Blue Origin Thread (merged)

[Serpens Solidus]

Because chutes don't work on planets without an atmosphere...

[PB666]

7 minutes ago, Serpens Solidus said:

 

They have a point where the nominal deceleration forces or the landing accelerations are going to converge on each other at a critical point. If you think about it the sooner they fire the less drag force has time to operate, and the more hoover time steals thrust, but if they wait too late, then they are going have super high g-force accelerations which will destroy the rocket and the margin of error on landings go way down, to the millisecond range, to the point were wave action will either spare the core or destroy it just by changing height a meter or so. I don't know if this is their game plan or not, but looks like they are heading in that general direction.

Edited June 15, 2016 by PB666

[Kartoffelkuchen]

2 minutes ago, icedown said:

Why don't they use chutes to assist with the slow down? I would think that they would be worth more than their weight in propellant

Because chutes are heavy and cut into payload mass and are expensive. 

Because chutes cannot be "rapidly reused", as SpaceX wants it.

Because you can't efficiently steer with parachutes towards the landing location, which is required for reusability, at least on GTO missions where you have to land on a "tiny" platform in the ocean.

 

[Rdivine]

Just now, Kartoffelkuchen said:

Because chutes are heavy and cut into payload mass and are expensive. 

Because chutes cannot be "rapidly reused", as SpaceX wants it.

Because you can't efficiently steer with parachutes towards the landing location, which is required for reusability, at least on GTO missions where you have to land on a "tiny" platform in the ocean.

 

Why dont they build detachable chute modules and attach it on the side of the rocket? As it opens the chute, the ASDS will steer itself under the rocket instead. Problem solved. (Just kidding, its a kerbal idea).

[PB666]

56 minutes ago, Serpens Solidus said:

Because chutes don't work on planets without an atmosphere...

Fortunately earth does. The reason is because the rocket is traveling at an excessive speed with a high vertical component, too speedy to deploy until the landing burn has initiated.
They could make the diverter planes larger and extendable to increase drag. Thats about it.

You have to remember one thing, that rocket it plowing through variable upper level winds and surface winds. The slower it goes through there, the more off-target it will be, the faster it goes the closer its natural trajectory will be to the barge, if you could have drogue chutes slow it speed down greatly then what happens is the upper level ambient drift requires more lateral burn fuel to correct for any off course and equipment such as RCS to keep course.

 

Edited June 15, 2016 by PB666

[KerBlammo]

So out of 4 GTO landing attempts 2 were successful (yay!) and 2 were not (boo!).  Both of the failures were apparently due to low thrust on an engine.  The GTO landings involve firing 3 engines during part of the landing burn (as opposed to 1 in the LEO landings).  This seems to indicate that the Merlin engines as currently designed have a fairly high failure rate after being restarted and are not (yet) fully reusable.

[Green Baron]

8 minutes ago, KerBlammo said:

So out of 4 GTO landing attempts 2 were successful (yay!) and 2 were not (boo!).  Both of the failures were apparently due to low thrust on an engine.  The GTO landings involve firing 3 engines during part of the landing burn (as opposed to 1 in the LEO landings).  This seems to indicate that the Merlin engines as currently designed have a fairly high failure rate after being restarted and are not (yet) fully reusable.

I'd rather put my bet on they calculated to the last drop of fuel + x. Or the high g-loads have an impact on the fuel system.

Could someone do the math for me, pls: 3 Merlins, sealevel, 25(don't know were i have that from pls. verify ?) tons = many g's.

 

[kerbiloid]

http://spaceflight101.com/spacerockets/falcon-9-v1-1-f9r/

3 * 654000 / (25000 * 9.81)= 8 g

[benjee10]

16 minutes ago, KerBlammo said:

So out of 4 GTO landing attempts 2 were successful (yay!) and 2 were not (boo!).  Both of the failures were apparently due to low thrust on an engine.  The GTO landings involve firing 3 engines during part of the landing burn (as opposed to 1 in the LEO landings).  This seems to indicate that the Merlin engines as currently designed have a fairly high failure rate after being restarted and are not (yet) fully reusable.

Except the Merlins go through a static fire test just days before launch and then perform perfectly, and IIRC the first landed stage was static fired successfully with the engines performing nominally. I imagine it's not the engines failing but some problem with the propellant feed - I'm sure starting up an engine while the entire booster is decelerating is quite a challenge in terms of what it must do to fuel distribution (especially when there's so little left in the tanks - must be sloshing around everywhere)

[wumpus]

41 minutes ago, Serpens Solidus said:

Because chutes don't work on planets without an atmosphere...

