The Real Soyuz Problem - Looking Past the Smoke and Flames

#180 by mikers on 30 Apr, 2008 12:07
I would welcome a separate thread if needed, Clongton, no need for caution. It's all good as far as I am concerned. The central 'Why' is an as good question as any. I am a (self-described) space exploration fan, but I do have my notions. Best to you all.
#181 by synchrotron on 30 Apr, 2008 13:49
Sorry for initiating a digression. I have a pernicious tendency to do that. I think the on-topic aspect of my point was that an approach that has too much focus on the program metrics (specifically cost and schedule) can lead to systemic corner-cutting and oversights that lead to failures during the performance of the mission. Energia's spacecraft industrial capacity is, in the words of their management, entirely consumed with the mass production of Progress and Soyuz vehicles. They are operating without margin in their schedule in a process, although an assembly operation, is considerably less automated than the production of Toyota Camrys.
So, what is the response of a highly competent workforce when something goes wrong due to schedule pressures? Something that has worked (apparently) quite well in the past? I would hope that I would have the cojones to use the Asoh defense (http://en.wikipedia.org/wiki/Asoh_defence), but there is real pressure on individuals to dilute the blame by implicating others outside of one's organization. One slick way to do this is to appeal to a conspiracy of unnamed and possibly foreign individuals, media, and organizations that can never fully accept or deny the blame. The buck doesn't stop anywhere and everyone gets to keep their job. And the underlying problem does not get fixed.
The CAIB and the Rogers Commission are exemplary cases of performing the right kind of failure assessment with transparency. Good short-term outcomes if perhaps a little slow due to our level of risk aversion. But sustainable broad spectrum change in the day-to-day way we do space exploration has been elusive.
I'd like someone to tell RSA/Energia, "You're smart, you're good at what you do, you f**ked up. Calm down, we still want to work with you. Now tell me what went wrong, and what you're going to do about it."
#182 by Rusty_Barton on 30 Apr, 2008 16:36
South Korean Spacefarer Hospitalized With Back Pain
http://www.aero-news.net/index.cfm?ContentBlockID=a658dfc9-c659-42f9-8d01-f398e0fd329e

SKorea's first astronaut in hospital with back pain blamed on steep re-entry
http://canadianpress.google.com/article/ALeqM5hUyZ2vJ-IV9_AWX-mSt88bq8yBtQ
#183 by Blackstar on 30 Apr, 2008 16:57
Quote
synchrotron - 30/4/2008 8:49 AM
The CAIB and the Rogers Commission are exemplary cases of performing the right kind of failure assessment with transparency. Good short-term outcomes if perhaps a little slow due to our level of risk aversion. But sustainable broad spectrum change in the day-to-day way we do space exploration has been elusive.

When I worked CAIB we considered the Rogers Commission the _wrong_ way to do an investigation.

Yeah, yeah, I'm digressing...
#184 by daj24 on 30 Apr, 2008 17:04
More landing information from Yi So-yeon via an article on MSNBC.com. She recounts being "rescued" by nomads and how they helped the capsule occupants. http://www.msnbc.msn.com/id/24386004/
#185 by on 30 Apr, 2008 21:08
Quote
Rusty_Barton - 30/4/2008 11:36 AM South Korean Spacefarer Hospitalized With Back Pain http://www.aero-news.net/index.cfm?ContentBlockID=a658dfc9-c659-42f9-8d01-f398e0fd329e SKorea's first astronaut in hospital with back pain blamed on steep re-entry http://canadianpress.google.com/article/ALeqM5hUyZ2vJ-IV9_AWX-mSt88bq8yBtQ
Not only can you have contused nerves, another situation likely here is rib fractures and intercostal nerve pulls (they often make themselves apparent with severe back pain). This is very common (and hard to diagnose - can't really see on an X-ray). Worse yet, you can get diaphragm spasms that start with every breath.
So ballistic reentry is not a valid failsafe mode. More of a last ditch, like deciding if you want to eject and possibly kill/maim yourself. Really feel for all of them, would not wish an 8g 2minute load on my worst enemy.
#186 by Bruce H on 02 May, 2008 19:20
Ever considered writing an editorial yourself Chris?
#187 by Chris Bergin on 02 May, 2008 19:41
Quote
Bruce H - 2/5/2008 8:20 PM

Ever considered writing an editorial yourself Chris?

