RocketLab Discussion Thread

[tater]

 

 

[.50calBMG]

So, neutron grew up a bit. 7m core, 7 methalox engines on the first stage, 8 tons of payload, propulsive landing, petal fairing.

[RCgothic]

I like the tension upper stage  and petal fairing idea. I do wonder how they get it to attach though.

A few digs at SpaceX, but I consider that by disposing of the upper stage they're not really solving the same problem, so they didn't really land for me.

Let's see that carbon fibre test again at cryo temperatures and then again during the heat of re-entry.

[tater]

Replay:

https://www.rocketlabusa.com/assets/Uploads/neutron-overview.jpg

 

 

 

 

Edited December 2, 2021 by tater

[Beccab]

4 minutes ago, RCgothic said:

A few digs at SpaceX, but I consider that by disposing of the upper stage they're not really solving the same problem, so they didn't really land for me

Well, it was quite informative. For example, I did not know that a mayor engineering concern in rocket design is resistance to swinging I-beams mid flight

[tater]

 

[sevenperforce]

1 minute ago, Beccab said:

Well, it was quite informative. For example, I did not know that a mayor engineering concern in rocket design is resistance to swinging I-beams mid flight

Hahaha, yes, that cracked me up.

I do rather like the design. Keeping the fairing on the first stage means more dV allocated to the first stage, though, because you need to get further out of the atmosphere. I wonder what the speed at boostback is like. I imagine that knowing reusable vs expendable payload gives us some idea.

Gas generator methalox is a solid choice. Lots of digs at both SpaceX and Blue, I thought. 

I can’t tell if the upper stage is a Rutherford or an Archimedes. Going with a Rutherford would mean better prop density.

[Spaceception]

13 minutes ago, sevenperforce said:

I can’t tell if the upper stage is a Rutherford or an Archimedes. Going with a Rutherford would mean better prop density.

Vacuum optimized Archimedes, as per their site https://www.rocketlabusa.com/launch/neutron/

So I suppose they're going for streamlined reuse and reflight by using the same propellants on the first and second stage.

[RCgothic]

14 minutes ago, sevenperforce said:

I can’t tell if the upper stage is a Rutherford or an Archimedes. Going with a Rutherford would mean better prop density.

But dissimilar propellants and the thrust division doesn't work out well.

7MN on the 1st stage to ~1/40thMN (26kN) on the second stage doesn't seem right.

[RCgothic]

A few more thoughts. So this is a Falcon 9 competitor.

Neutron weighs 480t and can put 15t in LEO expendable or 8t RTLS. That's a payload mass fraction of 1/32 or 1/60th depending.

Falcon 9 weighs 550t and can put 22.8t in LEO expendable or 11t RTLS. That's a payload mass fraction of 1/24 or 1/50.

So F9's quite a bit better in terms of propellant to payload, despite Neutron's higher energy propellants and ultra-lightweight construction. That surprised me. I think I was expecting a little more.

Methane's cheaper than RP1, so that's definitely a point in Neutron's favour.

Expendable second stage - F9US is a significant proportion of an F9 flight. But it uses the same tooling as F9 1st stage. Neutron's ultra-lightweight upper stage doesn't really scream "low cost", and avionics won't be cheaper. Doesn't obviously share tooling with the 1st stage. Unclear how Neutron US would be cheaper TBH.

Do we have any idea on cost, Archimedes Vs Merlin?

Edited December 2, 2021 by RCgothic

[Nightside]

1 hour ago, RCgothic said:

A few more thoughts. So this is a Falcon 9 competitor.

Neutron weighs 480t and can put 15t in LEO expendable or 8t RTLS. That's a payload mass fraction of 1/32 or 1/60th depending.

Falcon 9 weighs 550t and can put 22.8t in LEO expendable or 11t RTLS. That's a payload mass fraction of 1/24 or 1/50.

So F9's quite a bit better in terms of propellant to payload, despite Neutron's higher energy propellants and ultra-lightweight construction. That surprised me. I think I was expecting a little more.

Methane's cheaper than RP1, so that's definitely a point in Neutron's favour.

Expendable second stage - F9US is a significant proportion of an F9 flight. But it uses the same tooling as F9 1st stage. Neutron's ultra-lightweight upper stage doesn't really scream "low cost", and avionics won't be cheaper. Doesn't obviously share tooling with the 1st stage. Unclear how Neutron US would be cheaper TBH.

Do we have any idea on cost, Archimedes Vs Merlin?

Well the 2nd stage is completely protected from airflow by the fairing, so that might simplify it's design a bit.

[CatastrophicFailure]

1 hour ago, RCgothic said:

So F9's quite a bit better in terms of propellant to payload, despite Neutron's higher energy propellants and ultra-lightweight construction. That surprised me. I think I was expecting a little more.

I wonder if, perhaps, they’re counting more on cost optimizations on the reuse side as the major benefit. Beck has said before that for him, reuse is more about launch cadence than outright savings. Since it’s built from the ground up for reuse, Neutron may have a much easier process from landing pad back to launch pad. Quicker turnaround means less boosters needed overall so production can focus on mass-producing the upper stage. 

