New Glenn 9x4 discussion

[sstli2]

This discussion about capabilities is hard to do in the abstract. You really need to dig down into the details.

I've been working on a spreadsheet, that I first debuted on the New Armstrong thread, to try to put together an estimate of what the delta-V capabilities of New Glenn actually are. Indeed, there are plenty of assumptions baked into this spreadsheet, and changing any of them by even a bit moves the numbers substantially.

I've linked the latest version of my spreadsheet here. I don't want to put too much emphasis in the numbers I came up with, because there are huge error bounds around them. Instead, I'd highlight some general conclusions from this exercise:

- The current 7x2 is dominated by gravity losses, and it's hard to get the TWR or the TWR/gravity loss curve into a place where that wouldn't be the case¹. I think the current performance is probably severely limited by this.
- The 7x2 with engine upgrades reverses these gravity losses and brings the performance closer to nominal.
- No amount of tinkering with assumptions gets the 9x4 performance close to 70 tons. Best I can do is 50 tons. Therefore, I believe that the target capability of 70 tons to LEO is based on performance characteristics (engine thrust improvements, dry mass optimizations) that are beyond what has been publicly shared to date.

Feel free to play around with this spreadsheet or the assumptions, if you think you can do a better job than I did.

¹ Interestingly, underfilling the propellant tanks also eliminates much of these gravity losses, and also brings the performance closer to nominal.

[spacenut]

Huge rocket for the engine thrust and size.  How much do anyone think they can improve the thrust of the engines? 

[sstli2]

Huge rocket for the engine thrust and size.  How much do anyone think they can improve the thrust of the engines?

The RD-180, the prototypical ox-rich staged combustion engine, has almost exactly double the chamber pressure of the BE-4.

And while thrust depends on a whole host of other factors as well, there's a lot of room there for scaling. They are targeting 640,000 lbf in 2026, but even a thrust level of 750,000 lbf or more I think is well within feasibility.

That said, I don't think this is a matter of increasing thrust indefinitely. The 9x4 is by my estimate a TWR of ~1.3 and a hypothetical 750k lbf BE-4 would imply a TWR of over 1.6. You're well past the point of diminishing returns with respect to gravity losses and what New Glenn really needs at that point are larger tanks to hold more propellant and improve the mass fraction and thus payload to orbit.

[deltaV]

I've linked the latest version of my spreadsheet here.

Thanks for that spreadsheet.

IIUC your 9x4 GS2 model has a propellant mass fraction (PMF) of 244/(244+48) = 84%. That's a lot worse than usual for hydrogen stages, e.g. Centaur V is 54/59.4 = 91%, SLS core stage is 987/(987+98) = 91%, EUS is 129/(129+14)=90%, and Delta IV common booster cores were 1-27/226 = 88%. Why the low PMF?

[sstli2]

I've linked the latest version of my spreadsheet here.

Thanks for that spreadsheet.

IIUC your 9x4 GS2 model has a propellant mass fraction (PMF) of 244/(244+48) = 84%. That's a lot worse than usual for hydrogen stages, e.g. Centaur V is 54/59.4 = 91%, SLS core stage is 987/(987+98) = 91%, EUS is 129/(129+14)=90%, and Delta IV common booster cores were 1-27/226 = 88%. Why the low PMF?

I have a very high dry mass estimate (30 tons) to start with, and then I take the aft portion of that (15 tons) and double it for the 4 engines versus 2 engines. That might be overly conservative as some portion of the aft dry mass is the thrust structure and RCS thrusters.

I'll make two changes. I'm going to revise down the base stage dry mass to 25 tons. And I'll revise the aft dry mass scaling factor to 1.75 instead of 2 from 7x2 to the 9x4. After doing so, I have a PMF of 87%, which should be less of an outlier. The URL should auto-update with these changes.

Edit: Also fixed a formula issue with recovery propellant estimate. This counter-acts the above changes, so the broader themes still hold.

[DreamyPickle]

Are there any near-term payloads that don't fit on 7x2?

For a very long time a big problem for the development of larger rockets was the lack of demand. Falcon Heavy only flies once or twice a year and is comparable to 7x2.

What is the business case for Blue investing in a larger rocket?

[sstli2]

Are there any near-term payloads that don't fit on 7x2?

For a very long time a big problem for the development of larger rockets was the lack of demand. Falcon Heavy only flies once or twice a year and is comparable to 7x2.

What is the business case for Blue investing in a larger rocket?

It's not a question of "doesn't fit", it's a question of how much fits. Both Amazon LEO and AST SpaceMobile would benefit from it. Artemis refueling ops would also benefit from it.

Falcon Heavy presumably doesn't fly often because the logistics of launching and recovering it were too onerous for the regular cadence of Starlink.

[DanClemmensen]

Falcon Heavy presumably doesn't fly often because the logistics of launching and recovering it were too onerous for the regular cadence of Starlink.

For constellations, the metric is cost per satellite for identical satellites. F9 with ASDS recovery is cheaper per-satellite than FH for Starlink V2 mini. If FH were more effective for V2 mini, they would have improved the logistics.

