Commercial Crew Schedule Analysis

[docmordrid]

Space News adds some details

>
Boeing didn't elaborate on the nature of the problem, but other sources, including social media postings several days before the official statement, claimed that a hydrazine valve in the propulsion system failed to close properly at the end of the test, causing the propellant to leak.
>

Aerojet, in an October 2016 release about an earlier set of hot-fire tests of the thruster, touted the use of "innovative" valves in the launch abort engines.
>

From AJR's PR (linked above)

>
The LAEs, designed by Aerojet Rocketdyne, include a fuel valve and oxidizer valve, which were developed and tested under the company's Commercial Crew Transportation Capability (CCtCap) subcontract to Boeing.
>
"These innovative valves successfully enabled the engine to demonstrate precise timing, peak thrust control and steady-state thrust necessary during a mission abort. This testing culminates a year of dedicated hard work by the LAE Integrated Product Team at Aerojet Rocketdyne," said Aerojet Rocketdyne CEO and President Eileen Drake.
>

[Comga]

It seems to me both commercial crew vehicles have a more complicated path to achieving LOC numbers compared to shuttle since they have to stay attached to ISS for 6 months, no?  Shuttle had a smaller window of opportunity for damage in space.

And what did they say other than "Danger Will Robinson!"?
Or is that off topic for this discussion of schedule? 
If so, do you have a link to an appropriate discussion thread?

You could actually look at the document    Start reading on page 22, which is page 26 of the pdf file.

Thanks for the document. I did read it. It will take more than a virtual keyboard on a phone to compose a real comment.

However, just one first:
OMB found 4 elements of the review process have five ways (one applying different criteria to Boeing and SpaceX) for Loss of Crew probabilities to disparate and mostly arbitrary targets, and says that if the contractors don’t meet the goal, they can apply for a “waiver” with even less definition. It will be at that point that the program decides if human spaceflight is still worth significant risk.

It is odd that OMB, an organization formed to perform numerical analysis  (“Budget”) didn’t examine the single number at the center of the issue: 270. NASA wants the LoC <1/270. From where does that strange number come. (It was 1000 back at the start of Constellation IIRC.)  It appears that it’s twice the number of Shuttle flights. (2x135) setting the goal at exactly 4 times better than the track record of the Shuttle.

Has this been elucidated to the knowledge of anyone here?

An interesting aspect of the MMOD issue is that China has bolstered their position as one of only two countries that can launch humans by filling LEO with debris to a level that may be intolerable to others.

Doesn’t that make the use of the Shuttle number even MORE curious?
Unless it’s an amazing coincidence and the derivative even more obscure.

Here’s a suggestion: For every year of schedule delay the LOC target number can be recomputed downward because there will be that many fewer flights over the life of the ISS.

If commercial crew was certified in 2016 there would have been ~18 flights. If P(LoC)=1/270 a linearized model (OK for a small number of low probability trials. Please don’t start on this.) says the cumulative probability of one LoC is ~18/270=1/15.

At present certification may occur 5 years before the end of the planned life of the ISS. That means about 10 planned USCTV flights. With 10 flights the same cumulative risk could be had with a P(LoC) of 1/150. 

If certification delays the flights 2 more years  there will be only 6 flights. (Assuming they can keep the ISS going without these flights) so the same sum could be had with P(LoC)=1/90.

So schedule CONTRIBUTES to safety.
Inversely, it is the old lesson NASA knows well: Systems are safer the longer you DONT fly them.

[deruch]

https://directory.eoportal.org/web/eoportal/satellite-missions/s/space-fence - I strongly recommend if interested in MMOD. Space Fence is an ongoing project to upgrade the capability of tracking MMOD.



There are _lots_ more unobserved particles of course that this will not be able to track, but it's at least a start.

When I saw this graph I was struck by this thought: I wonder if the Chinese had not shot down Fengyun-1C, would the Dragon and Starliner spacecraft be meeting NASA's LOC/LOM requirements now?

The answer to this is trickier than you might think.  It depends on which orbital debris environment model Dragon and Starliner are using for LOC/LOM requirements.  The ORDEM2000 model was the established model at the time requirements were being crafted and wouldn't include debris from that event, while the ORDEM 3.0 model was crafted later and would.  Which environment would be written into a particular vehicle's requirements would depend on the result of negotiations between NASA and that particular provider.

Ultimately, it depends on how each vehicle's requirements are written.  Whether a particular environment model is reflective of reality is a separate discussion that doesn't have any bearing on meeting requirements.

Wasn't the Fengyun-1C in a polar orbit?  While there will be some minor overlap, I don't think that particular event will have very much impact on CC vehicles while at or going to/from ISS.

[edzieba]

It seems to me both commercial crew vehicles have a more complicated path to achieving LOC numbers compared to shuttle since they have to stay attached to ISS for 6 months, no?  Shuttle had a smaller window of opportunity for damage in space.

And what did they say other than "Danger Will Robinson!"?
Or is that off topic for this discussion of schedule? 
If so, do you have a link to an appropriate discussion thread?

You could actually look at the document    Start reading on page 22, which is page 26 of the pdf file.

