Funding for a domestic liquid engine in the National Defense authorization bill

[MATTBLAK]

I'd like to see the RS-68A become a regenerative nozzle engine. Apparently, that project would cost the better part of $1 billion but would give the Delta IV (all variants) about 10% percent in performance boost. If the hydrocarbon engine issue becomes too complex and the Atlas V goes away, the Delta family still has significant variants to do many missions.

No offense, but that sounds like a terrible investment. In a sane world, AJR would rather develop RS-25E and sell it to both ULA and NASA.

RS-25E has nowhere near the thrust of RS-68A. An RS-68 regenerative would have the high thrust plus an efficiency that was moved much closer to RS-25. RS-68A is a rather heavy engine, though. And of course, RS-68 Regen would make a good engine for the SLS corestage, probably negating the need to build the RS-25 Expendable at all. Commonality of propulsion units across more than one family of launchers could have cost benefits in the long run and bring SLS performance up. The difference in Isp between the RS-25E and the RS-68 Regen would only be about 5% percent, if that.

[USFdon]

PWR said that a regen version of the RS-68 would have a vac ISP of around ~420 secs (at least according to the official nasa sponsored rebuttal to DIRECT). 

[vulture4]

The RS-25 uses tube-wall construction, which is horribly expensive to fabricate. Moreover, taking a reusable engine and making it expendable is insane. The RS-68 Regen would use modern channel-wall construction, much more practical. It would be interesting to see if it can be made reusable.

[93143]

PWR said that a regen version of the RS-68 would have a vac ISP of around ~420 secs (at least according to the official nasa sponsored rebuttal to DIRECT). 

I believe that depends on whether you fix the injector design flaw or not.

The RS-25 uses tube-wall construction, which is horribly expensive to fabricate. [...] The RS-68 Regen would use modern channel-wall construction, much more practical.

The SSME Block III (and presumably by extension the RS-25E) uses channel-wall construction.

[deltaV]

It only makes sense to invest ~$1 billion in a domestic RD 180 if doing so would be cheaper than the alternatives. The obvious alternative is to retire Atlas and use Delta and Falcon. If Falcon works as Elon hopes then most of the launches will be SpaceX launches within 5 years and the high cost of Delta won't matter. On the other hand if Falcon proves unsuitable (e.g. due to poor reliability) then the high cost of Deltas would presumably make building a domestic RD 180 save money eventually. But what if thinks work out somewhere between those extremes? How much Delta use is required before a domestic RD 180 saves money? Just how much more expensive is Delta than Atlas?

[Lobo]

I'd like to see the RS-68A become a regenerative nozzle engine. Apparently, that project would cost the better part of $1 billion but would give the Delta IV (all variants) about 10% percent in performance boost. If the hydrocarbon engine issue becomes too complex and the Atlas V goes away, the Delta family still has significant variants to do many missions.

No offense, but that sounds like a terrible investment. In a sane world, AJR would rather develop RS-25E and sell it to both ULA and NASA.

RS-25E has nowhere near the thrust of RS-68A. An RS-68 regenerative would have the high thrust plus an efficiency that was moved much closer to RS-25. RS-68A is a rather heavy engine, though. And of course, RS-68 Regen would make a good engine for the SLS corestage, probably negating the need to build the RS-25 Expendable at all. Commonality of propulsion units across more than one family of launchers could have cost benefits in the long run and bring SLS performance up. The difference in Isp between the RS-25E and the RS-68 Regen would only be about 5% percent, if that.

The biggest obstacle to standardizing is probably USAF.  They have their RS-68 engine and probably won't want to be tied to a NASA engine.  So I doubt there'd actually be any headway in that area.

