IPB

Welcome Guest ( Log In | Register )

7 Pages V  < 1 2 3 4 > »   
Reply to this topicStart new topic
Nozomi in perspective, Revisiting the causes of failure
pandaneko
post Nov 2 2011, 09:22 AM
Post #16


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

(I realise that I forgot to look at the earlier reference number. I will deal with it seperately.)

page 10

・After the termination of OME burn a close command was sent out following the sequence to liquid and gas system latching valves (LV2, LV5, and LV6) and the indicator showd "closed".

・After this a series of sequence including spin down and attitude change was excecuted without any problem.

2) Mulfunction situation

From these points above the mechanism of insufficient delta V is estimated to be as follows.


① Despite the fact that LV2's indicator is showing open status not enough pusher gas is being supplied to the oxidiser tank. From the relationship between the estimated value of empty tank volume change and and pressure change it is estimated that the supplied amount was 2% of the LV2 complete close case.


②Oxidiser tank pressure decreased.


③ Not enough oxidiser was supplied to OME, leading to an inefficient burn, resulting in OME propulsion drop.


④ Because of this unexpected propulsion drop OME firing was terminated as the maximum burn time was reached before attaining the planned delta V.


3) Where faults took place

Insufficient delta V during TMI was due to the decrease in the capacity to supply pusher gas to the oxidiser tank. Possible cause of this decrease is thought to be hitches relating to the check valve CV2 and the latching valve LV2 in the propulsion system. In this particular case it has been shown that this insufficiently opened passway occurred not with CV2 but with LV2 from the following reasons.

① The pressure value returned to normal as a command was sent out from the ground to open LV2.


② The output of the acceleration meter indicated that there was a shock as the LV2 valve opened and also there was a vibration as the gas flew as a result.


(4) Operation after this and the status of LV2


1) Evaluation of the status of the propulsion system

In order to make up for the insufficient delta V during TMI two additional corrective OME burns (TMI_C1, TMI_C2) were conducted and these were carried out normally. We also carried out an evaluation test on the OME propulsion and supply system in order to check on approapriateness of the corrective burns and subsequent health of the propulsion system and we are satisfied that propulsion system returned to normal.

a) Evaluation of OME propulsive power

We obtained, from the acceleration of OME burn and tank pressures data, OME propulsive power and its specific impulse as well as the amount of propellant flow. These characteristics were then sorted out according to the regulator pressure (P2). The result showed that during the subsystem burn tests (SFT) and during the early stage after the launch TMI_C1 and TMI_C2 both showed almost similar characteristics, meaning that there was no trouble with the OME burns. (Schematic II-1-8)

end of page 10

(actually, part of the very last paragraph spills into page 11, P)
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 3 2011, 09:36 AM
Post #17


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

for your information the reference for the (2) on page 9 is (schematic II-1-4)


page 11

cool.gif Evaluation of OME supply system

We examined, in terms of OME functioning time, the pressure loss ratio of the gas and liquid systems due to OME burn during SFT and after the launch. All other indicators except that for the oxidiser system during TMI showed almost similar behaviours and we did not find any trace of characteristics change due to material deterioration and comcluded that the supply system was healthy. (schematics II-1-9 and II-1-10)

2) Operation of LV2

With Nozomi's propulsion system LV2 and CV2 prevent the reverse flow of vapour from mixing with hydrazine and leading to explosion. For an explosion to actually take place the vapour must condense upstream and a certain minimum amount must stay there.

However, it was confirmed that there was enough electrical power available from the track record of operation and that it was possible to maintain, due to the relaxation of the allowable tank temperature range, the temperature of the valve module A (VM-A) (schematic II-1-1) 10 degrees higher than that of the NTO tank.

Therefore, we concluded that even if LV2 is open we will be able to be free from above trouble with this newly acquired temp control leading to a dual safety mechanism with the use of CV2.

For this reason, the operation thereafter was such that both LV1 and LV2 were kept open all the time so as not to cause LV mulfunctioning.


2. History etc. of selecting LV2 for Nozomi

As shown above it became clear from the analysis of the telemetry data that the oxidiser gas system's latching valve LV2 developped a mulfunction. This LV2 in question had been selected and passed the verification test as shown below.