Thanks for the document. I did read it. It will take more than a virtual keyboard on a phone to compose a real comment.

However, just one first:
OMB found 4 elements of the review process have five ways (one applying different criteria to Boeing and SpaceX) for Loss of Crew probabilities to disparate and mostly arbitrary targets, and says that if the contractors don’t meet the goal, they can apply for a “waiver” with even less definition. It will be at that point that the program decides if human spaceflight is still worth significant risk.

It is odd that OMB, an organization formed to perform numerical analysis  (“Budget”) didn’t examine the single number at the center of the issue: 270. NASA wants the LoC <1/270. From where does that strange number come. (It was 1000 back at the start of Constellation IIRC.)  It appears that it’s twice the number of Shuttle flights. (2x135) setting the goal at exactly 4 times better than the track record of the Shuttle.

Has this been elucidated to the knowledge of anyone here?

An interesting aspect of the MMOD issue is that China has bolstered their position as one of only two countries that can launch humans by filling LEO with debris to a level that may be intolerable to others.

Here’s a suggestion: For every year of schedule delay the LOC target number can be recomputed downward because there will be that many fewer flights over the life of the ISS.

The MMOD risk increases proportional to the duration a vehicle is in orbit, but is 'reset' every time that vehicle lands and is inspected/refurbished. The number of individual flights does not affect the per-flight probability.

[speedevil]

The MMOD risk increases proportional to the duration a vehicle is in orbit, but is 'reset' every time that vehicle lands and is inspected/refurbished. The number of individual flights does not affect the per-flight probability.

The proportionate increase is only if the heatshield is not covered in orbit, and is not monitored.

As one obvious example, monitoring for impacts and doing a once-over with the arm before release would reduce any additional risk, and a cover, such as the trunk of dragon 2 would remove any concern, other than the largest most easily noticed impacts.

The 'reset' argument misses the point unless it is cumulative effect of multiple impacts, because if you have one vehicle destroying damage inducing event every ten years, your total loss of crew does not go down if you've replaced the vehicles 50 times in the meantime.

It gets a lot safer if you monitor in orbit, and have a couple of ways down.

[SWGlassPit]

Wasn't the Fengyun-1C in a polar orbit?  While there will be some minor overlap, I don't think that particular event will have very much impact on CC vehicles while at or going to/from ISS.

A polar orbit ensures that the debris cloud that rains down through ISS altitudes carries sufficient encounter velocity to be very damaging.  Every orbital plane intersects every other orbital plane -- thus, every debris generation event has the ability to affect every operational vehicle in the appropriate altitude range.

[Jcc]

The MMOD risk increases proportional to the duration a vehicle is in orbit, but is 'reset' every time that vehicle lands and is inspected/refurbished. The number of individual flights does not affect the per-flight probability.

The proportionate increase is only if the heatshield is not covered in orbit, and is not monitored.

As one obvious example, monitoring for impacts and doing a once-over with the arm before release would reduce any additional risk, and a cover, such as the trunk of dragon 2 would remove any concern, other than the largest most easily noticed impacts.

The 'reset' argument misses the point unless it is cumulative effect of multiple impacts, because if you have one vehicle destroying damage inducing event every ten years, your total loss of crew does not go down if you've replaced the vehicles 50 times in the meantime.

It gets a lot safer if you monitor in orbit, and have a couple of ways down.

And of course the heat shield is covered on orbit for both Starliner and Dragon.

[edzieba]

The MMOD risk increases proportional to the duration a vehicle is in orbit, but is 'reset' every time that vehicle lands and is inspected/refurbished. The number of individual flights does not affect the per-flight probability.

The proportionate increase is only if the heatshield is not covered in orbit, and is not monitored.

As with the latter years of STS, just monitoring the heatshield does you little good if you need to use that heatshield at some point. CC vehicles (like any other vehicle docked at the ISS) act as 'lifeboats' for the duration of their stay. That means that if any MMOD is detected, you are not in a situation where a portion of the ISS crew has no evacuation route until an additional vessel can be launched.
Covering the heatshield reduces risk (though makes visual inspection more difficult, likely no effect on acoustic impact detection) but does not eliminate it.

[speedevil]

As with the latter years of STS, just monitoring the heatshield does you little good if you need to use that heatshield at some point. CC vehicles (like any other vehicle docked at the ISS) act as 'lifeboats' for the duration of their stay. That means that if any MMOD is detected, you are not in a situation where a portion of the ISS crew has no evacuation route until an additional vessel can be launched.
Covering the heatshield reduces risk (though makes visual inspection more difficult, likely no effect on acoustic impact detection) but does not eliminate it.

Implicit in the above replied to comment was that there was enough resource to sustain life until next vehicle could arrive.
Clearly, if you actually need to use it, and don't have it, you're screwed.

Covering the heatshield reduces the risk enough that the LOC number due to that is not within the same order of magnitude as 1:270.

[gongora]

Tweet from Brendan Byrne:

I'm told a visit next week to Kennedy Space Center by @VP is a "no go." He was expected to give an update on Commercial Crew program.

