HLV Alternative - Kerolox Core

[Ben the Space Brit]

As has been widely reported (and discussed), the Orlando Sentinel has reported (here) that amongst the options Administrator Bolden's HLV review are looking at is a 8.4m kerosene/LOX core stage with five RD-180 engines on the core and a 4 x J-2X hydrolox upper stage.  The LV is also shown (see illustration below) with two liquid-fuelled outrigger boosters, possibly Atlas-V CCBs.

There are some interesting 'wrinkles' to this design proposal.  Firstly, the upper stage is not suitable to act as an EDS, so a seperate EDS "third stage" is required.  Secondly, using this vehicle as a HLV would break the symmetry between CLV ad CaLV that is one of the supposed selling points of the Ares Launch System.  Could this proposal be intended to work in concert with an evolved Atlas-V CLV?

This thread is to discuss the advantages, disadvantages and the possible politics of this proposal (that something tells me might end up being known as 'Ares-V-RP1' or something similar).

Just to get the ball rolling, I have a few questions and alternate options that may or may not have been considered by the MSFC assessment team but might be worth mentioning.

1) The Saturn-IB was essentially a cluser of a Juno and eight Redstone cores all bound together.  Could something similar be done with the Atlas-V CCB to fast-track the kerolox core - say, seven cores, Angara-7 style, with only the outriggers fitted with engines?

2) Could the four-seg shuttle SRMs be used instead of the Atlas-V CCB outriggers?

3) Re.: the EDS; Is there any reason why the ULA ACES-41 could not be used (creating symmetry with the Atlas-V Phase 1)?

On the previous thread on this subject, a launch-to-LEO payload of ~100t was predicted.  What variations on the lunar mission archetecture could using this vehicle allow (1.5 launch? EOR? LOR?)? How much would adding a second pair of Atlas-V CCB outriggers improve the vehicle's performance?

NB: A previous thread on this subject was removed, but I don't know why.  If there is some reason why this thread is inappropriate, I'd be grateful if Andy, Chris or the moderators could let me know for future reference.

[libs0n]

Secondly, using this vehicle as a HLV would break the symmetry between CLV ad CaLV that is one of the supposed selling points of the Ares Launch System.  Could this proposal be intended to work in concert with an evolved Atlas-V CLV?

My money would be on a commercial crew to ISS program utilizing the Atlas V as its prime booster as being the near term NASA manned system being worked on behind the scenes, and the synergistic advantage going to the HLV concept that is coupled with that.  Orion's delivery would be up in the air at this point; HLV booster launch might even be the presumed method taken.

As I said in the last thread, if the mission is flexible path, then that entails a mission mass that is unknown at this point and would be ever changing as the exploration program progressed.  If this HLV concept is proposed as a scalable system wherein changing the number of Atlas boosters would change the launch system's payload capabilities, than that would more lend itself as a HLV to the flexible path than a rocket with a more fixed payload amount.

[zaitcev]

On this topic, Ed Kyle posted an article discussing other alternatives:
 http://www.spacelaunchreport.com/liquidhllv.html
RS-84 is not mentioned.
-- Pete

[wdobner]

I too have a few newbie questions about this proposal.

1). By "5 RD-180 engines" they mean 5 turbopump/machinery sets, but there would be 10 combustion chambers and nozzles, right?

2)  The RD-180 is a regenerative design, right?  I know the RD-170 under the Energia looks like a regenerative engine, but the RD-180 looks a lot like a pair of RS-68s to me.

3)  Is there even an outside chance the design could use 5 or more RS-84s under this thing instead of the RD-180?  It's  not that I buy into the Russian engine paranoia, I'd like to see them head in the general direction of reusability.  It also seems to me that if we develop the RS-84 for our HLV then it'd potentially have applications with the Atlas or Falcon.  Certainly with the Ares V Lite having the RS-68 Regen modifications to be accounted for in development time and money then we could potentially fund and complete an RS-84 development program in with roughly the same time and money.  Or could the Kerolox HLV start flying on RD-180s and then be modified to use RS-84s when their development program was completed?

4)  Why does it appear that they want to use foam insulation around the RP-1 tanks on the Kerolox HLV?  Since RP-1 is not cryogenic then wouldn't a more conventional aluminum body be better? 

5) Is there a chance at least the first stage could be made reusable?  With 4 J-2s in the second stage it sounds like it will stage at relatively a relatively low and slow point in its flight.  Would it be possible to develop a lightweight TPS which would protect the first stage without sacrificing undo amounts of payload?