(1) History of LV2 selection

1) Method of valve selection

Valves for space use are little produced domestically. Therefore, in the case of Nozomi valves that met neccessary specs were procured from overseas. With this particular valve there was know-how related subtle technical information relating to its design and manufacturing. Therefore, details of its fine structure were not available and it was also prohibited to disassemble the valve in Japan.

LV2 was procured from a US manufacturer with excellent track records in space use. It was converted by the manufacturer, at the request of the then Institute of Space and Astronautical Sceinces (ISAS) so that it conformed to the structure of Nozomi by adding a status monitor (LVDT).

(from here on the remaining text spills into page 12)

LV2 is based on the combination of two components which have substantial flight records. As mentioned above its delails are not available and what was done was to investigate the potential imaginable risks invloved upon selecting this particular valve. Detailed information of this investigation is shown in the table II-2-1.

end of page 11

P
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 4 2011, 09:50 AM
Post #18


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 12

2) Building in of LV2

We had debated about whether we should carry LV2 on Nozmi and our conclusion was that we should do so from the viewpoint of preventing a massive reverse flow of oxidiser vapour. LV2 is placed in the propulsive system at the downstream side of the regulator after the forking out of fuel and oxidiser pipes and between the check valve CV2 and the oxidiser tank (schematic II-1-1). Its function is to prevent the mixing of fuel and oxidiser at the gas side of the system.

With this particular propulsive system we are employing a dual redundancy system of accident prevention together with the CV2 placed in the upstream. This dual redundant system with LV2 and CV2 does not mean we did not have confidence in any of these valves. Rather, it was employed to increase reliability to an even higher stage against the reverse flow of vapour which might lead to a fatal accident, damaging the probe.

During the development stage of Nozomi in 1994 an accident occurred to a US Mars probe (Mars Observer) and it was lost. Its cause was estimated to be an accidental mixing of oxidiser and fuel vapours in the gas system pipeline. Reverse vapour flow is likely to lead to a total loss of a satellite and this point was also taken into consideration.


3) Adding LVDT

About whether we should install a valve status monitor LVDT with the LV2 our conclusion was that LVDT constituted an important source of information for the steady operation of LV2, reducing the pressure on the operating team and the possibility of human errors.

In the case of Nozomi, on the other hand, there was a factor of influence at the time of valve selection arising from the merger and acquisition of the US valve manufacturer in that it meant the addition of LVDT to a valve without a status monitor but with an ample proven record of operation.

Generally speaking the reliability evaluation of a design change to an operationally proven part has not been established just as in the case of newly developped parts. However, our judgement was that this was going to be based on an operationally proven valve and the design change was going to be introduced by a reliablbe US manufacturer and consequently the risks involved in this design change were small enough. For this reason we selected the LV2 with LVDT added to it.

As shown in the table II-2-1 the risk of valve mulfunction after the desing change was thought to be similar to that of operationally proven valves.

end of page 12

P

表Ⅱ-2-1に示すように、バルブ開閉ができなくなるリスクでは、改造品といえども、そのリスクは搭載実績品と同様であると判断されていた。
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 5 2011, 09:30 AM
Post #19


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 13


4) Structure of LV2

LV2 structural outline is shown by the schematic II-2-1. Inside the valve there are 2 electromagnets (one each on the right and left of the schematic) and they activate the inner piece (some kind of intermidiate piece, P).

いての独自の追加試験の検討
By letting an electrical current flow through the close side electromagnet (on the left) the inner piece is pushed in and the flow is stopped (state of the schematic) and by letting an electrical current flow through the open side electromagnet (on the right) the inner piece will leave the plug (or seperate from the plug, P) and in that state there occurs a small pressure difference between the up and down streams and this differential pressure pushes the plug and the flow will start.

Also, as shown on the schematic II-2-1, LVDT on the right will monitor the open/close status of the valve by detecting the position of the inner piece. Those original valves with track records of being used with satellites did not have this LVDT. Design change by adding this LVDT meant that the plug and the inner piece were seperate items.

This arrangement of having a seperate structure had been proposed by the US valve manufacturer in an answer to the concern that the addition of LVDT meant a longer inner piece and a resulting small positional erorr might influence the open/close operation of the valve.