Falcon9 is only planned to work on Earth.  As far as I know, SpaceX has no plans to land on planets with zero atmosphere (Mars has .5%? of Earth's).  We go through this many times (and SpaceX started with trying to use parachutes on Falcon[1]).  For any break between 0-100% parachute landing, SpaceX figures on 100% retrorockets, 0% parachutes.  Some things they have done to improve vertical landing:

Paddle controls:  Pretty much real life a real-life cross between the KSP AV-R8 and AIRBRAKES part.  I wonder if they might make them bigger for these "hotter" landings.  Expect any additional aerodynamic means to slow descent will involve something like this.

Three engine descent: A single engine should have a TWR ~3, meaning significant gravity losses.  Three engines stopping the rocket on a dime will have much less gravity losses, at least until they turn the descent over to a single engine for the final landing.  I'm guessing they still need to work out fuel margins, and that is what doomed today's launch (cutting from three engines to one).

The Soyuz combines both parachute and retrorockets.  I suspect the parachute is sized such that the cosmonauts can survive without the retrorocket, leaving a parachute so large the thing often drifts *way* off target.  Falcon9 routinely hits the barge, often where "x marks the spot".

[PB666]

47 minutes ago, KerBlammo said:

So out of 4 GTO landing attempts 2 were successful (yay!) and 2 were not (boo!).  Both of the failures were apparently due to low thrust on an engine.  The GTO landings involve firing 3 engines during part of the landing burn (as opposed to 1 in the LEO landings).  This seems to indicate that the Merlin engines as currently designed have a fairly high failure rate after being restarted and are not (yet) fully reusable.

They push each engine past the rating, and there is no margin for error at those rates of deceleration, even off by 5% on one engine is a crash.

[sojourner]

1 hour ago, icedown said:

Why don't they use chutes to assist with the slow down? I would think that they would be worth more than their weight in propellant

You'd be wrong.

[icedown]

Tired of the flame

Edited June 15, 2016 by icedown

[KerBlammo]

24 minutes ago, benjee10 said:

Except the Merlins go through a static fire test just days before launch and then perform perfectly, and IIRC the first landed stage was static fired successfully with the engines performing nominally. I imagine it's not the engines failing but some problem with the propellant feed - I'm sure starting up an engine while the entire booster is decelerating is quite a challenge in terms of what it must do to fuel distribution (especially when there's so little left in the tanks - must be sloshing around everywhere)

True enough for the most part.  I should have added that "assuming the low thrust was not due to propellant starvation... " blah blah blah.

As far as the 1st landed refire goes, there were "thrust fluctuations" on one engine.  At the time Musk speculated that it was due to debris ingestion and everyone seems to have taken that as a fact although AFAIK that was never officially confirmed.  Indeed that static fire lasted less than 2 seconds and appeared identical to a previous launch pad abort (although again there was never any official word on how long the test was expected to last or whether it was terminated early due to the previously mentioned "thrust variations").

As far as the pre-flight static fire goes there is a world of difference between restating after a few seconds of operation on the pad and restarting after a launch with the vibrations, g-forces and thermal gradients that go with it.

[linuxgurugamer]

49 minutes ago, KerBlammo said:

So out of 4 GTO landing attempts 2 were successful (yay!) and 2 were not (boo!).  Both of the failures were apparently due to low thrust on an engine.  The GTO landings involve firing 3 engines during part of the landing burn (as opposed to 1 in the LEO landings).  This seems to indicate that the Merlin engines as currently designed have a fairly high failure rate after being restarted and are not (yet) fully reusable.

You seem to forget that this is a brand new process, and they are still learning.  So 3 out of 5 successful landing attempts seems a pretty good success ratio to me, giving the limits of the technology

[sojourner]

1 minute ago, icedown said:

Some information would be nice rather than just a personal remark

The reasons have already been discussed.

[KerBlammo]

2 minutes ago, linuxgurugamer said:

You seem to forget that this is a brand new process, and they are still learning.  So 3 out of 5 successful landing attempts seems a pretty good success ratio to me, giving the limits of the technology

I know its a brand new process.  What I am saying is that it doesn't look like they are where they want to be yet as far as engine design goes.

 

[linuxgurugamer]

Just now, KerBlammo said:

I know its a brand new process.  What I am saying is that it doesn't look like they are where they want to be yet as far as engine design goes.

 

Let's see, one failure due to fuel starvation, one failure due to a failed landing leg.  Neither of which is a problem with the engine design

[PB666]

Just now, icedown said:

Some information would be nice rather than just a personal remark

OK, first they release the payload at 8800 km/h which is 2444.444 meters per second.

To deploy droque chutes at that altitude would have to have a huge crossectional area, 100s of meters appart, as it comes into the atmosphere the bindings would have to be incredibly strong.

