ULA almost Kerbals it on Cygnus launch

[GeneCash]

The Atlas V first stage cut off 6 seconds short, and the Centaur thus had to burn another minute and two seconds.

Also, the first stage RD-180 went to a fuel rich burn which dropped the ISP about 2.5% while raising the thrust. This means the Centaur got dropped off "low and slow" and it had to pitch way up to fight gravity losses.

The Centaur was supposed to do an 11 second half-throttle retrofire to deorbit, and end up having a couple tons of propellant left over. Instead, it ran out of propellant 3 seconds into the burn.

This means it had about 1.08 seconds of burn time left at MECO. If the first stage had shut down 1 second earlier, Cygnus wouldn't have made the planned orbit. If it had shut down 1.3 seconds earlier, perigee would have stayed in the atmosphere. Oops.

See the numbers here: http://spaceflight101.com/by-the-numbers-how-close-atlas-v-came-to-failure-in-this-weeks-cygnus-launch/

More info at: http://spaceflight101.com/potential-performance-hit-suffered-by-atlas-v-a-closer-look-at-the-data/

This also means MUOS-5 gets delayed a week while ULA and the Navy go WTF? http://www.spaceflightinsider.com/organizations/ula/oa-6-anomaly-causes-delay-muos-5-mission/

Just goes to show even the big boys using full MechJeb have their off days...

 

[fredinno]

2 hours ago, GeneCash said:

The Atlas V first stage cut off 6 seconds short, and the Centaur thus had to burn another minute and two seconds.

Also, the first stage RD-180 went to a fuel rich burn which dropped the ISP about 2.5% while raising the thrust. This means the Centaur got dropped off "low and slow" and it had to pitch way up to fight gravity losses.

The Centaur was supposed to do an 11 second half-throttle retrofire to deorbit, and end up having a couple tons of propellant left over. Instead, it ran out of propellant 3 seconds into the burn.

This means it had about 1.08 seconds of burn time left at MECO. If the first stage had shut down 1 second earlier, Cygnus wouldn't have made the planned orbit. If it had shut down 1.3 seconds earlier, perigee would have stayed in the atmosphere. Oops.

See the numbers here: http://spaceflight101.com/by-the-numbers-how-close-atlas-v-came-to-failure-in-this-weeks-cygnus-launch/

More info at: http://spaceflight101.com/potential-performance-hit-suffered-by-atlas-v-a-closer-look-at-the-data/

This also means MUOS-5 gets delayed a week while ULA and the Navy go WTF? http://www.spaceflightinsider.com/organizations/ula/oa-6-anomaly-causes-delay-muos-5-mission/

Just goes to show even the big boys using full MechJeb have their off days...

 

This is why you always need extra margin

[GeneCash]

55 minutes ago, fredinno said:

This is why you always need extra margin

Well, ULA has always made a huge deal every Cygnus launch saying "we've got so much extra margin, we've got a full hour window! Unlike those SpaceX losers!"

So this time they pretty much used it all. I'm hoping some sort of post-mortem comes out, but I'm also pretty sure it'll be kept under wraps as proprietary or ITAR secret information.

Also, it's interesting almost no one is covering this story. As far as I know, Spaceflight 101 is the only one that went "hey waitaminnit, those numbers don't match! Summat's WRONG!" and did any digging. I'd love to see URLs of other stories on this.

[fredinno]

43 minutes ago, GeneCash said:

Well, ULA has always made a huge deal every Cygnus launch saying "we've got so much extra margin, we've got a full hour window! Unlike those SpaceX losers!"

So this time they pretty much used it all. I'm hoping some sort of post-mortem comes out, but I'm also pretty sure it'll be kept under wraps as proprietary or ITAR secret information.

Also, it's interesting almost no one is covering this story. As far as I know, Spaceflight 101 is the only one that went "hey waitaminnit, those numbers don't match! Summat's WRONG!" and did any digging. I'd love to see URLs of other stories on this.

