(page 51)

B.8 Examples of gas supply system in long duration flight with 2 liquid propellant system

(please refer to original schematics for this page"

(a) Hayabusa

Stainless diaphragm is installed so that propellant is supplied to the engine without fail on touch-down. As a result we could prevent the flow of oxidiser vapour in the piping system.

(with the piping schematic for Hayabusa top left character string is "fuel system" and top right "oxidiser system", corresponding lower ones are fuel tank and oxidiser tank) (String with an arrow is "stainless diaphragm"

( HTV

Manned mission requirement meant that valves are installed in cascade and a buffer tank is also installed in order to alleviate the pressure rise in the gas supply piping system. As a result, the flow of fuel vapour in the system is prevented.

(schematic captions are similar here, and an arrow starts from "buffer tank", also valves 1 and 2, 3, and 4 are indicated below buffer tank, above fuel tank.)

(end of page 51)

(page 52)

B.8 Examples ofgas supply system in long duration flight with 2 liquid propulsion system (continuation)

© Cassini (NASA Saturn space probe)

Source ：T. J. Barber, R. T. Cowley, “Initial Cassini Propulsion System IN- Flight Characterization”, AIAA 2002-4152

Many pyro-valves are installed in order to prevent oxidiser vapour flow in the piping system


(d) Messenger (NASA Mercury space probe)

Source ： San Wiley, Katie Domer,“Design and Development of the Messenger Propulsion Sytem”, AIAA 2003-5078

Separate piping is arranged for fuel and oxidiser distribution in order to prevent propellant vapours migrating in the piping system.

(end of page 52)

(page 53)

B.8 Examples of gas supply piping system in long duration flight using 2 liquid propellant system (continuation)

(e) NEAR (NASA asteroid space probe)


Source ： S. Wiley, G. Herbert, L.Mosner,“Design and Development of the NEAR Propulsion Sytem”, AIAA 95-2977

CV- and V- valves are arranged so as to prevent flow of oxidiser vapour in the piping system

(f) Mars Observer (NASA Mars space probe)

Source ： Carl S. Guernsey, “Propulsion Lessons Learned from the Loss of Mars Observer”, AIAA 2001-3630

CV-valves and pyro-valves are placed to prevent oxidiser vapour flow in the piping system.

Most convincing cause of failure is said to be that oxidiser at the cold portion of the gas supply system caused condensation and liquefaction and this flew into the fuel side when the pyro-valve was opened, leading to explosive reaction.

(end of page 53)

(page 54, main text portion only)

C.1 Current understanding of OME status analysis

Phenomena inside a burner are complex and there are many models. However, we need further efforts in trying to model "cooling" before we can quantitatively discuss absolute values. Naturally, analysis of stress distribution and destruction probability is useful in discussing burner's structural strength and heat resistance. However, even then, we need further efforts before we can discuss values of quantitave and non-quantitative aspects.

We have been following two research themes as per below as our long term projects even before Akatsuki was designed. We are hoping that the burn data we obtained thanks to the failure of Akatsuki outside the designed range of parameters will be useful in increasing the accuracy of our burn analysis model.

Burn analysis

We analyse steady state of burn inside a burner. At this point in time we are satisfied that our model can reproduce test results more or less faithfully by giving it a film cooling (FC) vanishing point. Our next long term projects will be about modelling FC and non-steady state analysis.

Strength analysis

Here, we try to evaluate the destruction probability by analysing the thermal stress inside the burner using burner temp. distiribution obtained in tests burns and burn analysis.

