totm nov 2023 SpaceX Discussion Thread

[sevenperforce]

6 minutes ago, sh1pman said:

If you double the dry mass, stage delta-v drops from 3.9 km/s to 3.5 km/s if the first stage is expended, or from 2.9 km/s to 1.9 km/s with RTLS (I've spent all this time doing the calculations!).

That's a lot. S1 dry mass is critical to their ability to recover and reuse boosters.

Oh, obviously. I certainly don't suggest doubling the dry mass of the first stage. I'm saying that increases in the mass of recovery components has no significant impact to liftoff TWR.

[sh1pman]

1 minute ago, sevenperforce said:

 I'm saying that increases in the mass of recovery components has no significant impact to liftoff TWR.

Yeah, it's very hard to even make a dent in liftoff TWR, but I was replying to this point:

1 hour ago, sevenperforce said:

If S2 starts with 109 m/s more, it can push a 6% larger payload to GTO, or it can push the same payload to a 22% higher (!!!!) apogee.

My point is that extra dv from uprated engines may be partially or even completely negated by increased dry mass of both stages, especially if they're going for S1 reuse. 

[CatastrophicFailure]

6 hours ago, mikegarrison said:

Haven't changed that much?

No, not really. Still a tube that holds a pressure difference less than one ATM, with a couple hundred or so degrees of temperature shift, in a relatively benign environment. 

The pressure tank in question must contain many hundreds of atmospheres, at cryogenic temperatures, surrounded by pure oxidizer.

This isn’t a SpaceX thing, per say, hyperbole aside, ULA or RocketLab, etc, would have the same claim to such an argument if they’d designed it. 

If it works... 

[sevenperforce]

16 minutes ago, sh1pman said:

My point is that extra dv from uprated engines may be partially or even completely negated by increased dry mass of both stages, especially if they're going for S1 reuse. 

All four titanium grid fins together are probably less than a 100 kg increase over the Al ones. And the larger grid fins' ability to increase drag on landing and increase lift coefficient means they need less propellant, so they can burn more prop on ascent, which means m0/m1 increases rather than decreases.

 

Edited May 11, 2018 by sevenperforce

[tater]

Regarding the Merlin:

Quote

In terms of performance upgrades, we also have a number of those. The Merlin engines, the engine thrust is going to increase by approximately 8%, to 190,000 pounds of thrust at sea level. We think there's probably a little more room there, maybe going up to 10% or so. As well as some small increases in specific impulse of a few seconds. So both the efficiency of the engine and the thrust of the engine have increased. While not increasing, we haven't made any material change to the mass of the engine. So the thrust-to-weight of the engine is getting clearly incredible at this point. It was already the highest thrust-to-weight engine in the world and now it's got even... So. The vacuum version of Merlin increased in thrust by about 5%, to 210,000 pounds-force. Sorry, to 220,000 pounds-force. But we will be de-throttling this engine on this first flight, to assess the vibration increase in the environment, so it will currently be operating at its old thrust level. Just throttling down, essentially. It's a new engine operating at 5% below its rated thrust. So we'll be operating it at 210,000 pounds of thrust. But that's something we expect to increase by 5%, maybe 10% down the road. We are, again, very careful about the level of expansion of the thrust of the engines.

 

(from the transcript of the call yesterday)

[sh1pman]

32 minutes ago, sevenperforce said:

All four titanium grid fins together are probably less than a 100 kg increase over the Al ones. And the larger grid fins' ability to increase drag on landing and increase lift coefficient means they need less propellant, so they can burn more prop on ascent, which means m0/m1 increases rather than decreases.

 

During S1 reentry most of the fuel is spent for boost-back (for RTLS) and braking (ASDS) burns, which are done in higher atmosphere, where the aerodynamic drag is negligible, so these burns won't benefit from better grid fins. They may decrease the landing burn, but it's already much shorter than previous burns. So the fins might shorten the suicide burn from 5s to 4s (three-engine burn), which won't really save you that much fuel. 

