Pros and Cons, Dragon V2, CST-100 and Dream Chaser

[TrevorMonty]

SNC are lucky in that they have shuttle to learn from, hopefully the DC will be easier to reuse.

[Darkseraph]

SNC isn't vertically integrated for one. Costs them a lot to build a new Dreamchaser for that reason.

I think vertical integration is over-hyped as an advantage. Orbital is the complete opposite of being vertically integrated (it contracts nearly everything out to other companies), yet it only charged about 13% more than SpaceX to deliver the same amount of mass to the space station - and it had the advantage of being very well established, which means it can command a higher price than a startup like SpaceX.

Unless you're producing huge volumes of product, it actually don't make much sense to make everything in house. Basically, it means having to pay for huge amounts of expensive machinery and facilities that are not spread over many different high volume products.

[guckyfan]

I think vertical integration is over-hyped as an advantage. Orbital is the complete opposite of being vertically integrated (it contracts nearly everything out to other companies), yet it only charged about 13% more than SpaceX to deliver the same amount of mass to the space station - and it had the advantage of being very well established, which means it can command a higher price than a startup like SpaceX.

They do charge more and they don't offer the very valuable downmass SpaceX does.

[dglow]

SNC isn't vertically integrated for one. Costs them a lot to build a new Dreamchaser for that reason.

I think vertical integration is over-hyped as an advantage. Orbital is the complete opposite of being vertically integrated (it contracts nearly everything out to other companies), yet it only charged about 13% more than SpaceX to deliver the same amount of mass to the space station - and it had the advantage of being very well established, which means it can command a higher price than a startup like SpaceX.

That 13% delta is for CRS. What does Orbital quote for an Antares comsat launch?

[Ronsmytheiii]

I think vertical integration is over-hyped as an advantage. Orbital is the complete opposite of being vertically integrated (it contracts nearly everything out to other companies), yet it only charged about 13% more than SpaceX to deliver the same amount of mass to the space station - and it had the advantage of being very well established, which means it can command a higher price than a startup like SpaceX.

They do charge more and they don't offer the very valuable downmass SpaceX does.

They do offer higher payload volume than SpaceX, and if NASA ever wanted down mass from OSC they have a version of Cygnus that can do it. SpaceX only does down mass right now because it is "Free" for NASA as Dragon is a capsule anyhow.

Plus with the new commercial crew providers, there will be more down mass opportunities.

[dglow]

They do offer higher payload volume than SpaceX, and if NASA ever wanted down mass from OSC they have a version of Cygnus that can do it.

Hang on. Orbital has a Cygnus that returns to Earth... intact?

[arachnitect]

They do offer higher payload volume than SpaceX, and if NASA ever wanted down mass from OSC they have a version of Cygnus that can do it.

Hang on. Orbital has a Cygnus that returns to Earth... intact?

There's a proposal to do an inflatable heatshield demo on a Cygnus. Then perhaps evolve that demo into a return capability for Cygnus.

It's in the powerpoint stage.

Google "NASA HIAD"

[dglow]

I see. So Orbital 'has' a version of Cygnus that returns down mass in the same way I 'have' a BFR... running in KSP, on my laptop.   

[guckyfan]

My understanding was, that it is a NASA experiment on deceleration in the high atmosphere with Cygnus as the carrier vehicle. No connection to possible intact return of Cygnus.

[Nomadd]

 What the companies charge is what they thought they could bid and still get the contract. It doesn't mean that's what the missions cost them. Why would that say anything about the advantages of vertical integration? SpaceX could very well be spending less than half what Orbital does to get these missions up. That's no reason for them to bid less than they can get.

[Jarnis]

I see. So Orbital 'has' a version of Cygnus that returns down mass in the same way I 'have' a BFR... running in KSP, on my laptop.   

Well, I think their plans are closer to reality, but you do have a point.

[clongton]

Any discussion of Cygnus is off topic. Read the thread title.

[Burninate]

What is it about re-certification that makes it so complex that it becomes more efficient to build a new Dragon-V2 every time as opposed to a few and recycling them.

SpaceX and NASA are, if SpaceX is selected, going to have to come up with a certification process for a new Dragon V2.  Same as the other possible providers.  Just developing that process itself (not talking about executing on it) is going to be a complicated and lengthy endeavor.  To be able to include a reusable Dragon V2 as part of the CCtCAP proposal, a separate re-certification process would need to be included as a part of that proposal, or a modified version of the new capsule certification process.  Either would introduce substantial additional complexity in the CCtCAP timeline.

