GSLV D3 Launch Failure

All threads that are locked or marked for deletion will be moved to this forum. The topics will be cleared from this archive on the 1st and 16th of each month.
krishnan
BRF Oldie
Posts: 7342
Joined: 07 Oct 2005 12:58
Location: 13° 04' N , 80° 17' E

Re: GSLV D3 Launch Failure

Postby krishnan » 18 Apr 2010 16:23

http://sify.com/news/gslv-mission-scien ... gbhha.html

Thiruvananthapuram: A team of Indian space scientists has established that the indigenously built cryogenic engine had ignited for a second during the failed GSLV mission on Thursday.

"This took place for a second and then the fuel supply to power turbo got blocked. The Indian Space Research Organisation chairman and we knew this on Thursday but then we wanted to be doubly sure about it. And now this has been substantiated with the data. By all means this is a great achievement," said a senior scientist who did not wish to be identified.

Rocket launch failure will not delay moon mission: ISRO chief

Senior space scientists said they have with them records to prove that compression had taken place and the cryogenic engine had ignited.

The GSLV D3 blasted off with a GSAT 4 satellite around 5 p.m. on Thursday from Sriharikota in Andhra Pradesh but then stopped emitting signals within minutes.

A high-level meeting began here Saturday to review the flight data of Thursday's failed rocket mission.

Flight analysis of failed rocket mission begins

The meeting is taking place at the Vikram Sarabhai Space Centre (VSSC) here, a unit of the ISRO.

ISRO Chairman K. Radhakrishnan, who arrived here on Saturday night, is leading the discussion.

"The two-day meeting has also decided to constitute a failure analysis committee, where complete reasons would be listed of the failure of the mission. Finer details of the probable causes of failure would be analysed thread-bare. This committee would be headed by our chairman and would have sittings here and in Bangalore and in a month the report would be ready," added the space scientist.

It was at ISRO's centre at Valiyamala, in the city suburbs, that the cryogenic project took shape from the design stage to the assembly stage.

PratikDas
BRFite
Posts: 1927
Joined: 06 Feb 2009 07:46
Contact:

Re: GSLV D3 Launch Failure

Postby PratikDas » 18 Apr 2010 16:31

"While the problem might have happened during the ignition period I'm not convinced that the problem is with the ignition system itself or the engine environment being in vacuum.

Since ours is a staged combustion system, I assume the critical task of ignition takes place in the preburning chamber. This chamber should provide some degree of isolation from the nozzle especially when it is gushing with expanding O2 and H2, assuming the pumps worked and didn't incur some fault"

I had written the text above when I wanted to check ArunKumar's engine schematic and then found Krishnan's post above. End of discussion.

Gagan
BRF Oldie
Posts: 11209
Joined: 16 Apr 2008 22:25

Re: GSLV D3 Launch Failure

Postby Gagan » 18 Apr 2010 16:32

Guys here are thinking that ISRO's understanding of the technology is fragile, that somehow ISRO is not sure that the engine they have developed will work in vaccume or not. :rotfl:

Point is that out of 180 odd countries in the world and some 4 dozen odd space programs around, the guys at ISRO develop a cryo engine, successfully test it several times on the ground, not only test it, but build it from materials available within the country. These guys ensured that the nation's ability to develop the materials needed for such a project are produced in house.

There is a failure alright, but it is not because there is a lack of understanding of the science or that there is a problem with the construction of the engine. This probably a failure of some critical component. Or a minor design problem (Like the agni-3 first launch) that'll be identified and rectified.

Already the engine developed is lighter than the russian engine, has more growth potential, and is bleeding edge tech as far as contemporary cryos is concerned.

But the bigger failure is the fact that ISRO can't convince jingos and media idiots of its abilities even after proving themselves multiple times. I bemoan the fact that our top PSUs - the ISROs and the DRDOs and HALs don't employ a TFTA, suave and canny media manager to pander to the silliness of the media and the jingos alike.

PratikDas
BRFite
Posts: 1927
Joined: 06 Feb 2009 07:46
Contact:

Re: GSLV D3 Launch Failure

Postby PratikDas » 18 Apr 2010 16:42

Gagan wrote:...
But the bigger failure is the fact that ISRO can't convince jingos and media idiots of its abilities even after proving themselves multiple times. I bemoan the fact that our top PSUs - the ISROs and the DRDOs and HALs don't employ a TFTA, suave and canny media manager to pander to the silliness of the media and the jingos alike.

While some might be questioning the whole process, I surely am not trying to belittle ISRO Gagan. I hope that's ridiculously obvious from my posts.

We could all just wait for press releases and applaud, and BR could go to sleep. In the meanwhile we're all trying to make sense of what might have happened. Now that I've seen Krishnan's post, I'm quite happy to the STFU.