Edited May 11, 2018 by sh1pman

[sevenperforce]

4 minutes ago, tater said:

Regarding the Merlin:

"The Merlin engines, the engine thrust is going to increase by approximately 8%, to 190,000 pounds of thrust at sea level. We think there's probably a little more room there, maybe going up to 10% or so. As well as some small increases in specific impulse of a few seconds."

(from the transcript of the call yesterday)

Uptick in isp is expected simply due to pressure thrust. Higher propellant flow means higher chamber pressure, which means bigger pressure drop at SL. Engines will be underexpanded even at liftoff. Another second or two of isp could be gained by adding slightly larger engine bells, but I doubt they have space for that (and it would throw a massive wrench into production).

By my numbers, the Merlin 1D now has a SL TWR of 183 and a vacuum TWR of 198. If they can push it up to 10% more than before, they can hit a TWR of 201.9:1.

10 minutes ago, sh1pman said:

During S1 reentry most of the fuel is spent for boost-back (for RTLS) and braking (ASDS) burns, which are done in higher atmosphere, where the aerodynamic drag is negligible, so these burns won't benefit from better grid fins.

Au contraire. The bigger grid fins allow AoA that approaches a 1:1 body-lift glide ratio, so that they essentially only have to boostback halfway and can "fly" the remaining distance. That's for RTLS. For ASDS landings with boostback, the same thing is true; better glide ratio means a smaller boostback trajectory adjustment to hit the droneship. Larger grid fins can also handle more re-entry heat, which means potentially shorter entry burns.

10 minutes ago, sh1pman said:

They may decrease the landing burn, but it's already much shorter than previous burns. So they may decrease the landing burn from 5s to 4s (three-engine burn), which won't really save you that much fuel. 

They can decrease the total landing dV. Suppose terminal velocity for the Al-fin stage is 290 m/s and terminal velocity for the Ti-fin stage is 260 m/s. Not only is that a savings of 30 m/s on the landing burn, but that means a lighter stage, so a better glide ratio and lower entry heating and a smaller boostback burn and exponentially more propellant deliverable to the second stage at staging.

[tater]

If they want to try a 24 hour turn around, this is necessarily a RTLS flight, it takes a few days to RTB from the ASDS landing zone.

[kerbiloid]

Spoiler

If they replace those four expandable legs with a single titanium spike with a sharp steel bit, they can just stick the rocket into the sand (on land or on barge).
Maybe this would additionally save several tons.

 

[sevenperforce]

6 minutes ago, tater said:

If they want to try a 24 hour turn around, this is necessarily a RTLS flight, it takes a few days to RTB from the ASDS landing zone.

Would they forego the static fire? Or would they do the static fire with the payload integrated?

If 24 hours is measured from T-0 to T-0, then we have the following sequence:

• Launch to landing: <1 hr
• Landing to safing and acquisition: 1 hr
• Return to strongback: 3 hrs
• Return to pad; prop load: 3 hrs
• Static fire and post-fire review: 1 hr
• Return to hanger: 1 hrs
• Payload integration: 4 hrs
• Return to pad, prop load: 3 hrs

Just about 7 hours of buffer in there.

[tater]

Unless he means 24 hours touching the booster, not actual time, which seems more likely.

Ie: Start the clock when the crane starts taking the booster off the ASDS. Move to HIF, integrate payload, move to pad.

2 minutes ago, sevenperforce said:

Would they forego the static fire?

Static fire has been to characterize how things work, I always assumed they'd dump static fires altogether at some point, they can always shut down before they let the clamps go.

[sevenperforce]

5 minutes ago, tater said:

Unless he means 24 hours touching the booster, not actual time, which seems more likely.

Ie: Start the clock when the crane starts taking the booster off the ASDS. Move to HIF, integrate payload, move to pad.

Static fire has been to characterize how things work, I always assumed they'd dump static fires altogether at some point, they can always shut down before they let the clamps go.

I suppose that high-margin, low-performance missions could do a 5-second hold down at launch and have the computer automatically give a go-no-go.

Though I'm sure static fire data review has been lengthy up to this point.

