kind, typically found on the lines coming out from the helium tank, presumably solenoid operated, and this has not appeared until page 11, anyway. And, rather abrupt appearance of the valve of this type, I think, is meant for the integirity of measuring the flow rates only. CV-F is the second type and is the only one of its kind in this report.

(page 11 +)

(this is about CV-F and its schematic in detail, trying to show the passage of gas through CV-F when the valve is closed)

The picture inside the circle is the cut-off section of the closed contact area. The rectangular block at the centre is the sealing material, made of polymer. The character string insdie it reads "dispersion". The gas is coming from the left (downstream), moving beyond the sealing material. Immediately underneath the sealing material is shown part of the moving part of CV-F.

To the left of this rectangular block of sealing material they talk about two types of gas, all coming from the left, and the one at top is "transmission gas" and its flow is indicated by larger arrows. The character just before the block reads "melting" (and by the way the character string immediately after the block reads "transmission").

The character string just below the transmission gas reads "leak gas". Its flow is indicated by smaller arrows. The character string immediately after the leak gas and to the left of the block reads "small pores", meaning the gap between the sealing material and the moving valve.

There are two chunks of character strings to the right of the centre block.

The first block at top reads "phenomenon which shows transmission of gas through the valve seat sealing material", corresponding to the transmission gas whose character is shown to the left of the block at similar height.

The character block below the first one reads "phenomenon which shows the passage through the gap between the movning valve and the valve seat sealing material", corresponding to the leak gas escaping upstream through the gap.

(end of page 11 +)



pandaneko
(page 12)

2.2.1 Estimating the propulsive substance flow speed across CV-F (revised analysis)

We measued the flow speeds using real propulsive substance and obtained following knowledge.

(Here, this new knowledge is layed out in two boxes, and in-between them there is a table of comparison)

(the first box contains;)

The actual flow speed across CV-F was found to be larger than the value given by the leak model where all of the propulsive substance going through CV-F was equated to "leaking amount" (leak model) (See comparison of A and C in the table below)

((On the other hand, I think, P) They were more or less equal across stop valves)

Therefore, it seemed to suggest that the propulsive substance flow speed across CV-F via transmission is much larger than our previous assumption and we had to revise our estimation by equating all of the flow to the amount due to transmission only (transmission model) (see column cool.gif

(This is frustrating. In all of my translation of page 12 the character "B" has turned into a smiley as far as I can see. Please mentally replace three smilies on this page with "B", surprised that the entry here is intact even if this note has been added as a last measure, P)

Table: Comparison of propulsive substance flows speeds and the measued values (across CV-F, P)

(Here again, the table structure was lost due to C/P, so I will explain the table structure first. There are 4 columns and 3 rows. The first entry in the row 1, column 1 (R1C1) is CV-F )

R1C2: A) Design (leak model※1)
R1C3: cool.gif Revised analysis (transmission model※1)
R1C4: C) Measured values

R2C1: Oxidant @ saturated vapour pressure
R3C1: Fuel @ saturated vapour pressure

R2C2: 2x10‐8 mg/s R2C3: 3x10‐5mg/s R2C4: 0.8x10‐5mg/s
R3C2: 2x10‐10 mg/s R3C3: 1x10‐10mg/s R3C4: 7x10‐8mg/s ※2

※1:See section B.5 for details of each model
※2:Actual fuel flow speeds were measued using valves similar to CV-F since the number of available units was limited

(the second box contains what follows, P)

As for oxidant the measued value of flow speed © was larger than the estimated value used at the time of design (A) and was of the same order of magnitude as that given by the revised model (transmission model)

As for fuel (A) and (cool.gif values were of the same order and they were smaller by one order of magnitude compared with the value ©. However, the fuel flow speed is small enough even using the © value (see section B.6). From this we concluded that as for oxidant transmission effects were dominant in the current case we were looking into.

(end of page 12)
pandaneko
(page 13)

(with this page I may make 2 mistakes, neither of which should be serious, however, P)

2.2.2 Estimating the flow volume of the propulsive substance

Here, we re-calculate (and show the graphs, P), using the measued oxidant flow speed across CV-F and stop valve, the volume of the vapour of the propulsive substance (oxidant here, P) which flows within the gas supply piping system of Akatsuki.

We see that the amount of oxidant which flows into section D (downstream of CV-F) during 6,000 hours (time between tank charged up and VOI-1) is larger than pre-flight prediction by two places (here, first likely mistake, two digits?, P). Incidentally, the estimate for the fuel flow is shown in section B.6.

(Before moving on I should mention the layout of this page. First schematic shows the sections as follows (from oxidant tank to fuel tank).

Section A: between oxidant tank and CV-O. Here, oxidant vapour is saturated including its tank.
Section B: between CV-O and stop valve
Section C: between stop valve and CV-F and helium gas flows into this section from its tank
Section D: between CV-F and fuel tank and this is where the sliding part of CV-F exists


After this there are two rows of 3 graphs each. Left hand graph corresponds to section D, middle graph to C, and right hand graph to B, and the same order with the second row.

The first row is the amount of oxidant in mg. The second row is the partial (?, my second mistake likely here) pressure of oxidant in MPa)

(with each of the graphs horizontal is the time scale up to 6,000 hours and the blue solid line is the re-calculated value and the dotted blue line is the pre-flight estimate)

It is our understanding that since section D contains the emptied volume of the fuel tank and its (section's, P) overall volume is far larger than that of other sections, larger by 3 places, or digits, the difference between partial pressures is maintained and this results in continued oxidant flow across CV-F.