[edkyle99]

On this topic, Ed Kyle posted an article discussing other alternatives:
 http://www.spacelaunchreport.com/liquidhllv.html
RS-84 is not mentioned.
-- Pete

RS-84 doesn't, and never will, exist.  RD-180 does.  RS-84, a prospective reusable engine, was canceled five years ago.  The heavy lifter contemplated here would not need a reusable engine.

Answering one of the preceding questions, yes, seven RD-180 engines means 14 combustion chambers fed by seven turbopumps.  And yes, RD-180 is regenerative.  It is also a staged combustion engine, which makes it much more efficient than, for example, the Saturn V F-1 engines or the Falcon 9 Merlin 1C engines, etc.

Here are some prospective numbers [updated 11/27/09] for an "Evolved Atlas" with an 8.4 meter diameter core.  Interestingly, the two booster stages don't want to be fully loaded with propellant.  For the version below, the boosters are only loaded 74% full!

If two J-2X engines power the upper stage, and if the upper stage initial T/W is allowed to be 0.7, the numbers work out as follows.

2xCCB Boosters (Combined, 1xRD180 each)
M_propellant_usable = 420 tonnes
M_total = 462 tonnes
Thrust combined = 780 tonnes
ISP_avg = 329 sec
Tburn = 168 sec

Core (5xRD-180)
M_propellant_usable = 1,323 tonnes
M_total = 1,423 tonnes
Thrust = 1,950 tonnes
ISP_avg = 329 sec
Tburn ~= 305 sec (assuming throttle down to 50% at T+60 sec)

Stage 2
M_propellant_usable = 235 tonnes
M_total = 262 tonnes
Thrust = 266 tonnes
ISP = 448 sec
Tburn = 397 sec

Payload Fairing = 10 tonnes
Payload = 118 tonnes
Ideal Delta-V = 9,208 m/s

Vehicle GLOW = 2,275 tonnes
Liftoff Thrust = 2,730 tonnes
Liftoff T/W = 1.2

[Will]

Interesting work, Ed. Am I correct in assuming that if you wanted to optimize for TLI without a third stage, you'd want to make the core somewhat larger and the second stage somewhat smaller for a given GLOW?

It would be interesting to see how a ten meter RS-68 core would work with a pair of Atlas CCB boosters.

And how four Atlas CBC boosters would work with an 8.4 meter RS-68 core.

[sdsds]

Interesting work, Ed. Am I correct in assuming that if you wanted to optimize for TLI without a third stage, you'd want to make the core somewhat larger and the second stage somewhat smaller for a given GLOW?

It would be interesting to see how a ten meter RS-68 core would work with a pair of Atlas CCB boosters.

And how four Atlas CBC boosters would work with an 8.4 meter RS-68 core.

I really like this design, and would like to explore some variations, but had a bit of trouble reproducing the delta-v result.

What amount of delta-v are you getting from the second phase of the core burn, after the boosters have been jettisoned?  If the core were burned at full thrust there wouldn't be any fuel available at that point, so at what point does throttling start?  Also, when does the fairing jettison take place?

Finally, regarding the core diameter:  Does the core diameter effect "ideal delta-v" at all? 

[edkyle99]

Interesting work, Ed. Am I correct in assuming that if you wanted to optimize for TLI without a third stage, you'd want to make the core somewhat larger and the second stage somewhat smaller for a given GLOW?

It would be interesting to see how a ten meter RS-68 core would work with a pair of Atlas CCB boosters. 

And how four Atlas CBC boosters would work with an 8.4 meter RS-68 core.

Strangely enough, a TLI-optimized version wants a slightly *heavier* upper stage and core stage, with even more propellant offloading from the strap on boosters!   With this big upper stage, the rocket can only boost about 35 tonnes to TLI by itself.  It really needs a smaller TLI stage.  A Delta 4 Heavy upper stage carried as part of the LEO payload would be able to boost 45 tonnes or more to TLI.  A heavier TLI stage would be able to accelerate 50 tonnes toward the moon.  This all assumes that the second stage can be restarted to provide part of the TLI burn.

I would rather not see the strap-on boosters.  All seven RD-180s might fit beneath a 10 meter core. 

ESAS looked at a variant with boosters powered by RD-170 type engines - essentially two RD-180s per booster.  (The boosters were essentially Zenit stages.  Note, BTW, that Taurus 2 will use Zenit-diameter tanks built in Ukraine).  More liftoff thrust - more payload, etc.