[sh1pman]

19 minutes ago, sevenperforce said:

Au contraire. The bigger grid fins allow AoA that approaches a 1:1 body-lift glide ratio, so that they essentially only have to boostback halfway and can "fly" the remaining distance. That's for RTLS. For ASDS landings with boostback, the same thing is true; better glide ratio means a smaller boostback trajectory adjustment to hit the droneship. Larger grid fins can also handle more re-entry heat, which means potentially shorter entry burns.

Can you really boostback halfway? During RTLS the stage comes down almost vertically, overshooting at first, ad pitching to negative angles to hit the pad. If you do a shorter boostback burn, you'll just end up coming down vertically, but several kilometers away from the pad. Won't have much time to cover any significant distance with gliding.

Quote

They can decrease the total landing dV. Suppose terminal velocity for the Al-fin stage is 290 m/s and terminal velocity for the Ti-fin stage is 260 m/s. Not only is that a savings of 30 m/s on the landing burn, but that means a lighter stage, so a better glide ratio and lower entry heating and a smaller boostback burn and exponentially more propellant deliverable to the second stage at staging.

Let's say you're saving 30 m/s from the landing burn. If your stage is 1% heavier, it'll have 22 m/s less dv after staging. So it kinda cancels out.

Edited May 11, 2018 by sh1pman

[tater]

^^^ that image grossly compresses the downrange distances.

[CatastrophicFailure]

14 minutes ago, sevenperforce said:

Would they forego the static fire?

If it’s just put a payload into space, one could argue it’s just had a very thorough static test fire... 

[sevenperforce]

28 minutes ago, sh1pman said:

Can you really boostback halfway? During RTLS the stage comes down almost vertically, overshooting at first, ad pitching to negative angles to hit the pad. If you do a shorter boostback burn, you'll just end up coming down vertically, but several kilometers away from the pad. Won't have much time to cover any significant distance with gliding.

The graphic is completely not to scale. The stage is quite far downrange on even high-margin RTLS missions.

Here's Elon talking about using the grid fins to fly back:

Spoiler

 

Since the audio is bad, here's a transcript:

Quote

Elon: "The new grid fins will be, should be capable of taking a scorching, and being fine. They'll also have significantly more control authority, so that should improve the reusability of the rocket. It will actually improve the payload to orbit by being able to fly at a higher angle of attack and use the aerodynamic elements of the rocket to effectively fly, it does actually have an L/D of probably 1, flying at the right angle of attack."

In the Zuma mission, the booster was 60 km high and doing 1.7 km/s downrange at staging. If it burned off that 1.7 km/s, it would fall straight down into the ocean. To get back to its origin point, it would need to burn an additional 1.7 km/s, if you look at patched conics. But with the new grid fins, it may only need to do a 1 km/s kick back toward the shore, and let the glide ratio carry it the rest of the way to the pad.

19 minutes ago, CatastrophicFailure said:

If it’s just put a payload into space, one could argue it’s just had a very thorough static test fire... 

Heh.

In all seriousness, the idea is non-destructive testing and component stress. Those nine engines take a beating during launch and landing; the easiest way to verify that they will start up and function normally is to fire them up for a few seconds.

Edited May 11, 2018 by sevenperforce

[sh1pman]

6 minutes ago, tater said:

^^^ that image grossly compresses the downrange distances.

It's just a scheme. Still, the first stage doesn't spend that much time in the dense atmosphere to cover any significant distance with gliding alone. It helps to steer it the right way and achieve terminal velocity faster, though.

[tater]

Looks like for RTLS landings, the booster separates between 30 and 40 km downrange.

Downrange ASDS landings have MECO and sep at 40 to nearly 100km downrange.

RTLS boosters sep at far lower velocities, too. Well under 2 km/s.

 

[sh1pman]

31 minutes ago, tater said:

Looks like for RTLS landings, the booster separates between 30 and 40 km downrange.

30-40 km downrange and what, 100 or so km apoapsis after boostback? That feels pretty vertical to me.

Stolen from reddit:

Edit: the graphic shows that it goes even higher than 100 km, maybe 130 or so.