Gagan
BRF Oldie
Posts: 11209
Joined: 16 Apr 2008 22:25

Re: GSLV D3 Launch Failure

Postby Gagan » 18 Apr 2010 16:56

Oh sir-ji,
I had written this post of mine well before your post came around. It it just that by the time I finished my post and hit the post button, your post appeared. :)

Most certainly not directed at you.

This is just a general post aimed at jingos, DDM

SSridhar
Forum Moderator
Posts: 23993
Joined: 05 May 2001 11:31
Location: Chennai

Re: GSLV D3 Launch Failure

Postby SSridhar » 18 Apr 2010 17:55

Igniting a cryogenic engine is not exactly an easy task, as is anything to do with cryogenic engines. Before the main engine starts, the pre-burner in the staged combustion cycle has to be started. Proper ratio of the fuel & oxidizer has to be ignited to start the turbine. It will be a low initial flow of the fuel & oxidizer which will stabilize after the turbines achieve the steady state. If proper ratio of the fuel & oxidizer is not maintained, the ignition may not be sustained. Sometime may even lead to explosion. There are myriad possibilities for failures to happen. Vapour pockets for example. It is a tough time for ISRO and one has to wish them well. They have faced similar or worse situations before and have come out stronger, like in the ASLV disaster, for example.

Shankar
BRFite
Posts: 1905
Joined: 28 Aug 2002 11:31
Location: wai -maharastra

Re: GSLV D3 Launch Failure

Postby Shankar » 18 Apr 2010 17:56

This is what happened me thinks

The main engine did not ignite at all
The steering engines ignited

so the rocket started tumbling into ocean

while there is facility for vacuum testing the steering engine similar facility for main engine for C-12 do not exists and there lies the problem for non ignition

Engine ignition under vacuum is bit unpredictable since it takes place in few mbar pressure or high vacuum environment where flame front separation -convective heat transfer is minimal
where as in ground test the engine ignites at 1000 mbar pressure -both are not same

It happened because some facility that was required was not sanctioned by who ever and today ISRO is paying the price

The facility is costly -may be several hundred crore -you have to pump out tons of hot exhaust gases quickly to simulate ignition altitude . Am studying the phenomena now shall post tomorrow with more details about the facility required -lest not even ask why it was not provided in the first place for such a critical project

SriKumar
BRF Oldie
Posts: 2024
Joined: 27 Feb 2006 07:22
Location: sarvatra

Re: GSLV D3 Launch Failure

Postby SriKumar » 18 Apr 2010 18:24

Sanjay M wrote:Somebody posted this in another forum:

Image



What is the source of this data? I am guessing that someone watched the youtube video of the launch broadcast on Doordarshan TV and (painstakingly) jotted down the data and plotted it.

*IF* the above plot is accurate, then:
The data stream (for 2-3 seconds) at 3:08 and 3:20 and 3:25 are interesting. The altitude is increasing while the velocity is flat- not decreasing. There had to be some motive force during all these durations of time. It is only after 3:30 that the altitude starts to decrease.

By the way, I find it amazing that ISRO shows all these graphs live on TV. I dont know of anyone else who broadcasts it to the world- live (well, is there is anyone else?). Talk about open-ness &/or confidence.

sum
BRF Oldie
Posts: 10046
Joined: 08 May 2007 17:04
Location: (IT-vity && DRDO) nagar

Re: GSLV D3 Launch Failure

Postby sum » 18 Apr 2010 19:14

A team of Indian space scientists has established that the indigenously built cryogenic engine had ignited for a second during the failed GSLV mission on Thursday.

"This took place for a second and then the fuel supply to power turbo got blocked. The Indian Space Research Organisation chairman and we knew this on Thursday but then we wanted to be doubly sure about it. And now this has been substantiated with the data. By all means this is a great achievement," said a senior scientist who did not wish to be identified.

Great news if true...now only why the question of blocking of supply to turbo needs to be debugged. Means that no major design failure in ignition part!!

Gagan
BRF Oldie
Posts: 11209
Joined: 16 Apr 2008 22:25

Re: GSLV D3 Launch Failure

Postby Gagan » 18 Apr 2010 19:41

Hai re!

GAGAN payload is at the bottom of the sea !
:cry:

BTW the confirmation that the CUS ignited after all - Jingo khush hua.

PratikDas
BRFite
Posts: 1927
Joined: 06 Feb 2009 07:46
Contact:

Re: GSLV D3 Launch Failure

Postby PratikDas » 18 Apr 2010 20:50

Would this work?