(2) Contents of LV2 verification

1) Verification method for LV2

Starting from valves used for scientific satellites those valves used for space use are almost all imported from overseas. Consequently, verification of the selected valves are carried out by those overseas manufacturers and the Japanese procurers. In the case of Nozomi too, this verification was carried out by both US manufacturer and in Japan as shown below.

a)LV2 verification method

① Contents of verification requested by the Japanese side to the US valve manufacturer

・ Verification of design validity

・ Verification of manufacturing validity concerning the shipped flight items

・ Flight track records

Flight tracking records include, if possible and approapriate, concrete mission names and usage. We regarded the usage as important because if the past programmes' mission duration and environment had been different it might be difficult to count it as a proven track record.

② Contents of verification conducted by the Japanese side

・ Confirmation of the justification of the verification methods for each of the items which relate to the verification of the validity of design and manufacturing

(up until here, translation from page 13, but the circle 2 continues into page 14 as follows, P)

- 14 -

・ Carrying out a consolidated system test in the state in which the items were actually built into the satellite

・ Investigating into the possibility of our own additional tests with those items whose flight tracking records could not be confirmed and where details of the verification methods were not available.

end of page 13

P
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 6 2011, 09:03 AM
Post #20


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 14

cool.gif History of LV2 tests

Confirmation tests of LV2 were carried out in two major parts. One is the design quality confirmation test where another test piece different from the flight piece was fabricated and offered to testing. Second is the confirmation test for flight model manufacturing approapriateness, which is conducted on the flight model. Their outlines are shown below and the ground test history of LV2 is carried on the table II-2-2.


① Design quality confirmation test

This is the test whereby design validity is confirmed. In addition to the test conducted by the US valve manafacturer another test was carried out in Japan.

The valve manufacturer conducted their own test using two test pieces, LV2 and a similarly designed LV1 together with HLV. In Japan we used a spare part, common to both LV1 and LV2, and conducted a confirmation test on the adaptability to the oxidiser (NTO) environment.

We had been given a report from the manufacturer that the valve in question had resistance to NTO and ours was carried out seperately to confirm this report using the flight piece of the valve. The number of valve actions during this test is shown on the table II-2-3.

② Confirmation test for the manufacturing validity of the flight model

This was the test conducted on the flight model. Confirmation was sought with the part built into the satellite system for its health. Below is the number of delivered pieces.

・LV1、LV2 : one each
・LV1、LV2 common piece: two
・HLV: one and HLV spare part: one

2) Action history of LV2

The number of counted valving actions during TMI with the valve in question was 42nd since the day of the delivery and 6th from the day of launch. It was confirmed that all valving actions prior to that had been normal. LV2 action history is shown on the table II-2-4.

end of page 14

P
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 9 2011, 09:23 AM
Post #21


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 15

3.Estimating the causes of mulfunction

(1) Fault Tree Analysis (FTA)

We conducted an FTA with the following events at tree top in order to estimate where LV2 mulfunction occurred.

・ A command had been sent from the ground for opening of LV2 and LVDT did show that LV2 was open. However, in reality the opening was minimal and the flow rate was very small. The result of this FTA is shown on the table II-2-5. We at JAXA did exchange information with the US valve manufacturer after the accident. However, as mentioned earlier, the information relating to the valve structure etc was limited in availability.

We summarised the possible mulfunction candidates based on the FTA analysis as shown under (2) below.


(2) Possible mulfunction candidates

1) Bad sliding motion of the plug

One strtucural problem with LV2 is that the plug and the inner piece are separated physically. In addition, there is a possibility of the material surface of the sliding component inside the valve suffered from a fletching wear due to the valve motion and this led to the issue of valve material's compatibility with the oxidiser.

It is thought possible that these two factors might have led to bad sliding motion of the valve.


(note 6) : fletching wear

surface damage arising from repeated slidings at the sliding area


a) Valve structure


We did conduct a repeat evaluation on the past flight records and had confirmation that valves with this particular material had flown many times in the past. However, these valves had a monostructure of the plug and the inner piece and the valve was forced to activate with an electromagnet.


With respect to the valve that we used its (valve) open mode was similar to that of so called check valves in that it opens passively when there is a diffential pressure before and after the plug. The driving force arising from this differential pressure was smaller compared with the force originally ensured by an electromagnet and the valve could have suffered if the plug's sliding force resistance (note 7) had increased.