It reminds me of aerobraking at Duna. If your lander is going straight down (like F9 RTLS), you won't aerobrake at all, even using a spaceplane with great gliding ratio. But if you set the periapsis to something like 20 km, eventually you'll bleed off most of your orbital velocity, and lifting body will be very useful for maneuvering and keeping the lander from falling down. So, IMO, titanium grid fins will be useful for slowing down during high-energy ASDS landings, where the core spends more time in the lower atmosphere, but they won't be much better than aluminium fins during RTLS (which is required for 24-hour turnaround).

Edited May 11, 2018 by sh1pman

[tater]

27 minutes ago, sh1pman said:

30-40 km downrange and what, 100 or so km apoapsis after boostback? That feels pretty vertical to me.

I'm just saying that the other graphic makes it look like the rocket literally never gets "feet wet" off the beach.

[sevenperforce]

32 minutes ago, sh1pman said:

30-40 km downrange and what, 100 or so km apoapsis after boostback? That feels pretty vertical to me.

Stolen from reddit: <snip>

First of all, we know that Elon said the new grid fins give it a high body lift glide ratio, which allows S1 to deliver more energy to S2.

Second, the four RTLS missions on the graphic you sent were with Al grid fins, so of course it had to come in nearly vertical, because there was no glide ratio.

Plus, these had lower velocities at staging, and weren't as far downrange, as a Block 5 booster will be.

Here's a graphic showing how a Block 5 RTLS might look:

As you can see, staging is farther downrange due to B5's higher thrust. The green line shows the required trajectory if the boostback burn did all the work; the red line shows the effect of using glide ratio.

 

[tater]

Yeah, still more vertical than ASDS, but some number of extra km downrange.

Also, if the sep velocity is higher, they are perhaps still using the same boostback props, but more is used to kill velocity, and the "downrange" for return stretches via aerodynamics since the props were used to kill velocity.

[sh1pman]

28 minutes ago, sevenperforce said:

First of all, we know that Elon said the new grid fins give it a high body lift glide ratio, which allows S1 to deliver more energy to S2.

Second, the four RTLS missions on the graphic you sent were with Al grid fins, so of course it had to come in nearly vertical, because there was no glide ratio.

Plus, these had lower velocities at staging, and weren't as far downrange, as a Block 5 booster will be.

Here's a graphic showing how a Block 5 RTLS might look:

As you can see, staging is farther downrange due to B5's higher thrust. The green line shows the required trajectory if the boostback burn did all the work; the red line shows the effect of using glide ratio.

 

That looks to me like a 1-2 second shorter RTLS burn, at most. It may actually be longer, because staging will happen further downrange, and possibly at higher speed. Remember, you still need to burn off 2+ km/s of prograde velocity first. 

Edited May 11, 2018 by sh1pman

[magnemoe]

1 hour ago, tater said:

Unless he means 24 hours touching the booster, not actual time, which seems more likely.

Ie: Start the clock when the crane starts taking the booster off the ASDS. Move to HIF, integrate payload, move to pad.

Pretty sure its time touching the rocket. 
24h turnover will give issues if rocket need servicing. 

Same as buses who has to run the same route over and over with no pause, if route take longer because of stuff like roadwork will cause more and more delays during the day.

As for static fire they will keep it as its nice for safety, they might drop it for starlink as they can loose some of them. 

[CatastrophicFailure]

42 minutes ago, magnemoe said:

Pretty sure its time touching the rocket. 
24h turnover will give issues if rocket need servicing. 

Same as buses who has to run the same route over and over with no pause, if route take longer because of stuff like roadwork will cause more and more delays during the day.

As for static fire they will keep it as its nice for safety, they might drop it for starlink as they can loose some of them. 

That 24-hour turnaround won’t be standard practice at all. It’ll (probably) be a one-time “stunt” just so SpaceX can demonstrate that they can. There’s really no need for that quick a turnaround even in a perfect world, and then there’s the regulatory hurdle of how many launches per x they’re allowed in the first place. 

I suspect such a demonstration might go up at Pad 39 first and 40 the second, or vice versa too.