Image

vina
BRF Oldie
Posts: 6046
Joined: 11 May 2005 06:56
Location: Doing Nijikaran, Udharikaran and Baazarikaran to Commies and Assorted Leftists

Re: GSLV D3 Launch Failure

Postby vina » 18 Apr 2010 20:56

There had to be some motive force during all these durations of time


Yes.. Something that Sir Issac Newton described very well.. Something called inertia.

Rahul M
Forum Moderator
Posts: 17016
Joined: 17 Aug 2005 21:09
Location: Skies over BRFATA
Contact:

Re: GSLV D3 Launch Failure

Postby Rahul M » 18 Apr 2010 20:58

:lol:

Sanjay M
BRF Oldie
Posts: 4892
Joined: 02 Nov 2005 14:57

Re: GSLV D3 Launch Failure

Postby Sanjay M » 18 Apr 2010 21:54

Gagan wrote:But the bigger failure is the fact that ISRO can't convince jingos and media idiots of its abilities even after proving themselves multiple times. I bemoan the fact that our top PSUs - the ISROs and the DRDOs and HALs don't employ a TFTA, suave and canny media manager to pander to the silliness of the media and the jingos alike.



Well, when I look at a Youtube video of even a SpaceX launch, and I see onboard cameras giving a marvelous view from the vehicle as it travels all the way up its trajectory, I wonder why ISRO can't do this like all the established space program do. Instead we end up looking at some measly plotgraph of the whole thing. Is this part of being "3% of NASA's budget" again? There are some things that aren't worth skimping on.

SSSalvi
BRFite
Posts: 666
Joined: 23 Jan 2007 19:35
Location: Hyderabad
Contact:

Re: GSLV D3 Launch Failure

Postby SSSalvi » 18 Apr 2010 21:57

SriKumar wrote:
Sanjay M wrote:Somebody posted this in another forum:

Image



What is the source of this data? I am guessing that someone watched the youtube video of the launch broadcast on Doordarshan TV and (painstakingly) jotted down the data and plotted it.

*IF* the above plot is accurate, then:
The data stream (for 2-3 seconds) at 3:08 and 3:20 and 3:25 are interesting. The altitude is increasing while the velocity is flat- not decreasing. There had to be some motive force during all these durations of time. It is only after 3:30 that the altitude starts to decrease.

By the way, I find it amazing that ISRO shows all these graphs live on TV. I dont know of anyone else who broadcasts it to the world- live (well, is there is anyone else?). Talk about open-ness &/or confidence.



Here is the source of numerical data: ( As guessed could be from the YouTube Video )

Time (s) velocity (km/s) altitude (km) event
262.8 4.113 127.5
263.8 4.138 127.8

290.4 GS2 Shutoff

293.0 GS2 Separation

298.0 4.898 134.9
299.0 4.898 135.1
300.0 4.898 135.3
301.0 4.897 135.5

304.9 CUS Ignition

308.0 4.895 136.5
309.0 4.895 136.7
310.0 4.895 136.8

317.0 4.893 137.6

329.0 4.892 138.3

373.0 4.899 134.9

379.0 4.901 133.7

446.5 4.951 107.4

505.0 5.023 65.9

sanjeevpunj
BRFite
Posts: 971
Joined: 04 Sep 2009 13:10

Re: GSLV D3 Launch Failure

Postby sanjeevpunj » 18 Apr 2010 22:07

ISRO should provide special images from its unlimited resources whenever they have a major test in the offing.The media would love to lap up whatever is dished and please be reminded that not all in the press are ignorant of science, some do love reporting about scientific events too.All depends on what info ISRO wants to release as de-classified info, and what info they are keeping secret from the press for reasons of security.Glossy photos, videos definitely can and should be provided by ISRO to enhance its own image. Look at NASA's website....even kids love to go there they have a special section for kids.Give me a job in ISRO to help with the media, I will surely come up with some good ideas.

Gagan
BRF Oldie
Posts: 11209
Joined: 16 Apr 2008 22:25

Re: GSLV D3 Launch Failure

Postby Gagan » 18 Apr 2010 22:12

They probably have bandwidth problems in transmitting live video from the launcher... :P

But they did invest in an el-cheapo way to try and overcome this handicap - they commissioned a not so smooth graphic of the PSLV at various stages of separation! :roll: Oh boy! I feel like pulling my hair out of my head everytime DD transmits that animation in the middle of the launch.

And point to note is that they have only done the PSLV graphic so far, the one for the GSLV-2 and -3 are still in the making.

ISRO if you are reading this post, please please for gods sake put those cameras on all your launchers. This will also help you to know exact status of the launcher etc, as well as the jingos will be happy.

For instance, if they had a camera positioned strategically enough, one never knows they just might have seen more of what went wrong with the CUS in this instance.