The are traveling down to the barge just under Mach speed, Even at half or 1/3 mach the rate of air acceleration around the chute causes air to have to exceed the speed of sound, which places a 'air hammer' on the chute and would tear it off the rocket and destroy the fabric and the bindings. So deploy that chute you really need to cut the velocity down to <130 m/s which you have to fire the retros anyway.

Finally if you did both of these correctly, unless the winds aloft were perfectly predictable, your rocket lands a few 100 meters from the barge in the atlantic.

[kerbiloid]

13 minutes ago, wumpus said:

Three engines stopping the rocket on a dime will have much less gravity losses

With 8-9 g F9 with 3 Merlins lands as Titan II accelerates.

This also means triple stress for structure and triple price of error.

[KerBlammo]

1 minute ago, linuxgurugamer said:

Let's see, one failure due to fuel starvation, one failure due to a failed landing leg.  Neither of which is a problem with the engine design

No.  I am talking about the 4 GTO  attempts of which 2 failed due to "low thrust" on one one engine.  In neither case has SpaceX said it was due to fuel starvation (that might be the case, but no-one outside of SpaceX knows if that is true).  What I am saying is that unless it really is due to fuel starvation (which again is unknown to us) then they seem to have a problem with engines performing as expected after a launch.

 

[wumpus]

1 minute ago, kerbiloid said:

With 8-9 g F9 with 3 Merlins lands as Titan II accelerates.

This also means triple stress for structure and triple price of error.

Huh?

You do realize that rockets weigh less when landing then when launching?  The only way the rocket could reach launch stress is if it weighed well over 1/10th the launch weight (and remember the whole upper stage isn't there when landing).

Price of error remains the same.  There are nine rocket engines on the booster regardless of however many are firing on the way down.  Loss of one booster either way.  My guess is that if there is room for the fuel for a single engine landing, they will go that way.  If not, then a three engine landing it is.

[kerbiloid]

16 minutes ago, PB666 said:

Finally if you did both of these correctly, unless the winds aloft were perfectly predictable, your rocket lands a few 100 meters from the barge in the atlantic.

And probably slant, and in general case - onto a slant surface (with another - and random -  slant direction). With 40 m height and chutes as sails this means great chances to overturn.

9 minutes ago, wumpus said:

You do realize that rockets weigh less when landing then when launching?

9 Merlins stop working when ~100 t payload (100 t of 2nd stage + 10 t cargo + 20-25 t of 1st stage dry mass) is still on top of the 1st stage, T/W = 9 * 700 / 120 = 5 g.
1 Merlin then runs with 22-25 t nearly empty 1st stage, and even this makes 3 g.
8-9 g are definitely greater.

9 minutes ago, wumpus said:

Price of error remains the same

It can't be the same because dV/dT changes.

Edited June 15, 2016 by kerbiloid

[wumpus]

5 minutes ago, kerbiloid said:

And probably slant, and in general case - onto a slant surface (with another - and random -  slant direction). With 40 m height this means great chances to overturn.

9 Merlins stop working when ~100 t payload is still on top of the 1st stage, T/W = 9 * 700 / 120 = 5 g.
1 Merlin then runs with 22-25 t nearly empty 1st stage, and even this makes 3 g.
8-9 g are definitely greater.

F=ma * 

the acceleration is higher, but the mass is lower (100*5>25*9).  The force is therefore less and you have less stress on the structure (I'm surprised it is that close.  Note that it has ~500t of force** just sitting on the pad and then taking off with a TWR of 1.3? makes it go even higher).

* at least one professor at my college refused to teach this equation (I didn't have him) and my later professors had to apologize about using it.  Any reason it might be controversial (I'm guessing it has to be taken as an axiom, where most other things in physics can be derived).

** yes, those are metric tons and not a unit of force.  Convert to imperial tons (which should be within the single significant digit) for units of force (I *think* tons are force, and never did any funky conversions between 2000lbs to one 1 ton, but try to avoid doing science in imperial).

[Damien_The_Unbeliever]

1 hour ago, Streetwind said:

Probably by throttling up the other two engines. Depends entirely on how big the thrust shortfall is. Running 5-10% above spec for two or three seconds probably doesn't faze the Merlins, especially if they shut down right after anyway.

Reddit suggests that they started working on this after SES-9 already, which may have had similar issues (unconfirmed). Which is why Elon calls it "underway".

So, assume one of the engines is contributing 0% of the thrust it's meant to, and it's not the centre engine. If you boost the other two engines to compensate, you have quite a lot of asymmetry to deal with. During normal (ascent) usage, you don't have to deal with this because it's always assumed that (given we're not talking about the centre engine, again) there's another engine opposite each engine and so we don't have to deal with any torque issues.

Do you design your entire set of engines to have far more gimbal to cope with that situation, or accept the loss?