(end of text portion of page 54)

(schematics' translation will have to follow tommorrow as I am not exactly sure how to translate this at the moment. Some characters, even blown up, are illegible, P)

(page 54, schematics translation, or rather my understanding of them, only)

(there are two parts, I think, top and bottom and the top schematic seems to be a CG from CFD simulation and there are 6 boxed characters, boxes 1.2. and 3 from left to right above boxes 4.5. and 6 in the same order. Each box is aparently is a model and the character for it is placed at the bottom. There are other characters relating to the model name hovering above it, which I will place after the model name)

box 1: secondary microglobular model- secondary crush, microglobules
box 2: vapourisation model- N2H4 gas, N2H4 liquid drop, N2O4 gas, N2O4 liquid drop
box 3: thermal analysis model:
box 4: primary globular model- N2H4 injector mouth, N2O4 injector mouth, N2O4 liquid drop, N2H4 liquid drop, and an additional character with this box reads "virtual liquid membrane fan" with an arrow

box 5: film cooling liquid membrane vanishing point model (formation of formulae for experiments)
box 6: burn or combustion model- burn gas, N2H4 gas, N2O4 gas, vapour phase burn

(here above, I must say I do not know what they are talking about, P)

(Bottom half is as follows)

box 1: measured temp. distribution during test burn (there is a photo just underneath it)
box 2: thermal analysis (arrow is pointed to this from above)
box 3: stress analysis (arrow is pointed to this from above)

(after this, there is a strange looking structure and just underneath it)

box 4: stress distribution (no arrow as such)
box 5: destruction probability evaluation (and this box is pointed at with an arrow)

(after this there are 2 circled numbers + a note)

1st circled number: evaluating in unit of elementary volume (volume standard destruction probability)
2nd circled number: evaluating in unit of elemental area (area standard destruction probability)

NOTE: strength depends on surface processing methods

(end of schematics on page 54)

(page 55)

C.2 Observation of breakage of the broken burner

Here, we show the results of our surface observation of the burner which was destroyed during test burn (part 1). As a result, we confirmed the existence of the breakage starting point and did not find any clear material fault around it, indicating that the breakage was not due to a simple manufacturing error.

(there are 4 photos which are numbered as follows)

a) low resolution photo
material surface of the observed spot
c) blow up of the region around starting point
d) blow up of the region around starting point

(end of page 55)

(page 56)

C.3 History of accerleration and angular velocity during the latter half of VOI-1

Here, we looked at the effects suffered by OME.

In order to compare test burn data with that from the 1st investigation meeting we carry the latter as follows.

(no need for translation, except perhaps that; )

left vertical axis is angular velocity (deg/s) and right vertical is acceleration (m/s/s), horizontal is the time line relative to OME start taken at zero second.

On the graph itself, blue is angular velocity around X axis, red Y axis, yellow around Z axis. Solid line is accerleration.

(end of page 56)

(page 57)

C.4 Propulsion characteristics of the broken burner

Here, we compared the estimated (from acceleration telemetry data) propulsive power with that measued during the test burn (part 1) after thruster nozzle breakage.

Behaviour during VOI-1 (telemetry data);

• 152 seconds after the start of burn a step like power fall is observed, and after that 2 more step like power changes are observed.

Behaviour during test burn (part 1)；

• In this test burn we also observed a step like power drop as was observed at the time of breakage during VOI-1

• In other examples of breakage there was a case where cracks opened up along the circumferential direction. If burn were to continue with this kind of breakage it might develop further, suggestive of the power change observed during VOI-1

(after this there are 2 graphs and one photo)

graph 1: VOI-1 propulsion (estimated from acceleration) (vertical is power (N) and horizontal is time line as with page 56)

(there is a character string astriding both graph 1 and graph 2 (both lines pointed at with an arrow from this string) and it says;)

propulsion after VOI-1 step change and that of test burn upon breakage are more or less in agreenment

graph 2: Shorter nozzled burner was used in test burn (part 1), leading to smaller steady propulsion (time line is real)

(caption pointing to the vertical precipitation reads;)

emergency stop was enforced in ground test

photo: Crack example at nozzle breakage (almost half the circumference) (fluorescent immersion defect search with another broken burner, broken in another test burn)

Therefore, we can explain the behaviour of power after 152 seconds of VOI-1 start as resulting from burner breakage.