I don't think I'd be any good at it, to be honest. I'd prefer to continue to gain experience with my current style of reporting on space flight, which is a big learning curve on this subject matter.
#188 by Oersted on 04 May, 2008 17:56
Quote
Jorge - 30/4/2008 3:34 AM
However, the *overall* record of the Soyuz of bringing the crew back alive is equivalent to that of the shuttle. (2 fatal accidents in 98 flights, vs 2 in 122 for the shuttle; 4 fatalities out of 237 person-trips vs 14 out of 733 for the shuttle, 87 consecutive successful crew returns for Soyuz (Soyuz 12 to Soyuz TMA-11) vs 87 for the shuttle (STS-26 to 113)).

Though it must certainly count for something that the Soyuz losses happened in the very earliest "test flight phase" (as I've also noted in another thread). One can reasonably argue that the design has worked very well after the initial faults were ironed out.
#189 by Jorge on 04 May, 2008 18:01
Quote
Oersted - 4/5/2008 12:56 PM

Quote
Jorge - 30/4/2008 3:34 AM
However, the *overall* record of the Soyuz of bringing the crew back alive is equivalent to that of the shuttle. (2 fatal accidents in 98 flights, vs 2 in 122 for the shuttle; 4 fatalities out of 237 person-trips vs 14 out of 733 for the shuttle, 87 consecutive successful crew returns for Soyuz (Soyuz 12 to Soyuz TMA-11) vs 87 for the shuttle (STS-26 to 113)).

Though it must certainly count for something that the Soyuz losses happened in the very earliest "test flight phase" (as I've also noted in another thread).

It counts for something, but not as much as people like to think. It counts only inasmuch as the intervening years have allowed flight experience to accumulate.

Number of flights is the metric that matters, not number of years.

I find myself constantly dismayed that people are either unable or unwilling to grasp this simple concept.
#190 by synchrotron on 04 May, 2008 18:40
Quote
Jorge - 4/5/2008 5:01 PM

Quote
Oersted - 4/5/2008 12:56 PM

Quote
Jorge - 30/4/2008 3:34 AM
However, the *overall* record of the Soyuz of bringing the crew back alive is equivalent to that of the shuttle. (2 fatal accidents in 98 flights, vs 2 in 122 for the shuttle; 4 fatalities out of 237 person-trips vs 14 out of 733 for the shuttle, 87 consecutive successful crew returns for Soyuz (Soyuz 12 to Soyuz TMA-11) vs 87 for the shuttle (STS-26 to 113)).

Though it must certainly count for something that the Soyuz losses happened in the very earliest "test flight phase" (as I've also noted in another thread).

It counts for something, but not as much as people like to think. It counts only inasmuch as the intervening years have allowed flight experience to accumulate.

Number of flights is the metric that matters, not number of years.

Surely you are making use of hyperbole with this comment. Trend analyses are clearly valuable for extrapolating expected outcomes. At even probability, fatalities becoming more frequent over time is clearly worse than fatalities that are decreasing over time.

Likewise, the increasing frequency of Soyuz re-entry failures recently is a cause for concern.
#191 by Jorge on 04 May, 2008 18:50
Quote
synchrotron - 4/5/2008 1:40 PM

Quote
Jorge - 4/5/2008 5:01 PM

Quote
Oersted - 4/5/2008 12:56 PM

Quote
Jorge - 30/4/2008 3:34 AM
However, the *overall* record of the Soyuz of bringing the crew back alive is equivalent to that of the shuttle. (2 fatal accidents in 98 flights, vs 2 in 122 for the shuttle; 4 fatalities out of 237 person-trips vs 14 out of 733 for the shuttle, 87 consecutive successful crew returns for Soyuz (Soyuz 12 to Soyuz TMA-11) vs 87 for the shuttle (STS-26 to 113)).

Though it must certainly count for something that the Soyuz losses happened in the very earliest "test flight phase" (as I've also noted in another thread).

It counts for something, but not as much as people like to think. It counts only inasmuch as the intervening years have allowed flight experience to accumulate.

Number of flights is the metric that matters, not number of years.

Surely you are making use of hyperbole with this comment.

No. I don't think you understood my comment. And don't call me Shirley.

Quote
Trend analyses are clearly valuable for extrapolating expected outcomes.

Of course. But trends are measured over flights, not over time.

Quote
Likewise, the increasing frequency of Soyuz re-entry failures recently is a cause for concern.

Of course. But that is still a per-flight trend, not a per-time trend.
#192 by David BAE on 04 May, 2008 18:52
It must also count for something vs Soyuz that the two losses with shuttle were both faults that were not with the orbiter at the top of the fault tree.

