Nozomi in perspective, Revisiting the causes of failure Options

[pandaneko]

http://www.mext.go.jp/b_menu/shingi/uchuu/...ts/04061101.pdf

above for ease of reference

Page 64

Fig. III-1-5 Signal reception strength changes from Nozomi received at Usuda (Japan) station's (60m dish)

Unlike normal recption which shows a tendency that strength is max. in the middle and amplitude is min. in the middle we find on 25 April, instead, that strength is decreasing to the right and amplitude is monotonously increaseing.

This corresponds to the situation in which automatic (or autonomous) Earth searching is not being activated (or obeyed).

Characters on the graph is red says "at normal times".

end of page 64

page 65

Fig. III-1-6 Temporal changes recorded by the Solar Proton Detector

Character string in red on the graph says "Time of accident (estimate)".

end of page 65

page 66

Fig. III-1-7 Strength changes since (or after ) year 2000 in the number (?) of high energy particles suffered by Nozomi

Graph above : Average number of counts per every spin (about 8 seconds) of the satellite in a day.

Character string in red says "020421 (data saturated)".

Graph below: Cumulative flux (cmXX2/str) since (or after) year 2000.

end of page 66

P

[pandaneko]

http://www.mext.go.jp/b_menu/shingi/uchuu/...ts/04061101.pdf

above for ease of reference

page 67

Fig. III-1-8 1 bit comms. (communication using autonomous function)

We could regard this to be a table, but for intuitive simplicity I will be descriptive as follows:

top 3 squares from left to right are:

1. setting a question by a command.

2. judgement by autonomous function.

3. Beacon ON command.

Long horizontal line is "Beacon ON" and the short dotted line just below it says "Corresponds to "Beacon OFF" and the character string below says "YES" and the dotted line is "judgement".

end of page 67

page 68

Table III-1-2 Predicted and actually measured values of data on 2 May and 1 bit comms. (operation by ON/OFF beacon signals)

(here, I regard this to be a C5 and R21 matrix and what follows refers only to the tabular entries)

C2R1: Data on 24 April

C3R1: Estimated value on 2 May

C4R1: Actually measued value

C5R2 to R16 is common and it says"Normal except the heater and the common power source".

C5R17: Measued value on 3 May.

C5R18 to R21: cooled to temp. below freezing temp..

end of page 68

P

[pandaneko]

http://www.mext.go.jp/b_menu/shingi/uchuu/...ts/04061101.pdf

above for ease of reference

page 69

Fig. III-2-1 FTA relating to the cause candidates for "not being able to switch to telemetry mode"

(Here, I want to do a few things (experiments) to make sure that what I have been doing is not a waste in the sense of translation quality (?).)

(Here, if I am right, you should be looking at a table (please refer to the original PDF), not a proper table in the sense of it being square in shape, but still retaining its remote hope for being one. My only regret is that I have given R1 to the top header line's row number as you will see in my translation. Actual table elemtns (apart from header line) could have been named from R1 ( In my notation of what folows R1 is denoted as R2, I am afraind), instead of R2 against each column number) (I will also experiment with lines used for squaring off areas used for character strings)

C1R1: Tree top event

C2R1: Primary cause

C3R1: Secondary cause

C4R1: Tertiary cause

C5R1: Judgement reasons

(After this you must be seeing a thick horizontal line separating the header entries above it and the table contents which follows below the horizontal demarcation line)

C1R2: No reception desipite the satellite being in TLM ON mode

C2R2: X 100: Data processing unit mulfunction (X means negative, I think)

(This "X 100: Data processing unit mulfunction" is based on my onw notation and I will seek your comments on the reproduction quality of this particular entry (C2R2) as a member of the larger table. For instance, ":" is simply a short vetical line for seperation between 100 and what follows. For that matter, there is another vertical line in the original text between X and what follows it. Can you see all these features with your 8 bits OS codnig system)

(OMG!, I have lost what I had in mind for further explanation..., but let me continue anyway)

C2R3: X 200: TMX mulfunction

C2R4: X 300: TCI mulfunction

(row numerbs are effective numbers only, not corresponding to the finest row numbers on the extreme right column)

C3R2: (there is a horizontal connection from 100) X 110: Command autonmation sequence mulfunction