(end of page 57)

(page 58)

C.4 Propulsion characteristics of the damaged burner (continuation)

Here below, we show photos of the flight model burner and the burner which was damaged during part 1 of our test burn and the comparison of propulsion characteristics before and after the damage.

As for the lateral propulsion we conducted computer fluid dynamics (CFD) estimates using the shape data of the damaged burner. The result was more or less in unison with the lateral propulsion estimated from the telemetry data of the X-axis angular velocity measured at the time of mulfunction during VOI-1.

(here after, there is a computer output, entitled;)

Estimation of the lateral propulsion of the damaged burner (example of CFD results)

1st display: vertical is pressure (Pa) and pressure distribution is horizontal
2nd display: vertical is Mach number and horizontal is Mach number distribution
3rd display: vertical flow (m/s) speed and horizontal is flow vector


photo 1: that of the burner that actually flew
photo 2: that of the burner which was damaged during part 1 of the test burns


Propulsion characteristics of the flight model and damaged burners

(here, I use my table concept, 4 rows and 5 columns)

R1C2 and 3 combined: Flight model burner
R1C4 and 5 combined: Burner after the damage during test burn (part 1)

R2C2: At the time of VOI-1 start
R2C3: after 152 seconds of OME burn in VOI-1
R2C4: test burn data
R2C5: estimation from CFD

(Here below, I attempt to create the rest of the table, in the same order, vertical and horizontal, and I need to put a slash between entries, I think)

Propulsive power (N): 476 / 300 / 315 / 307


Lateral propulsive power (N): 0 / 5〜20 / （no available data） / 14

(end of page 58)

(page 59)

C.5 Looking at reducing re-ignition impacts

We are currently conducting re-ignition experiments using a damaged burner with a view to igniting OME once again

(here, two photos sandwiching a larger arrow)

Left photo: burner after nozzle damage

Larger arrow reads: cases occurred where thrusters were totally lost immediately upon re-ignition

Right photo: burner which developped further damage upon re-ignition (we confirmed an existence of a penetrating crack using immersion defects serach test)




We therefore think that the burner may not withstand the impact upon re-ignition. At the time of this reporting we do not know the state of the damaged burner in orbit (such as a penetrating crack etc)

Looking at reducing the re-ignition impact

Our current view, upon examining the re-ignition impacts, is that by starting oxidiser injection earlier (100 to 200 milliseconds) there is a possibility that impacts may be reduced. We are now discussing the operation including that of carrying out this manuever on the real burner in flight.

(end of page 59)

Here, I forgot to translate a small thing. With angular velocity we are supposed to refer to section C.3. Thus, angular velocity above should be "angular velocity (see section C.3)"

Also, I forgot to translate the graph on page 59. The vertical axis is the propulsion (N). The horizontal axis is the time or duration of oxidiser pre injection in milliseconds. The slanted caption inside the graph says "possibility of reducing re-ignition impacts"

Two more pages to go! Actually, the last two pages are rather interesting.

P

(page 60)

C.6 Looking at the possibility of realising continuous burn of OME

It is thought that during VOI-1 the closure of CV-F led to O/F ratio going out of the design range and this resulted in OME being affected (damage likely)


Therefore, we are now talking about maintaining the right O/F ratio even under the restriction of CV-F closure. This takes into account the next nearest approach to sun and we are currently conducting test burns with blowing down on fuel and oxidiser tanks.

During part 1 of the manuever test burns there was no change (damage?, P) in the burner and the burn was completed normally.


All this means that we will be operating outside the planned operational range of O/F ratio and we are now discussing if we can go ahaed with its execution.

