Blue Origin Thread (merged)

[wumpus]

5 hours ago, Streetwind said:

From what we currently know, a typical FH scenario would see both side boosters return to landing pads, while the center booster touches down on a droneship.

Of course, whether that plan survives actual implementation, we can only speculate. First they need to get this super-delayed thing off the ground for once.

The reason for my question is that the data I had for MEC (main engine cutoff) for stage 1 of Falcon9 was 2km/s for recovery and ~3.5-4 km/s for non-recovery situations.  I plugged in the numbers for Falcon Heavy and became convinced that the booster was going to be going faster than the non-recovery situations (although quite possibly with the needed fuel reserve).

 I was curious to see if the Falcon upper stage was designed with such a high delta-v because they needed to get the non-recoverable (and paying) missions up and running first (so even splitting on the delta-vs for non-recoverable made sense) or they had issues recovering a booster traveling faster than 2km/s.  I'm guessing both and that recovering the center booster will be quite a challenge (although far easier than recovering the upper booster.  I think they at least pretended to have plans for that at first).  I'd still expect more delta-v to be in the lower stage, but I admit that KSP has taught me absolutely nothing about engine bells and design for sea-level vs. vacuum and that would be a huge consideration to how you break your stages.

I'm still shocked that 90% of the engines do 25% of the delta-v.  Even with the tyranny of the rocket equation, I would expect it to be greater than 50%, and closer to 66-75%.  I'm guessing that the lousy mass-efficiency that kerbals have has been teaching me wrong and that reserve fuel simply is that heavy (did KSP help kill the parachute idea?  Some sort of hybrid parachute retro-rocket combination (Soyuz does this on a small scale) would make a lot of sense [more for Blue Origin, who have less need of steering], but was eventually dropped).

[PB666]

14 minutes ago, wumpus said:

The reason for my question is that the data I had for MEC (main engine cutoff) for stage 1 of Falcon9 was 2km/s for recovery and ~3.5-4 km/s for non-recovery situations.  I plugged in the numbers for Falcon Heavy and became convinced that the booster was going to be going faster than the non-recovery situations (although quite possibly with the needed fuel reserve).

 I was curious to see if the Falcon upper stage was designed with such a high delta-v because they needed to get the non-recoverable (and paying) missions up and running first (so even splitting on the delta-vs for non-recoverable made sense) or they had issues recovering a booster traveling faster than 2km/s.  I'm guessing both and that recovering the center booster will be quite a challenge (although far easier than recovering the upper booster.  I think they at least pretended to have plans for that at first).  I'd still expect more delta-v to be in the lower stage, but I admit that KSP has taught me absolutely nothing about engine bells and design for sea-level vs. vacuum and that would be a huge consideration to how you break your stages.

I'm still shocked that 90% of the engines do 25% of the delta-v.  Even with the tyranny of the rocket equation, I would expect it to be greater than 50%, and closer to 66-75%.  I'm guessing that the lousy mass-efficiency that kerbals have has been teaching me wrong and that reserve fuel simply is that heavy (did KSP help kill the parachute idea?  Some sort of hybrid parachute retro-rocket combination (Soyuz does this on a small scale) would make a lot of sense [more for Blue Origin, who have less need of steering], but was eventually dropped).

No, it quite simple, those engines start out at 1.2 g of six parts , five go to gravity (hoovering), 1 goes to forawrd momentum, this goes up to 800 km/hr about 1.6 g (not factoring in drag) exactly at which they throttle down to about 1.4 g. They are not at a 45 degrees until well above Max Q thus alot of the thrust is wasted. 

 

You guys remember  what i said about lifting off from a mountain in ecuador, this is the reason, you can really burn up with high g engines at lift off because of the atmosphere, if you start at a higher altitude you can begin to make the turn to horizontal much more quickly. The thrust you burn down you mostly loose, the thrust wou burn sideways is yours to keep. At a high elevation you can accelerate to max q much more quickly, reducing the hoovering time. 

[sevenperforce]

20 minutes ago, PB666 said:

At a high elevation you can accelerate to max q much more quickly, reducing the hoovering time. 

Yeah, one of the benefits of living on a mountain in Ecuador is much less dust, greatly reducing hoovering time.

Sorry, couldn't resist.

43 minutes ago, wumpus said:

The reason for my question is that the data I had for MEC (main engine cutoff) for stage 1 of Falcon9 was 2km/s for recovery and ~3.5-4 km/s for non-recovery situations.  I plugged in the numbers for Falcon Heavy and became convinced that the booster was going to be going faster than the non-recovery situations (although quite possibly with the needed fuel reserve).