SriKumar
BRF Oldie
Posts: 2024
Joined: 27 Feb 2006 07:22
Location: sarvatra

Re: GSLV D3 Launch Failure

Postby SriKumar » 18 Apr 2010 23:21

vina wrote:
There had to be some motive force during all these durations of time


Yes.. Something that Sir Issac Newton described very well.. Something called inertia.


Yes, but you may want to also re-visit the energy part of it.

You forget the increase in the potential energy as the altitude increased during this time. Inertia can keep it going, but the velocity cannot remain constant while the altitude *increases* (Earth is still pulling it down). Velocity has to reduce. (Hint: increased altitude = increase in potential energy. KE (=velocity) therefore has to reduce in the absence of a motive force. Inertia alone is not enough).
Last edited by SriKumar on 18 Apr 2010 23:49, edited 1 time in total.

SriKumar
BRF Oldie
Posts: 2024
Joined: 27 Feb 2006 07:22
Location: sarvatra

Re: GSLV D3 Launch Failure

Postby SriKumar » 18 Apr 2010 23:49

SSSalvi wrote:
Time (s) velocity (km/s) altitude (km) event
262.8 4.113 127.5
263.8 4.138 127.8

290.4 GS2 Shutoff

293.0 GS2 Separation

298.0 4.898 134.9
299.0 4.898 135.1
300.0 4.898 135.3
301.0 4.897 135.5

304.9 CUS Ignition

308.0 4.895 136.5
309.0 4.895 136.7
310.0 4.895 136.8

317.0 4.893 137.6

329.0 4.892 138.3

373.0 4.899 134.9

379.0 4.901 133.7

446.5 4.951 107.4

505.0 5.023 65.9

Interesting. The 3-digit resolution on the velocity does not seem enough, but after CUS ignition, it gains altitude (at seemingly constant velocity-if there were 4 digits we would know) upto atleast 3:10. So, there is a motive force. After that there is a big gap (ISRO would have all the gap data) and at 3:17 the velocity reduces while the altitude increases (inertia). By this time, CUS probably shut off.

Sanjay M
BRF Oldie
Posts: 4892
Joined: 02 Nov 2005 14:57

Re: GSLV D3 Launch Failure

Postby Sanjay M » 18 Apr 2010 23:51

http://beta.thehindu.com/sci-tech/techn ... 402907.ece

Indigenous cryogenic engine didn't fail to ignite: scientists
S. Anandan

Senior Indian Space Research Organisation (ISRO) scientists, who met at the Vikram Sarabhai Space Centre in Thiruvananthapuram to examine the reasons for the failure of the GSLV-D3 mission with indigenous cryogenic upper stage, on Sunday ascertained that contrary to initial reports the cryogenic stage had doubtless ignited in the vacuum of the space.

After deliberating on the performance parameters of the cryogenic stage (the third stage) of the unsuccessful GSLV development flight last Thursday, they arrived at the conclusion that the mission failed after the fuel turbo pump that supplied fuel to the cryogenic engine had stopped working a second after ignition. ISRO chairman K. Radhakrishnan chaired the two-day meeting.

“The data clearly shows that combustion [of the cryogenic engine fuel, liquid hydrogen at minus 253 degree Celsius, and the oxidiser, liquid oxygen at minus 183 degree Celsius] had indeed taken place. The rocket's acceleration had increased for a second before it drifted off the designated flight path. Indications are that the turbine that powered the fuel turbo pump had somehow failed. [The propellants are pumped using turbo pumps running around 4,000 rpm.] There could be various reasons for its failure,” a senior ISRO scientist told The Hindu.

The ISRO will now constitute a ‘Failure Analysis Committee' to close in on the exact reason for the failure. It will come out with its report by May-end, following which the national experts' panel, constituted to review and give clearance to the GSLV-D3 mission, will examine the report. The ISRO Chairman will brief the Prime Minister on Tuesday on what had gone wrong with GSLV-D3.

Shankar
BRFite
Posts: 1905
Joined: 28 Aug 2002 11:31
Location: wai -maharastra

Re: GSLV D3 Launch Failure

Postby Shankar » 19 Apr 2010 00:05

when you put an ignition source in hydrogen oxygen mixture surely it will ignite in a batch mode - that does not call for any high end rocket science .After second stage separation the ignitors supplied sparks to the explosive propellant mixture and it ignited
Problem started after wards -the flame could not sustain itself it extinguished as it got separated from the nozzle walls and quenched out the explosive mixture of unburnt propellants built up and that might have caused an explosion causing loss of telemetry data . The explosion could have been caused by the hot exhaust of steering engines getting into the unburnt propellant cloud coming out of nozzle

The vernier engines did ignite and that's why the tumble effect .Both the engines tried to right the heavy stage and payload as it nosed down

