Commercial Crew - Discussion Thread 1

[Robotbeat]

His point wasn't that Dragon V1 met all of what NASA wants, it's that it could've certainly been done.

[enkarha]

Correct. Too bad Boeing didn't use that for their parachute drop tests.

What more would be learned from that that wasn't learned from using a boilerplate capsule?

You don't waste expensive hardware on potentially destructive tests unless you absolutely have to.

Tell that to SpaceX. They based their parachute drop test article around an actual pressure hull and dropped it from a helo. Twice.
Somehow that didn't bother them.

On the other hand you have Jim who became somewhat upset when QC suggested that the Boeing drop test article was constructed mainly of wood, sheet-metal and styrofoam. Big difference between a for-real pressure hull and a glorified wooden model.

What's the difference in a drop test?

[Space Ghost 1962]

You make some good points. But nevertheless, the CST-100 is only a mockup at this point as you pointed out and SpaceX has a capsule that has already been tested. So they are ahead in terms or real hardware. They have had a few years to plan out the life support issue.

To me, this is more than a mockup.

Correct. Too bad Boeing didn't use that for their parachute drop tests.

Wasn't the spacex drop test vehicle a modified v1?

Pretty much. Having seen it myself in person, it wasn't a mockup.

Ok, the idea here is high vs low fidelity representation. Yes, as QG earlier posted, you can use a ton of bricks for an extremely low fidelity test.

Such allows you to prove a subsytem, component, or assembly as being able to function under the intended capabilities but is not in the use case for qualification, and even further from certification .

Why we approach higher fidelity tests is that we find out more in the use case, and that historically has always brought surprises of small through large variety.

To boil this ocean storm of niggling posts down, my concerns of flying these three vehicles are as follows"
  1) CST100 - the vendor has a history, albeit past, of constructing, testing, ... certifying such a vehicle in the past. They then can be counted on to do so again, but because they did so not on a FFP before (as me jousting with Jim got at), the could go long on schedule, long on budget, and there were ways of bringing schedule in with longer costs, which is among. If we have an imperative to do so, then as YG has implied, its not really the intent of CC but more like Orion LEO only, which plays into Congress going "why are we funding two capsules". The reason is that they have held off actual flight hardware level high fidelity work, choosing to couple it with the program requirements verification/certification/other of the massive part of the program, such that in the flood with NASA they use own advantage of high rate of effective communication to achieve best effect of historical interaction - their advantage.
 2) Dragon - the vendor has a contemporaneous capsule program which is current and flying with up to date flight history on a valid HR vehicle. They seek to revise the vehicle can carry as much of the flight history, especially launch and entry validation into this. The key lack of abort validation is the biggest event they have to conquer. But they don't have any idea yet how much they are going to be fire hosed during qualification/certification and the response rate you can't fall behind on - a significant NASA concern. However, they have all along been doing high fidelity tests/fixtures, unlike others, so they come to the game with much better position of addressing schedule/program risk than even Boeing. The issue is about being able to communicate it effectively, and in addressing it such that NASA does not doubt the results. Since they don't trust past practices e.g. they are often skeptical of NASA/"old space" to euphemize here, the ability to accomplish this is more complicated as they wish to preserve more modern advantages. Which does not assure. Yet in the end may be more effective at bringing in schedule than others fear. Keep in mind that they can retest and get to certification a lot faster than Boeing can, due to starting on actual hardware many years earlier.
 3) DC - has the burdens of both of the above, but possesses significant advantages from being the most current of HSF vehicles being contemporaneously done with Shuttle. Meaning they are in a hybrid position with Boeing/SpaceX for in between reasons. My fear for them is that they underbid given all that there is to do, and that too much rides on too few tests, in essence like Boeing.

Bottom line - SpaceX will be fine even if NASA nervous, just like last time. Boeing will way exceed schedule/budget and get cut a "mulligan" or fifty. DC will have a "white knuckle ride" but can do it.