[GuessingEveryDay]

Can't wait for the Hazegrayart video. That's going to be fun.

[sevenperforce]

Hanging the upper stage on the quadrisymmetric petal fairing reminds me of hanging the Soyuz core on the four Korolev cross boosters.

Time to do some math. What do we know?

• 8 tonnes to LEO reusable
• 15 tonnes to LEO expended
• 1.5 tonnes to Mars or Venus (assumed reusable)
• 480 tonnes GLOW
• 5.96 MN liftoff thrust
• 7.53 MN peak thrust
• 1.11 MN second-stage thrust
• 320 seconds SL Isp

We can draw some inferences from that. First of all, the 5.96 MN liftoff thrust cannot be the full SL thrust of all 7 engines. SL engines grow in thrust as they ascend because underexpanded engines increase in specific impulse as the external pressure drops, but a ratio of 1.26:1 is vastly too high; that would put the vacuum specific impulse of the SL-optimized engines at 404 seconds which is obviously way higher than you can get from methane. I'm not sure why they would launch without full thrust, though.

If we divide the 7.53 MN peak thrust by 7, that gives 1,076 kN per engine on the first stage, slightly less than the stated thrust of the upper-stage vacuum engine, so that checks out. Beck said that Archimedes is a 1-MN engine, and while that may be a round number, let's do the math anyway. Growth from 1,000 kN to 1,076 kN for the SL engine and 1,110 kN for the vacuum engine means the SL-engine gets 344 seconds in vacuum and the vacuum engine gets 355 seconds. That math checks out. The Aeon 1 and Aeon 1V are also GG-methalox and are expected to get 310 s and 360 s, respectively. The closed expander methalox M10 (vac) is projected to get 362 s, and Raptor's FFSC gives it 330-350 for SL and 380 for vacuum. The SL Raptor only gets a 6% increase from SL to space, lower than the 7.6% increase I'm estimating for Archimedes, but that's to be expected because the SL Raptor has such high chamber pressure that it has less to gain.

 

[JoeSchmuckatelli]

4 hours ago, tater said:

Replay:

https://www.rocketlabusa.com/assets/Uploads/neutron-overview.jpg

 

 

 

 

The reentry profile shown in the vid looks very Kerbal.  Or rather, like some of my reentries where I fogot to put a stage separator between my command capsule and my rocket body and just used the engine as a heat shield (some of which actually worked!).  They demonstrate an almost entirely tail-first flight.

Contrast this with the SX belly flop.

The SX approach makes sense to me as you have a lot of surface 'sharing the load' as it uses the atmosphere to slow it down to terminal and then landing velocity.  OTOH, the narrow end-first design looks like it would not be as efficient at using the atmosphere to slow the rocket.  (Analogy of throwing a javelin sideways vs properly)

 

Anyone think their method looks doable - or have I missed something?

[RCgothic]

4 minutes ago, JoeSchmuckatelli said:

The reentry profile shown in the vid looks very Kerbal.  Or rather, like some of my reentries where I fogot to put a stage separator between my command capsule and my rocket body and just used the engine as a heat shield (some of which actually worked!).  They demonstrate an almost entirely tail-first flight.

Contrast this with the SX belly flop.

The SX approach makes sense to me as you have a lot of surface 'sharing the load' as it uses the atmosphere to slow it down to terminal and then landing velocity.  OTOH, the narrow end-first design looks like it would not be as efficient at using the atmosphere to slow the rocket.  (Analogy of throwing a javelin sideways vs properly)

 

Anyone think their method looks doable - or have I missed something?

Looks reasonable. Same as F9 and Superheavy basically.

[JoeSchmuckatelli]

12 minutes ago, RCgothic said:

Looks reasonable. Same as F9 and Superheavy basically.

Thanks - I wasn't aware of the reentry profile of either (although, tbh, I assumed neither actually reenter (or leave) the atmosphere, figured the second stage did the atmospheric exit. 

[Beccab]

13 minutes ago, RCgothic said:

Looks reasonable. Same as F9 and Superheavy basically.

As well as electron and, to a certain extent, NS

[RealKerbal3x]

2 minutes ago, JoeSchmuckatelli said:

Thanks - I wasn't aware of the reentry profile of either (although, tbh, I assumed neither actually reenter (or leave) the atmosphere, figured the second stage did the atmospheric exit. 

Falcon 9 goes above 100km in a normal flight profile, not sure about Superheavy given that stage separation for that happens sooner.

The velocities involved are way lower than an orbital re-entry, so only minimal heat shielding is required coming in engines-first.

[tater]

 

[CatastrophicFailure]

Ahem... 

Beck has officially replaced Musk as "Billionaire most likely to become a Bond Villain."

Also:

 

[sevenperforce]

I made it in KSP. Looks pretty good. Works all right. One of the issues is that you can't do the separation until you're out of the atmosphere, which leads to some inefficiencies.