Me thinks problem is with nozzle design - some modifications must have been carried out compared to earlier cryo acceleration blocks we have used to save weight -possible reduction in length/weight expansion ratio involving rise to higher than acceptable expansion ratio

we need a longer expansion nozzle and a high altitude test set up

Shankar
BRFite
Posts: 1905
Joined: 28 Aug 2002 11:31
Location: wai -maharastra

Re: GSLV D3 Launch Failure

Postby Shankar » 19 Apr 2010 00:10

the term ignition means sustained ignition not momentary ignition as highlighted in telemetry data - the combustion chamber and exhaust nozzle design that is why so complex and time consuming involving amongst other things CFD analysis simultaneous analysis of heat and mass transfer and variable fluid properties in different combustion and regenerative cooling zones

Shankar
BRFite
Posts: 1905
Joined: 28 Aug 2002 11:31
Location: wai -maharastra

Re: GSLV D3 Launch Failure

Postby Shankar » 19 Apr 2010 00:16

Turbo pump failure is a possibility but it should have happened at start that is in priming stage when it can overheat but once started the chances of it stopping is much lower since it is already cold fill of fluid and rotating at the designated speed of 38000 rpm

Sanjay M
BRF Oldie
Posts: 4892
Joined: 02 Nov 2005 14:57

Re: GSLV D3 Launch Failure

Postby Sanjay M » 19 Apr 2010 01:42

http://www.deccanherald.com/content/645 ... -fail.html

About Rs 335 crore was spent on the GSLV-D3 launch. The advanced communication satellite GSAT-4 cost Rs 150 crore. It was also experimental. It is pointed out that ISRO should not have risked launching a costly and experimental satellite in a test flight, however certain it was of the success. The view is not entirely wrong.

Craig Alpert
BRFite
Posts: 1440
Joined: 09 Oct 2009 17:36
Location: Behind Enemy Lines

Re: GSLV D3 Launch Failure

Postby Craig Alpert » 19 Apr 2010 04:34


vina
BRF Oldie
Posts: 6046
Joined: 11 May 2005 06:56
Location: Doing Nijikaran, Udharikaran and Baazarikaran to Commies and Assorted Leftists

Re: GSLV D3 Launch Failure

Postby vina » 19 Apr 2010 07:56

SriKumar wrote:You forget the increase in the potential energy as the altitude increased during this time. Inertia can keep it going, but the velocity cannot remain constant while the altitude *increases* (Earth is still pulling it down). Velocity has to reduce. (Hint: increased altitude = increase in potential energy. KE (=velocity) therefore has to reduce in the absence of a motive force. Inertia alone is not enough).


What energy saar?. I know only 9th grade Fizzycs . Energy busienss was in the 10th standard I think . From NCERT Fyzzics book, I remember only one thing out of all that and that is .

v = u + at .. I can remember simple stuff like that and plug numbers.

Okay. Now you say this.
The data stream (for 2-3 seconds) at 3:08 and 3:20 and 3:25 are interesting

So, u at 3.08 secs is around 5:02 KM/S , so assuming that the engine had not ignited, what should be the velocity after 2 seconds ?.

v = 5.02 - g*2 = 5.02 KM/S - (9.8/1000 Km/s^2)*2 sec .. Now you have (9.8/1000) is approx 0.01 (in fact less, coz you have to account for g at 400 kms or so above earth surface , but as an approx it should be fine) .. So in 2 secs, the velocity will be 5.00 k/ms , in 5 secs a 0.05 km/s decrease.

Now poor old me with tired thick eyeglasses can barely see beyond my nose and even in my best days, I wouldn't have been able to make out a 0.02 change in a graph of that scale .. All that I can make out is that the line is essentially flat. Of course you have better eyes than me, but I suspect it would be difficult to find trends in 2 to 5 sec interval in a graph of that scale, unless you have some secret facility to zoom in on those particular intervals in question via a feed to ISRO's data centers.

Sanjay M
BRF Oldie
Posts: 4892
Joined: 02 Nov 2005 14:57

Re: GSLV D3 Launch Failure

Postby Sanjay M » 19 Apr 2010 08:47

In another forum, some people were commenting about "ullage" - ie. the potential for gaseous bubbles to get swallowed by the fuel turbopump in a zero-G environment, thus causing it to cavitate and be destroyed. But this is something that's supposed to be taken care of through proper pressurization of the propellant tanks.