NASASpaceflight covered it. But the thing is, no one really cares that much since

1. It's not SpaceX

2. It wasn't even a partial failure- the mission was still successful.

[DonaldPK51]

There might be a blurb in one of SpaceflightNow's live coverage updates citing an "anomaly" but considering they're mostly "in bed" with ULA, I'm not surprised they didn't cover it. They run ULA and Orbital ATK ads on a frequent basis, too.

[kiwi1960]

OK...... who at Squad was behind this launch?

WAIT,.... I know, it wasn't a Monkey, it was either a famous person or a KSPTV dude.... Scott Manley or E.J.

I reckon it couldn't have been Scott.... he is (ahem) way too perfect.... (ok, stop laughing....)

So it was probably E.J. ... probably attached another 200 parts to the payload....

 

 

Edited March 28, 2016 by kiwi1960

[Scotius]

Well, stuff happens Most important part is that bird is in space and able to reach its target.

[YNM]

Really ? Hmm... I wondered whether the second stage burn was too long. Now it's confirmed...

[Dfthu]

I was watching the livestream and i thought the centaur was pitching too high up. Its a good thing it worked though!

[cantab]

This calculation shows that Centaur had to generate 3.1m/s of delta-v for every 1m/s of CCB deficiency when only looking at the early engine cutoff.

Interesting. In KSP I've often found that making the lower stage a bit smaller and the upper stage a bit larger helps the overall payload fraction. But from this it shows there's a limit, since the upper stage may end up wasting a disproportionate amount of delta-V. It also shows something we know but sometimes forget - that rocket design and ascent profile affect dV to orbit, it's not a fixed figure.

[Theysen]

20 minutes ago, cantab said:

 

Interesting. In KSP I've often found that making the lower stage a bit smaller and the upper stage a bit larger helps the overall payload fraction. But from this it shows there's a limit, since the upper stage may end up wasting a disproportionate amount of delta-V. It also shows something we know but sometimes forget - that rocket design and ascent profile affect dV to orbit, it's not a fixed figure.

This is based on the very low TWR modern upper stages have in order to make the whole rocket more efficient. Therefore you need a higher release velocity delivered by the core stage so the upper stage doesn't have to fight against gravity that much and does not need to waste that huge amount of fuel to fight gravity. Remember, real scale earth needs ~ 7.8km/s for orbital velocity so you don't just fall back to earth. This is a lot of velocity the Centaur had to compensate for.

 

Anyway. Sounds like someone tried TestFlight at ULA

Edited March 28, 2016 by Theysen

[magnemoe]

3 hours ago, cantab said:

 

Interesting. In KSP I've often found that making the lower stage a bit smaller and the upper stage a bit larger helps the overall payload fraction. But from this it shows there's a limit, since the upper stage may end up wasting a disproportionate amount of delta-V. It also shows something we know but sometimes forget - that rocket design and ascent profile affect dV to orbit, it's not a fixed figure.

This is true, an constraint in KSP is the small planet diameter who demands pretty high TWR if you use an small first stage, In my experience you want to keep twr above 0.5 if you go for upper stage below 1800 m/s. Lower and you will start to deorbit. 
On earth you have more time but also need more dV

[kerbiloid]

Probably somebody occasionally dragged the maneuver node with  mouse.

[fredinno]

5 hours ago, cantab said:

 

Interesting. In KSP I've often found that making the lower stage a bit smaller and the upper stage a bit larger helps the overall payload fraction. But from this it shows there's a limit, since the upper stage may end up wasting a disproportionate amount of delta-V. It also shows something we know but sometimes forget - that rocket design and ascent profile affect dV to orbit, it's not a fixed figure.

That's pretty much what Falcon 9 does, which is why it's able to get so much more Delta V out of a small rocket- Atlas uses the Centaur even though it is underpowered, as RL-10s are expensive, and it likely saved a lot on development cost, not designing an entirely new upper stage...

[ModZero]

So, this is the year for almost-failing it? (Referring to launch pad aborts of a Soyuz and Falcon, and the Briz-M that blew up instead going to the graveyard like a good zombie).