(note 7) : sliding force resistance (please note that this sentence astrides pages 15 and 16)

The resistance against the sliding motion involving two objects in contact. If this force becomes large

- 16 -

then objects become harder to slide against each other, requiring a larger force for sliding.

end of page 15

P
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 10 2011, 09:51 AM
Post #22


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 16

cool.gif oxidiser resistance of the material

Material combination of the sliding parts is austenite (spelling unsure, P) stainless steel at the sliding area of the plug, and ferrite staeinless steel for the valve body. Ferrite stainless is magnetic by nature and is used extensively with electromagnetically activated valves. On the other hand it is slightly less resistant against NTO compared with non magnetic austenite stainles steel.

It is thought possible that fletching wears would have led to corrosion of the valve body area in question after being filled with NTO, itself further leading to an increase in the sliding friction.

2) Gritting of the plug into the opening area

It is thought possible that the plug which had been pressed against the opening area for a long time might have gritted into the opening, making the plug less mobile. It is also thought that from the status of the lowering of the oxidiser tank pressure a small amount of helium gas was still being supplied to the oxidiser tank and that in the event of plug gritting this might have led to , with a high possibility, the complete closure of the sealing area.

With respect to LV2, the duration in which it was kept closed immediately prior to TMI operation was not especially long compared with similar operational duration with other launches. For that reason it is thought that LV2 did not suffere from the same trouble. For your information the history of LV2 use in operation is shown on the table II-2-6.


3) Plug mis-allignment

The valve in question had been subjected to Quality Test (QT) and undergone more than 1,000 times of open/close operation. All this was monitored using LVDT monitor and also pressurised heilum gas flow confirmed the valving actions. Also, quantitative flow volume tests had been put into action before and after this open/close test and it was confirmed that everything was normal as designed.

For these reasons we consider that there was no design problem with the clearance between the valve body and the plug.

In order for this particular mulfunction to take place we could suspect, as its cause, bad manufacturing of the valve. However, the valve in question operated normally 42 times before mulfunction including the period immediately after launch and there was no sign of bad manufacturing.

Furthermore, the valve continued to function normally even after the occurrence of the mulfunction and we believe that the valve is not a culprit.


4) Glitching by foreign pieces

The diamterwise clearance between the plug and the valve supporting body is such that if foreign pieces larger than the gap migrate into

栓とバルブ本体の直径クリアランスは、より大きい異物がバルブ内に入り込むと噛

- 17 -

the area inside the valve they may cause gritting there. However, the pre-launch gas filling was done using a filter whose mesh is smaller than the clearance dimension and it is thought unlikely that there were foreign pieces of that size present inside the piping system and that foreign pieces were the culprit.

Also, generally speaking, once gritting takes place it is rare for the system to resume normal operation immediately. However, in this case we are looking at the function of LV2 returned to normal after the hitch and consequently it is thought highly unlikely that this phenomenon happened to LV2.

end of page 16

Go to the top of the page
 
+Quote Post
pandaneko
post Nov 11 2011, 09:09 AM
Post #23


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 17

5) Temporary immobility (or solidification) by material growth (or some kind of crystal formation?)

If there had been residual water inside the propulsion system it could have been possible that the reaction of fuel and oxidiser led to the formation of ammonium nitrate. However, this possibility has been discarded as extremely low for the following reasons.


・ dryness at the time of propellant filling was very good (condensation temperature being minus 55 degrees C or below)


・ Had there been a mixing with fuel the check valve upstream would have caused mulfunction


・ It would have shown up in a short span of time (a few days)


(3) Result of estimating for the causes of mulfunction

If we are to summarise what we have been talking about so far, we think that the causes of LV2 mulfunction are due to, as shown in 1) of (2), the fact that the valve was susceptive to increased sliding friction of the plug given that the valving was powered by the differential pressure by the seperate inner piece and the plug

and also the fact that the sliding surface caused a fletching wear leading to corrosion by the oxidiser environment, further leading to an inceased sliding friction.