Is the 2 km/s vs 4 km/s break taking into account RTLS vs downrange recovery?

I have a sneaking suspicion that fuel reserve, not re-entry speed, is the break point for the core. SES-9 seemed to survive re-entry just fine; from what I've been able to tell it only failed because it lacked sufficient fuel reserves for a one-engine landing burn and had to go with a three-engine landing burn, which failed either because it was uncontrollable or because it didn't even have enough reserve fuel for that. I cannot think of any reason for a re-entry burn unless they are doing an exhaust shield.

[PB666]

21 minutes ago, sevenperforce said:

Yeah, one of the benefits of living on a mountain in Ecuador is much less dust, greatly reducing hoovering time.

Sorry, couldn't resist.

Is the 2 km/s vs 4 km/s break taking into account RTLS vs downrange recovery?

I have a sneaking suspicion that fuel reserve, not re-entry speed, is the break point for the core. SES-9 seemed to survive re-entry just fine; from what I've been able to tell it only failed because it lacked sufficient fuel reserves for a one-engine landing burn and had to go with a three-engine landing burn, which failed either because it was uncontrollable or because it didn't even have enough reserve fuel for that. I cannot think of any reason for a re-entry burn unless they are doing an exhaust shield.

I'm going to give some actual number about 6650 km/hr is around 1847 meters per second the vertical velocity is only around 1500 m/s they turn more sharply once over 2000 km/hr thus much of that was obtained earlier in the burn. If you burn more its going to be almost all in the horizontal, which means that your horizontal traverse on retro and horizontal error or retro will be much greater. Your barge on the other hand is setting in the Atlantic trying to hold a GPS position. If you come in at a 45 you only really need to kill vectors that create error which is a smaller proportion than if you reenter at a 10 degree angle with 2 or 3 times the horizontal velocity (factor in that the atmosphere pressure at height is not perfectly uniform).  Remember the idea that the most efficient place to burn at apogee, they are gaining 1500 m/s how many seconds to reach apoapsis. Considering that w2r at horizontal vector of 1847 is 2-3%% gravity, and at 4km/s is about 25% of gravity, which means your vertical slows down much more slowly you have to burn much later and at a higher altitude, and you have much less control over where you reenter. The fuel reserve also needs there for an engine failure, because they will need more burn time to reach MECO energy and need more fuel for the back-burn. At 4 km/second, if you have an engine loss a barge landing is impossible. If the Stage value is 100 times that of fuel, then you don't care about fuel if you are recycling and can make first stage slightly larger and less efficient.

Couple of problems I see in going over the flight data, Max Q comes well after Mach X-over, their payload nose-cone is not ideal, and it appears that boundary collapse is creating alot of turbulence around the plume area. Here is an area of improvement, if they give the nose cone a more Sears-Haack shape, then their Max-Q would come at a lower altitude, but allowing them to turn more quickly and efficiently. Because of the expense of the fairings and nose-cone they are going for non-ideal shapes. Presumably if they can find a way to recycle it make them much more cheaply these they will have better flight dynamics. They could make the first stage smaller and second stage larger, or burn 1st stage higher and more horizontal 

The economics you need to understand is this.

Nose cone (independent of first stage) cost, weight, turnaround cost and time.
Fairing (almost independent of first stage) cost, weight, turnaround time cost and time.
Stage 1 (can vary in size or height, made slightly wider, slightly more fuel)  initial (fixed) cost per unit size, recycling (variable) cost per unit size, fuel cost (variable) and cost per size increase. Stage 2 (can vary in size or height, " ", "" ), cost per unit, fuel cost variable.
Stage 3 (payload) variable, but effects nose cone and fairings. Adjustment cost for wider payloads versus economies of scale for larger defined fairings and nose cone.

 

 

The way the payload is set up the launch and second stage can be more robust, launch phase can carry more fuel, even have one or two more engines. If the rocket is truely recyclable they can go with the expensive but much less turbulent engines, which would allow for a higher max-q (Max-q is the compresssion of thrust and atmosphere on the most weight sensitive part of the rocket, and if you have turbulent thrust then the atmospheric component has to be lower).

[wumpus]

1 hour ago, sevenperforce said:

Yeah, one of the benefits of living on a mountain in Ecuador is much less dust, greatly reducing hoovering time.

Sorry, couldn't resist.

Is the 2 km/s vs 4 km/s break taking into account RTLS vs downrange recovery?