(there is one large graph)

(Left vertical scale is oxidiser tank pressure (MPa))
(Horizontal is fuel tank pressure (MPa))
(Top horizontal is O/F ratio)

(Square on the graph (inside the graph, upper right) is flight plan range)
(Dotted lines: result of blow down operation preparatory test)
(Yellow band sandwiched between dotted lines: range which is expected during orbital manuever)

(there are two character sets on the graph with arrows pointing from them)

(top set reads) "conditions at the start of preliminary (or, preparatory,P) operation of blow down test"

(bottom set reads) "conditions at the end of blow down operation test"
(please note there is no preliminary or preparatory here, P)

(end of page 60)

(page 61) ( final page of this 3rd Akatsuki report)

C.7 Looking at the method whereby oxidiser may be jettisoned

If we find orbital re-insertion by OME is not possible, we will then have to turn to insertion by RCS. Since RCS is a one liquid engine we will need to reduce the craft's inertial mass by jettisoning oxidiser.

We are currently conducting preliminary experiments in order to come up with approapriate methods ( of freeze prevention etc) of oxidiser purge

Here below, we show an example from temp. changes at various points on OME during such preliminary tests.

(with the graph on the left, vertical is temp. (deg C) and horizontal is time)
(arrowed peaks are "injection")

(with the photo points of temp. measurement are indicated by A, B, and C, and these are reflected on the graph, of course, P)

(main text continues as follows)

In this example, we see that oxidiser purge may be possible without it freezing if we use pulse injection of duration less than 1 second. However, we will continue with our experiments in view of the possibility that the burner is totally damaged.

As for the contribution by oxidiser purge to the propulsive power the resulting specific impulse, even in the ideal case of perfect expansion, will be in the order of a few tens of seconds and its contribution to delta V is thought to be minimal.

(end of page 61!)

I learnt just now from a local newspaper that Akatsuki will fire its OME for 2 seconds on 7th this month to see what it is like. This will be followed by 20 seconds firing on 14th while attidue is maintained by smaller thrusters. This is meant to move Akatsuki through a distance of 5,000 km.

If all these are OK, then OME will be used again in November this year for insertion in 2015.

I also learnt a few weeks ago and this really is pleasant and surprising news to me, at least, that Hatsune Miku is flying with Akatsuki, possibly with 2 more vocaloids because media seem to be talking about 3 metal plates fixed as some kind of supporting structure on Akatsuki!

P

according to Akatsuki_JAXA tweets, people is cheering at the Sagamihara operation center. It looks like the first engine firing was successful (just how successful remains to be seen..)

JAXA HP is saying;

OME was fired for 2 seconds at 11:50 (JST) on 7 September as planned in order to establish quantitatively external disturbances (such as lateral propulsion) and the telemetry data is now being analysed.

The 20 seconds firing planned for 14th will be used to verify the attitue control logic system.

P

Pandaneko, can you confirm the Google translation of today's JAXA release that the measured acceleration was less than expected?

Yes, I confirm that the measued acceleration was less than expected. As a result the planned firing of 20 seconds on 14th will be shortened to 4 seconds in order to re-check the status of OME. Apparently, there is no change to Akatsuki after the first firing.

What does all this mean? Someting fell off again before the first firing? Oxidiser leak? Perhaps, helium, less of it remaining?

P

if the "bell" of the thruster is physically damaged, gases would not expand the way they should, which could explain the "loss of acceleration"

according to the NASAspaceflight forum the engine provided only 13 p.c. of the expected thrust. very bad news...

My apologies. Next firing will be "about" 5 seconds, not 4 seconds.

P

I am no longer sure if Akatsuki will be able to be of any use because today's Yomiuri newspaper says;

"the propulsion measued at the first firing was one ninth of the expected value"

I do not know how much fuel there still is left, but my gut feeling is that Akatsuki will run out of fuel. Sad...

P

they can still make it to venus using the RCS thrusters, but the mission will be shorter

re second firing test on 14th, same as the 1st, propulsion less than expected (no value given in today's JAXA release). JAXA will think about what to do next based on these lower values.

They may announce something more concrete.

P

...thanks for the update, Pandaneko.

The (possibly) good news here is that the engine is still capable of generating at least SOME delta-V, so instead of merely jettisoning it & relying entirely on the RCS some productive maneuvers seem possible at first glance.