Do we know the cost for 7x2 and 9x4 for Amazon LEO sats?

To normalsize, you must factor in the delivered user data rate per satellite. If a V3 delivers 20x the BW of a V2 mini, a Starship delivering 60 Starlink V3 would be about 40 times the delivered BW of an F9 delivering 30 V2 mini

[ZachF]

This discussion about capabilities is hard to do in the abstract. You really need to dig down into the details.

I've been working on a spreadsheet, that I first debuted on the New Armstrong thread, to try to put together an estimate of what the delta-V capabilities of New Glenn actually are. Indeed, there are plenty of assumptions baked into this spreadsheet, and changing any of them by even a bit moves the numbers substantially.

I've linked the latest version of my spreadsheet here. I don't want to put too much emphasis in the numbers I came up with, because there are huge error bounds around them. Instead, I'd highlight some general conclusions from this exercise:

- The current 7x2 is dominated by gravity losses, and it's hard to get the TWR or the TWR/gravity loss curve into a place where that wouldn't be the case¹. I think the current performance is probably severely limited by this.
- The 7x2 with engine upgrades reverses these gravity losses and brings the performance closer to nominal.
- No amount of tinkering with assumptions gets the 9x4 performance close to 70 tons. Best I can do is 50 tons. Therefore, I believe that the target capability of 70 tons to LEO is based on performance characteristics (engine thrust improvements, dry mass optimizations) that are beyond what has been publicly shared to date.

Feel free to play around with this spreadsheet or the assumptions, if you think you can do a better job than I did.

¹ Interestingly, underfilling the propellant tanks also eliminates much of these gravity losses, and also brings the performance closer to nominal.

I've done calcs and seen similar things, my guess is that the 7x2 and 9x4 performance numbers of 45t and 70t both need prop densification to reach those goals. This will increase first stage propellant by ~8%.

[JEF_300]

Has Blue ever clarified exactly which propellants they are going to supercool? Because I had been assuming just the LOX and LNG, but "densified" LH2 could pick up a fair bit of the slack here, especially when compounded with everything else.

[ZachF]

Has Blue ever clarified exactly which propellants they are going to supercool? Because I had been assuming just the LOX and LNG, but "densified" LH2 could pick up a fair bit of the slack here, especially when compounded with everything else.

Hydrogen melts at 14K and boils at 20K which is a pretty tiny temp window so my guess is probably not

[jongoff]

Has Blue ever clarified exactly which propellants they are going to supercool? Because I had been assuming just the LOX and LNG, but "densified" LH2 could pick up a fair bit of the slack here, especially when compounded with everything else.

Hydrogen melts at 14K and boils at 20K which is a pretty tiny temp window so my guess is probably not

Some of my ex-NASA KSC friends used to work on LH2 densification. It's hard, but you can get an even bigger density increase (up to 20% IIRC) than you can with LOX or Methane, and given LH2 tanks typically being >3x the size of the LOX tank, the juice may very well be worth the squeeze.

~Jon

[deltaV]

I'll make two changes. I'm going to revise down the base stage dry mass to 25 tons. And I'll revise the aft dry mass scaling factor to 1.75 instead of 2 from 7x2 to the 9x4. After doing so, I have a PMF of 87%, which should be less of an outlier. The URL should auto-update with these changes.

Edit: Also fixed a formula issue with recovery propellant estimate. This counter-acts the above changes, so the broader themes still hold.

Thanks for those updates.

If we add 4267 m/s (that’s the delta vee from a 200 km 28.5 degree inclination LEO to GEO) to the target delta vee in sstli2’s updated spreadsheet it gives a 9x4 payload of *negative* 9 tonnes to GEO. It’s possible to tweak the spreadsheet to get a positive payload but I don’t see a plausible way to get the 14 tonnes that Blue claimed without also getting much more than the 70 tonnes to LEO Blue claimed. I reached the same conclusion using my own back-of-the-envelope calculations, so this isn’t just a bug with sstli2’s spreadsheet.

The best way I see to resolve this discrepancy is if the 9x4 numbers Blue quoted are for a vehicle with more than two stages. The most natural third stage for 9x4 would be something similar to Vulcan’s Centaur V but using BE-7s (probably based on cis-lunar transporter and/or Blue Moon tech) (see also hkultala's post that I recently cross-posted), but such a stage would likely give us the opposite problem of performing too well direct GEO relative to LEO. So maybe they’re assuming a lower performing but easy to develop third stage such as something hypergolic using Blue Ring tech, a lightly modified Blue Moon mark 1 (which is smaller than optimal), or Impulse Space’s Helios.

[deltaV]

Are there any near-term payloads that don't fit on 7x2?

For a very long time a big problem for the development of larger rockets was the lack of demand. Falcon Heavy only flies once or twice a year and is comparable to 7x2.

What is the business case for Blue investing in a larger rocket?

It's harder for outsiders to estimate business cases than mass to orbit but here are some guesses anyway.

1. In-space assembly is possible but inconvenient and expensive. It's often easier for customers to launch fewer bigger payloads. This helps with Blue's internal projects such as Artemis and moon base. It also helps encourage external customers with a big payload to design for both 9x4 and Starship instead of targeting Starship alone.