I wonder if the fuel turbopump itself could have been faulty?

juvva
BRFite
Posts: 306
Joined: 20 Oct 2008 17:34

Re: GSLV D3 Launch Failure

Postby juvva » 19 Apr 2010 09:06

From Arun Kumar's diagram on page 5, the H and O2 pumps are driven by the turbine. How are the pumps driven before ignition? battery driven?

vina
BRF Oldie
Posts: 6046
Joined: 11 May 2005 06:56
Location: Doing Nijikaran, Udharikaran and Baazarikaran to Commies and Assorted Leftists

Re: GSLV D3 Launch Failure

Postby vina » 19 Apr 2010 09:33

But this is something that's supposed to be taken care of through proper pressurization of the propellant tanks


Raakits and Mijjiles usually have something called ullage motors (small motors that give a small push before the main engines ignite). Remember, under acceleration, however small, the raakit is NOT under zero g, and liquid will be "normal" like on earth. In zero g, due to surface tension, the liquid tends to form into a sphere (disclaimer, me know only 9th std Fyzzics.. for more details, e-mail to Bade Saar or AmberG please)

Don't know if the Yindoo CUS is designed with ullage motors. Doesn't look to be the case. Maybe taking ullage motors from the Agni TD/ Older Raakits and attaching it to the Musharraf of the CUS to give a small "dhakka" before stage ignition could do the job. Quick, easy, tickao and low risk , if that indeed is the problem .. Or maybe, since the interstage is a truss, fire the CUS just before the 2nd stage cuts out , venting through the interstage , preventing coasting and letting Sir Newton's inertia and weightlessness remain in his grave!. Even quicker, dirtier and more tickao, no need for any modifications, just a software program change, which any DOO can do in 2 hrs flat with testing included.

Pressurization is usually done to keep the structural integrity of tanks. For eg, if you take bottle of softdrink that comes in PET bottle and close your mouth fully over the opening and drink, the sides will buckle due to air pressure. Similar thing happens in Raakit in atmosphere. So you need to pressurize it. Even aircraft fuel tanks are pressurized.

Narad
BRFite
Posts: 854
Joined: 04 Jan 2010 15:15

Re: GSLV D3 Launch Failure

Postby Narad » 19 Apr 2010 09:53

It is established that it was indeed the turbo pump that malfunctioned.

Senior Indian Space Research Organisation (ISRO) scientists, who met at the Vikram Sarabhai Space Centre in Thiruvananthapuram to examine the reasons for the failure of the GSLV-D3 mission with indigenous cryogenic upper stage, on Sunday ascertained that contrary to initial reports the cryogenic stage had doubtless ignited in the vacuum of the space.

After deliberating on the performance parameters of the cryogenic stage (the third stage) of the unsuccessful GSLV development flight last Thursday, they concluded that the mission failed after the fuel turbo pump that supplied fuel to the cryogenic engine had stopped working a second after ignition. ISRO Chairman K. Radhakrishnan chaired the two-day meeting.


http://beta.thehindu.com/sci-tech/technology/article402907.ece

ramana
Forum Moderator
Posts: 54513
Joined: 01 Jan 1970 05:30

Re: GSLV D3 Launch Failure

Postby ramana » 19 Apr 2010 10:16

Is the turbo pump driven by the vernier motors?

vina
BRF Oldie
Posts: 6046
Joined: 11 May 2005 06:56
Location: Doing Nijikaran, Udharikaran and Baazarikaran to Commies and Assorted Leftists

Re: GSLV D3 Launch Failure

Postby vina » 19 Apr 2010 10:26

Is the turbo pump driven by the vernier motors?

By the pump driven by turbine, which is turned by the gases from the pre combustion chamber. .

For pump to fail, the flow has to be blocked for some reason,like say a bubble in the lines to the pump, the fluid in the pump cavitates (another reason to pressurize the lines and fuel in the tanks, to raise the onset of cavitation by raising local pressures well above vaporization pressure) or some mechanical breakdown like an impeller breaking off, bearing failing, seal failing, whatever.

The "fluid mech" part is easy and the instrumentation will easily catch it (pressure sensors in fuel lines and tanks) and rpm counter and shock sensor on turbine. But if it is a mechanical breakdown, I guess it could be bit more difficult .

Somehow I think "fluid mech" chances are much higher (and also easier to fix, read previous post about ullage motor or starting engine while "dhakka" is still there) I dont think this Yindoo upper stage is a restartable engine. It is fire up and burn until empty engine. Only in a restartable engine you would go to extreme paranoid lengths and consider all nine yards to ensure a restart .

Narad
BRFite
Posts: 854
Joined: 04 Jan 2010 15:15

Re: GSLV D3 Launch Failure

Postby Narad » 19 Apr 2010 10:36

Any news about the performance of new mission computer onboard??

juvva
BRFite
Posts: 306
Joined: 20 Oct 2008 17:34

Re: GSLV D3 Launch Failure

Postby juvva » 19 Apr 2010 10:46

vina wrote:
Is the turbo pump driven by the vernier motors?

By the pump driven by turbine, which is turned by the gases from the pre combustion chamber. .