But, for fun, if there -would- be a standardization on the RS-25E between SLS and one EELV for USAF/DoD, then take the Delta IV core, keep just the 5m DCSS, replace the RS-68A with RS-25E, and have the core capable of mounting a number of GEM-60 boosters.  Make a Delta II out of Delta IV essentially.  The regen nozzle on the RS-25E could handle being completely surrounded by GEM-60's.  This LV couldn't liftoff without some GEM-60's, so it's base model would probably have 2 or 4 GEM-60's.  But with 8 or 10 of them, it might be able to match or better D4H performance, and probably for cheaper (as I don't think GEM-60's are very expensive).  There'd be just one single core too, rather than the 4 different Delta IV cores they have to build now (Medium, heavy left, heavy right, and heavy center). 
That RS-25E has great ISP, and a single one would give a very long burn with the Delta IV, as it would be burning about half the propellant of an RS-68A.  The 5m DCSS would stage very high, probably much higher than it would for D4H, giving it perhaps more propellant for lofting payloads BLEO.
That higher ISP engine burning for much longer, should give some interesting performance.  Should be a -relatively- easy modification to hardware and  infrastructure.

But, it's hard to say how long those changes would take to pay back from any somewhat lower coasts that a standardized and shared RS-25E might produce.  I doubt USAF would move off of RS-68.

[Lobo]

It only makes sense to invest ~$1 billion in a domestic RD 180 if doing so would be cheaper than the alternatives. The obvious alternative is to retire Atlas and use Delta and Falcon. If Falcon works as Elon hopes then most of the launches will be SpaceX launches within 5 years and the high cost of Delta won't matter. On the other hand if Falcon proves unsuitable (e.g. due to poor reliability) then the high cost of Deltas would presumably make building a domestic RD 180 save money eventually. But what if thinks work out somewhere between those extremes? How much Delta use is required before a domestic RD 180 saves money? Just how much more expensive is Delta than Atlas?

Interestingly, every time Jim has weighed in on the issue of a possible future ULA down select to just one EELV, he's always indicated that Delta IV would be dropped.  This has always puzzled me, because they already have an all-US EELV, which is already flying in a heavy configuration.  That would seem like the most logical downselect, but Jim indicates it is not.  So I wonder what the heck is the problem with Delta IV that they would never downselect to it?, and would choose Atlas with a Russian engine over it?   And why they keep currently flying it if it's just an expensive, undesirable LV?
Why aren't they pushing to cut loose Delta IV right now if it's such boat anchor??

[baldusi]

It only makes sense to invest ~$1 billion in a domestic RD 180 if doing so would be cheaper than the alternatives. The obvious alternative is to retire Atlas and use Delta and Falcon. If Falcon works as Elon hopes then most of the launches will be SpaceX launches within 5 years and the high cost of Delta won't matter. On the other hand if Falcon proves unsuitable (e.g. due to poor reliability) then the high cost of Deltas would presumably make building a domestic RD 180 save money eventually. But what if thinks work out somewhere between those extremes? How much Delta use is required before a domestic RD 180 saves money? Just how much more expensive is Delta than Atlas?

Interestingly, every time Jim has weighed in on the issue of a possible future ULA down select to just one EELV, he's always indicated that Delta IV would be dropped.  This has always puzzled me, because they already have an all-US EELV, which is already flying in a heavy configuration.  That would seem like the most logical downselect, but Jim indicates it is not.  So I wonder what the heck is the problem with Delta IV that they would never downselect to it?, and would choose Atlas with a Russian engine over it?   And why they keep currently flying it if it's just an expensive, undesirable LV?
Why aren't they pushing to cut loose Delta IV right now if it's such boat anchor??