2km/s was presumably used in the recent downrange landing.  3.5 (or more) km/s is what you get if you burn the landing reserves.   You will need many more reserves if you have to brake 4km/s.  - Note that these numbers appear fishy: if the fuel reserves for an empty booster can provide either 1.5 km/s up for an empty booster + a full upper stage or 2km/s arresting down velocity of an empty upper stage (plus some more due to gravity overcome on the way up), something doesn't add up.  This was pretty much the point of my comment: the MEC given for Falcon 9 seems absurdly low, and seems to be an issue if Falcon Heavy has to do anything similar.  My guess is they are going much faster during MEC.

2 hours ago, PB666 said:

No, it quite simple, those engines start out at 1.2 g of six parts , five go to gravity (hoovering), 1 goes to forawrd momentum, this goes up to 800 km/hr about 1.6 g (not factoring in drag) exactly at which they throttle down to about 1.4 g. They are not at a 45 degrees until well above Max Q thus alot of the thrust is wasted. 

Hopefully Falcon Heavy can get that fraction up to at least 1.3 or more (although 1.4 would probably be nudging max-Q right away, but the whole point of Falcon Heavy is to throttle down the center anyway).  There is a reason I'm such a fan of SRBs in KSP (more thrust!).  Get a pad out in Ecuador and strap some COTS SRBs to the side! (no comment about the politics of having three landing pads in the Amazon and various non-reused SRBs landing there as well).

I started digging for data on my original comment assuming that I could make snarky comments about strapping SRBs to the sides of the Falcon Heavy upper stage (it still looks like a good idea to me, and would presumably fix a bunch of landing issues).  But the real issue is that while KSP might have taught me a ton about orbital dynamics, it has taught me very little about launching into orbit (of course, putting most of my time in before 1.0 doesn't help at all).  Maybe I should download Orbiter.

[Spaceception]

360 video of the SpaceX barge landing!!!

 

[CatastrophicFailure]

18 minutes ago, Spaceception said:

360 video of the SpaceX barge landing!!!

 

The noise! The NOISE! Never realized just how far it bounced to the side, either. Is there a better quality one anywhere?

 

 

 

Edited April 29, 2016 by CatastrophicFailure

[Frozen_Heart]

W..wait... There is videos where you can look around now!!!

[Spaceception]

3 minutes ago, CatastrophicFailure said:

The noise! The NOISE! Never realized just how far it bounced to the side, either. Is there a better quality one anyone?

 

 

 

It was from SpaceX, so... no?

Just now, Frozen_Heart said:

W..wait... There is videos where you can look around now!!!

Yep Crash course made one too!

[CatastrophicFailure]

10 minutes ago, Frozen_Heart said:

W..wait... There is videos where you can look around now!!!

And rockets that land on boats. And robotic invasion force on Mars. What a time to be alive, eh?

ETA: I just pulled it on on my iThing™ instead of the PC. Never actually tried one of those 360 videos like that before. Brings a whole new perspective. 

 

Edited April 29, 2016 by CatastrophicFailure

[StrandedonEarth]

18 minutes ago, Spaceception said:

360 video of the SpaceX barge landing!!!

That was cool, nice little sideslide on the landing. I noticed a seagull gettin the heck outta there before the Falcon landed!

It's about time they released that vid. I wonder if they have or will ever release a vid of the guys coming aboard to safe it and weld the shoes on.

A vid of the falcon punch (crash landing) would be nice too...

[PB666]

4 hours ago, wumpus said:

2km/s was presumably used in the recent downrange landing.  3.5 (or more) km/s is what you get if you burn the landing reserves.   You will need many more reserves if you have to brake 4km/s.  - Note that these numbers appear fishy: if the fuel reserves for an empty booster can provide either 1.5 km/s up for an empty booster + a full upper stage or 2km/s arresting down velocity of an empty upper stage (plus some more due to gravity overcome on the way up), something doesn't add up.  This was pretty much the point of my comment: the MEC given for Falcon 9 seems absurdly low, and seems to be an issue if Falcon Heavy has to do anything similar.  My guess is they are going much faster during MEC.

Hopefully Falcon Heavy can get that fraction up to at least 1.3 or more (although 1.4 would probably be nudging max-Q right away, but the whole point of Falcon Heavy is to throttle down the center anyway).  There is a reason I'm such a fan of SRBs in KSP (more thrust!).  Get a pad out in Ecuador and strap some COTS SRBs to the side! (no comment about the politics of having three landing pads in the Amazon and various non-reused SRBs landing there as well).

I started digging for data on my original comment assuming that I could make snarky comments about strapping SRBs to the sides of the Falcon Heavy upper stage (it still looks like a good idea to me, and would presumably fix a bunch of landing issues).  But the real issue is that while KSP might have taught me a ton about orbital dynamics, it has taught me very little about launching into orbit (of course, putting most of my time in before 1.0 doesn't help at all).  Maybe I should download Orbiter.

