Hacker News Reader

Comment by foobarqux | Hacker News Reader

> Don't fret the failures.

Except when it kills people (though this accident didn't).

Fair point. Although I'd say: fret the suffering of people, not the fact that something really difficult ended up being difficult.

Not to slight your friend, but his was an unmanned launch to resupply the ISS, it's relatively common, comparatively easy, and successful (1 other failure in just under 80 missions).

No disrespect intended, but this is a disruption for the ISS, a loss (hopefully insured) for Planetary Resources, a real setback for Orbital... there's no point in sugar coating anything, it sucks. There's just no point in dwelling on it, either. RCA and move forward.

The failure rate in manned space missions is approximately 1/50 ("failure" as in "it explodes and everybody dies"). (For example, the Space Shuttle had 2/135 failures.)

The unmanned missions have a higher failure rate, IIRC ~10%, but that number probably includes prototype and old models.

So a 2/80 failures is not so bad. It sucks when it explodes, but going to space is difficult.

That 1/50 failure rate is debatable, and depends on your definition.

For example, Soviet/Russian manned flight, "Soyuz", had 131 flights (123 Soyuz + 2 Voskhod + 6 Vostok) [1][2] and 2 accidents, resulted in death of 4 astronauts (1 + 3).[3]

However, both of these accidents occurred before 1972, and both were highly experimental flights. Komarov's was the VERY FIRST flight of Soyuz, and before Soyuz 11 accident, they were descending without space suits, as the space inside the cabin was so tiny.

So you can also say that there were 2 accidents in the first 10 Soyuz flights - they were still finalizing the technology - and no accidents in the remaining 113.

In addition, both of these accidents occurred during the landing, so there was NO accidents during manned-space take-off in Russian/Soviet flights.

[1] http://en.wikipedia.org/wiki/List_of_Russian_manned_space_mi... [2] http://en.wikipedia.org/wiki/List_of_Soviet_manned_space_mis...

[3] http://en.wikipedia.org/wiki/List_of_spaceflight-related_acc...

> "there was NO accidents during manned-space take-off in Russian/Soviet flights."

Soyuz T-10-1 burned on the launch pad, although the launch escape system saved the lives of the cosmonauts on board.

http://en.wikipedia.org/wiki/Soyuz_7K-ST_No._16L

Thanks - I stand corrected, I didn't know about that accident.

Still, it wouldn't be a "failure" under GP's original definition: "failure" as in "it explodes and everybody dies" :)

It should also be noted that Komarov died in first manned Soyuz spacecraft launched - not the first Soyuz. In other words, Komarov's Soyuz-1 - and naming system isn't obvious in Russian space school - wasn't analogous to Columbia first flight as the very first flight of any STS.

And in Soyuz they also had other pretty close calls. Diving into athmosphere hatch-forward because engine module didn't separate in Shatalov-Volynov-Eliseev-Khrunov group flight... Near premature separation of said engine module in the Intercosmos flight with Afghani cosmonaut... Several ballistic landings - with accelerations up to about 20 g's...

Still nobody died. Perhaps in no small part because of overbuilding the ship and also changing the ship design along its long history, learning on mistakes. Soyuz spacecraft is the real workhorse.

There was another unsuccessful launch, when second stage rocket hasn't properly separated from the third. Everyone survived, but they have experienced pretty high acceleration and landed just 21 minutes after the launch.

http://en.wikipedia.org/wiki/Soyuz_7K-T_No.39

Also here is a nice video that shows: "How Soyuz Works".

https://www.youtube.com/watch?v=AVvgpKt5uCA

Basically how cosomaunts train in Star City to how the rocket is built, take off and all the way till it is about to dock with the ISS, including how launch escape works etc.

The reliability and resilience of the system is just a marvel of engineering.

Or when it costs billions of dollars from NASA.

The mission was insured, and it didn't cost "billions".

but it did cost millions to insurance company

If the insurance company is doing its job right, this was priced into its rates.

Then it cost millions to a lot of people.

Which is understood by all those people when they purchase insurance. This isn't news. Spreading the risk around is why people buy insurance in the first place.

Then the failure wasn't free. This customer and all other customers paid for this failure. Other customers, who may have done more and more expensive work to avoid failure, are stuck paying for this failure. Insurance thus incentivizes a race to the bottom -- using other customers to finance ones own risk.

Billions, not millions.

One launch is dozens of millions, not billions.

Insurance != free. Nor is paying a claim.

That means future cost of buying similar insurance will rise.

Hopefully they had Nationwide insurance because they have accident forgiveness policies. Your rates don't go up after your first launch accident.

are only cargoes insured or is the entire launch vehicle insured? I am curious as to how an insurance company assess risk in these launches, let alone what the starting point for any insurance would be relative to the value of the rocket and cargo.

NASA's CRS contract with Orbital is for 8 flights for $1.9 Billion. Today's flight was supposed to be number 3 of 8.

It remains to be seen if they will be able to complete the rest of their contract. The cause of the accident will need to be investigated, changes to future vehicles may be necessary, and right now it appears as though they damaged their launch pad severely.

The CRS contract includes development costs, though. The marginal cost of one mission isn't anywhere near 1.9B/8.

Sort of. COTS was supposed to be the development costs, CRS was supposed to be the launches. Orbital got $288 Million for development of Antares/Cygnus from COTS.

From wikipedia:

> "COTS is related but separate from the Commercial Resupply Services (CRS) program.[3] COTS relates to the development of the vehicles, CRS to the actual deliveries. COTS involves a number of Space Act Agreements, with NASA providing milestone-based payments. COTS does not involve binding contracts. CRS on the other hand does involve legally binding contracts, which means the suppliers would be liable if they failed to perform. Commercial Crew Development (CCDev) is a related program, aimed specifically at developing crew rotation services. It is similar to COTS-D. All three programs are managed by NASA's Commercial Crew and Cargo Program Office (C3PO)."

Personally, I am worried about Orbital's ability to complete the remaining flights for CRS. This is a very unfortunate setback for them.

Oh, true, my bad. I conflated COTS and CRS... Can't keep the acronyms straight. Not to mention CCDev vs CCtCaP, etc etc...

I think it's important to realize that this is a NASA contractor, not NASA itself. Furthermore, Congress & the President have forced NASA into purchasing these transport contracts by killing off the Space Shuttle.

A $200Million loss is better than the best of all possible Space Shuttle's outcomes.

The Space Shuttle which costs between $450 million [1] and $1.5 billion [2] per launch depending on who you ask and still blows up 1.5% of the time.

[1] http://www.nasa.gov/centers/kennedy/about/information/shuttl... [2] http://sciencepolicy.colorado.edu/admin/publication_files/re...

Reply on news.ycombinator.com