AFAIUI, Delta IV flow design is sort of botched. The cores need a lot of work at the HIF and the pad. They use the pad for solid installation and vehicle integration testing. Thus, no only they need a lot of pad time, but also need a lot of HIF time. And need an infrastructure at the pad that's not far from a VAF. In other words, they have the worst schedule and cost of horizontal integration and vertical integration.
They have two completely different upper stages. The avionics were ancient. They can only be shipped by boat (AV can be airlifted in an AN-124). The lead time is three years (vs two for Atlas V). They don't have a true Common Core, nor will they ever have. And they have to pay a lot to Boeing per core.
As of right now, with the RS-68A, they have the Common Core for M, (4,2), (5,2) and (5,4), and they could implement it for (4,4). But still have the three of Heavy and if they were to cover the whole Atlas V range they would have to develop (5,6) and (5,8) which would require different cores again.
Some things are being solved, like Common Avionics (based of the Atlas V ones), and the Common Upper Stage (again, leveraging a lot of Centaur technology). But the rest of the issues are sort of impossible to solve.
Say that you actually wanted to down select to Delta IV, you'd need both a new HIF AND a new PAD + VAF since they can't get the necessary throughput. Current Atlas V could cover it. But if they needed more cadence for Atlas V, they would just need a VAF+MLP.
They would still need two different upper stages until they develop Common Upper Stage. They would have to add the (5,6) and (5,8) core designs (with a 48month lead time for IOC). They would lose the option of airlifting cores and be dependent on a single vessel for transport. They would lose the option of accepting orders at 24month lead time. They won't be able to switch cores for (5,6), (5,8) and Heavy missions among themselves and the rest. The whole flow would be more expensive.
Overall, as Jim said, an Atlas V Phase II would be a much better alternative. Even a new RD-180 factory would be cheaper and lower risk than moving everything to Delta IV. In fact, if SpaceX actually make an inroad on the DoD missions, I wouldn't be surprised if Delta IV were dropped.

[Lobo]

It only makes sense to invest ~$1 billion in a domestic RD 180 if doing so would be cheaper than the alternatives. The obvious alternative is to retire Atlas and use Delta and Falcon. If Falcon works as Elon hopes then most of the launches will be SpaceX launches within 5 years and the high cost of Delta won't matter. On the other hand if Falcon proves unsuitable (e.g. due to poor reliability) then the high cost of Deltas would presumably make building a domestic RD 180 save money eventually. But what if thinks work out somewhere between those extremes? How much Delta use is required before a domestic RD 180 saves money? Just how much more expensive is Delta than Atlas?

Interestingly, every time Jim has weighed in on the issue of a possible future ULA down select to just one EELV, he's always indicated that Delta IV would be dropped.  This has always puzzled me, because they already have an all-US EELV, which is already flying in a heavy configuration.  That would seem like the most logical downselect, but Jim indicates it is not.  So I wonder what the heck is the problem with Delta IV that they would never downselect to it?, and would choose Atlas with a Russian engine over it?   And why they keep currently flying it if it's just an expensive, undesirable LV?
Why aren't they pushing to cut loose Delta IV right now if it's such boat anchor??

AFAIUI, Delta IV flow design is sort of botched. The cores need a lot of work at the HIF and the pad. They use the pad for solid installation and vehicle integration testing. Thus, no only they need a lot of pad time, but also need a lot of HIF time. And need an infrastructure at the pad that's not far from a VAF. In other words, they have the worst schedule and cost of horizontal integration and vertical integration.
They have two completely different upper stages. The avionics were ancient. They can only be shipped by boat (AV can be airlifted in an AN-124). The lead time is three years (vs two for Atlas V). They don't have a true Common Core, nor will they ever have. And they have to pay a lot to Boeing per core.
As of right now, with the RS-68A, they have the Common Core for M, (4,2), (5,2) and (5,4), and they could implement it for (4,4). But still have the three of Heavy and if they were to cover the whole Atlas V range they would have to develop (5,6) and (5,8) which would require different cores again.
Some things are being solved, like Common Avionics (based of the Atlas V ones), and the Common Upper Stage (again, leveraging a lot of Centaur technology). But the rest of the issues are sort of impossible to solve.
Say that you actually wanted to down select to Delta IV, you'd need both a new HIF AND a new PAD + VAF since they can't get the necessary throughput. Current Atlas V could cover it. But if they needed more cadence for Atlas V, they would just need a VAF+MLP.
They would still need two different upper stages until they develop Common Upper Stage. They would have to add the (5,6) and (5,8) core designs (with a 48month lead time for IOC). They would lose the option of airlifting cores and be dependent on a single vessel for transport. They would lose the option of accepting orders at 24month lead time. They won't be able to switch cores for (5,6), (5,8) and Heavy missions among themselves and the rest. The whole flow would be more expensive.
Overall, as Jim said, an Atlas V Phase II would be a much better alternative. Even a new RD-180 factory would be cheaper and lower risk than moving everything to Delta IV. In fact, if SpaceX actually make an inroad on the DoD missions, I wouldn't be surprised if Delta IV were dropped.