Beck said that the upper stage being in tension makes it the lightest stage in history. Obviously it's not the lowest-mass stage ever; he probably means it has the best mass ratio for a methalox stage. I believe Centaur has a dry mass fraction of around 9% and F9US has a dry mass fraction of around 4% so if they can get to ~5.5% or so, they would be doing a really good job.

That prompts another question, and some more maths. They say they can deliver 1.5 tonnes to Mars or Venus, each which require about 3.55 km/s for the transplanetary injection. I'm going to go out on a limb and say that's probably flying in expendable mode (although if it's not I'm sure the maths will tell us soon enough). We don't know the dry mass of the stage and we don't know the staging velocity. However, the rocket equation does tell us that you need a propellant fraction of around 64% to get 3.55 km/s with 355 s of isp, so for an interplanetary injection, the residual props in LEO are 64% of the total mass in LEO. Using our 5.5% number, we can estimate that the total mass at staging was 18.2 * dry mass + 1.5 tonnes. Similarly, with an expendable launch to LEO, the total mass at staging was 18.2 * dry mass + 15 tonnes. We can assume that staging velocity is the same because a difference of 13.5 tonnes is going to be negligible in comparison to the total mass of the upper stage and the dry mass and residuals of the lower stage.

If we take the required Δv from staging to LEO as Δv_LEO, then the rocket equation looks like this in each situation (where m is dry stage mass):

LEO expendable:  Δv_LEO = 3483 m/s * ln( ( 18.2*m + 15000 kg ) / ( m + 15000 kg ) )

TPI expendable:  Δv_LEO + 3550 m/s = 3483 m/s * ln( ( 18.2*m + 1500 kg ) / ( m + 1500 kg ) )

Conveniently, you can combine these two equations to eliminate Δv_LEO and solve for m.

3483 m/s * ln( ( 18.2*m + 15000 kg ) / ( m + 15000 kg ) ) = 3483 m/s * ln( ( 18.2*m + 1500 kg ) / ( m + 1500 kg ) ) - 3550 m/s

ln( ( 18.2*m + 15000 kg ) / ( m + 15000 kg ) ) = ln( ( 18.2*m + 1500 kg ) / ( m + 1500 kg ) ) - 1.0192

And using Wolfram to solve (because I hate logarithms) we find that the dry mass of the stage is 4.6 tonnes. Algebra gives us the following projections:

• Stage 2 dry mass: 4.6 tonnes
• Stage 2 propellant: 79.1 tonnes
• Stage 2 methane: 17.6 tonnes
• Stage 2 LOX: 61.5 tonnes
• Stage 2 tank volume: 41.5 cubic meters CH4 + 53.9 cubic meters LOX = 95.4 cubic meters total

Does this pass the sniff test? Yes. A spherical tank with an internal volume of 95.4 cubic meters would have a diameter of 5.66 meters, which fits well within the visual range of what we see from the small upper stage tank inside the 7-meter fairing. It also allows for a nice sporty upper-stage TWR of about 1.2 gees.

These numbers suggest that for an expendable launch, Δv_LEO = 5.63 km/s. For a reusable launch, where the payload is only 8 tonnes, Δv_LEO = 6.91 km/s. This gives us yet another piece of information: staging velocity in an expendable launch is 1.28 km/s greater than in a reusable launch. These numbers also allow us to estimate expendable payload to GTO, which is 4.5 tonnes. It cannot make GTO flying reusably (which is probably why he kept talking so much about constellations).

 

So what else do we know? If total upper stage wet mass is 83.7 tonnes and the total vehicle mass is 480 tonnes, then the lower stage obviously has a wet mass of around 396 tonnes. Using the rocket equation, the reserve propellant for boostback and landing is around 31.6% of the total mass of the stack at separation (since burning it would deliver an extra 1.28 km/s).

Burn time for the upper stage is around 248 seconds, plus throttled flight time, but since throttled flight time doesn't really factor into gravity drag, we'll ignore that. The T/W ratio is good and so we can estimate that the area under the curve for gravity drag (accounting for centrifugal acceleration) is about 1/3 of the total time * 1 gee, or about 810 m/s. So if the upper stage needs 7.8 km/s to reach orbit and is carrying 6.91 km/s and loses 0.81 km/s to gravity drag, it must stage at 1.7 km/s.

 

[sevenperforce]

Playing around with my KSP version -- it does look like they have done a decent job of solving the quintessential "point one way on the way up and the other way on the way down" problem. 

[DDE]

Dumb question, but can you do a Block D-style donut drop tank with methalox? I see some space under the fairing wasted.

[sevenperforce]

10 minutes ago, DDE said:

Dumb question, but can you do a Block D-style donut drop tank with methalox? I see some space under the fairing wasted.

It's easy enough to do a drop tank for any propellant type. However, the upper stage engine nozzle is almost certainly radiatively cooled, so it needs to be open to space to work properly.