Oh, and I completely disagree with Jim. Sometimes when I buy on eBay/internet, the more trusted vendor disappoints - my strategy is to study real time feedback carefully. Jim, it's a "big data" world that didn't exist before - get with the program! The smarter you are at absorbing the fire hose of inbound data, the better the outcome. Wished we had this 20-30 years ago, fewer people dead, fewer bad decisions, and better schedules/budgets. But it doesn't come for free, very taxing to apply.

Judge now in aerospace by the application of oceans of data, where trickles/creeks ran before. 1,000,000x Shuttle/EELV.

[yg1968]

You make some good points. But nevertheless, the CST-100 is only a mockup at this point as you pointed out and SpaceX has a capsule that has already been tested. So they are ahead in terms or real hardware. They have had a few years to plan out the life support issue.

To me, this is more than a mockup.

I forgot about that image. Thanks for reminding me. They should have used that for their unveiling. 

[SWGlassPit]

Correct. Too bad Boeing didn't use that for their parachute drop tests.

What more would be learned from that that wasn't learned from using a boilerplate capsule?

You don't waste expensive hardware on potentially destructive tests unless you absolutely have to.

Tell that to SpaceX. They based their parachute drop test article around an actual pressure hull and dropped it from a helo. Twice.
Somehow that didn't bother them.

On the other hand you have Jim who became somewhat upset when QC suggested that the Boeing drop test article was constructed mainly of wood, sheet-metal and styrofoam. Big difference between a for-real pressure hull and a glorified wooden model.

That was an answer to *some* question.  It wasn't an answer to mine.  I'll repeat and clarify: for a parachute drop test, what is the benefit of dropping a (not cheap!) bare prototype pressure shell instead of a boilerplate capsule that more accurately simulates the mass and outer mold line of the actual flight article?

[ncb1397]

Btw, is not oxygen but CO2 poisoning one of the problems. The other being humidity control (with water extraction being, of course, the hard one).

20,000 PPM CO2 is considered safe. 70,000 PPM + is potentially fatal. Do the calculation on how long it will take 1 astronaut to generate that amount of CO2 in a 350 cubic foot cabin at STP.

[baldusi]

You make some good points. But nevertheless, the CST-100 is only a mockup at this point as you pointed out and SpaceX has a capsule that has already been tested. So they are ahead in terms or real hardware. They have had a few years to plan out the life support issue.

To me, this is more than a mockup.

Correct. Too bad Boeing didn't use that for their parachute drop tests.

Wasn't the spacex drop test vehicle a modified v1?

Pretty much. Having seen it myself in person, it wasn't a mockup.

Ok, the idea here is high vs low fidelity representation. Yes, as QG earlier posted, you can use a ton of bricks for an extremely low fidelity test.

Such allows you to prove a subsytem, component, or assembly as being able to function under the intended capabilities but is not in the use case for qualification, and even further from certification .

Why we approach higher fidelity tests is that we find out more in the use case, and that historically has always brought surprises of small through large variety.

If the mass simulator was correct, the outer mold line was correct, and the release mechanism was correct, what else is there to learn from using elements that will be passive?
The Dragon and Cygnus experience have shown that the pacing item has been software and interfaces certification. With human piloted crafts, that problem is compounded. Only other pacing item I can think of, is the LAS. But Aerojet tested and retested the engines even before SpaceX had fired their first SuperDraco. The pressure vessel testing article has a long time. And it does have some innovations. For example, it has no welds. It is formed out of a single billet of aluminum and then machined on the outside.
Oh! And Aerojet is trying to 3D print the LAS engine's Nozzle and MCC in just three parts. So there is a lot of innovation in CST. It just not on making things that might seem cool, but on actually lowering costs, reducing complexity and increasing reliability.
Sure, I might personally like SpaceX method better (I'm a BSD lover), but Boeing's approach is extremely professional. In fact, its CCDev1/2 and CCiCap performance have been the best performers.
And please stop spreading the argument about extra cost. It's a firm fixed price contract and nobody have their financial backing. And its clear from their price that they padded their numbers with a lot of margin. But if they had to actually put their own money, some executive's head might roll but they won't fault on a contract with Uncle Sam.