C3R3: X 120: Data bus mulfunction

C3R4: X 210: Command recption circuit mulfunction

(horizontal line connection from 200)

C3R5: X 220: Reley device mulfunction

C3R6: X 310: TCI itslef only going bad

C3R7: X 320: Power cut-off

(now we move on to the next column, which is long)

C4R2: X 111: autonomous (or automatic) sequence DISABLE

C4R3: X 112: Devices shrank back to IPL mode

C4R4: X 211: Device internaland +5V system mulfunction

C4R5: X 212 Responsible componetnts mulfunction

C4R6: X 221: Device internal and +V12 sytem mulfunction

C4R7: X 222: Responsible components mulfunction

C4R8: X 311: Responsible components mulfunction

(Herre after column 5 entries, only row numbers)

C5R2: We sent the same sequence of commands as REAL, but we could not confirmt its outome.

C5R3: after starting up the devices once again, we sent the same related sequence command, but we could not confirm its outcome.

C5R4: We were able to confirm that regular outcome of the command XPA ON and OFF and there was no reason why some other commands copuld not be accepted and executed acccordingly.

C5R6: Ditto

C5R7: Ditto

C5R8: Same as R5

C5R9: Same as R6

C5R10: Power source (CI-PSU) for TCI is the same and is commonly used for the heater control system. Therefore, it is accepted that the POWER OFF led to freezing of propulsive materials and further leading to inability of keeping the Earth in view.

Also, POWER OFF can explain all othe events.

C5R11: Analyses later showed that the time ellapse between POWER OFF and actual freezing was 2 hours at maximum, but this itself does not negate any of the operational sequence.

end of page 69

P

[pandaneko]

http://www.mext.go.jp/b_menu/shingi/uchuu/...ts/04061101.pdf

above for ease of refernce

page 70

Fig. III-2-2 FTA relating to cause candidates for "CI-PSU not being put into ON as required" (part 1 of 2)

(In what follows you will see the sort of difficulties I am having with this particular figure and with others on several fronts. Unlike all other figures so far this figure turned out to be a picture. I was very pleased with this finding and copied it and tried to paste it into this page for ease of translation for myself. It did not work. It perhaps is this forum's policy not to allow simple picture pasting?

So, I ended up handwriting everything within this figure for retyping in here. I do know that I can open up more than one internet pages for work on my display unit, but in reality I will find it difficult to cope with. Anyway, back to this picture figure in question. Because it is a picture, obviously photo-reduced from a much larger original figure which was converted into a picture,
characters on it are very corrupt. You may say enlarge them, but if you do the same corruptnes is enlarged. So, my translation will be sometimes bad, to say the least.)

(Here it goes, and without using row numbers, and the top header line first)

C1: Tree top event
C2: Primary reason
C3: Secondary reason
C4: Tetiary reason
C5: 4th reason
C6: Reason for judgment

(Now, I am coming on to the main figure entries and start using row numbers in line with the number of rows in C6, which will act as reference row numbers without disruption or vacancy between entries in this C6 column. Unlike previous page, R1 here refers to the very 1st row entry, ignoring the header line above them.)

C1R1: Circle: CI-PSU cannot be put into ON state.

(This leads on to 4 tree elements (or squares in the picture if you can recognise it as such), or branches in C2 as follows. Also, a vertical line pointing down from the intersection between C1R1 and C2R1 ppoints to the next figure to follow on page 71.)

C2R1: Cross: 100: PSU limitter?, or resetter?, or reactor? bad action (or failure, or failing). (Here, ? means "uncertain in my recognition of the characters to be read")

(You may actually find more reasons for them to be something more meaningful as my translation continues)

C2R7: Cross: 200: PCU-CU-PSU + (plus) somehing I cannot recognise at all

C2R8: Triangle: 300: CI-PSU bad action (or failure)

C2R12: Cross: 400: CI-PSU device + something I cannot recognise

C3R1: Cross: 110: PCU whole function mulfunctioning

C3R2: Cross: 120: Commanding function mulfunctioning

C3R4: Cross: 130: Limitter? function not functioning

C3R8: Cross: 310: Voltage control not functioning

C3R10: Triangle: 320: Secondary system not functioning

(Actually, just thinking about the picture property of this figure, I am beginning to suspsect that most of the 1.09 Meg of this report might be coming from these pictures. Perhaps not, because these are one layered B/W pictures...)

