Blue Origin Thread (merged)

[magnemoe]

10 hours ago, sgt_flyer said:

seems SpaceX showed their Dragon 2 touchscreen UI at GDC (for docking / RCS control mode)

i hope there's some kind of autopilot for normal operations (or that the pilot will have a better training than the guy who flew the demo ;)). especially if there's no tactile feedback, the kind you could have with at least digital sticks (analog sticks would require some kind of RCS impulse frequency control) - with those digital sticks, you could at least feel the clicking point, even with gloves.

www.reddit.com/r/spacex/comments/4avhuw/dragon_2_software_demo_gdc

Im pretty sure they use autopilot as standard, the UI is mostly for supervising autopilot and you can take over and do it manual if needed. Probably the reason its no analogue controls as the system is unlikely to get used. 

[Frozen_Heart]

3 hours ago, Nibb31 said:

The difference between those tests and what SpaceX is doing is that those are mandatory certification tests, and although there is a lot of breakage and fire, the tests are actually successful.

SpaceX's method is entirely different. They favor rapid prototyping instead of software modelling techniques.

But SpaceX would lose the rocket even if they didn't do these tests. Real world tests will always be more accurate than any computer simulation.

[Hcube]

21 hours ago, Nibb31 said:

Most aerospace companies do this sort of testing in computer simulations. Only when the design is frozen, and when they are 100% confident, do they actually build a working prototype. You don't see Boeing or Airbus blowing stuff up to see if it works.

The A400M does crash during testing though

 

(I know it's not relevant, no need to start a thread war about this, it's just funny)

[wumpus]

22 hours ago, CatastrophicFailure said:

AFAIK, SpaceX has yet to reveal what actually went wrong, too (too much velocity/not enough fuel, mechanical error, computer error, etc), so there's quite alot of speculation being thrown around right now.

I would be rather impressed if they knew (and had reasonable confidence in) knowing what actually went wrong.  It seems the machine was coming down out of spec (known in advance), and couldn't handle the landing.  Judging from the landing video (real time) and the hole in the barge, I can only assume that they burned too little/too late on the landing burn (there was still a visible plume when it hit, so presumably it still had fuel and thrust left in the tank).  Note that I'm still assuming the falcon came down hard enough to punch the hole, from memory of the livestream it might have done such only because the rocket had melted/weakened the steel underneath (with presumably 3 times the power as usual).

Basically, you have a lot of tolerances the didn't stack up right and weren't expected to stack up right (assuming the rocket was out of spec for amount of fuel for its velocity).

Basically it comes down to two things, plus a few external conditions.

The rocket's "altimeter" vs. *exactly* how high off sea level it was (same sensors determine velocity).

The rocket's thrust vs. precisely needed/requested thrust (note that if there is an error above, then the requested trust won't quite be right.  Also expect this to be an iterative process (nearly all control circuits rely on feedback) the falcon may not have time for on this landing.

The external conditions would be how far the rocket and barge may have floated off course (due to wind/currents), requiring any more additional fuel isn't going to help things.  Space-x may have been counting on the booster to slow down even more in the atmosphere.  Since it weighed less than the usual booster, it certainly should have slowed down more, but they didn't have the data until now.

What I saw on the livestream was a booster coming down and the livestream cutting out.  It wasn't falling terribly fast (considering it was probably supersonic a few of seconds earlier*).  I'm guessing it came down to a few milliseconds off on the landing burn and the plume melted a pilot hole in the deck.  The falcon just isn't going to land boosters after launching GTO vehicles.

* 4g of deceleration should go from mach 1 to 0 in under 10 seconds.  With 3 engines running, the throttle range should be something like 3.5g to 7.5g (assuming minimum thrust on one engine gives a TWR=1.5 (we know it is >1) and subtracting gravity).

[Lukaszenko]

7 hours ago, Nibb31 said:

The difference between those tests and what SpaceX is doing is that those are mandatory certification tests, and although there is a lot of breakage and fire, the tests are actually successful.

SpaceX's method is entirely different. They favor rapid prototyping instead of software modelling techniques.