 

In CRS-8 to be exact. Their final stage 1 velocity was 6680 km/h. which is 6680 megometer/hour which is 6680/3.6 meters per second = 1856 meters per second, at apogee it is lower. I have the precise flight numbers I just haven't presented them yet. The numbers do not need to be precise since there is a conversion of kinetic to thermodynamic energy in post staging, but as the vehicle levels out and adds more horizontal velocity to circularize those differences become meaningless because almost all the kinetic energy is in the horizontal vector. I am not sure when it is trajectory that the backburn took place, but its vertical ascent of at least 1000 m/s would have left <1500 m/s of horizontal velocity, given no payload and the fact that most of its velocity came in the last few second of flight, the ISP as at max, it would have required little fuel. I think the primary concern was to establish a trajectory that intercepted the barge, given drag. A lower descent angle means more energy is lost via drag, a higher descent angle means more control with regard to target landings.

SRBs are problematic for vertical ascents, because you have the risk of hitting ground targets (which for the most part game we laughably ignore launch pad damage). The 800 kmh means you could add more fuel and use two SRBs that blew off at 800. This could mean higher and faster second stage deliveries, but also barge is going to be further offshore, and back-burn will be more intense.

 

 

[tater]

That vid is awesome.

[AngelLestat]

Wow.. that was a nice gift to watch.
Next:  matrix effect!  

[Frozen_Heart]

Looks like the Falcon Heavy 1.2 can still manage 54 tons to LEO despite no crossfeed. Falcon 9 has been upped to $62 million however.

https://www.reddit.com/r/spacex/comments/4h3yjg/spacex_pricing_payload_capabilities_changed_for/

 

[Elukka]

I struggle to understand how these numbers are possible and I'm wondering if there isn't a mistake. They have the F9 match the Delta IV Heavy in LEO payload, despite carrying a lesser propellant load and much lower isp engines.

We know the Falcon's engines have a thrust-to-weight ratio well in excess of any other engine in operation (and possibly any engine in history) and that its structure is most likely lighter than that of most rockets. However, the only way I can see those numbers is if the F9 has always been much more capable than we've been let on, and the true payload figure for expendable missions was never previously published. I'd still have trouble reconciling it with the Delta IV Heavy. It would imply the effect Falcon's lower dry mass is so significant it outweighs the reduced isp and propellant load.

Edited April 30, 2016 by Elukka

[Rakaydos]

15 minutes ago, Elukka said:

 I'd still have trouble reconciling it with the Delta IV Heavy. It would imply the effect Falcon's lower dry mass is so significant it outweighs the reduced isp and propellant load.

Dude, low dry mass is EVERYTHING in the rocket equation.

So the Falcon 9 has high TWR engines, true, but it also has extremely dense (though slightly lower ISP) propellant, shifting the fuel mass/dry mass ratio even furthur.

[Frozen_Heart]

35 minutes ago, Elukka said:

I struggle to understand how these numbers are possible and I'm wondering if there isn't a mistake. They have the F9 match the Delta IV Heavy in LEO payload, despite carrying a lesser propellant load and much lower isp engines.

We know the Falcon's engines have a thrust-to-weight ratio well in excess of any other engine in operation (and possibly any engine in history) and that its structure is most likely lighter than that of most rockets. However, the only way I can see those numbers is if the F9 has always been much more capable than we've been let on, and the true payload figure for expendable missions was never previously published. I'd still have trouble reconciling it with the Delta IV Heavy. It would imply the effect Falcon's lower dry mass is so significant it outweighs the reduced isp and propellant load.

People are suggesting that these numbers are for an upcoming planned upgrade such as V1.3, as they seem off for V1.2. Though I really can't see how they could boost performance further.

Also if there is another upgrade coming then Falcon Heavy will be delayed yet another year or two. :/

[PB666]

11 minutes ago, Elukka said:

I struggle to understand how these numbers are possible and I'm wondering if there isn't a mistake. They have the F9 match the Delta IV Heavy in LEO payload, despite carrying a lesser propellant load and much lower isp engines.

We know the Falcon's engines have a thrust-to-weight ratio well in excess of any other engine in operation (and possibly any engine in history) and that its structure is most likely lighter than that of most rockets. However, the only way I can see those numbers is if the F9 has always been much more capable than we've been let on, and the true payload figure for expendable missions was never previously published. I'd still have trouble reconciling it with the Delta IV Heavy. It would imply the effect Falcon's lower dry mass is so significant it outweighs the reduced isp and propellant load.

How high and how fast does delta IV deliver. Falcon-9 launch only delivered to 1855 mps and iirc 30 km up. It lets the second stage do the brunt of dV. 