That is to say that the multiplier effect by these two factors was the most likely cause of the failure.

end of page 17

P
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 12 2011, 09:38 AM
Post #24


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 18

Ⅲ.About the mulfunctions of comms. and thermal control systems

1. Environment in which mulfunctions took place

(1) Outline of the power source system

Unlike all other scientific satellites which had adopted a centralised power source Nozomi was the first satellite to use a decentralaised power source. Within the Nozomi power system there were 15 power sources and one of them was the Common Instrument Power Supply Unit (CI-PSU) which was used to supply power to a multiple number of commonly used systems (equivalent to power source 2 in fig. III-1-1).

This CI-PSU receives power from solar cells and batteries placed on the primary side and supply power to 10 secondary side subsystems such as the telemetry command interface (TCI) for the X-band transmitter (TMX) and the heat control circuit (HCE).

For your information, the structure of CI-PSU is shown by the schematic III-1-2 and the table III-1-1.

(2) About the operation from the time of losing signals until the day of mulfunction taking place

The mulfunction was first detected during the operation on 25 April 2002 and the location of Nozomi at that time is shown by the schematic III-1-3. Also, the operational sequence from a day earlier (24 April 2002) prior to losing signals until the day of detecting the mulfunction is shown on the schematic III-1-4. This operational sequence tells us the following story.


① At 18:05 on 25 April 2002 (UTC) signals came in in the beacon mode despite the expectation that the transmission mode had changed to the telemetry mode (note 8). We then sent a command from the ground so that the mode changed to the telemetry mode, but this was not successful. From all this it is obvious that the status remaining was such that we could not switch the mode.

(note 8) : telemetry mode

mode in which telemetry information from the satellite is carried on the satellite wave

(note 9) : beacon mode

satellite is emitting waves, but these waves are not carrying telemetry information.


② We had anticipated an attitude change (about one degree) at 09:00 on 25 April (UTC). However, from the reception level of the satellite waves on 25 April it is estimated that this attitude change did not take place (fig.19-III-1-5). It is also estimated that the command sent from the ground for the attitude change did not succeed.

From all above it is estimated that the attitude change remained impossible.

end of page 18

P

Go to the top of the page
 
+Quote Post
pandaneko
post Nov 13 2011, 09:04 AM
Post #25


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 19

(3) Space environment at the time of mulfunction being detected

A solar flare of class X occurred on the western area of the Sun's surface on 21 April 2002 with a magnitude of X 1.5 (X-ray flux of 0.15mW/m in the vicinity of the earth). Nozomi received a direct hit of high energy particles associated with this flare on 22 April. As the strength of high energy particles it was the largest since the launch of Nozomi and it is estimated that it was one magnitude higher in energy than anything that Nozomi had experienced earlier. This magnitude is estimated to be such that you normally expect this magnitude in interplanetary space only once or twice during any one solar period (11 years).

For your information the temporal variation of solar proton monitoring by the device on board Nozomi is shown on the graph III-1-6. Also, the high energy particles strength variation experienced by Nozomi since 2000 is shown on the graph III-1-7.


(4) "One bit comms." and grasping of the phenomenon by the autonomous function etc.

It became possible to obtain the data neccessary for the operation of Nozomi (table III-1-2) by establishing a measurement method using Nozomi's autonomous function (note 10). For instance, if you wish to measure the temperature of the inside of the satellite, the beacon is turned either on or off, depending on whether the temp. is above or below the temp. set by the command sent from the ground. (1 bit comms, hereafter) (grapgh III-1-8)

As a result, it became clear that CI-PSU remained in the state of being off. Also, since the possibility of the primary circuit going wrong is extremely low it is estimated that the protection circuit against excessive current flow resisted the CI-PSU from becoming on.

Under these circumstances we expect following phenomena.


① Since TCI is in the off mode a command cannot be issued to TMX. For this reason, it remained impossible to switch between beacon and telemetry modes.


② Since the heat control circuit (HCE) is in the off mode all the heaters inside the satellite controlled by this device are also in the off mode. For this reason, the temp. of the inside of the satellite decreases rapidly to that temp. which is determined by the heat generated by the satellite, sun angle, and the distance to the sun.

Consequently, as shown by the data obtained (table III-1-2) the inner temp. of the satellite decreased below the freezing point of the fuel and attitude control became impossible.