Successful or not, the reason they are done is because until that point nobody knows for sure, no matter how much money was spent on simulations. And that's what I was responding to, the claim that a prototype isn't built until 100% confident. And also to the claim that aerospace companies don't blow stuff up. They do. There is NEVER 100% confident, even after simulations AND tests. They do software, they do wind tunnels, they do stress, they do certification, they do actual missions, they do actual missions with people...and yet excrements still breaks and things explode and people die. 

Likewise, if a broken wing can be called a successful test, then so can a broken rocket and a hole in a barge, depending what exactly SpaceX was trying to get out of it.

I'd also venture to say that the total cost of this particular test was orders of magnitude less than an engine blade-out test. 

[CatastrophicFailure]

All good points, but also purely speculation. The whole reason SpaceX is attempting "unlikely" landings like this one is data acquisition so they can know for sure, not just speculation. As far as I know they have telemetry from the booster until the last moments, and with that they've been able to piece together exactly what went wrong with the previous barge landings (ran out of hydraulic fluid, etc). I'm just curious to know what the actual cause was here. 

 

2 hours ago, wumpus said:

The falcon just isn't going to land boosters after launching GTO vehicles.

Eventually launches like this will be handled by the Falcon Heavy, which is definitely intended to be recovered, but the situation will be quite similar. The core stage will be going very fast, very far from land, albeit with enough fuel for a boostback burn for better aiming. This flight was a perfect opportunity to practice procedures & such for that kind of recovery, even if any recovery its self was highly unlikely (just as the first barge landing attempt was). It's how they learn.

[DerekL1963]

1 hour ago, Lukaszenko said:

Likewise, if a broken wing can be called a successful test, then so can a broken rocket and a hole in a barge, depending what exactly SpaceX was trying to get out of it.

Apples and oranges to some degree.

Don't get confused by the broken wing because that's not the point of the test, the expectation of the certification test is that at a given load the wing will not fail.  (That's why the cheers when the wing reaches 150% loading, it passed the test.)   The test-to-fail is somewhat separate, private, test by the builder to ensure that they didn't build the wing too strong.  Too strong means too heavy which means the aircraft will operate at a weight penalty, something the customers won't like very much,

This is somewhat different than the most recent test by Space-X, where they went into the test knowing there was an exceptionally high chance of hull loss.

[sevenperforce]

According to this article, SpaceX will attempt another barge landing after the April 8 ISS resupply run.

I'm not sure if this is because the weight of the BEAM plus the Dragon 1 is too great for the first stage to RTLS. It might be. More likely, though, the stage would be capable of RTLS but they want to try and actually stick a barge landing, both for publicity reasons and for more practice. They've already gotten a RTLS perfectly on the first attempt; failing a RTLS landing would be very bad for publicity and give them very little useful data, whereas failing a barge landing would be less damaging and give them more data. Plus, if the first stage has enough fuel for RTLS, then landing on a barge instead would give them more practice landing under a range of weights. It might even be easier because T/W will be lower. Even though they have very good estimates of remaining fuel and so forth, the inherent variability means that the computer onboard the stage have to assess the exact conditions and velocity and altitude and use that to precisely time the suicide burn, and those equations are going to continue to need tweaking.

2 hours ago, wumpus said:

Judging from the landing video (real time) and the hole in the barge, I can only assume that they burned too little/too late on the landing burn (there was still a visible plume when it hit, so presumably it still had fuel and thrust left in the tank).  Note that I'm still assuming the falcon came down hard enough to punch the hole, from memory of the livestream it might have done such only because the rocket had melted/weakened the steel underneath (with presumably 3 times the power as usual).

Basically, you have a lot of tolerances the didn't stack up right and weren't expected to stack up right (assuming the rocket was out of spec for amount of fuel for its velocity).

 

What I saw on the livestream was a booster coming down and the livestream cutting out.  It wasn't falling terribly fast (considering it was probably supersonic a few of seconds earlier*).  I'm guessing it came down to a few milliseconds off on the landing burn and the plume melted a pilot hole in the deck.  The falcon just isn't going to land boosters after launching GTO vehicles.

* 4g of deceleration should go from mach 1 to 0 in under 10 seconds.  With 3 engines running, the throttle range should be something like 3.5g to 7.5g (assuming minimum thrust on one engine gives a TWR=1.5 (we know it is >1) and subtracting gravity).