If all of that's true, then why the heck is ULA screwing around with Delta IV at all?  If it's such a kludge that it could never be downselected to, could never take over for Atlas, and is so expensive and takes so much longer to process...why is it flying?  It just sucks up a bunch of extra resources and taxpayer money, and provides -no- redundancy for Atlas.  Where it seems Atlas provides full redundancy for Delta IV, and could launch all but the occasional D4H mission.  And it could duplicate that with a new 5m wide body Centaur upper stage (accodring to Jim.  Atlas Phase I essentially).  So Delta IV just gets some token payloads to keep operational as some sort of make-work program??

Why even operate it, if it's whole point of existing (redundancy) isn't even there?  Cut the losses and retire it now then.
Let's get a US-made RD-180 replacement in development ASAP, as well as a 5m WBC upper stage.  The money saved by retiring Delta IV and closing it's pads would probably pay for them, as you mentioned.  And as you said, if more Atlas capacity is needed on the East coast, another VIF and ML could be constructed for shared use of the LC-41 pad, as it was basically designed for anyway.  The LC at VAFB is adequate for the number of combined Atlas and Delta West coast launches I would think.  The increased production rate/usage of a US-made RD-180 replacement by Atlas flying all ULA missions would also help offset it's development and likely higher per unit costs.

I'm curious to know from Jim or others in the know, why Delta IV is flying at all then.  Politics?  make-work pork?

Unless NASA wants to pay for it, I don't think Atlas Phase II is getting off the back burner.  It's not needed for USAF/DoD.  Atlas Phase 1 can handle D4H payloads, and thus handle all USAF/DoD payloads.  Only need one RL-10 variant then, probably the RL-10C-B as there's enough room for that big nozzle under a 5m upper stage.  (with a possible move to NGE or RL-60 later, if RL-10's are retired all together at some point.)

Maybe we need a thread for this discussion?  ;-)

[butters]

AFAIUI, Delta IV flow design is sort of botched. The cores need a lot of work at the HIF and the pad. They use the pad for solid installation and vehicle integration testing. Thus, no only they need a lot of pad time, but also need a lot of HIF time. And need an infrastructure at the pad that's not far from a VAF. In other words, they have the worst schedule and cost of horizontal integration and vertical integration.

How is the Atlas flow any better than the Delta flow? For Atlas, the entire vehicle has to be stacked in the VIF, rolled out to the pad, launched, and the MLP rolled back into the VIF before the next vehicle can begin stacking. Delta splits the flow between HIF and MST, permitting the flow to be pipelined if necessary, with one team preparing the next vehicle in the HIF while another is stacking the current vehicle in the MST. Delta minimizes vertical processing, which is more complex than working in the horizontal.

It was my understanding that Delta is more expensive primarily because of the cost of RS-68 vs. RD-180 and perhaps also because it's a substantially larger airframe with more sophisticated fluid systems for handling large quantities of liquid hydrogen.

It seems that ULA is paying about $10M apiece for RD-180. They are undoubtedly paying quite a bit more for each RS-68.

[Jim]

1.  How is the Atlas flow any better than the Delta flow? For Atlas, the entire vehicle has to be stacked in the VIF, rolled out to the pad, launched, and the MLP rolled back into the VIF before the next vehicle can begin stacking.

2.  Delta splits the flow between HIF and MST, permitting the flow to be pipelined if necessary, with one team preparing the next vehicle in the HIF while another is stacking the current vehicle in the MST. Delta minimizes vertical processing, which is more complex than working in the horizontal.

1.  It can do a launch every 45 days.  There is also offline testing in the ASOC.

2,  There is no testing capability in the HIF.  The solids and payload are still installed at the pad.  Delta still lot of work at the pad.  Only launch mount and upperstage mate are done in the HIF (there is no other work done).  Launch intervals are longer than 60 days.