C4R2: Cross: 121: Logic circuit for commanding failure

C4R3: Cross: 122: "something" ON command output failure

C4R4: Cross: 131: detection register (or resistance) "something" failure

C4R6: Cross: 132: OP amp. gain increase

C4R7: Cross: 133: Comparator failure

C4R9: Cross: 311: Switching TR failure

C4R10: Cross: 312: PAM control failure (I may be wrong here)

C4R11: Triangle: 321: One or more components failure

C5R11: Triangle: 321A: Accidental failure

C5R12: Cross: 321B: Bad soldering of components (Here, if you find a smiley after 321, it is a capital byee)

C6R1: Apart from the yet-to-be done "batch something" after the accident there is no mulfunction outside CI-PSU On/Off function

C6R2: Other commands by the common F?PCA (On/Off in other systems) are functioning properly and it is hard to imagine that only an arbitary function is affected.

C6R3: Ditto

C6R4: In the line in question there is a built-in function whereby FMT, or TMT, or FET switch is turned off in the event that a limit is exceeded on comparison of detected current value and the limit value (or reset value)

If the detection register or resistance fails on "release (infinite impedance as I suspect)" then secondary output is terminated in theory. However, in this case in hand there was an imperfect secondary output reported.

C6R5: This will only occur if the input register (or resistance) fails, or on "something register" "release (infinite impedance?)" fails

However, register failure in a very minute current circuit is unthinkable.

C6R6: 4 system functions are all integrated into the same comperator layout and it is very hard to imagine that only a particular function failed.

C6R7: This is "something" inside the satellite and it is very hard to imagine that it failed accidentally after 4 years in service in deep space.

C6R8: This can be denied in view of the secondary output.

C6R9: Here, do not forget that we are talking about 17 ceramic condensors.

C6R10: Failure mode of the soldered parts after years of service in space use is not "something", but "relaese (and I suspect this to mean infinite impedance)

C6R11: Same as C6R7 and it may be that the said "something" may be WHN... I do not know what WHN may mean, though.

end of page 70

P

[Paolo]

thanks panda, I look forward to your translations every day!
by the way, is there anybody who wants to collect them in a proper document? I would do it myself, but I'm too lazy...

[pandaneko]

thanks panda, I look forward to your translations every day!
by the way, is there anybody who wants to collect them in a proper document? I would do it myself, but I'm too lazy...

Thanks, Paolo. If that somebody is watching these posts regularly you may eventually come into contact with him (or her) and in fact you may jointly pose further questions to me for action.

I might also add at this stage, that I will be wanting to explain, in the season's spirit, why I am bothered at all by this issue of 8 bits or 16 bits coding. I just hope that Admin will allow me to spend some time (space, and not a lot at all) on it as an exceptional posting. It is after all Merry Christmas, is it not? P

[elakdawalla]

I was thinking that a way to make good use of pandaneko's translations might be to set up a multi-author blog and have volunteers match original slides to his translations as blog posts. This is something we'd be able to set up on the Society website after our upcoming redesign, but I'm pretty sure we could also just ingest an existing wordpress blog if someone wanted to get started on the project before.

[pandaneko]

http://www.mext.go.jp/b_menu/shingi/uchuu/...ts/04061101.pdf

above for ease of reference

page 71

(This page is even worse than page 70 in that the extent of photo-reduction is much larger as the number of rows in the last column is now 19, compared with 11 on page 70, making almost all of entries impossible to decipher, but I will do my best.

In the meantime, I forgot to mention something with page 69. There is an arrow pointing upward from C5R11 to C5R10.)

Fig. III-2-2 FTA relating to cause candidates for "CI-PSU not being put into ON as required" (part 2 of 2)

(This figure is effectively a 6C and 19 R matrix. Header entries from C2 to C6 are the same as on page 70. C1 header is continuation, I think and there is no entry in C1 column at all.)