[arachnitect]

Btw, is not oxygen but CO2 poisoning one of the problems. The other being humidity control (with water extraction being, of course, the hard one).

20,000 PPM CO2 is considered safe. 70,000 PPM + is potentially fatal. Do the calculation on how long it will take 1 astronaut to generate that amount of CO2 in a 350 cubic foot cabin at STP.

No.

5,000 ppm is the maximum allowable average over 8 hours. Studies show that mental impairment starts closer to 2,000 ppm.

4 astronauts would have 5 or so hours more like 1 or 2 from when you pulled the ventilation line out at the launch pad. They should try not to move around much.

Just because the average NSF member is willing to go to space in a garbage bin with the lid taped on doesn't mean its a good idea.

[AncientU]

Btw, is not oxygen but CO2 poisoning one of the problems. The other being humidity control (with water extraction being, of course, the hard one).

20,000 PPM CO2 is considered safe. 70,000 PPM + is potentially fatal. Do the calculation on how long it will take 1 astronaut to generate that amount of CO2 in a 350 cubic foot cabin at STP.

As an x-submariner, I know the painful (screaming headaches) concentration is much below the 'fatal' level.  The reference below shows that this severe symptomatic threshold is 2,000ppm, with 1,000 ppm as normal indoor air.  So, do your calculation for a delta of 1,000 ppm instead of your suggested 50,000 (a factor of 50 more restrictive with one astro, 350 with seven).

The levels of CO2 in the air and potential health problems are:

    250 - 350 ppm – background (normal) outdoor air level
    350- 1,000 ppm - typical level found in occupied spaces with good air exchange.
    1,000 – 2,000 ppm - level associated with complaints of drowsiness and poor air.
    2,000 – 5,000 ppm – level associated with headaches, sleepiness, and stagnant, stale, stuffy air.  Poor concentration, loss of attention, increased heart rate and slight nausea may also be present.
    >5,000 ppm – this indicates unusual air conditions where high levels of other gases could also be present. Toxicity or oxygen deprivation could occur. This is the permissible exposure limit for daily workplace exposures.
    >40,000 ppm - this level is immediately harmful due to oxygen deprivation.

http://www.dhs.wisconsin.gov/eh/chemfs/fs/carbondioxide.htm

Edit: What arachnitect said.

[Robotbeat]

Btw, is not oxygen but CO2 poisoning one of the problems. The other being humidity control (with water extraction being, of course, the hard one).

20,000 PPM CO2 is considered safe. 70,000 PPM + is potentially fatal. Do the calculation on how long it will take 1 astronaut to generate that amount of CO2 in a 350 cubic foot cabin at STP.

so, 2% and 7%. I'll assume it's by mass, a conservative assumption. In 350 ft^3 at 1.2kg/m^3 density there is 11kg of air. Average person expels roughly 1kg of CO2 a day, so... 0.22 kg is the limit for safe, .77kg is limit for fatal. Better do fast rendezvous! Astronaut has just a bit over 5 hours at safe levels, and 18.5 hours before fatal levels. "Just" bring a few scuba rebreather scrubber cartridges and put them in front of the recirculation fans (which Dragon already has for ISS).

[ncb1397]

Btw, is not oxygen but CO2 poisoning one of the problems. The other being humidity control (with water extraction being, of course, the hard one).

20,000 PPM CO2 is considered safe. 70,000 PPM + is potentially fatal. Do the calculation on how long it will take 1 astronaut to generate that amount of CO2 in a 350 cubic foot cabin at STP.

so, 2% and 7%. I'll assume it's by mass, a conservative assumption. In 350 ft^3 at 1.2kg/m^3 density there is 11kg of air. Average person expels roughly 1kg of CO2 a day, so... 0.22 kg is the limit for safe, .77kg is limit for fatal. Better do fast rendezvous! Astronaut has just a bit over 5 hours at safe levels, and 18.5 hours before fatal levels. "Just" bring a few scuba rebreather scrubber cartridges and put them in front of the recirculation fans (which Dragon already has for ISS).