[USFdon]

Won't the HIF become kind of redundant (save for the Heavy) if they move to centaur based upper stages? I was under the impression that they have to be stacked.

[Hauerg]

Can somebody pls explain why  I keep reading that even for an "US" version of the RD-180 "parts" from Russia are needed?
If this is true, then what is the point of (more expensive) US production?
Thanks.

[AncientU]

Can somebody pls explain why  I keep reading that even for an "US" version of the RD-180 "parts" from Russia are needed?
If this is true, then what is the point of (more expensive) US production?
Thanks.

References that you've seen?  I am also interested in the 'spare parts' that seem to be needed to avoid impacting the planned launches... is this the 'parts' to which you refer?  It does seem that these parts are beyond the scope of the Warehouse full of RD-180s. 

Maybe we are talking about two different things, though...

[baldusi]

AFAIUI, Delta IV flow design is sort of botched. The cores need a lot of work at the HIF and the pad. They use the pad for solid installation and vehicle integration testing. Thus, no only they need a lot of pad time, but also need a lot of HIF time. And need an infrastructure at the pad that's not far from a VAF. In other words, they have the worst schedule and cost of horizontal integration and vertical integration.

How is the Atlas flow any better than the Delta flow? For Atlas, the entire vehicle has to be stacked in the VIF, rolled out to the pad, launched, and the MLP rolled back into the VIF before the next vehicle can begin stacking. Delta splits the flow between HIF and MST, permitting the flow to be pipelined if necessary, with one team preparing the next vehicle in the HIF while another is stacking the current vehicle in the MST. Delta minimizes vertical processing, which is more complex than working in the horizontal.

First, is the issue of cost. Atlas V infrastructure is cheaper than Delta IV for a given launch rate. Delta IV needs two big building and Atlas only one. The upkeep is higher for Delta IV infrastructure.
Second is the issue that Atlas V arrives pretty complete from the factory. Thus, a lot of time is saved at the pad. This is a flow design issue.
Third, one extra MLP could accelerate things. One VIF+MLP doubles the launch rate. Which is way cheaper than HIF+VAF+Pad.
It's difficult to point to a single issue. Is more like a series of small issues that get compounded. The fact is that if Delta IV were to launch a lot of commercial satellites, that could be integrated on the HIF while a DoD is in the pad, it could work a lot better. But it isn't cheap enough for that. And as I understand it, the RD-180 contract was renegotiated and increased the prices significantly. It's the whole package the big cost. If the average launch costs are 400M, and an RS-68A is 40M (which I believe I'm exaggerating), is still a small difference.

[TrevorMonty]

This may have been covered elsewhere. 

How does blockage of RD180 effects non DOD launches scheduled for Atlas?. DC has booked a 2016 flight, not sure about CST100 plus there maybe NASA flights booked for next few years. In interests of national security would DOD reserve these Atlas LVs for themselves.

[Lobo]

First, is the issue of cost. Atlas V infrastructure is cheaper than Delta IV for a given launch rate. Delta IV needs two big building and Atlas only one. The upkeep is higher for Delta IV infrastructure.
Second is the issue that Atlas V arrives pretty complete from the factory. Thus, a lot of time is saved at the pad. This is a flow design issue.
Third, one extra MLP could accelerate things. One VIF+MLP doubles the launch rate. Which is way cheaper than HIF+VAF+Pad.
It's difficult to point to a single issue. Is more like a series of small issues that get compounded. The fact is that if Delta IV were to launch a lot of commercial satellites, that could be integrated on the HIF while a DoD is in the pad, it could work a lot better. But it isn't cheap enough for that. And as I understand it, the RD-180 contract was renegotiated and increased the prices significantly. It's the whole package the big cost. If the average launch costs are 400M, and an RS-68A is 40M (which I believe I'm exaggerating), is still a small difference.