(end of page 13)
pandaneko
(page 14)

(here, halfway down the page there is a schematic of transparent CV which was used for the experiments. What is coming in from the left is he gas plus fuel vapour, and from the right he gas plus oxidant vapour. You can see green and red rings in the transparent part of the CV-F. Green corresponds to the sliding area (part) of the CV-F and red to the polymer sealing portion. One caption to the left of this schematic, just below it reads "salt formation confirmed")

2.2.3 Evaluating the possibility of salt crystal formation in the region of CV-F

We saw from sections 2.2.1 and 2.2.2 that Akatsuki's failure might be related to the possibility of fuel and oxidant vapours mixing in the area of D which includes the sliding portion and downstream of CV-F

and that in that case the reaction products could include solid crystals (forming) in the downstream of CV-F.

Therefore, we set out to conduct experiments to see if this actually happens, by making a transparent CV-F and confirmed the formation of crystal salts near the seal and sliding areas as the reaction product.

For your information the chemical reaction is as follows.

6N2H4 + 3N2O4 will lead to 9N2 + 12H2O (ordinary burn reaction) and 4NH4NO3 + 2N2H4 + 3N2 (crystal salt formation)

(end of page 14)
pandaneko
(page 15)

(here, apart from texts, you have one schematic, actually a pictuture, I think, of the transparent CV-F and a symbolic representation of fuel and oxidant vapour flow into CV-F. With the picture the transparent CV-F is enclosed in a dotted square block. With the symbolics he gas and fuel vapour coming in from above and he gas and oxidant vapour from below)

(in addition, there is a table at the bottom which gives the value of the pushing force required to operate the valve before and after salt formation. As I fear that the table structure will be lost I will explain it as follows.

(There are 4 rows. Top row's character reads the force I just mentioned above. Second row is the numbered valve operating sessions. Here, with this table they did valve operation 10 times. 3rd row gives the force required for the valve operation before salt formation. 4th row gives the force required for the valve operation after salt formation.

With 1st, 3rd and 5th sessions (highlighted in yellow) the forece greater than 400gf was applied and the valve still failed to operate) (here below, I am pasting, with my fingers crossed!)

2.2.4 Observation of the possibility of CV-F closure by salt formation

Since it was confirmed in section 2.2.3 that salt crystals do form in the vicinity of CV-F we conducted further experiments as follows.

We supplied he and fuel, he and oxidant vapours to the closed CV-F from both upstream and downstream directions of CV-F and let the salt crystals form inside the CV-F. We then, after a certain lapse of time, tried to see how valving operation might be affected. This valving operation was carried out using the same experimental setup used in section 2.2.3.

By confirming salt formation (visible confirmation) after in-orbit time simulation (taking into account of the oxidant flow) we tried to push the valve directly into CV-F. Prior to salt formation the pushing force (equivalent to cracking pressure) required was about 160 gf.

3 times out of the 10 times we tried the force larger by 2.5 times the usual pushing force (160 gf) failed to operate the valve, confirming the closed state of CV-F.

(here below, only the numbers


Before salt formation: 160 160 160 160 160 157 158 156 160 162
After salt formation: 400>※ 160 400>※ 160 400>※ 170 160 170 160 180

Note: 400>※ means that valve failed to operate even with the force of up to this magnitude

(end of page 15)

(I am feeling very upset, looking at the figures in this table, 3 times out of 10!!!, no wonder Akatsuki did not succeed, P)
pandaneko
(page 16)

2.2.5 Design policy on oxidant flow and our current understanding based on new knowledge

Design policy

1. Explosive pressure rise inside the piping system due to mixing of the vapours of propulsive substance must be avoided

2. Estimation of fuel vapour flow both upstream and downstream across the valve(s) (CV-F as such is not mentioned here, P) will be derived based on the "leak model" where oxidant vapour passes through an equivalent orifice calculated from the measued value of the standard gas (helium) flow

Our current understanding

1. There are two types of mechanism for the gas flow across the valves. They are as follows and depending on the valve structure there can be cases where the latter will play a more dominant role.

• Gas flows out through minute gaps (leak)
• Gas pases through the inside of the polymer valve seal material (transmission)

2. In particular, with oxidant it was found vital to consider not just the leak speed using the equivalent orifice but also transmission

3. There is a possibility that mixing of oxidant and fuel vapours may create salt crystals

4. Created salts may lead to the closure and/or mulfunctions of the valves

*Research into past mulfunctions regarding oxidant flow shows examples in section B.7

(end of page 16)
pandaneko
(page 17)

2.2.6 Other spacecraft in the past

Japanese satellites up until now, Akatsuki included, have been designed based on the policy described in section 2.2.5.

Given the new insight/knowledge/understanding we conducted rough quantitative estimations as to whether JAXA's past 2 liquid propulsion engines might have caused valve closures due to salt formation. Our conclusion is that no other JAXA satellites could have developped such mulfunctions for the reasons listed below herewith (Nozomi's case will be mentioned later)


Reasons:

• Even with 2 liquid engines on board those satellites whose operational periods are short (such as earth satellites) we cannot find evidence of the effects of oxidant flow on valve functions before operations end

• With all other satellites other than above mentioned satellites the effects of fuel and oxidant mixing are found to be small enough due to the number of valves, valve positions and piping layouts

Section B.8 will show examples of gas provision piping systems used with JAXA satellites whose 2 liquid propulsion engines' operational periods are long. Also, some examples from overseas will be shown.

(end of page 17)
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2024 Invision Power Services, Inc.