It is parts per million, so it is by volume or molar. For this purpose, they are interchangeable. If an astronaut consumes 19 cubic feet of oxygen per day, he generates 18 cubic feet of CO2 per day or  .75 cubic feet per hour. This represents a .75/350 hourly increase of the proportion of CO2 or 2142 ppm/hour.

To clarify a bit:
0-20,000: no noticeable effects or very little incumberance
20,000-70,000: symptoms of CO2 intoxication
70,000 +: CO2 poisoning, loss of consciousness, death, etc.

It will reach the 20,000 level in 9 hours and the 70,000 mark in 32 hours.

Soyuz' fast rendevous is 6 hours.

[spacetraveler]

If it's such an incredibly overwhelming priority to get an American crew launch option, they can stick some seats and suits in a cargo Dragon and launch in December.

Sorry but this has to be the most uninformed statement in this entire thread.

[Atomic Walrus]

Correct. Too bad Boeing didn't use that for their parachute drop tests.

What more would be learned from that that wasn't learned from using a boilerplate capsule?

You don't waste expensive hardware on potentially destructive tests unless you absolutely have to.

Tell that to SpaceX. They based their parachute drop test article around an actual pressure hull and dropped it from a helo. Twice.
Somehow that didn't bother them.

On the other hand you have Jim who became somewhat upset when QC suggested that the Boeing drop test article was constructed mainly of wood, sheet-metal and styrofoam. Big difference between a for-real pressure hull and a glorified wooden model.

That was an answer to *some* question.  It wasn't an answer to mine.  I'll repeat and clarify: for a parachute drop test, what is the benefit of dropping a (not cheap!) bare prototype pressure shell instead of a boilerplate capsule that more accurately simulates the mass and outer mold line of the actual flight article?

The benefit of using a higher fidelity structure is that you'd be able to gather experimental data on the effect of the landing on the structure.  Mechanical shocks aren't easy to model, so there could be some benefit there.  Of course,  a test engineer is always trying to balance the incremental benefit against the incremental cost.  If you've got spare structures around, might as well use one of them and get the benefit.  If you have to fabricate one and you don't expect to mitigate much risk with the test, you go with a cheaper route.  Probably also worth noting that using an actual pressure hull isn't necessarily a huge benefit if the intended design deviates significantly.

[ncb1397]

Btw, is not oxygen but CO2 poisoning one of the problems. The other being humidity control (with water extraction being, of course, the hard one).

20,000 PPM CO2 is considered safe. 70,000 PPM + is potentially fatal. Do the calculation on how long it will take 1 astronaut to generate that amount of CO2 in a 350 cubic foot cabin at STP.

As an x-submariner, I know the painful (screaming headaches) concentration is much below the 'fatal' level.  The reference below shows that this severe symptomatic threshold is 2,000ppm, with 1,000 ppm as normal indoor air.  So, do your calculation for a delta of 1,000 ppm instead of your suggested 50,000 (a factor of 50 more restrictive with one astro, 350 with seven).

The levels of CO2 in the air and potential health problems are:

    250 - 350 ppm – background (normal) outdoor air level
    350- 1,000 ppm - typical level found in occupied spaces with good air exchange.
    1,000 – 2,000 ppm - level associated with complaints of drowsiness and poor air.
    2,000 – 5,000 ppm – level associated with headaches, sleepiness, and stagnant, stale, stuffy air.  Poor concentration, loss of attention, increased heart rate and slight nausea may also be present.
    >5,000 ppm – this indicates unusual air conditions where high levels of other gases could also be present. Toxicity or oxygen deprivation could occur. This is the permissible exposure limit for daily workplace exposures.
    >40,000 ppm - this level is immediately harmful due to oxygen deprivation.

http://www.dhs.wisconsin.gov/eh/chemfs/fs/carbondioxide.htm

Edit: What arachnitect said.

I'm using submarine standards. The following PDF describes one sumbarine standard upper limit used by the Canadians of 1.75% or 17,500 ppm. Canadian Submariners need to think while an astronaut in a cargo vehicle would be doing absolutely nothing as it is controlled remotely. This scenario though applies to a dragon v2 without a crew life support system or to a vehicle with a broken or non-functioning life support system.