My experience in metallurgical engineering is very limited, but I am extremely doubtful that the engine plume would significantly weaken the strength of the steel. There's just not enough time. It was probably energetic enough to blast off particles (like a sandblaster that is actually a flamethrower) but weakening the tensile strength of the metal is just not likely.

30 minutes ago, CatastrophicFailure said:

Eventually launches like this will be handled by the Falcon Heavy, which is definitely intended to be recovered, but the situation will be quite similar. The core stage will be going very fast, very far from land, albeit with enough fuel for a boostback burn for better aiming. This flight was a perfect opportunity to practice procedures & such for that kind of recovery, even if any recovery its self was highly unlikely (just as the first barge landing attempt was). It's how they learn.

Yeah, I think they will be able to stick landings like this in the future, either immediately or once they have additional data.

[Hcube]

9 minutes ago, sevenperforce said:

My experience in metallurgical engineering is very limited, but I am extremely doubtful that the engine plume would significantly weaken the strength of the steel. There's just not enough time. It was probably energetic enough to blast off particles (like a sandblaster that is actually a flamethrower) but weakening the tensile strength of the metal is just not likely.

Obligatory "Jet fuel can't melt steel beams !" 

On a more serious note i agree 100% with that, there's no way the plume (even if it's pure O2 oxidizer) could melt the barge in one or two seconds... The stage probably simply hit the barge hard

Edited March 18, 2016 by Hcube

[sojourner]

At this stage (no pun intended) I believe the main goal in SpaceX's landing attempts is to retrieve intact first stages.  Based on that I am guessing that they will still try to get approval for another RTLS.  Remember that the last RTLS only got approved about 2 days before launch.

[wumpus]

32 minutes ago, Hcube said:

Obligatory "Jet fuel can't melt steel beams !" 

On a more serious note i agree 100% with that, there's no way the plume (even if it's pure O2 oxidizer) could melt the barge in one or two seconds... The stage probably simply hit the barge hard

Yes, I just don't remember thinking it came down that fast (maybe the plume was much longer and that is all I saw).  The main reason I brought it up was I didn't think the rocket hit all that hard, there was a hole, and there was this extremely hot fire nearby.

[sevenperforce]

45 minutes ago, wumpus said:

Yes, I just don't remember thinking it came down that fast (maybe the plume was much longer and that is all I saw).  The main reason I brought it up was I didn't think the rocket hit all that hard, there was a hole, and there was this extremely hot fire nearby.

The "punching through" aspect is what makes me think it hit hard. It may have still been supersonic. I know we appear to see a plume hovering, but that's probably poor frame rate more than anything else.

[DerekL1963]

59 minutes ago, sevenperforce said:

The "punching through" aspect is what makes me think it hit hard. It may have still been supersonic. I know we appear to see a plume hovering, but that's probably poor frame rate more than anything else.

I suspect the punch through may have had as much to do with the high lateral velocity and the hard touchdown as it does with raw vertical velocity.

[Mitchz95]

1 hour ago, DerekL1963 said:

I suspect the punch through may have had as much to do with the high lateral velocity and the hard touchdown as it does with raw vertical velocity.

I doubt a lateral impact would have left such a clean hole in the barge. It would have looked more like a scar going in one direction.

[fredinno]

4 hours ago, sevenperforce said:

According to this article, SpaceX will attempt another barge landing after the April 8 ISS resupply run.

I'm not sure if this is because the weight of the BEAM plus the Dragon 1 is too great for the first stage to RTLS. It might be. More likely, though, the stage would be capable of RTLS but they want to try and actually stick a barge landing, both for publicity reasons and for more practice. They've already gotten a RTLS perfectly on the first attempt; failing a RTLS landing would be very bad for publicity and give them very little useful data, whereas failing a barge landing would be less damaging and give them more data. Plus, if the first stage has enough fuel for RTLS, then landing on a barge instead would give them more practice landing under a range of weights. It might even be easier because T/W will be lower. Even though they have very good estimates of remaining fuel and so forth, the inherent variability means that the computer onboard the stage have to assess the exact conditions and velocity and altitude and use that to precisely time the suicide burn, and those equations are going to continue to need tweaking.

My experience in metallurgical engineering is very limited, but I am extremely doubtful that the engine plume would significantly weaken the strength of the steel. There's just not enough time. It was probably energetic enough to blast off particles (like a sandblaster that is actually a flamethrower) but weakening the tensile strength of the metal is just not likely.