Big unknown here is whether the thrust vector is still aligned properly (thinking nozzle damage here) or perhaps even stable & predictable.

I recall contingency plans to jettison excess oxidizer, but I believe that fuel not burned in the main engine can be burned in the RCS. (The engineering details are far from clear to me, because chemically speaking oxidizer should either be required or not required, regardless of what engine is being used.) If however, the excess fuel is somehow usable by the RCS then the question becomes where it can be used most efficiently in terms of changing the course of the spacecraft. If the OME is 87% less efficient that originally planned, then unless the RCS is even less efficient that that, it probably doesn't make sense to burn any of the remaining fuel in the OME.

the OME is a bi-propellant thruster, i.e. it generates gas by the spontaneous combustion of two liquids (hydrazine fuel and nitrogen tetroxide oxidizer).
RCS thrusters are mono-propellant thrusters, i.e. they generate gas by decomposing hydrazine on a catalyst bead. Think of your car's catalyser: that is a remote relative of this technology.
This is why RCS thrusters need no oxidizer.

Edit: the OME nominally provides 500 N of thrust, and should now provide about 1/9th of that, i.e. about 55 N.
There are two families of RCS: one providing 23 N and the other 3 N (for roll attitude control only, probably not usable for trajectory control).
I am not sure that using the OME would be a good idea. it would probably use too much hydrazine to provide too little thrust (I don't have info on the fuel consumption of the different thrusters but I assume that that of the OME would be larger than that of the RCS)

A Mainichi Daily News release: Venus probe unlikely to enter orbit fit for atmospheric observation

Am I right to assume that aerobraking, Magellan style, isn't plausible, if worst comes to worst and the engine is kaput?

Magellan was already in orbit, and very gradually lowered that orbit a bit at a time with atmospheric friction that was within tolerable limits. To bleed off enough speed all at once to make orbital insertion at the rate Akasuki is going - no, you would be fried. BUT ... if you could manage to get into an orbit with the thrusters, then I don't know, you might look into using aerobraking to get into a lower and more favorable orbit.

Yeah, I guessed as much; one would need an actual heat-shield to do aerocapture.

If I understand correctly the google translation of this Mainichi article (in Japanese), Akatsuki could enter an elliptical Venus orbit using the RCS with a period between one week and 90 days. Pandaneko can you confirm?
I wonder: what science could be accomplished on such a distant orbit? and would it be stable? I guess it would get large solar perturbations...

Yes, what the paper says is as follows.

1. It is no longer possible to insert Akatsuki into the planned orbit.

2. If they use RCS the orbit will be 90 days elliptic. Original plan was 30 hours per rotation.

3. In order to try November 2015 inerstion attempt they will have to fire the engine for 400 seconds in November this year.
They have not yet decided which engine to use and the decision will be made by the end of this month.

4. Akatsuki was developped to observbe super rotation at the cost of JPY 250 times 10 to the power of 8.

5. The only consolation is that all of the deviecs for observation are OK and healthy.

6. If they use RCS then it will take 72 times longer per rotation and the maximum distance will be 10 times further away.
They are therefore looking at optimum observational conditions for their devices.

This means I paid about 2 Macs worth of tax money for Akatsuki. I am willing to pay more.

P

it's going to be RCS in the end...
http://www.jaxa.jp/projects/sat/planet_c/index_e.html

Thanks, Paolo. Today's Asahi newspaper here says the same thing. In addition, it says the total mass of whatever is to be jetissoned is 64kg. It does not say oxidiser or fuel or both.

P

Another newspaper here says it is the oxidiser and it is going to be jetissoned some time this month.