As seen in Figure 1, the first protocol was monitored without any purification over 13.25 hours
with a final average concentration in all of the compartments at the end of the protocol was
1.34%. Specifically, CO2 concentrations of 1.39% were seen in the motor room and WSC, 1.34%
in the senior accommodation space and control room, and 1.20% and 1.37% in the junior rates
accommodation space and junior ranks mess, respectively. This increase was not unexpected, as
the Standard suggests that with 50 crew a 1% CO2 concentration would be reached in 7.7 hours. It
was projected that with a crew complement of 59 that the upper limit of 1.75% would be reached
in 13.5 hours. This calculation has been based upon the prediction guidelines identified in BR
1326, whereby an initial concentration of 0.2% CO2 with no air purification is assumed. The
calculation, as identified in BR 1326, is also based also upon an average respiration rate of
24L/man/hour and a total breathable volume of 1129 m3 (39870 ft3).  The findings have shown
that after 13.25 hours under patrol conditions the recommended ceiling of 1.75% CO2 was not
reached, even without the aid of purification assistance.

http://www.dtic.mil/cgi-bin/GetTRDoc%3FAD%3DADA473000&rct=j&frm=1&q=&esrc=s&sa=U&ei=XR5AVPiKHsrCsATEuYDwDQ&ved=0CCgQFjAD&usg=AFQjCNHq0nb-a7j8WTyg2Tw_9zW4wmMs8w

[Space Ghost 1962]

You make some good points. But nevertheless, the CST-100 is only a mockup at this point as you pointed out and SpaceX has a capsule that has already been tested. So they are ahead in terms or real hardware. They have had a few years to plan out the life support issue.

To me, this is more than a mockup.

Correct. Too bad Boeing didn't use that for their parachute drop tests.

Wasn't the spacex drop test vehicle a modified v1?

Pretty much. Having seen it myself in person, it wasn't a mockup.

Ok, the idea here is high vs low fidelity representation. Yes, as QG earlier posted, you can use a ton of bricks for an extremely low fidelity test.

Such allows you to prove a subsytem, component, or assembly as being able to function under the intended capabilities but is not in the use case for qualification, and even further from certification .

Why we approach higher fidelity tests is that we find out more in the use case, and that historically has always brought surprises of small through large variety.

If the mass simulator was correct, the outer mold line was correct, and the release mechanism was correct, what else is there to learn from using elements that will be passive?

For proving a subsystem - nothing. For addressing systems qualification - everything. Qualification feeds certification.

During Shuttle, a lot was learned in the step up to high fidelity. In fact, if they had trusted this earlier instead of being overruled by the supposed insights of certain people, things would have turned out very differently. It was surprising the things that were found out so late in the program that had "impact".

The Dragon and Cygnus experience have shown that the pacing item has been software and interfaces certification. With human piloted crafts, that problem is compounded.

Neither firms software groups am I remotely comfortable with BTW. This I don't worry about with Boeing.

Only other pacing item I can think of, is the LAS. But Aerojet tested and retested the engines even before SpaceX had fired their first SuperDraco.

I have concerns about GNC. I have concerns about stability with both capsules. Trust SpaceX's models of control and stability more here.

The pressure vessel testing article has a long time. And it does have some innovations. For example, it has no welds. It is formed out of a single billet of aluminum and then machined on the outside.

This I have concerns for. Don't think that the materials and the tests are sufficient. Would have preferred the parachute/airbags as an all up test with representative pressure vessel. Duh. Look at Orion for "surprises" here.

Oh! And Aerojet is trying to 3D print the LAS engine's Nozzle and MCC in just three parts. So there is a lot of innovation in CST.

Yes. And they also could do additive manufacturing to improve cost/strength in pressure vessel / capsule other. I agree, these are good.

It just not on making things that might seem cool, but on actually lowering costs, reducing complexity and increasing reliability.

Where is the traceability of these through Apollo, the last capsule? Then I'd feel better about that. Not seeing it.