BTW, this rocket (with two 1xRD-180 strap on boosters) could lift more than 125 tonnes to LEO if three J-2X engines powered the upper stage - in that case the upper stage would gain mass while the core and boosters lose mass.  A fourth J-2X would only buy a few more tonnes to LEO, so probably wouldn't be worth adding.

 - Ed Kyle

[edkyle99]

I really like this design, and would like to explore some variations, but had a bit of trouble reproducing the delta-v result.

What amount of delta-v are you getting from the second phase of the core burn, after the boosters have been jettisoned?  If the core were burned at full thrust there wouldn't be any fuel available at that point, so at what point does throttling start?  Also, when does the fairing jettison take place?

Finally, regarding the core diameter:  Does the core diameter effect "ideal delta-v" at all? 

I've updated my model results above, so recheck them.  I'm assuming core throttle down to 50% at T+60 sec, which is roughly what an Atlas 5 core does.  A rocket like this might not throttle down that much due to the core mass, but this is all an approximation!

With a 118 tonne payload, I get the following delta-v breakdown
.
Core Plus Booster Phase:  2225 m/s
Core-Only Phase:  2727 m/s
Stage 2 Phase:  4256 m/s
Total Delta-V:  9208 ms

Payload fairing jettison would probably be able to take place near the end of the core stage burn.  I've modeled it as happening at essentially the same time as the core/upper stage staging event.

A wider core would have a slight effect on drag losses and propellant mass fractions.  My simple model doesn't go to that level of detail. 

 - Ed Kyle

[Downix]

I would rather not see the strap-on boosters.  All seven RD-180s might fit beneath a 10 meter core. 

My thinking on this is that a 5 RD-180 central core would be a good lifter as/is, but for heavier loads using the Atlas V common core as boosters would give an edge on the lift.  I would not even limit my thinking to just 2, and could easily see 4, 6, or even incase of super-heavy lifts an 8-booster configuration. 

The beauty of this design is the commonality of growth, you get a baseline core configuration which is a lifter by itself, good as a direct shuttle replacement, which you can then grow to fit your need.  Would do even better if the upper-stage is as flexible, using either one or two J2X or AJ-60.  (I'd prefer the AJ's personally, due to lower cost of operation)

[Will]

Interesting work, Ed. Am I correct in assuming that if you wanted to optimize for TLI without a third stage, you'd want to make the core somewhat larger and the second stage somewhat smaller for a given GLOW?

It would be interesting to see how a ten meter RS-68 core would work with a pair of Atlas CCB boosters.

And how four Atlas CBC boosters would work with an 8.4 meter RS-68 core.

Strangely enough, a TLI-optimized version wants a slightly *heavier* upper stage and core stage, with even more propellant offloading from the strap on boosters! 

Sounds like a design that could benefit from cross-feed!

[tnphysics]

Oh yeah, the Core using RD-180s is SSTO with a useful payload.

[Proponent]

A wider core would have a slight effect on drag losses and propellant mass fractions.  My simple model doesn't go to that level of detail. 

Could yo tell us a bit about your model?  I'd be particularly interested in the approach to losses due to gravity or other factors.

[edkyle99]

A wider core would have a slight effect on drag losses and propellant mass fractions.  My simple model doesn't go to that level of detail. 

Could yo tell us a bit about your model?  I'd be particularly interested in the approach to losses due to gravity or other factors.

It's just straight up rocket equation number crunching on a spreadsheet.  I end up with an ideal delta-v number and don't try to figure drag, gravity, and other losses.  Instead, I estimate that 9,200 m/s is enough to reach LEO with a kerosene booster, 9,300 m/s with a hydrogen booster, both accounting for losses.  It is a broad approximation, but it is, I suspect based on comparison with real rocket numbers, within a few percent of what is actually possible.

 - Ed Kyle

[edkyle99]

The beauty of this design is the commonality of growth, you get a baseline core configuration which is a lifter by itself, good as a direct shuttle replacement, which you can then grow to fit your need.  Would do even better if the upper-stage is as flexible, using either one or two J2X or AJ-60.  (I'd prefer the AJ's personally, due to lower cost of operation)

A core only with five RD-180 engines topped by a twin J-2X powered second stage would be able to lift 85 tonnes or so to LEO, which meets the Augustine Committee's minimum criteria for HLLV.  That's one less RD-180 than Atlas 5 Phase 2 (which can do 75 tonnes to LEO), but the trade off is two J-2X engines rather than four RL10s.  With only one J-2X, payload drops below 70 tonnes.