For pump to fail, the flow has to be blocked for some reason,like say a bubble in the lines to the pump, the fluid in the pump cavitates (another reason to pressurize the lines and fuel in the tanks, to raise the onset of cavitation by raising local pressures well above vaporization pressure) or some mechanical breakdown like an impeller breaking off, bearing failing, seal failing, whatever.

The "fluid mech" part is easy and the instrumentation will easily catch it (pressure sensors in fuel lines and tanks) and rpm counter and shock sensor on turbine. But if it is a mechanical breakdown, I guess it could be bit more difficult .

Somehow I think "fluid mech" chances are much higher (and also easier to fix, read previous post about ullage motor or starting engine while "dhakka" is still there) I dont think this Yindoo upper stage is a restartable engine. It is fire up and burn until empty engine. Only in a restartable engine you would go to extreme paranoid lengths and consider all nine yards to ensure a restart .


But this means that ignition should take place before the turbine spins and spins the pumps? Is this a correct understanding?
If so, how are pumps driven before ignition?

vina
BRF Oldie
Posts: 6046
Joined: 11 May 2005 06:56
Location: Doing Nijikaran, Udharikaran and Baazarikaran to Commies and Assorted Leftists

Re: GSLV D3 Launch Failure

Postby vina » 19 Apr 2010 11:31

But this means that ignition should take place before the turbine spins and spins the pumps? Is this a correct understanding?
If so, how are pumps driven before ignition?


Yes.. A small amount of liquid hydrogen and lox are burnt in the pre-burner /gas generator and that spins the turbine . Now this is different from the ignition in the main combustion chamber of the rocket.

Initial starting - Tanks are pressurized and so a small amount of LH2 & LOX can flow to the gas generators under pressure and burnt . Also in some cases, there is a secondary system like hydrogen peroxide that delivers high temperature steam to get the turbine spinning initially , sort of like the battery and electric motor to start the engine of your car.

This picture from wikipedia gives a great illustration
Image

Shankar
BRFite
Posts: 1905
Joined: 28 Aug 2002 11:31
Location: wai -maharastra

Re: GSLV D3 Launch Failure

Postby Shankar » 19 Apr 2010 12:31

the propellant t supply system in GSLV is like this

apart from main turbo pump there is also a small booster pump which feed liquid at positive pressure to the main turbo pump

The turbine of the booster pump is operated by stored gas at high pressure
The hydrogen liquid is stored at 2 bar and is pushed upto 6 bar in the booster pump and then to gas generator and then to turbine of main pump which takes the fluid to around 60 bar plus operating at 38000 rpm plus

so the chance of turbo pump freezing up after start is remote because it always get gas at sizable positive pressure and the liquid is subcooled(saturated at 2 bar and then when compressed to 6bar at the booster pump it get sub cooled for all practical purpose)

it has been tested separately and as integrated stage many times

so I think it is quenching of the ignited gas that caused the failure because we do not have a high altitude simulation chamber for main engine
The approximate cost of such an infrastructure ma be 100 crore main cost going for the huge injector which has to pump out the hot light weight exhaust very quickly to maintain the low atmospheric pressure of few mbar as expected in the ignition altitude of 60 km and above

For C25 such is planned for and under implementation

vina
BRF Oldie
Posts: 6046
Joined: 11 May 2005 06:56
Location: Doing Nijikaran, Udharikaran and Baazarikaran to Commies and Assorted Leftists

Re: GSLV D3 Launch Failure

Postby vina » 19 Apr 2010 13:34

Shankar wrote:the propellant t supply system in GSLV is like this

apart from main turbo pump there is also a small booster pump which feed liquid at positive pressure to the main turbo pump

The turbine of the booster pump is operated by stored gas at high pressure
The hydrogen liquid is stored at 2 bar and is pushed upto 6 bar in the booster pump and then to gas generator and then to turbine of main pump which takes the fluid to around 60 bar plus operating at 38000 rpm plus

so the chance of turbo pump freezing up after start is remote because it always get gas at sizable positive pressure and the liquid is subcooled(saturated at 2 bar and then when compressed to 6bar at the booster pump it get sub cooled for all practical purpose)


Shankarovitch.. Are you sure ?. I find the bolded part difficult to believe. There is definitely a booster pump/secondary system to feed the pre-burner with pumped fuel (i think a pure pressure fed loop there wont do), but I doubt the small booster delivers any fuel/oxygen to the main pumps, except maybe during start-up.Coz, if you do, you need to handle ALL the mass flow rate through the engine via the boosters and that will take a huge amount of energy/stored gas.

I think the booster will be used for startup and then bypassed completely once the engine starts is running.