P

thanks for the info Pandaneko. do you have a link?

and here is the usual very detailed release (Japanese PDF only) about the two September firings. as I understand it, the nozzle must have broken off and the engine is only providing 50 N of thrust instead of 350 N.
given the propulsion system plumbing schemes. only oxidiser will have to be vented overboard (hydrazine for the RCS comes from the very same tank as hydrazine for the OME). I guess the simplest way of venting oxidiser will be through the OME

Thanks, Paolo

I will start translating the new document from tommorrow on. Links to local papers will be useless, I think, as they are all in Japanese. There are papers in English here, such as;

The Japan Times
The Asahi Evening News
The Mainichi Daily
The Daily Yomiuri (not sure if this really exists, but I am 90% certain that it does)

P

thanks. I didn't dare to ask
I think you can skip the first pages, which seem to just repeat thing we already know about the missed orbit insertion, clogged valve etc.

Yes, I agree completely. However, we want to know contents list, at least. That is shown below.

Contents list

1. Events up until now. Discussing orbit control with the nearest approaches to sun

2. Summary up to the third meeting

2.1 Understanding the phenomena during VOI-1
2.2 Trade offs in operating for Venus reunion

3. Discussions for re-firing of OME

3.1 Searching for conditions for mitigating the re-firing shocks
3.2 Mitigating the re-firing shocks using the damaged burner
3.3 Rehearsing for orbital test firings
3.4 Results of OME firings
3.5 Orbiting after test firings

4. (I have lost the character string that ought to be here, or perhaps it was not there in the first place)

4.1 Operating by jettisoning the oxidiser
4.2 Discussing the long duration firing of RCS
4.3 Orbit plans from now on

5. Scheduling for nearest to sun approach orbit control

6. Summary of this report

End of contents list page

I think I will start with 2.2. from tommorrow.

P

Page 6:

2.2 Trade offs in Venus reunion operation

We reported at the third meeting that we will be carrying out test firings in orbit in order to verify the in orbit status of OME . Our plan was such that we will be choosing one of the following two options based on the result of these test firings and carry out the required orbital change in November 2011 at the time of nearest sun approach.


０．OME test firings in orbit

Several test firings were to verify the status of OME and confirm our ability to control attitude during OME firings.

ＯＭＥ：Orbital main engine (2 liquids and class 500N）
ＲＣＳ： Attitude control system (1 liquid and class 23N×4）

Option1： where we can use both OME and RCS

Step １．Nearest sun approach orbital control:

around Nov 2011 or June 2012

Step ２．Venus reunion manouver (4 day orbit):

around Nov 2015 (Venus reunion)

Step ３．Insertion into observational orbit (from 4 day orbit to 30 hours orbit):

around Nov 2011 or June 2012


Option 2： where we cannot use OME

We will carry out delta V using only RCS engines

Step 1．Jettisoning of oxdiser:

(during Oct 2011 as reported in the press)

2．Nearest to Sun approach (NSA) orbit control:

just before nearest sun orbital control

3．Venus orbit reunion manouever：

after Nov 2011 at NSA position and around Nov 2015 (Venus reunion)

end of page 6

P

Page 7

3. Discussions for OME refiring

3.1 Searching for conditions for mitigating ignition shocks

We estimated that OME had been broken at the time of VOI-1 in December last year. Therefore, we carried out tests and experiments with a view to understanding OME behaviour upon reignition in order to insert the probe into originally planned orbit.

(after this there are two photos horizontally (as it were) and in between them a large arrow with a character string in it)

Caption for the left photo says;

Burner after nozzle breakage (existence of through cracks confirmed by penetration flow detection method)

Character string in the arrow says;

cases occurred where thruster is wholly lost immediately upon ignition

Caption for the right photo says;

Burner whose breakage continued after re-ignition

Given continued breakage upon re-ignition we searched for conditions in which ignition shocks are mitigated.


What we found quantitatively is that re-ignition shocks can be mitigated if we inject oxidiser slightly ahead of fuel injection and by keeping fuel temp at a relatively high level before injection. These findings that we should not inject fuel and oxidiser simultaneously were realistically the best operational option we could find from our ground tests for Okatsuki in orbit.

In carrying out our experiments we first constructed a metal thruster whose parametric performance we could monitor with high accuracy and then applying the found conditions to ceramic thrusters.