Quote
Jorge - 4/5/2008 1:50 PM

And don't call me Shirley.

:laugh:
#193 by Blackstar on 04 May, 2008 18:56
Quote
synchrotron - 4/5/2008 1:40 PM
Surely you are making use of hyperbole with this comment. Trend analyses are clearly valuable for extrapolating expected outcomes. At even probability, fatalities becoming more frequent over time is clearly worse than fatalities that are decreasing over time.

Likewise, the increasing frequency of Soyuz re-entry failures recently is a cause for concern.

But you have to float the number, so that recent experience is much more heavily weighted than historical experience.

There was an interesting little hint at this in a recent article in Space News about a major launch insurance deal. A comsat company just bought a package deal for insurance for a number of upcoming launches. They are insured a different rates for different rockets (I'd have to go look up the specifics). Anyway, the worst deal is for the Proton. I believe that they're insured at a rate of 10.5% if they choose to launch on a Proton, meaning that they have to pay a premium of 10.5% of the overall value of the launch if they fly on the Proton. However, after seven successful Proton launches, the rate drops to 8.5%. That's the insurance industry's way of saying that they are taking into account recent experience over past historical experience.

Of course, when you're talking launch vehicles, none of this is absolute. There's a certain amount of perception involved in making these determinations, in part because it is so hard to put quantifiable values on things. When I was working on CAIB, I came up with the brilliant idea of doing a sidebar in the final report on launch vehicle reliability. But then I started talking with some people who had looked into this in the past and they said that it gets really difficult to make these kinds of calculations (heck, I talked to the USAF safety and mission assurance guys and got dispirited really fast). For instance, if you launch two vehicles in a row, but one of them is using updated flight software, do you count them as identical vehicles? At what point have modifications been introduced into the hardware (and software) to such a point that you should count it as a new version? And how do you even find out if those changes have been made? You can't, because a lot of that information is known only inside the companies building the rockets.

So beyond very crude categories, you cannot apply statistics to this kind of stuff. For all rocket launches N is a low number. You look at it and you apply common sense and gut instincts and say things like "the recent trend is not good..."
#194 by Spacenick on 04 May, 2008 21:06
Speaking of changes, it should be noted that the early Soyuz were very different from the current ones, unlike the Shuttle Soyuz made many incremental advances, especially Soyuz 1 lead to a lot of redesign.. However, that makes the last two incidents nothing less a cause of concern. On the other hand, this also means, that as long as Russia is willing to fund Soyuz flights (and I'm pretty sure in the event that it couldn't with the ISS still operational someone like Europe would pay) the likelyhood of loosing Soyuz as a vehicle is very small and I'm sure the current problem will be found as it might be caused by some gradual design change as well or maybe some kind of production problem.
#195 by extropiandreams on 05 May, 2008 00:29
First of all, a new vehicle wouldn't solve anything. problems are solved by flying and not by theory. There can't be an alternative because it's simply riskier to fly a new vehicle than a soyuz. assembly of new soyuz spacecraft are halted, and they are investigating it. but what's the big deal ? they cosmonauts survived and they are trying to correct it. the fact that the soyuz is able to survive such problems is just a proof of the systematic reliability of the system. they have enough money, they have quite a lot of experience, i would even say on par with the us, so what ? does anybody really believe that the dragon is more reliable ? or the shuttle ? perhaps they'll even send a extra soyuz up there, i've heard they are even discussing it. My audi a6 limited me to 40 mph last time on the high way, so such problems are happening everywhere. but does that that mean that german companies do not know what they do ?
#196 by Andrewwski on 05 May, 2008 00:58
Quote
extropiandreams - 4/5/2008 8:29 PM

First of all, a new vehicle wouldn't solve anything. problems are solved by flying and not by theory.

I don't think people are calling for a new vehicle. And problems are solved by flying AND theory. Applying theory during flight. I hate to keep referencing Challenger and Columbia, but were those problems solved by simply flying over and over? Nope. A lot of planning and redesign was put in after both accidents.

Quote
assembly of new soyuz spacecraft are halted, and they are investigating it. but what's the big deal ? they cosmonauts survived and they are trying to correct it.

What's the big deal? The propulsion module didn't separate, the capsule was pointed sideways, and the crew is fortunate they survived. If that propulsion module didn't separate, or the capsule didn't correct it's attitude, things could have gotten real bad real fast. And there still was a ballistic re-entry, something that shouldn't happen anyway.

And are they trying to correct it? To some extent, yes, I won't doubt that. But Perminov isn't giving me that feeling of confidence right now.