 - Ed Kyle

[Downix]

The beauty of this design is the commonality of growth, you get a baseline core configuration which is a lifter by itself, good as a direct shuttle replacement, which you can then grow to fit your need.  Would do even better if the upper-stage is as flexible, using either one or two J2X or AJ-60.  (I'd prefer the AJ's personally, due to lower cost of operation)

A core only with five RD-180 engines topped by a twin J-2X powered second stage would be able to lift 85 tonnes or so to LEO, which meets the Augustine Committee's minimum criteria for HLLV.  That's one less RD-180 than Atlas 5 Phase 2 (which can do 75 tonnes to LEO), but the trade off is two J-2X engines rather than four RL10s.  With only one J-2X, payload drops below 70 tonnes.

 - Ed Kyle

What if you use 4 RL-10's instead?  Or 4 AJ-29/60's?  (I know, Kerolox on US is vorboten)

[edkyle99]

A core only with five RD-180 engines topped by a twin J-2X powered second stage would be able to lift 85 tonnes or so to LEO, which meets the Augustine Committee's minimum criteria for HLLV.  That's one less RD-180 than Atlas 5 Phase 2 (which can do 75 tonnes to LEO), but the trade off is two J-2X engines rather than four RL10s.  With only one J-2X, payload drops below 70 tonnes.

 - Ed Kyle

What if you use 4 RL-10's instead?  Or 4 AJ-29/60's?  (I know, Kerolox on US is vorboten)

Assuming the same upper stage T/W as Atlas V 401 (with two RL10 engines for LEO missions), four RL10s atop a 5xRD-180 stage would only lift 37 tonnes to LEO.  Six RL10s would do much better at 49 tonnes to LEO.

Four RL60 upper stage engines (these would have been liquid hydrogen engines BTW) would lift 70 tonnes, about the same as the single J-2X alternative mentioned in my previous message.

I'm not familiar with "AJ-29", but an interesting all-kerosene alternative to consider would be to use something like one half of an RD-180 to power a second stage.  An all kerosene/LOX rocket with five RD-180s on the first stage lifting a 1/2 RD-180 type powered (single chamber) second stage (or, heck, a pair of something like the SpaceX Merlin Vacuum engine) would be able to lift 45 to nearly 50 tonnes to LEO.

 - Ed Kyle 

[Downix]

A core only with five RD-180 engines topped by a twin J-2X powered second stage would be able to lift 85 tonnes or so to LEO, which meets the Augustine Committee's minimum criteria for HLLV.  That's one less RD-180 than Atlas 5 Phase 2 (which can do 75 tonnes to LEO), but the trade off is two J-2X engines rather than four RL10s.  With only one J-2X, payload drops below 70 tonnes.

 - Ed Kyle

What if you use 4 RL-10's instead?  Or 4 AJ-29/60's?  (I know, Kerolox on US is vorboten)

Assuming the same upper stage T/W as Atlas V 401 (with two RL10 engines for LEO missions), four RL10s atop a 5xRD-180 stage would only lift 37 tonnes to LEO.  Six RL10s would do much better at 49 tonnes to LEO.

Four RL60 upper stage engines (these would have been liquid hydrogen engines BTW) would lift 70 tonnes, about the same as the single J-2X alternative mentioned in my previous message.

I'm not familiar with "AJ-29", but an interesting all-kerosene alternative to consider would be to use something like one half of an RD-180 to power a second stage.  An all kerosene/LOX rocket with five RD-180s on the first stage lifting a 1/2 RD-180 type powered (single chamber) second stage (or, heck, a pair of something like the SpaceX Merlin Vacuum engine) would be able to lift 45 to nearly 50 tonnes to LEO.

 - Ed Kyle 

whoops, AJ-26/59, my goof.  The re-badged NK-43 russian engines.  More thrust than the J-2X, not as much isp however.

[madscientist197]

Half RD-180 = RD-190 i.e. the Angara first stage engine. Probably would require alterations to be airstartable.

[Downix]

Half RD-180 = RD-190 i.e. the Angara first stage engine. Probably would require alterations to be airstartable.

A *lot* of alterations.  Would be cheaper/faster to use the AJ-26/59 (NK-43) which is already air-startable and has higher performance than the RD-191 as/is.