Best re-ignition schock mitigating conditions we found from our 195 ground tests are as follows.

OME injector temp: higher than 150 deg C

OME fuel valve temp: 65 to 74 deg C *
OME piping temp: 57～68 deg C*
Oxidiser pre-emptive injection by: 100～400ms

*: Too high temps also leading to re-ignition schoks

end of page 7

P

Page 8

3.2 Ignition schock mitigated burn tests using damaged burners

We used thus obtained optimum conditions for ignition schock mitigation and carried out multiple burn tests using damaged burners. In view of next operational phase we conducted repeated short duration re-ignition tests.

Examples of continued or progressing damage or damage propagation

(hereafter there are three photos horizontally, from left to right. Please, for ease of reference, look at above link to see actual photos)

Caption for the leftmost photo says;

(initial state) burner damaged near the throat

Caption for the first arrow says;

re-ignition
↓
there are cases where damage progressed

Caption for the middle photo says;

burner whose damage progressed

Caption for the second arrow says;

re-ignition after damage progression
↓
there are cases where total loss occurred

Caption for the right photo says;

totally damaged burner

From these tests we observed that even using mitigation conditions we still had cases where damage still continued to propagate. Given these test results and our desire to insert the probe into the desirable orbit using OME we decided that we should conduct in-orbit ignition tests by giving due considerations to the mitigation conditions and using probe's OME and decide whether we might be able to make use of OME as desired.

end of page 8

P

Above for ease of reference

Page 9

3.3 In-orbit test firing rehearsal

In order to satisfy ground obtained conditions for mitigating ignition schoks we, in addition to temp control using heaters, changed the probe attitude (solar array angle) so that these conditions are met in orbit.

What is shown below is the comparison between the control parameters required of the probe and the prediction and in-orbit resulting values. As a result we were able to confirm that we will be able to satisfy these conditions for mitigation in orbit.

(after this there are three entries, something on the left and a graph on the right and these are followed by a table at the bottom. A lot of captions are stuck on top of another for the first entries and my translation are as follows)

Control parameters required of the probe (lefthand entry, P)

(captions are from top to bottom)

Fuel supply pressure
Oxidiser supply pressure
Fuel cut-off valve action timing
Oxidiser cut-off valve action timing
OME piping temp (fuel side)
OME piping temp (oxidiser side)
OME propellant valve temp (fuel side)
OME propellant valve temp (oxidiser side)
OME injector temp

Comparison of predicted values and in-orbit measurements for rehearsal (righthand entry, P)

(captions from top to bottm are;)

OME injector temp in-orbit measurement
OME propellant valve temp in-orbit measurement
OME piping temp in-orbit measurement
Analysis: OME injector temp
Analysis: OME propellant valve temp
Analysis: OME piping temp

(These are colour coded and being colour language blind I am simply ignoring to translate these different colours. Above reference should help, P)

(table is translated on CX RX basis)

C1R1: Temperature (perhaps, this was not here?)
C2R1: Required value
C3R1: Result of in-orbit rehearsal

C1R2: OME injector temp
C1R3: OME propellant valve temp
C1R4: OME piping temp

End of page 9

P

for something different, an interesting paper presented today at EPSC: In-flight observations performed by Akatsuki/IR2
note also Emily's tweet:

With the benefit of more than 140 characters I'll add that the phase curve had this odd double-humped shape near 0 phase that I don't recall ever seeing before in a phase curve, but admittedly I really haven't seen phase curves for anything but airless bodies.

Above for ease of reference

Page 10

3.4 OME test firings result (1/3)

We conducted test firings as planned on 7th and 14th of September. Each firing was about 2 and 5 seconds in duration.