Baldusi,

First, thanks for the information.  I'm learning a lot in this discussion.
Secondly, this keeps begging my previous question more and more.  Why are we screwing around with Delta IV at all then?  If it's not even capable of being a redundancy to Atlas for assured access to space?  Other than Delta IV-heavy, which could be covered by Atlas Phase 1, what can Delta do that Atlas cannot?  Why do we keep throwing money at a system that's only good for -some- USAF and DoD missions, has a slower process flow, isn't nuclear rated, is more expensive, etc?
Why not just fly the one LV that could do everything Delta IV does, plus everything Atlas does anyway?

[Tea Party Space Czar]

Baldusi,

First, thanks for the information.  I'm learning a lot in this discussion.
Secondly, this keeps begging my previous question more and more.  Why are we screwing around with Delta IV at all then?  If it's not even capable of being a redundancy to Atlas for assured access to space?  Other than Delta IV-heavy, which could be covered by Atlas Phase 1, what can Delta do that Atlas cannot?  Why do we keep throwing money at a system that's only good for -some- USAF and DoD missions, has a slower process flow, isn't nuclear rated, is more expensive, etc?
Why not just fly the one LV that could do everything Delta IV does, plus everything Atlas does anyway?

It is all about redundancy - which is not a bad thing.  After Challenger, DoD was in a real bad place.  There were a lot of lessons learned on the military side.  Having all your eggs in one basket being one of them; a multiple launch system was envisioned.  That is how Atlas and Delta grew into what they are today.

If someone has a bad day you want to be able to fly quickly while an investigation to figure out what went wrong occurs.

I do think we as a nation can do better and we can modernize as we go (SpaceX is proving this).  The Atlas and Delta lines compliment and augment each other.

Respectfully,
Andrew Gasser

[Lobo]

Baldusi,

First, thanks for the information.  I'm learning a lot in this discussion.
Secondly, this keeps begging my previous question more and more.  Why are we screwing around with Delta IV at all then?  If it's not even capable of being a redundancy to Atlas for assured access to space?  Other than Delta IV-heavy, which could be covered by Atlas Phase 1, what can Delta do that Atlas cannot?  Why do we keep throwing money at a system that's only good for -some- USAF and DoD missions, has a slower process flow, isn't nuclear rated, is more expensive, etc?
Why not just fly the one LV that could do everything Delta IV does, plus everything Atlas does anyway?

It is all about redundancy - which is not a bad thing.  After Challenger, DoD was in a real bad place.  There were a lot of lessons learned on the military side.  Having all your eggs in one basket being one of them; a multiple launch system was envisioned.  That is how Atlas and Delta grew into what they are today.

If someone has a bad day you want to be able to fly quickly while an investigation to figure out what went wrong occurs.

I do think we as a nation can do better and we can modernize as we go (SpaceX is proving this).  The Atlas and Delta lines compliment and augment each other.

Respectfully,
Andrew Gasser

But it's -not-.  That's my point.  I thought that too for a long time.  And so was scratching my head when Jim and others kept saying there's no way they could down select to Delta IV.  Why not?  Just cut loose that LV that depends on Russian relations and supply, and fly the all- US one.  Seems like a no-brainer.
Because it appears for a variety of reasons, Delta is -not- redundant to Atlas V.  Atlas can do what Delta can, but not vice versa.  Maybe initially that was a reason USAF kept them both.  Probably Atlas V using RD-180 had something to do with it (in just such an event of a supply problem with Russia), as did the Boeing theft issue.
But certainly now it's clear Delta isn't backing up Atlas.  It's just splitting the government workload.  Ensuring two LV being underutilized instead of just one.  But without cross redundancy.
We may not have known that in the early 2000's, but we do now.  So again, what's the reason to still be flying Delta IV?  Wouldn't it be better to secure a US supply for an RD-180 replacement ASAP, to shore up the one nagging issue of Atlas V?  Now that it has a 12 year great track record and the reliability of two new LV's isn't a question like it was in the early 2000's?

[JBF]

The Atlas V 552 has never flown and the Atlas V heavies would require substantial launch pad infrastructure upgrades.