Other wise what you are saying that it is really a staged pump with one stage being (with a pressure rise of 4 bar) being driven by secondary source, but handling the full mass flow through the engine which I find very hard to believe.

so I think it is quenching of the ignited gas that caused the failure because we do not have a high altitude simulation chamber for main engine
The approximate cost of such an infrastructure ma be 100 crore main cost going for the huge injector which has to pump out the hot light weight exhaust very quickly to maintain the low atmospheric pressure of few mbar as expected in the ignition altitude of 60 km and above

For C25 such is planned for and under implementation

[/quote]

Yeah. The flame died. The task is to find the root cause. One of the reasons could also be due to the flame going out in the pre-burner causing a shutdown of the turbo pumps due to some instability in the starting sequence/engine control system problem/ some failure in switching from start/booster to main pumps in the feeds, or choking due to cavitation/ ullage gas whatever . The ISRO failure analysis committee will examine every possible reason threadbare I am sure.

Good to hear about the high altitude facility under implementation. My relative retired from Isro a few years ago.. So I dont get those nice ISRO brochures and phamplets anymore... I am sure ISRO would have published a nice brochure of that facility if under implementation. Do you have it if such a thing exists ? Please post excerpts if you have it.

Shankar
BRFite
Posts: 1905
Joined: 28 Aug 2002 11:31
Location: wai -maharastra

Re: GSLV D3 Launch Failure

Postby Shankar » 19 Apr 2010 13:54

yes -the booster pump which is operated by stored gas is for start up purpose only so that main pump gets fluid at suffice ant pressure at starting phase in a sub cooled state and vapor lock in suction line do not take place .Later on the 2 bar pressure which is maintained in the tanks is sufficient to keep the main pump operating and at that high propellant flow the lines get enough chilling so no gas formation can take place in the suction line to the main pump

The propellant tank pressure is maintained by helium gas from stored high pressure cylinder which keeps the ulage space pressurized through out the duration of flight /usage

as regards the entire flow through booster pump in later stages of flight that is after the main pump is operating - I am not sure maybe it can handle just it idles around or is by passed -give me some time

The high altitude test facility is for C25 since i is a new engine altogether - where as the CUSP has you know Russian shadow so maybe the HAT facility was not incorporated in the infrestructure

more detailed post in a few hours

Shankar
BRFite
Posts: 1905
Joined: 28 Aug 2002 11:31
Location: wai -maharastra

Re: GSLV D3 Launch Failure

Postby Shankar » 19 Apr 2010 14:30

The booster pump is always on line the full flow goes through it and boosted to 6 bar before fed to main pump suction
Initially is operates from stored gas and then from a feed from gas generator

The ulage motor as mentioned in the thread has a different function altogether

as the rocket coasts along for some time before ignition of the cryo stage it is in an inclined position and the liquid inside can be in any position and so can the gas

When the stage is in operation the axial thrust keeps the liquid at bottom and gas at top during coasting that is after a stage separation the situation becomes bit uncertain

To avoid the booster pump suction getting a partial gas feed the ulage motors are fired just before the main engine ignition to provide some axial thrust in the same direction as motion so that the liquid and gas inside tank occupies the designated position and booster pump get full liquid feed

Flame quenching after ignition can take place because of

low oxygen in the propellant mix
high hydrogen in the propellant mix

faulty ignitor rockets (which are essentially small solid rocket motors which ignites and spews hot gas into the propellant mix in the combustion chamber ) to start the self sustaining combustion process

vina
BRF Oldie
Posts: 6046
Joined: 11 May 2005 06:56
Location: Doing Nijikaran, Udharikaran and Baazarikaran to Commies and Assorted Leftists

Re: GSLV D3 Launch Failure

Postby vina » 19 Apr 2010 15:01

The booster pump is always on line the full flow goes through it and boosted to 6 bar before fed to main pump suction

This part I still dont buy the story fully. It is fine during start up but not during flight. The pressure of the gases exiting the gas generator turbine has to be above the chamber pressure and it will be around the same as that of the main turbo pumps output . And for that it means that allowing for the pressure drop over the turbine of the gas generator, the flow in to the pre burner has to be HIGHER than the pressure of the feed into the main combustion chambers.

I think during flight, some flow will get diverted from the main pumps output and there probably is another booster that gets it to the required pressure to feed into the pre combustion chamber.

Initially is operates from stored gas and then from a feed from gas generator
Yes. Now it makes sense. I cant imagine running a pump that would require huge amounts of energy for the duration of the flight using secondary energy

I think with all these complex plumbings and switch over, any thing that is not properly done/ unanticipated can result in back pressure / pressure shocks, that can throw things out of kelter. Maybe there were some problems in the switchovers . I would put my money on the light in the preburner going out and the main turo pump simply stopping as a consequence.


Return to “Trash Can Archive”

Who is online

Users browsing this forum: No registered users and 16 guests