(after this there are two graphs side by side)

(left graph is about accerleration (vertical axis in m/sxs) and the horizontal axis is time in second relative to OME firing start)
(wtih this graph (and also one on the right) the 1st firing is in depicted in red and the 2nd firing is in blue)

(leftmost character says:) RCS settling
(characters to the right and from top down are:) 1st firing and 2nd firing

(right graph is about Doppler monitoring (vertical axis is the increment in velocity in line of sight in m/s) and the time axis is the same relative lapse of time)

(characters here are top down and they read:) RCS settling, 1st firing, 2nd firing

We observed almost the same phenomena both in acceralation and Doppler monitoring during the two in-orbit firing tests.

• Priort to OME firings we fired 4 RCS thrusters for 3 seconds (total of about 70N) and as a result the RCS settling which moves propellant in the tank to the exit port side was carried out normally.

• The acceralation subsequently obtained by OME firings was about 1/9 of our estimation and the proplusive power was about 40N (We accepted these values as reasonable and reliable as the acceralation obtained during RCS operation was in agreenment with our estimation)

• Also, the line of sight increment in velocity as obtained from our Doppler monitoring was in agreenment with the acceralation data.

Therefore, we concluded from the telemetry data that OME propulsive power was significantly less than our initial expectation.

End of page 10

above for ease of reference

page 11

3.4 OME test firings result (2/3)

(there are two graphs here, sandwitching a schematic)

(caption for the graph on left is:)

Temperature of each section of the propulsive system during the 1st firing

Vetical axis says: OME propellant valve temp and fuel piping temp in deg C
Horizontal axis is time in min relative to OME start

(colour coding is as follows)

blue: OME propellant valve temp (oxidiser side)
red: OME propellant valve temp (fuel side)
pale brown: OME injector temp
pale green: OME piping temp (fuel side)

(With this and also with the right hand graph the dark vertical line nearest to the vertical axis denotes the start of OME firing)

(caption for the schematic is:)

piping system around OME

(and the crossed square at top right is oxidiser system
(and the crossed square at top left is fuel system

(caption for the graph on right is

Fuel piping temp (analysis and measured) at the timne of 1st test firing

vertical axis is temp in deg C and horizontal is time in min relative to OME start

(dotted red line is fuel piping (analysis based on 1/9 of propellant flow)
(thick solid green line is fuel piping
(thin solid green line is fuel piping (analysis with normal flow rate)

During the engine firing tests we monitored the status of the propellant supply system as follows.

• Valve status monitoring during the firings returned normal value.

• History of the temp of each section of the propulsive system during OME firings was in agreenment with our expectation (by detecting temp decrease due to the 20℃ propellant flow) and showed similar tendency both on 7th and 14th September (see graph on left)

• We conducted propellant flow rate sensitivity analysis with temp drop against the temp history of each section of the propellant supply system. We noted that the piping temp simulation with the assumed normal flow rate was in quantitative agreenment with the history of the piping temp in orbit. We therefore concluded that the propellant supply amount during the firings was more or less normal (see graph on right)

From above observations we conclude that the supply of propellant during the engine test firings was normal.

End of page 11

above for ease of reference

page 12

3.4 OME test firings result (3/3)

Evaluating the propulsive power generated by the test firings

In what follows we will try to estimate the force of the gas burning in free space pushing the bottom of Akatsuki by assuming that OME destruction continued to progress, leading to the total loss of the burner and that the bottom of Akatsuki is now planar.

(after this there are two simple graphics side by side)

(the caption between the two graphics reads) : OME burner damaged at the throat section

(the captions for the lefthand graphic read) :

OME damaged at the throat section, approx. 350 N of propulsive power is assumed

(the captions for the righthand graphic read) :

OME burner is completely lost.

We estimate the propulsive power to be approx. 50 N by assuming the burning gas expanding in an ideal isotropic entropy flow.

In reality, actual flow and the burning state must be more complicated. However, we are satisfied at the measured propulsive power if we also assume that test firings led to further OME damage and OME not functioning as a burner.

We therefore think that switching from OME to RCS will be more efficient if we compare accerlation abilities per unit mass of the fuel (specific impulse) .

End of page 12