C2R1: Circle: 500: CI-PSU XXX (here, XXX means impossible to recognise characters in this space)

(and entries in C3 all come from C2R1 and they are):

C3R1: Cross: 510: ?PS XXX (here, ? means there is something in this space which cannot be recognised)

C3R2: Cross: 520: S (or 5)PM?, XXX

C3R3: Cross: 530: EPT-SA XXX

C3R4: Triangle: 540: WS (or 5)-SA XXX

C3R6: Cross: 550: DR XXX

C3R7: Triangle: 560: TC XXX

C3R12: Triangle: 570: HCE XXX

C3R17: ***: 580: LVDT XXX Here, *** is a character string, not a circle nor triangle. I strained my eyes and I still could not recognise what it is)

C3R18: ***: 590: USO XXX

C3R19: ***: XXX: pressure sensor XXX

C4R1: Cross: 511: XXX ( and this XXX will be the same in all rows of this column)

C4R2: Cross: 521: XXX

C4R3: Cross: 531: XXX

C4R4: Cross: 541: XXX (and this and the next entry comes from Triangle: 540 in C3)

C4R5: Triangle: 542: single event upset

C4R6: Cross: 551: XXX

C4R7: Triangle: 561: XXX

C4R12: Triangle: 571: XXX

C4R17: ***:581: XXX

C4R18: ***: 591: XXX

C4R19: ***: 5A1?: XXX

C5R7: Triangle: 561A: total dose effect

C5R8: Triangle: 561B: single event latch up

C5R9: Triangle: 561C: latch up by discharge

C5R10: Triangle: 561D: accidental

C5R11: Cross: 561E: bad soldering job

C5R12: Triangle: 571A: total dose effect

C5R13: Triangle: 571B: single event latch up

C5R14: Triangle: 571C: latch up by discharge

C5R15: Triangle: 571D: accidental

C5R16: Cross: 571E: bad soldering job

(and the last column entries are all too bad. There are entries in R1 to R9, R11 to R14, and R16 to R19. It is possible to read a few characters in any of these entries, but this column, giving reasons, contains much larger number of character strings and it is no use to be able to read here and there and translation will be bound to be wrong)

end of page 71

This is perhaps a good time for me to mention the coding issue which has been tormenting me, a particularly acute problem if I am facing a table. I will try to be as succinct as possible.

My ancesters did not have characters and burrowed them from China (Thank you! China) Making the story brutally simplistic, Chinese characters are all pictures (hieroglyphic), made up with a lot of strokes. So, the usual 8 bit coding system cannot cope with the complexity. We therefore need 16 bits and I have no idea why they do it, but they here use the 16 bit system for coding (simpler) alphanumerics as well.

Result is confusing. 16 bit alphabets are OK as they are of the same shape, but simply "fatter" in appearance, but numbers look deceptively similar. My problem is tables. Are the numbers in the table 8 bit coded or 16 bit coded? For that matter, are the line elements, forming the frame of tables, are they what they look like, or not?

My assumption so far is that people who made these tables used 8 bit coding for alphanumeric entries. The only way to check is to copy a table (not a picture table, but text table) for experiment. I meant to do that yesterday, but faced with the picture table not wanting to be pasted here I did not managed to do it. I will try again when all of the translation work comes to an end.

Now, what follows is an extra for your Christmas conversations.

So, characters came from China and you want to write, say, "I eat an apple".

What my ancesters did was to put "I- like picture, Eat- like picture, Apple- like picture". They, however, made a fatal mistake in all this because "XXX-like pictures" were all chosen for their sounds, not understanding each is a picture. Chinese naturally pointed out that above sequence actually reads "You drive a cow". Here, of course, this is only by way of explanation.

So, my ancesters eventually allocated the right pictures in the right places. But, they did not like it at all. Too many strokes! So, they decomposed Chinese characters into components and came up with 2 phonetic sets of characters. Had they stayed with these two sets only my current problem should not be existing.

In reality they decided to retain a portion of these Chinese pictures as well. So, our current writing system is a mixture of phonetic and hieroglyphic characters. This is why 16 bit coding system is used here and how my dilemma started.

Merry Christmas from Pandaneko to all of my colleagues and all those space probes out there in deep space!

[pandaneko]

http://www.mext.go.jp/b_menu/shingi/uchuu/...ts/04061101.pdf

above for ease of reference

page 72

Fig. III-2-3 Bus system voltage and current at the time of flares

(The only caption here is indicated by red lines and it says):

No change is seen in the portions indicated (by lines in red)

end of page 72

page 73

Table III-2-1 Accidental failure probability in space use by component types (quality assured level at class S)

(This is a C4 R many matrix. The only captions to be translated are):

C3R1: How many years before accident? or Accident every how many years?