You "certify" that everything went alright, in industrial processes that are repetitive enough that you have well-established baselines for behavior.  That's not really the case with spaceflight.  The first capsule reflight test in the history of spaceflight is going to cost $100M+, as a baseline.  The second reflight test, another $100M.  Drawing statistically valid conclusions about failure rate and wear after 1 or 2 or 10 reflights is just not a thing you get to do, the numbers aren't there.

So instead of statistically valid empirical observation, we have to guess.  Nobody wants to guess, because if and when they're wrong, it's their head.  Guessing and then iterating frequently is the most efficient path, but then every change iterated ends the test of the old equipment and starts a new one with zero accumulated data.  In wartime , technologies move fast because there is money and accountability and willpower available to iterate frequently.  Not so much otherwise.

There's just not enough business, not enough flights, to do this sort of thing in a resilient manner at this time.  In one form or another, if we want a reliable vehicle, we have to commit to flying it *a lot* before we are mathematically able to understand that it's reliable, and how it ages.

NASA isn't willing to pay SpaceX to do that, to perform reentry testing-to-failure, the gold standard of engineering.  They are willing to pay SpaceX to construct a bunch of capsules in a controlled-environment assembly line, and fly each one individually... and it turns out this is way cheaper than what they've been doing up to this point, and for the moment cheaper than paying a private corporation to hit the accountability tennis ball from department to department until they have adequately diffused blame such that no individual's head is on the line in a reusable capsule failure, then paying a bunch of government departments to do the same, "checking" their work (see: Orion).  The reality is, there are too many unknown unknowns to simulate, and only testing and iteration can mostly eliminate rather than slightly reduce and launder our uncertainty.

When there are 10x as many flights as today, or 100x, then reusability economics really comes into its own, and this kind of testing is something you would be able to perform on things like bulk cargo often enough that nobody would miss the eventual failed flights.   Test to failure, see what failed, fix that, test to failure again, et cetera.

So reusability is a bit of an investment in the future.  Problem is, so is assembly line optimization, something a non-reusable capsule is going to do too, very well, when you present a demand for high flight rates; We are very efficient now at producing metal structures of roughly this shape for other purposes.   Perhaps we'll be in the 10^2 or 10^3 or 10^4 or 10^5 flights per year regime before that reaches its limits, I really can't say.

[clongton]

The first capsule reflight test in the history of spaceflight ... <snip>

I wouldn't be the first. Gemini-2 was reflown once. See http://www.space1.com/pdf/news1096.pdf
I believe at least one Soyuz was also reflown successfully, possibly twice, but I can't locate that reference at the moment.

[The man in the can]

The first capsule reflight test in the history of spaceflight ... <snip>

I wouldn't be the first. Gemini-2 was reflown once. See http://www.space1.com/pdf/news1096.pdf
I believe at least one Soyuz was also reflown successfully, possibly twice, but I can't locate that reference at the moment.

I don't know about the Soyuz capsule but TKS capsules were reflown.
http://www.astronautix.com/craft/tksva.htm
http://www.excaliburalmaz.com/0002_History.html

[Jimmy_C]

Yes but were they orbited twice while manned? AFAIK only the Shuttle has that distinction. Maybe Soyuz does too, but I haven't heard of any.

[vulture4]

As stated by M Sirangelo the DC has no blackzones, meaning it can abort anywhere in the launch and either land at a runway or achieve orbit.

To my untrained eyes I just found it not very convincing that DC can glide back to the launch site or the other side of Atlantic anytime during the ascent, regardless of LV failure modes. Both Shuttle and DC are unpowered gliders in the atmosphere. Even the orbiter, with its very high cross-range, has to do a long retro burn to RTLS.

I don’t know, but my gut feeling is that maybe he actually means DC can survive a ocean ditching, unlike the Shuttle.

How would it survive a water impact at 200Kt?

[Kryten]

Yes but were they orbited twice while manned? AFAIK only the Shuttle has that distinction. Maybe Soyuz does too, but I haven't heard of any.

No Soyuz has orbited twice, but the descent module from the 1983 launch failure (AKA Soyuz T-10-1) was refurbished and launched.