Quote
the fact that the soyuz is able to survive such problems is just a proof of the systematic reliability of the system.

Not in a million years. Three such problems among 12 TMA flights, and one toward the end of the TM program...all which are in recent memory? No way. Reliability would mean it would work correctly. It isn't, and we don't know that in the future this problem won't occur again, with more grave circumstances.

Quote
they have enough money,

Ha, nope. Money doesn't grow on trees.

Quote
they have quite a lot of experience, i would even say on par with the us, so what ?

Yeah, they have experience. We had experience with the shuttle before Columbia too. And did that prevent it?

Quote
does anybody really believe that the dragon is more reliable ? or the shuttle ?

That's not the point. Dragon doesn't even exist yet, there's no way to say. Comparing to the shuttle is comparing apples to oranges. The point is whether or not the appropriate measures are taken to ensure that this problem is actually solved. Because supposedly it was solved after TMA-10.

Quote
perhaps they'll even send a extra soyuz up there, i've heard they are even discussing it.

You "heard"? Where? And for what purpose? Is that purpose relative to this issue?

Quote
My audi a6 limited me to 40 mph last time on the high way, so such problems are happening everywhere. but does that that mean that german companies do not know what they do ?

This analogy makes no sense. I have no idea what you're even trying to say. There's a difference between your car being able to go over 40 mph and a capsule surviving descent from orbit at a speed of Mach 25. Again, apples to oranges.

I have no idea why you just want to shrug this off. Because that kind of attitude is the kind that will only lead to disaster.
#197 by Blackstar on 05 May, 2008 02:25
Russians considering spacewalk to investigate the Soyuz currently docked at ISS:

http://www.russianspaceweb.com/iss_soyuztma12.html#redocking

"A report on Novosti Kosmonavtiki web forum said that an emergency spacewalk by the station crew was considered to review the status of one of the pyrotechnic locks onboard Soyuz TMA-12, in case the ongoing investigation confirms that the device was a culprit in the failure of separation between the crew module and the propulsion module during the landing of the Soyuz TMA-11 spacecraft in May, as well the similar incident during the Soyuz-TMA-10 landing last fall."
#198 by hop on 05 May, 2008 05:06
Quote
Blackstar - 4/5/2008 7:25 PM

Russians considering spacewalk to investigate the Soyuz currently docked at ISS:

http://www.russianspaceweb.com/iss_soyuztma12.html#redocking

"A report on Novosti Kosmonavtiki web forum said that an emergency spacewalk by the station crew was considered to review the status of one of the pyrotechnic locks onboard Soyuz TMA-12, in case the ongoing investigation confirms that the device was a culprit in the failure of separation between the crew module and the propulsion module during the landing of the Soyuz TMA-11 spacecraft in May, as well the similar incident during the Soyuz-TMA-10 landing last fall."
According to previous statements from Gerstenmaier, a similar spacewalk was considered and rejected for TMA-11 after the TMA-10 incident.

See http://www.space.com/news/080422-nasa-russia-soyuz-update.html (previously mentioned )

Soyuz TM-9 had a somewhat eventful spacewalk to deal with damaged blankets on the DM: http://www.astronautix.com/flights/mireo6.htm
#199 by DmitryP on 05 May, 2008 10:10
Quote
Blackstar - 5/5/2008 6:25 AM

Russians considering spacewalk to investigate the Soyuz currently docked at ISS:

http://www.russianspaceweb.com/iss_soyuztma12.html#redocking

"A report on Novosti Kosmonavtiki web forum said that an emergency spacewalk by the station crew was considered to review the status of one of the pyrotechnic locks onboard Soyuz TMA-12, in case the ongoing investigation confirms that the device was a culprit in the failure of separation between the crew module and the propulsion module during the landing of the Soyuz TMA-11 spacecraft in May, as well the similar incident during the Soyuz-TMA-10 landing last fall."

Just to be clear it was not a report, but a post in the forum by one of the users.
http://www.novosti-kosmonavtiki.ru/phpBB2/viewtopic.php?p=309890#309890 . It actully said that the reallocation of Soyuz was cancelled due to the need to organize an unplanned EVA to dismantle (razborka) the pyro lock (in case it is confirmed that the same pyro lock malfunction on both TMA-10 and TMA-11. The official news report only sites Interfax news agency which is referring to the unspecified source in the space industry confirming the cancellation of reallocation based on the need to protect crew safety until the investigation is over.

PS: Shouldn't this discussion be moved to ISS-17 thread?