2. Many space use cases such as Starlink-style constellations require hundreds of launches of a 7x2-sized rocket. That's definitely possible - Falcon did it - but cutting the number of launches is likely helpful, especially if non-space folks get annoyed about rocket launches closing airspace all the time and pass new laws or regulations to limit airspace closures.

3. Full reuse will inevitably substantially reduce performance. So they may be building a slightly bigger rocket than they need today so that it will be the right size once they get full reuse working.

[sstli2]

I've done calcs and seen similar things, my guess is that the 7x2 and 9x4 performance numbers of 45t and 70t both need prop densification to reach those goals. This will increase first stage propellant by ~8%.

My spreadsheet includes densified LNG, although it does not include densified LOX. I updated it to include densified LOX at 1.225 mt/m3. I also refined the gravity loss/TWR curve. Payload figures look a little better, but still same overall theme.

[sstli2]

I went ahead and modeled a third stage. Spreadsheet URL updated. All of this remains assumption dependent, so here are the key assumptions:

- BE-7 at 460s specific impulse
- Mixture ratio of 5.8-to-1, comparable to the RL10
- Very mass efficient with a 91% PMF, comparable to the Centaur V (consistent with the prior leaked render of a carbon composite + truss third stage)
- 3 tons of LH2, 17.4 tons of LOX (similar in footprint to Blue Moon MK1), and a dry mass of only 2 tons

As expected, this does wonders for the 9x4 high-energy performance. I have about 28 tons to GTO, 17 tons direct-to-GEO, and 22 to TLI.

It also improves the LEO performance, bringing it to 67 tons. Granted, a 2 ton GS3 might have some issues with in-space propulsion such a large payload. At least on the ground, they could use a special payload adapter to run the payload structural loads around the GS3 and to GS2 tank walls.

As always, I provide the caveat to take all these numbers with a grain of salt, as minor changes in assumptions can affect them greatly. But this exercise is useful to identify broader themes. And one of the main themes here is that it would appear all 3 payload figures for the 9x4 - LEO, GEO, and TLI - are dependent on a 3rd stage.¹

¹ There are other possibilities, but they do not seem likely. The two ways to significantly improve the payload of all versions would be to increase specific impulse and to reduce dry mass on the upper stage. The former is not likely because they are increasing thrust on the BE-3U and specific impulse would run counter to that. The latter is not likely because well, it's difficult, and it's not clear that moving to monocoque would be more mass-efficient than orthogrid. Balloon tanks might, and carbon fiber tanks might, but it's not obvious either of those are in the plan for GS2 (unlike GS3).

[greybeardengineer]

Some of my ex-NASA KSC friends used to work on LH2 densification. It's hard, but you can get an even bigger density increase (up to 20% IIRC) than you can with LOX or Methane, and given LH2 tanks typically being >3x the size of the LOX tank, the juice may very well be worth the squeeze.

Is the 20% end of density increase still pure liquid or is the liquid/ice mixture called "slush LH2" involved? If so, what is the high end of subcooled liquid LH2?

[sstli2]

Some of my ex-NASA KSC friends used to work on LH2 densification. It's hard, but you can get an even bigger density increase (up to 20% IIRC) than you can with LOX or Methane, and given LH2 tanks typically being >3x the size of the LOX tank, the juice may very well be worth the squeeze.

Is the 20% end of density increase still pure liquid or is the liquid/ice mixture called "slush LH2" involved? If so, what is the high end of subcooled liquid LH2?

FWIW NASA has a whitepaper that covers LH2 densification: https://ntrs.nasa.gov/api/citations/20180000059/downloads/20180000059.pdf

[greybeardengineer]

FWIW NASA has a whitepaper that covers LH2 densification: https://ntrs.nasa.gov/api/citations/20180000059/downloads/20180000059.pdf

No that paper describes LH2 subcooling to reduce/delay boil off and extend mission duration. No density increase figures or graphs are included and no mention of that at all as a goal.

[deltaV]

And one of the main themes here is that it would appear all 3 payload figures for the 9x4 - LEO, GEO, and TLI - are dependent on a 3rd stage.¹

For GEO I definitely agree. For LEO you're likely right but I'm not yet convinced we've proven it. I don't see how to rule out the possibility that the LEO figure is with two stages and GEO is with a third stage that's not optimized. Many changes in assumptions affect LEO and GEO performance roughly proportionally so the GEO/LEO ratio is a lot harder to reconcile with modeling than LEO alone is.

¹ There are other possibilities, but they do not seem likely. The two ways to significantly improve the payload of all versions would be to increase specific impulse and to reduce dry mass on the upper stage.

The other other possibility that seems at all plausible to me is if the second stage is actually using some form of parallel staging, e.g. something similar to what's commonly done in first stages such as Falcon Heavy or asparagus staging. (Of course such an arrangement would arguably have 3+ stages depending on your definition of "stage".) Edit: but this is inconsistent with the renders on https://www.blueorigin.com/es-MX/new-glenn/9x4 which show a traditional second stage.