(note 10) : autonomous function

the function which issues a particular set of commands which have been programmed in advance depending on the inside status of the satellite

- 20 -

(and in this instance, we obtained information by making the satellite execute the command which will turn off the power of the power amplifier of the X-band transmitter).

end of page 19

P
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 13 2011, 09:14 AM
Post #26


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


With above file I seem to be translating some entries which must be rather obvious to UMSF members. Please put up with this as these reports are meant for the Space Activities Commision (SAC) whose members include usually one civilian in order to reflect the opinions of the lay community. One such member as I recall from a few years ago was a newspaper cartoonist.

P
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 14 2011, 09:04 AM
Post #27


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 20

(5) Grasping events by recovery operation after the mulfunction

Grasping of satellite status, information gathering for the causes of mulfunction, and satellite operation after the mulfunction was carried out as follows.

1) Confirmation of the health of data handling unit (DHU) and command decorder (CMD)

A command was sent from the ground on 28 April 2002 and we succeeded in powering on and off of the X-band transmitter power amplifier (XPA)(XPA is normally ON). From this we concluded that DHU and CMD were operable normally.

2) Grasping probe status by "1 bit comms." using the autonomous function

During early part of May 2002 we managed to obtain probe status through "1 bit comms." using the autonomous function. As will be shown in (4) we discovered that the common power supply (CI-PSU) was in the state of being OFF and also it could not be made to switch into ON mode and that the satellite temp. was far below the freezing point of the fuel on board.


3 Loss of the beacon waves due to the recovery operation of the short circuited portion


This was the operation on 15 May 2002. By sending continuous ON commands to CI-PSU (about 100 times) we tried to burn out the faulty line, but we ended up in losing the beacon waves during this process. We carried out ground tests in order to find out why beacon waves were lost and we estimated that the beacon was made into OFF state because when TCI is started up it also meant that TMX's ON and OFF commands are simultaneously issued leading to erroneous action of TMX relay circuit.

It was also felt possible that the wrong command issued by TCI (such as TMX ON/OFF) in response to the ON command to CI-PSU might lead to TMX going back to ON mode again. Thus, we carried out beacon recovery operation by sending out a series of single ON commands.

4) Recovery of beacon waves

Beacon waves were recovered on 15 July 2002 after sending to CI-PSU a series of single ON commands (about 7,500 times). This confirmed our estimate shown in (3) above. It also became clear that the satellite power (primary power supplied to CI-PSU) was healthy and that CI-PSU were able to supply secondary power at least for a short period of time.

For this reason, it was thought likely that CI-PSU itself was healthy and that the power OFF state was caused by


- 21 -

the protection circuit against excessive current arising from the short circuiting problem within the secondary devices connected to CI-PSU.

end of page 20

P
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 15 2011, 09:07 AM
Post #28


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 21

5) Defreezing of the fuel tank and RCS thrusters

By the end of August 2002 thanks to the self generation of heat by on-board devices, distance to the sun, and improvements in the satellite attitude part of the propulsion system had been defrozen and in the upper part of September of the same year it became possible to control attitude and try small scale orbital control.

It has been confirmed since then that maintaining of the correct attitude helped to keep the fuel tank, attitude and orbital control thrusters from freezing. Owing to this recovery of attitude and orbital control thrusters we managed to succeed in the 1st Earth swingby operation on 20 December 2002, and also in the 2nd Earth swingby on 19 June 2003.

However, it was also confirmd that recovery of heater function was absolutely vital in defreezing the main thruster (required for orbital insertion around Mars) which was always on the shade side.


6) Recovery operation (operation for recovery of , P) CI-PSU and heater control function

Given that for orbital insertion it was vital to have CI-PSU and heating function back to normal we checked through the gournd tests that a continuous and rapidly-issued series of commands for CI-PSU recovery will not lead to action anomally by CI-PSU.

Based on this we started on 5 July 2003 to issue continuous ON-commands for CI-PSU so as to burn out the troublesome short circuited line on the secondary side. During this operatioh we managed to lose beacon waves. This recovery operation continued until 9 December 2003 without success. We therefore gave up the hope for orbital insertion.

7) Items confirmed through this recovery operation

Following the recovery operation explained as from 1) to 6) above we were able to confirm:

①DHUand CMD were operational normally.