The merlin 1d vacuum has an ISP of 348 and on the second stage has a burn time of 397 seconds at 934 kN.

The delta IV has 3140 x 3  kn for 242 seconds, and 3140 for 86 seconds, second has 110 kN for 1125 seconds. 

The 2nd stage has about 3 times the static forcextime capacity as the delta iv heavy. 

The difference it would appear is that the delta iv tries to get the payload as close to apogee as possible and the second stage circularizes, whereas the falcon launch delivers its stage out of the low performance atmosphere and lands then the second stage drives to apogee and cicularizes. 

 

 

 

 

[Elukka]

Well, I'm not unwilling to believe it if there's some further confirmation. It's just the figure is startlingly high.

Would be interesting if Falcon actually significantly outperforms a hydrolox rocket in terms of payload fraction.

We know they're working on a Raptor-based methalox second stage which would increase its performance but I doubt the numbers are based on that since that hasn't really even been formally announced yet.

Edited April 30, 2016 by Elukka

[wumpus]

1 hour ago, Rakaydos said:

Dude, low dry mass is EVERYTHING in the rocket equation.

So the Falcon 9 has high TWR engines, true, but it also has extremely dense (though slightly lower ISP) propellant, shifting the fuel mass/dry mass ratio even furthur.

Typically with an ISP advantage of ~400 vs. ~300 any dry mass advantage goes away.  I suspect that Delta IV heavy was pretty much an afterthought of the Delta family and really doesn't have the efficiency it should.

Remember, as far as "getting into orbit dry mass", the Falcon lower stages need a lot of fuel for backburns, and that is considered "dry mass" going up.  If the center lower stage needs 4km/s delta-v just on the way down don't expect a low "dry mass" and high delta-v going up.

[Rakaydos]

Just now, wumpus said:

Typically with an ISP advantage of ~400 vs. ~300 any dry mass advantage goes away.  I suspect that Delta IV heavy was pretty much an afterthought of the Delta family and really doesn't have the efficiency it should.

Remember, as far as "getting into orbit dry mass", the Falcon lower stages need a lot of fuel for backburns, and that is considered "dry mass" going up.  If the center lower stage needs 4km/s delta-v just on the way down don't expect a low "dry mass" and high delta-v going up.

For first stage landing, the "dry mass" of landing fuel is peanuts to the "dry mass" of the second stage. Losing the second stage increases the remaining DV dramatically.

[wumpus]

6 minutes ago, Rakaydos said:

For first stage landing, the "dry mass" of landing fuel is peanuts to the "dry mass" of the second stage. Losing the second stage increases the remaining DV dramatically.

On Falcon9 (and the side boosters) it is.  I'm less sure if you need 4km/s delta-v left in your booster (if the lower stage of a falcon9 can be a SSTO, this means that it needs roughly half its delta-v in reserve.  That is a lot of dry mass.)

On 4/29/2016 at 10:31 AM, PB666 said:

No, it quite simple, those engines start out at 1.2 g of six parts , five go to gravity (hoovering), 1 goes to forawrd momentum, this goes up to 800 km/hr about 1.6 g (not factoring in drag) exactly at which they throttle down to about 1.4 g. They are not at a 45 degrees until well above Max Q thus alot of the thrust is wasted. 

While fussing around checking these figured I noticed that the Falcon9 1.1 FT has a TWR (on the pad) of 1.28  This might be part of the reason they have managed to land this type of rocket (on the other hand, I think the one that landed and then fell over was a 1.1 non-FT).  It might only be the third decimal place, but it reduces the losses considerably.

[Rakaydos]

34 minutes ago, wumpus said:

On Falcon9 (and the side boosters) it is.  I'm less sure if you need 4km/s delta-v left in your booster (if the lower stage of a falcon9 can be a SSTO, this means that it needs roughly half its delta-v in reserve.  That is a lot of dry mass.)

I'm pretty sure they dont zero out their horizontal velocity on the core. The mission where they put a hole in the barge, the falcon 9 didnt do any breaking burn at all, rellying entirely on atmospheric effects to slow it down before the final landing burn. (which they didnt quite get right)

With the sme approach (and a bit more fuel, but not 4k d/v) the FH core can be brought to a barge landing.

[Kryten]

3 hours ago, Elukka said:

I struggle to understand how these numbers are possible and I'm wondering if there isn't a mistake. They have the F9 match the Delta IV Heavy in LEO payload, despite carrying a lesser propellant load and much lower isp engines.

LEO also isn't too fair a comparison; DIVH was designed and is typically used for high-energy orbits like direct GEO insertion, at which it would handily outperform F9.