Sure, I might personally like SpaceX method better (...

For completeness in this interrogation, WTF do you mean exactly?

I'm a BSD lover ...

(You have no idea at all of what you think you are stepping on with that comment BTW )

...), but Boeing's approach is extremely professional.

Never said otherwise. Just like in past programs I'll mention below.

In fact, its CCDev1/2 and CCiCap performance have been the best performers.

It depends on how you score performance. And concern.

And please stop spreading the argument about extra cost. It's a firm fixed price contract and nobody have their financial backing. And its clear from their price that they padded their numbers with a lot of margin. But if they had to actually put their own money, some executive's head might roll but they won't fault on a contract with Uncle Sam.

First, I've never claimed extra cost. I've claimed concern over FFP. Different. For example, some of the NASA claimed flexibility could vanish for no added cost. Duh. Happened before with Shuttle - will not specify so don't ask.

I'm quite familiar with them, Shuttle, Delta IV, and about a hundred other programs. My concerns I have specifically addressed. They have never done a HSF vehicle that has launched/returned humans without overruns both in budget and time. Its true that they could prove this wrong. And my concerns are very specific and well considered. This is a far simpler vehicle. But so was Orion supposed to be. As LockMart goes, so often goes Boeing ... duh.

Do not underestimate how much things have changed in even the last decade. So much so that to adapt to current improvements, there is a tendency to avoid traceability by all. Or, to abandon current improvements, and to presume traceability that doesn't exist either. Both vex me in the extreme. Am I communicating?

[PreferToLurk]

Original discussion was locked down in the SpaceX forums, because it was highly off topic. As a lurker who is here almost all the time, I should know better than to respond to off topic with more off topic...   

Anyway, if they want to continue the discussion, they can do so here. 

What about X37?  started off NASA, transferred to DOD, now flying.  Good luck getting solid cost figures out of DOD, but its definitely operational.

Also, the vast majority of these programs were "X" projects.  Designed to push the envelope and develop new technologies.  These programs are expected to run into cost overruns and technical delays.  If they knew how to design and build them on the first try, they wouldn't be X projects. 

NASA isn't asking for new technologies here.  They are asking for a service.  Get crew from point A to point B.  Sierra Nevada basically proposed an X project, SpaceX proposed an advanced version of their cargo vessel, and Boeing proposed a (mostly) plain vanilla capsule.  Can you guess who scored the highest and the lowest?

Pointing out past delays and cost overruns that Boeing has had in X projects is nothing more than red herring.

PreferToLurk, I am curious how Dream Chaser equates to an X-project by your own definition.

Hybrid engines, skid landing gear, automated aerodynamic landings.  What we have heard so far about the selection is that Dream Chaser posed the highest technical risk.  Its nothing like a true X-project, but it was the least proven design. I should have been more clear originally.

If anything, Dragon 2 approaches your definition more than SNC’s Dream Chaser.
Things like:
-Integrated and reusable LAS made possible by powerful and compact 3D printed engines.

Already highly tested (on the ground), and despite the 3D printing, they are pretty simple pressure fed engines.

-Precision land landings (ultimately).

You said it yourself -- ultimately.  not part of the proposal. The proposal has Dragon landing with chutes.

-High redundancy landing options (including abort modes).

Wait, how does this increase technical risk?

-“Off the shelf” non rad-hardened polling/voting avionics with triple redundancy

Already demonstrated on cargo dragon.

-- snip --

[rayleighscatter]

They have never done a HSF vehicle that has launched/returned humans without overruns both in budget and time.

Do not underestimate how much things have changed in even the last decade.

You said it well enough on your own.

[QuantumG]

The CST-100 pressure vessel is just another component. So far Boeing hasn't integrated anything.

[RanulfC]

The CST-100 pressure vessel is just another component. So far Boeing hasn't integrated anything.

Well they have actually integrated the paperwork pretty efficently

Randy

[erioladastra]

The CST-100 pressure vessel is just another component. So far Boeing hasn't integrated anything.

From public information that is what people would conclude.