I'm going to take this over last year's "everything explodes", though (I know, hyperbole).

[kerbiloid]

A glass is half full or half empty depending on is it SpaceX glass or not.

6 seconds less worked the 1st stage, 60 seconds more worked the 2nd stage, the payload is successfully inserted into its orbit. Almost fail.

One of six (or how many) stage landings finished successfully (and nobody knows whether that stage can be repaired), the next one crashed the barge. Almost success.

[Scotius]

And your point is...? Launching a satellite\spacecraft\probe is a well known, many times repeated process. Heck we are doing it for 50+ years now. If something goes wrong, it means someone made a blunder somewhere. Landing returning stage using its own power? No one tried this before. Everyone expect many failures - just like many early launches failed. We are glad for every little success that comes along. Even stage smashing into a barge and messing things up brings valuable experience and data.

[kerbiloid]

I'm afraid, the stage landing experiments were lasting since early 60s, yet with the same luck: yet none of them flied again.

(Except of Shuttle components of course, but that's another story.)

Edited March 29, 2016 by kerbiloid

[magnemoe]

3 hours ago, kerbiloid said:

I'm afraid, the stage landing experiments were lasting since early 60s, yet with the same luck: yet none of them flied again.

(Except of Shuttle components of course, but that's another story.)

What other non shuttle stage landings has gone on except probe landers on Moon and other bodies?
Yes it has been plenty of plans nothing more, 

[kerbiloid]

I mean, reusable boosters.
Landers, NASA's Morpheus , reusable Gemini and TKS re-entry modules are another thing.

[wumpus]

23 hours ago, fredinno said:

That's pretty much what Falcon 9 does, which is why it's able to get so much more Delta V out of a small rocket- Atlas uses the Centaur even though it is underpowered, as RL-10s are expensive, and it likely saved a lot on development cost, not designing an entirely new upper stage...

One big advantage in KSP is that the terrier and poodle rockets are cheap, light, and extremely efficient (how often IRL can you choose all three?  In KSP you only lose on high thrust).  You can load them up with delta-v until you run into issues of having enough TWR to circularize before heading past aposis (less of a problem with shallow launches and low TWR).  

Centaur is LH/LOX, so has spectacular efficiency at quite a cost.  It costs less to heave it more delta-v than to redesign it, so that's what they do.

[fredinno]

17 hours ago, kerbiloid said:

I'm afraid, the stage landing experiments were lasting since early 60s, yet with the same luck: yet none of them flied again.

(Except of Shuttle components of course, but that's another story.)

The Falcon 9 1st stage is a lot longer and more difficult to land.

[kerbiloid]

5 hours ago, fredinno said:

The Falcon 9 1st stage is a lot longer and more difficult to land.

I'm sure, it is.

But dry facts are: the payload of Atlas V is successfully delivered (and that's almost fail), yet none of the rocket stages have been reused again (and that's almost success).

That's what I call: SpaceX glass is always half full, any another one - half empty.

[Frybert]

Folks this thread is about the engine shutdown yet successful launch of ORB CRS-4. If you wish to discuss SpaceX's attempts to recover stages please post in the helpfully named SpaceX Vehicle Recovery and Reusability Thread.

[Mitchz95]

On 2016-03-28 at 4:38 PM, kerbiloid said:

A glass is half full or half empty depending on is it SpaceX glass or not.

6 seconds less worked the 1st stage, 60 seconds more worked the 2nd stage, the payload is successfully inserted into its orbit. Almost fail.

One of six (or how many) stage landings finished successfully (and nobody knows whether that stage can be repaired), the next one crashed the barge. Almost success.

I'm pretty sure that if the same thing happened to a Falcon 9, we would be calling an Almost Fail just as we are now.

And if ULA were the ones trying to land rocket stages, we would be cheering on every Almost Success just as we are now.

Anyway, the fact that Atlas V was able to recover from this says a lot about the quality of their rocket.

Edited March 30, 2016 by Mitchz95