[edzieba]

Covering the heatshield reduces the risk enough that the LOC number due to that is not within the same order of magnitude as 1:270.

As far as I am aware, MMOD Risk is still the primary component of LoC risk for Commercial Crew (at least as of the second quarterly ASAP meeting).

[whitelancer64]

*snip*

It is odd that OMB, an organization formed to perform numerical analysis  (“Budget”) didn’t examine the single number at the center of the issue: 270. NASA wants the LoC <1/270. From where does that strange number come. (It was 1000 back at the start of Constellation IIRC.)  It appears that it’s twice the number of Shuttle flights. (2x135) setting the goal at exactly 4 times better than the track record of the Shuttle.

Has this been elucidated to the knowledge of anyone here?

*snip*

The 1/270 LOC risk number comes from: 3x better than the Shuttle LOC risk of 1/90.

Constellation's 1/1000 LOC risk was always unachievable and NASA quickly realized that. The target LOC risk number for Constellation was reduced a few times and ended up at 1/270 when the program was cancelled. That number was later brought in to the Commercial Crew program.

[SWGlassPit]

The MMOD risk increases proportional to the duration a vehicle is in orbit, but is 'reset' every time that vehicle lands and is inspected/refurbished. The number of individual flights does not affect the per-flight probability.

The proportionate increase is only if the heatshield is not covered in orbit, and is not monitored.

As one obvious example, monitoring for impacts and doing a once-over with the arm before release would reduce any additional risk, and a cover, such as the trunk of dragon 2 would remove any concern, other than the largest most easily noticed impacts.

The 'reset' argument misses the point unless it is cumulative effect of multiple impacts, because if you have one vehicle destroying damage inducing event every ten years, your total loss of crew does not go down if you've replaced the vehicles 50 times in the meantime.

It gets a lot safer if you monitor in orbit, and have a couple of ways down.

And of course the heat shield is covered on orbit for both Starliner and Dragon.

The base heat shield, yes, well, mostly.  The entire outer mold line of the vehicle is a heat shield, though.  Even a large enough penetration of the backshell could be catastrophic if it allows ingestion of hot reentry gases.

[gongora]

Tweet from Michael Sheetz:

The event August 3 announcing updates to NASA’s @Commercial_Crew program moved to Johnson Space Center from Kennedy Space Center, a KSC source tells me.

It is unclear if @VP Pence will still attend.

Tweet from Eric Berger:

He will not.

[SMS]

https://www.nasa.gov/press-release/nasa-to-name-astronauts-assigned-to-first-boeing-spacex-flights

NASA to Name Astronauts Assigned to First Boeing, SpaceX Flights

NASA will announce on Friday, Aug. 3, the astronauts assigned to crew the first flight tests and missions of the Boeing CST-100 Starliner and SpaceX Crew Dragon, and begin a new era in American spaceflight. NASA Administrator Jim Bridenstine will preside over the event, which will begin at 11 a.m. EDT on NASA Television and the agency’s website.

NASA will announce the crew assignments for the crew flight tests and the first post-certification mission for both Boeing and SpaceX. NASA partnered with Boeing and SpaceX to develop the Starliner spacecraft to launch atop a United Launch Alliance Atlas V rocket and the Crew Dragon launching atop the Falcon 9 rocket, respectively.

{...}

[DigitalMan]

I hope they will be able to give us at least expected dates for Boeing/SpaceX flights 1 and 2, rather than the placeholders that have been published.

[Comga]

I hope they will be able to give us at least expected dates for Boeing/SpaceX flights 1 and 2, rather than the placeholders that have been published.

Hope all you want, but they clearly excluded what everyone wants to know.
How significant is which of the four astronauts gets assigned to which of the vehicles? 
Some of us have preferences.  In the end, though, we are much more interested in the vehicles. 
This appears to be another attempt to convey a sense of progress without anything really happening.

[gongora]

I hope they will be able to give us at least expected dates for Boeing/SpaceX flights 1 and 2, rather than the placeholders that have been published.

Hope all you want, but they clearly excluded what everyone wants to know.
How significant is which of the four astronauts gets assigned to which of the vehicles? 
Some of us have preferences.  In the end, though, we are much more interested in the vehicles. 
This appears to be another attempt to convey a sense of progress without anything really happening.

Maybe we should wait until after the event to complain about what they did/didn't say.

[Jcc]

According to the latest plans ( before the coming announcement) NASA is expecting Boeing to fly first, so crew announced for Boeing would fly first. But what if issues like the propellant leak delay Boeing and SpaceX is first, so that crew would go to the Station first. It would seem prudent to train all 4 astronauts on both vehicles, so they have flexibility on crew assignments.

[SWGlassPit]

According to the latest plans ( before the coming announcement) NASA is expecting Boeing to fly first, so crew announced for Boeing would fly first. But what if issues like the propellant leak delay Boeing and SpaceX is first, so that crew would go to the Station first. It would seem prudent to train all 4 astronauts on both vehicles, so they have flexibility on crew assignments.

Bit of friendly advice: don't put any stock in what any publicly released launch dates say regarding who flies first.  They're all still NET dates at best.  The answer won't be clear until you get much closer to the actual launch date.