Yeah, I think they will be able to stick landings like this in the future, either immediately or once they have additional data.

It's to test barge landings, Dragon is actually too small for F9FT, even with both cargo bays filled to the brim (increadably unlikely)

[Mad Rocket Scientist]

UPDATE: Looks like a DPL on April 8th for CRS-8.

[Motokid600]

How does SpaceX intend on dealing with Floridas 90 degree + temperatures in the summer?

Edited March 19, 2016 by Motokid600

[WuphonsReach]

1 hour ago, Mad Rocket Scientist said:

UPDATE: Looks like a DPL on April 8th for CRS-8.

And it was originally scheduled for Apr 4th, so it's been pushed back 4 days.

 

Quote

Launch time: 2043 GMT (4:43 p.m. EDT)
Launch site: SLC-40, Cape Canaveral Air Force Station, Florida

http://www.theverge.com/2016/3/18/11261546/spacex-iss-resupply-mission-april-8-falcon-9-beam

[WuphonsReach]

16 minutes ago, Motokid600 said:

How does SpaceX intend on dealing with Floridas 90 degree + temperatures in the summer?

90F is about 305 Kelvin.  50F is 283 Kelvin.  So 90F is only 8% hotter when measured on the absolute scale.  So it probably doesn't make much difference for LOX that is stored at 50-90 Kelvin (not sure what temperature that SpaceX is using).

[J.Random]

6 hours ago, Mitchz95 said:

I doubt a lateral impact would have left such a clean hole in the barge. It would have looked more like a scar going in one direction.

It doesn't look very clean, and the tear seems... triangular-ish. Came at an angle, poked a hole with one of the landing legs? Marmac 304's specs state "uniform deck strength of 4,500 lbs/sq. ft." which is roughly 22 tonnes/m^2. Almost empty stage would weigh what, 30 tonnes? Smaller holes are probably from shrapnel.

Edited March 19, 2016 by J.Random

[Hcube]

On 18/3/2016 at 6:15 PM, sevenperforce said:

 

I'm trying to quote @J.Random's quote here but it's not working and that quote up there won't go away... (That new forum though x_x )

 

Err...this comparaison is pretty much meaningless because it's not like the stage came down like a feather without any energy... It must have had A LOT of kinetic energy. The actual pressure equivalent was probably orders of magnitude higher than 22t/m²

 

[sevenperforce]

1 hour ago, Hcube said:

Err...this comparaison is pretty much meaningless because it's not like the stage came down like a feather without any energy... It must have had A LOT of kinetic energy. The actual pressure equivalent was probably orders of magnitude higher than 22t/m²

Yeah, the weight of the stage doesn't really matter. This is one of the time when the difference between weight and mass actually makes...a difference.

Instead of static weight, we are looking at two things: momentum and kinetic energy. If you want to know (roughly) whether the deck was going to give way, you would take the momentum (mass times velocity) and divide by the "crunch time" (i.e. the time it took the stage to crumple to a stop) to get the impulse force. Then you can divide by average contact area to get the pressure. This depends on some assumptions but can provide a first-order estimate of whether the forces exceeded the load bearing capacity of the deck. 

Now, to determine how far the stage would penetrate, you need to consider kinetic energy. In cases where there is little cohesion in the target material and the impactor is massively supersonic (e.g. a meteor impact or an antitank missile hitting a ceramic plate), you can use Newton's approximation, but that's not the case here. Instead, divide the kinetic energy of the stage by the compressive strength of the deck, then divide by the impact area. This gives you the distance across which a given amount of force produces work equal to the kinetic energy. Of course some kinetic energy is absorbed by the crumpling stage but this is still just first order. 

[Hcube]

... I don't see the point of doing this anyway. The stage *did* destroy the deck

[sevenperforce]

19 hours ago, Hcube said:

... I don't see the point of doing this anyway. The stage *did* destroy the deck

Well, it could give us a rough idea of how fast the stage must have been going.

Though obviously the stage will splat at lower speeds than the deck will boom, so...

[Motokid600]

I'm desperately looking for a specific video of the Orbcomm landing. I could've sworn i watched amateur footage that was zoomed in enough to capture the engines gimbaling, but I can't find it for the life of me.