②CI-PSU cannot be made to be ON.

③We cannot switch between telemetry and beacon modes.

④Loss of heating function led to fuel freezing and attitude control could not be achieved.

⑤Primary power supply voltage to CI-PSU was normal.

⑥CI-PSU can provide secondary voltage for a short time only.

From these reasons it was thought that CI-PSU was likely to be healthy and that with a high probability the short circuiting on the secondary device side led to the activation of the over-current protection mechanism, which in turn led to the OFF state of the power supply.

end of page 21

P
Go to the top of the page
 
+Quote Post
Paolo
post Nov 15 2011, 06:21 PM
Post #29


Senior Member
****

Group: Members
Posts: 1729
Joined: 3-August 06
From: 43° 35' 53" N 1° 26' 35" E
Member No.: 1004



thanks again for the translations, panda! I knew that they had regained trajectory control but I didn't know that the main engine remained unusable
Go to the top of the page
 
+Quote Post
pandaneko
post Nov 16 2011, 09:49 AM
Post #30


Member
***

Group: Members
Posts: 817
Joined: 17-April 10
From: Kamakura, Japan
Member No.: 5323



QUOTE (pandaneko @ Oct 23 2011, 06:12 PM) *


above for ease of reference

page 22

2.Estimated causes of mulfunction

(1) Fault Tree Analysis (FTA)

1) We conducted a fault tree analysis (FTA) with the fact at the top of the tree that we received no waves despite being in the telemetry mode (TLM_ON). The result of this particular FTA is shown on the schematic (or graph) III-2-1.


① As for the possibility of data processing unit mulfunctioning we discarded it because command(s?) were issued normally and we were able to obtain data.


② As for the possibility of TMX mulfunctioning we discarded it because it fails to explain why we could not change attitudes and also because there was no abnormal current consumption by TMX.


③The fact that TCI power was OFF, that is to say we could not keep CI-PSU to be ON can explain without contradiction why we could not change attitudes. It also explains why we could not swtich between beacon and telemetry modes.


2) We conducted another FTA, following one of the the results of above FTA, that is to say ③of 1) above, with the fact at the top of the tree that we could not keep CI-PSU power to be ON all the time. The result of this particular FTA is shown on the schematic (or graph) III-2-2.

There are three causes we could think of as follows.


① There was an anomally in the function whereby CI-PSU inside the satellite power source (PCU) is turned ON/OFF and this resulted in power not being available to CI-PSU.


② There was a mulfunction inside the primary side of CI-PSU and this meant that although the power was being supplied to CI-PSU the overcurrent protection mechanism worked and CI-PSU could not be brought to be ON.


③ There was a short circuiting inside the secondary side of CI-PSU and although the function of CI-PSU was normal the fact that there was an excessive current flowing downstream meant that protection mechanism activated within a few milliseconds, bringing the CI-PSU to be OFF.


3) Discussions about the result of above FTAs mentioned in section 2)

These are as follows.



As for ① of 2) above this was thought to be unthinkable because all other functions inside PCU were

- 23 -

normal.

(this particular paragraph continues into page 24 and is rather lengthy for it being translated as part of page 23. So, I stop here and will continue immediately after this as page 24)

end of page 23

P
Go to the top of the page
 
+Quote Post

7 Pages V  < 1 2 3 4 > » 
Reply to this topicStart new topic

 



RSS Lo-Fi Version Time is now: 29th March 2024 - 09:24 AM
RULES AND GUIDELINES
Please read the Forum Rules and Guidelines before posting.

IMAGE COPYRIGHT
Images posted on UnmannedSpaceflight.com may be copyrighted. Do not reproduce without permission. Read here for further information on space images and copyright.

OPINIONS AND MODERATION
Opinions expressed on UnmannedSpaceflight.com are those of the individual posters and do not necessarily reflect the opinions of UnmannedSpaceflight.com or The Planetary Society. The all-volunteer UnmannedSpaceflight.com moderation team is wholly independent of The Planetary Society. The Planetary Society has no influence over decisions made by the UnmannedSpaceflight.com moderators.
SUPPORT THE FORUM
Unmannedspaceflight.com is funded by the Planetary Society. Please consider supporting our work and many other projects by donating to the Society or becoming a member.