C4R1: Occurrence rate of short cicuiting mode (for information only)

Note: FIT is the number of failures in 10 to the power of 9 years

end of page 72

Page 73

Table IV-2-1 Measures to be taken for the cause candidates which this time have been rated low (With Nozomi measures have been taken)

(This is an easy C2R6 matrix.)

C1R1: Cause candidates
C2R1: Measures to be taken

C1R2: total dose
C2R2: choose components which are most approapriate for the environment in use

*: initial sign of total dose is a small increase in current and this itself will not easily lead to "lost functions". We must therefore ensure that current limitters will not react prematurely.

C1R3: single event upset (SEU)
C2R3:

1. regular patroling and provision of refreshening function
2. triple redundancy for important registers
3. robust control logic (eg. automatic discovery of imvalid data)

C1R4: latch up
C2R4: choose radiation hardened components

C1R5: charging up
C2R5: Get rid of conductive layers which are not earthed

C1R6: high voltage discharge
C2R6:

1. isolation of vulnerable systems (eg. surrounding by secondary ground, primary and secondary isolation)
2. installation of imprinted voltage variables

end of page 73

P

[Astro0]

That's a great story about the 16-bit coding
Thanks again for all the translation work this year.
A very Merry Christmas to you and your family.

[pandaneko]

http://www.mext.go.jp/b_menu/shingi/uchuu/...ts/04061101.pdf

above for ease of reference

page 75

Table IV-2-2 Means for seperating out failure causes and their characteristics

(This is a C3R6 regular matrix inluding the headers in row 1)

C1R1: Failure cause seperation means
C2R1: Meritts
C3R1: Demeritts

C1R2: Resistances
C1R3: Fuses
C1R4: Relays and limtter circuits
C1R5: FET switches and limitter circuits
C1R6: Ideal redundancies

C2R2: system is easy and repetitive activations are possible.

C2R3: Setting up is easy.

C2R4: Repetitive activations are possible.

also, cancelling of latching up is possible.

There is a possibility to save components from temporary shortciruiting.

C2R5: Repetitive activation is possible.

There is no limit to the number of activation.

also, cancelling of latching up is possible. There is a possibility to save components from temporary shortciruiting.

It is very easy to set the system to OFF side without fail at the time of power on.

C2R6: It is possible to accept, at least once and perfectly as well, every possible failure mode.

C3R2: Need heat resistance at times of short circuiting.

Given voltage drop we may find it difficutlt to use this as "load current" may fluctuate.

C3R3: Once activated it will remain in the same state forever.

We need to check anti-vivration characteristics at launch times.

Action possible region is generally fairly narrow in that it will function without being affected by a sudden surge etc, with a current which will not affect other devices.

C3R4: Composition is very complex.

Relaying system itself needs watching out for failures and there is a limitation on the number of possible actions that can be taken.

If the relay system is of a "latch type" it may not improve the situation if:

1. there is another breaker downstream and
2. response speed upstream is slow

C3R5: Composition is very complex.

We need to allow for:

1. some extent of voltage drop
2. some extent of heat generation

C3R6: Impact on heavy resources is largest (I have no idea what they are talking about, P)

If we want to introduce cross-redundancies it will lead to the system getting very complicated and we will have to be extremely careful in design and verification.

We will need to evaluate its usefulness against the failure rate of similar redundant systems if:

1. the vulnerability against failure is very localised and/or

2. if the failure rate in question is very low in the first place.

end of page 75

P

[pandaneko]

http://www.mext.go.jp/b_menu/shingi/uchuu/...ts/04061101.pdf

above for ease of reference

page 76

Table IV-2-3 Concrete examples of failure seperation methods

(Inside the top horizontally oblong square says):

Large characters and those enclosed inside ovals are the points of improvements made with Nozomi

(However, I do not find any large characters on this page and next. There are, however, characters shown in red toegther with lines in red as you will see)

(There are two system outlines on this and next page, one on left and the other on right)

(One on left here says): Nozomi's system outline : current

(the only other characters for translation here is inside the top of 3 boxes next to TCI and it says): "Pressure monitor".

(the other one on right here says): Nozomi's system outline with improvement plan (1): Protective resisters in case of Black Box failure.

(the only other characters for translation here is inside the top of 3 boxes next to TCI and it says): "Pressure monitor".

end of page 76

Page 77 (system outline continuued)

(One on left here says): Nozomi's system outline with improvement plan (2): Protective resisters in addtion to switching between CI and PSU ( as indicated by vertically long square in red) + 1 kg in weight ( and the characters in this square says):

"Power control motherboard (within TCI)"

(the only other characters for translation here is inside the top of 3 boxes next to TCI and it says): "Pressure monitor".

(the other one on right here says): Nozomi's system outline with improvement plan (3): Protective resisters in addition to individual power source for every device (+ 3 kg in weight)

(characters in the square above LVDT next to TCI-2 box in red says): "Pressure monitor".

end of page 77

P

[pandaneko]

end of page 77

I now realise that I have effectively come to the end of translation work for this particular document. There are some more pages including pages 79 to 81 which are glossary pages in English and Japanese and very useful for any future correction work.

However, just glancing at these pages I note with satisfaction that I have not made fatal mistakes in the choice of my wordings.

Other pages are irrelevant to the story of this failure and need not be translated.

For now I may take up to 10 days of break in view of the inevitable events coming up over the next 2 weeks or so. However, it does not mean that I will not be able to grab time to do what remains to be done even during this period. It is just inpredictable.

When I resume the first document will be the JAXA press release summing up the causes of failure.

P

[pandaneko]

When I resume the first document will be the JAXA press release summing up the causes of failure.

What follows is the press release. I will provide link info and any other info later. Also, I am not exactly sure if this will be useful and I may not be able to complete translation of the whole of this release this evening and in which case I will continue tommorrow on with the rest. Here, we go as follows.

About abandoning insertion of Nozomi into Mars circular orbit

10 December 2003, JAXA

We reported on above subject to Space Activities Comission (SAC) held today as follows.

1. Current status

1. Mars probe Nozomi (launched July 1999) (fig. 1) developped a mulfunction in the fuel supply system at the time of leaving the earth gravitational fields (20 December 1999) (mulfunction history shown in seperate paper 1) and we had to change the original orbital plan and the arrival had to be changed from October 2000 to December 2003.


• 1st Japanese Mars probe launched by ISAS, carrying instruments from Sweden, Germany, US and Canada, also in cooperation with France.

• Launched by M-V 3 solid fuel rocket on 4 July 1998 from Uchinoura Space Observation Centre of ISAS in Kagoshima prfecture.


• Main objectives: Interaction between Solar winds and upper utmosphere of Mars

(Martian magnetosphere, atmosphere, plasma composition, satellites)

• Nozomi has been in its final approach into Mars since June of this year. EAT is 14 December.


Fig. 1 Outline of Mars probe Nozomi (Planet


2. A mulfunction developped in April 2002 in the comms. and thermal protection systems, resulting in minimum amount of communication in addition to incapability to perform temperature control. The mulfucntion here refers to thoes events as shown in Fig. 2 whereby part of the series of the circuits meant to supply power from common sources of power to components developped short circuiting.


Fig. 2 Nozomi power supply, schematic outline

3. May 2002 we tried to stablise probe temp. by turning on instruments. Also, we made a restoration work by trying to burn out the short circuiting sections by directing currents. However, this operation resulted in total loss of communiction capability.

4. July 2002 we kept trying to recover comms. for two months based on the findings of trouble shootings and achieved a minimum level of communication capability.

5. August 2002 fuel temp. reached that of fuel defreezing temp. This was due to the decreasing distance between the probe and the Sun and also the heat generated by the instruments on board. From here on we managed to control fuel temp. by keeping the right probe orientation.

6. June 2003 we managed to insert Nozomi, using the bare minimum communication means, into its final transer orbit.

7. July 2003 we needed precise orbital determination and firing of the main engine. This meant that we had to restore thermal contorl system and the work began as required. Unfortunately, this work yet again led to the total loss of comms.

8. From July 2003 to today (December) we have been trying to burn out the short circuiting sections by directing currents to there, and this meant that the number of times we turned on the common power source reached 1.3 times 10,000,000,000.

We also tried to rewrite the ROMs on board in order to exclude the possibility of the onboard computers going bad to no effects. Based on this we returned the ROMs to their initial values and continued with "continous ON" operation, but there is no prospect of recovery on the night of 9 December.


9. All this made us persuade that insertion into Mars orbit was no longer possible and we began the process of work required to make sure Nozomi will avoid collision into Mars on the night of 9 December in accordance with an international agreenment (collision probability is less than 0. 1% as required)

For your reference current orbital plan with insertion in mind is shown in Fig. 3 and that of fly-by is shown in Fig. 4. As you can see from Fig. 3 continuing with the current orbital plan with a fly-by will mean just a slight acceleration by Mars gravity and lead to a clser approach to Mars.


NOTE) COSPAR planetatry protection policy and moral obligation to observe its policy



Fig. 3 Nozoni's original orbit

Fig. 4 Nozomi's Mars orbiting plan and its fly-by orbit


(This will continue, not by very much, though)

P

[pandaneko]

(This will continue, not by very much, though)
P

What follows is the translation of the "seperate paper 1" mentioned in the last entry. I am afraid I cannot paste the URL yet of this report as I do not wish to loose what has been copied for this entry. I will do that tommorrow.

Seperate paper 1

About giving up the hope of inserting Nozomi into Mars circular orbit



History of Nozomi's mulfunctions. Mulfunction of this time

20 December 1998

Escape from Earth gravity to the mulfunction in which not enough propuslive power was achieved


Causes:

Too much fuel had to be used because the valve upstream of the oxidiser tank had not been fully opened (note 1), leading to much less propulsion than expected. The valve in question was what had been specifically employed in order not to allow reverse flow of the fuel and oxidiser vapours upstream as precaution given the US Mars Observer's trouble.


Measures taken:

It was discovered that it was not going to be neccessary to close the above valve in order to stop the reverse flow given the amount of remaining fuel after Earth escape operation. It was therefore decided that we will open the above valve during the visible operational period on 21st and keep it opened during the rest of our operation.

Influences:

We had to give up our originally planned Mars insertion operation due to take place in mid October 1999 and had to delay it until sometime between end December 2003 to early January 2004. (Subsequent orbit optimisation led to the time of arrival to be 14 December 2003.)


26 April 2002: Signal came in as a beacon signal (*1) (trouble at this time)


Causes:

Short circuiting had developped in parts of the secondary circuit of the common system power source (CI-PSU) and it is estimated that this was due to the high energy particles in the wake of the massive solar flares on 22nd (maximum as far as Nozomi was concerned).

Influences:

We were no longer able to send data from Nozomi to the ground because there was no power. In addition, the thermal control circuit was not active. (It was later discovered at the end of April that the fuel had been frozen at the time of 26th.)

Measures taken:

3 May 2002:

We turned on instruments one by one so that satellite temp. could be improved. Thermal analysis told us that the left-alone- attitude operation will lead to natural defreezing in September.

15 May 2002:

We lost beacon wave reception due to continuous sending of "ON commands" to the sections concerned.


• It is thought that the X-band transmitter's relay circuit caused a mulfunction (to OFF) due to the imperfect start-up of the power source for the ICs meant for command distribution. (Both ON and OFF commands were issued at the same time)

• Ground tests showed that different relays will behave differently and this presented the possibility of recovering the beacon communication by issuing a one-off command meant specifically for the power source in question.

Measures taken (before report acceptance?):

15 July 2002:

After some 7500 trials beacon communication was back.

End August 2002:

Fuel defreezing temp. reached -> after this we kept attitude control for this state so that fuel freezing will not occur again.

20 December 2002:

1st Earth swing-by

19 June 2003:

2nd Earth swing-by

Since 5 July 2003:

Short circuit section's burn out operation by keeping CI-PSU permanently ON. During this process total loss of beacon communication on 9 July.

From 2 October 2003 to around 20 October 2003:

Rewriting of the contents of the memory on board in order to exclude the possibility of the mulfunction of the DHU on board (*2)


From 23 October 2003:

Re-starting the operation for burning out the short-circuiting sections by continuosly issuing commands by CI-PSU.


NOTE 1:

Looking into the causes of the valve mulfunction we now know that it is almost certainly due to the increased resistance to the sliding motion at the sliding section due to incompatibility of materials used at that section.


*1

Beacon state: Signals are sent out by the satellite, but no data is carried on these waves.


*2

DHU： the most important computer on board, for evrything.


End of this press release

P