I have spent considerable time reading and absorbing Dr. Michael D. Griffin's Remarks to the Space Transportation Association (STA) from Tuesday 22nd January 2008. I decided to wait and mull the remarks fully until I could put together a suitable and considered reply.
I haven't yet read any of the other feedback here on the site as I write this, so I apologise if I cover ground already well-trod. But as a leading representative of one of the teams who has placed itself in direct opposition to certain parts of the existing architecture, this is the first draft of an open letter which I intend to send to Michael Griffin. I would appreciate constructive comments from the members of the forum.
Dear Dr. Michael D. Griffin and the staff of the Constellation Program,
I would firstly like to say a big thank-you to Dr. Griffin for his recent remarks to the Space Transportation Association (STA) from Tuesday 22nd January 2008 . I can say, hand-on-heart, that I agree strongly with about three-quarters of his speech. While I may disagree with the remaining quarter, he makes many excellent arguments and provides extremely valuable insight to everyone into the process behind architecture selections in a program like this.
In particular, I concur with the general assessment that the US needs to rebuild its lost heavy lift launch capabilities we short-sightedly discarded in the 1970's. If we are ever to go to the moon in a truly efficient manner and, more importantly, if we are ever to attempt to expand out towards Mars and other worlds beyond, we need the ability to launch the vast bulk of hardware and resources which will be needed to establish a long-lasting space infrastructure throughout the 21st Century.
Dr. Griffin explains, in truly wonderful detail, the underlying requirements and reasoning which formed both the CAIB findings and the Congressional National Aeronautics and Space Administration (NASA) Authorization Act of 2005. This STA presentation makes for a very powerful and compelling read to anyone wishing to understand the background to NASA's new endeavors.
Obviously, I do have some differences of opinion regarding some specific choices which have been made so far in the specific field of launchers, which I will try to explain and compare in detail below. But I wish it to be very clear that I mostly agree with Dr. Griffin, and Congress, regarding the larger focus of the Vision For Space Exploration.
I agree that the Orion
Crew Exploration Vehicle (CEV) is the best near-term approach for NASA to primarily support the Lunar program, but also to support the existing International Space Station while commercial operations continue to mature their alternatives for LEO access for both cargo, and later crews.
I also agree that the Altair
Lunar Surface Access Module (LSAM) concept is the best general approach as an early-generation Lunar Lander for both Crew and Cargo use. It builds on valuable lessons learned during the amazing Apollo missions, while expanding the capabilities to a whole new level. It also offers an exceptionally robust concept upon which to build enhanced developments over many years into the future.
I further concur with Congress, that getting ISS operational with a minimum crew of six is essential for both scientific and political reasons, and that retiring the venerable Space Shuttle in 2010 when ISS is ready, is the best plan to allow us to fund the new program fully. I completely agree that we need a replacement for Shuttle, to launch the CEV operationally by no later than 2014 too. And that we need a heavy lift launcher for the large missions to the moon and Mars. Dr. Griffin points out many reasons behind the excellent logic of reusing existing Shuttle technology and hardware for the new launchers, as it benefits cost, schedule and the current workforce - a move which preserves all of the valuable knowledge gained over the past 25 years of that complex program.
While I have previously been a vocal and sometimes bitter critic of both Dr. Griffin and also of the current plans to build the two Ares launch vehicles, I wish my disagreements to be seen in the correct context - alongside the many elements I do agree with in the larger theater of the plans for the Vision for Space Exploration. Amazing as it may be to some, Dr. Griffin and I actually share many identical views on many of the basic elements of the new program. While there has been vitriol aplenty, I actually find I support NASA's Administrator far more than I oppose him.
Having outlined many of the elements which I do agree with, I will now concentrate on the specific subjects where I have a difference of opinion and will attempt to carefully explain the different thought processes behind them in the context of Dr. Griffin's remarks.
Dr. Griffin correctly states "Any system architecture must be evaluated first against the tasks which it is supposed to accomplish. Only afterwards can we consider whether it accomplishes them efficiently, or presents other advantages which distinguish it from competing choices."
He goes on to detail the specifics of Presidential policy and Congressional direction, which I will refrain from repeating here in the interests of brevity, but this is excellent reasoning which I would like to make clear forms the backbone of thinking behind the alternative architecture we have proposed in the DIRECT Team, not just the ESAS approach. The two approaches stem from very common ground.
From within the NASA Authorization Act of 2005, Griffin quotes the following: "The Administrator shall, to the fullest extent possible consistent with a successful development program, use the personnel, capabilities, assets, and infrastructure of the Space Shuttle program in developing the Crew Exploration Vehicle, Crew Launch Vehicle, and a heavy-lift launch vehicle."
We agree with this completely. However we suggest a different approach to executing it compared to NASA currently. While NASA has interpreted this "...Crew Launch Vehicle, and a heavy-lift launch vehicle"
instruction to mean two separate vehicles, we have interpreted it differently - to mean a single vehicle capable of flying in two different configurations capable of performing the two different duties. This is a fundamental, but subtle, difference in execution which has quite different results - although both can achieve the targets. I will attempt below, to detail why the "one-vehicle, two-configurations" method would offer NASA significant advantages over the "two-vehicle" approach.
Dr. Griffin continues "Finally, the new architecture must take advantage of existing Space Shuttle program assets "to the fullest extent possible"."
We have made the conscious effort to take this very literally with DIRECT, in that we are deliberately attempting to re-use the maximum possible amount of current Shuttle manufacturing, launch processing infrastructure and flight hardware that we can, with the least amount of new development, cost or new technology delaying the schedule, whilst also achieving all of the objectives and requirements of the three missions: ISS, Lunar and Mars programs. NASA has taken the different approach of using existing Shuttle manufacturing, launch processing infrastructure and flight hardware as a 'point of departure' to create two new vehicles which ultimately have very little real commonality to anything in the existing Shuttle infrastructure today.
Let me spend a moment here to compare all of these systems from this perspective.ESAS Ares-I and Ares-V:
While visibly similar, the Ares-I SRB's are being considerably changed in form, function and performance. The only common element will be the white cylindrical steel cases. All other elements, from parachutes to Aft Skirt, Nozzle to Thrust Vectoring System, Separation Systems to Recovery Parachutes - all are being replaced with brand new items custom-designed and manufactured for the new configuration.
Similarly, while covered with the same orange foam and while fabricated from the same Aluminum-Lithium alloys, neither the 5.5m diameter Ares-I Upper Stage, nor the 10m diameter Ares-V Core Stage or the 10m diameter Ares-V Upper Stage will be manufactured on the same fabrication hardware as the 8.4m diameter Space Shuttle External Tanks are today. The existing manufacturing equipment is mostly being mothballed.
Ares-V requires an engine which does not currently fly yet - an upgraded and human-rated version of the RS-68 used by the USAF Delta-IV program, but a '106%' version with improved performance from that flown today.
Both Ares-I and Ares-V will also need an all-new engine, the J-2X. While visibly bearing a resemblance to its predecessor, the Apollo J-2, the J-2X will essentially be an all-new engine. It is being designed using new techniques, will be built on brand-new manufacturing equipment because the original manufacturing facilities have not existed in more than 30 years and is being re-designed to be 27% higher performance. Additional scheduling pressure is created because this engine must be ready as soon as possible as it is the
crucial element determining the Ares-I schedule to 'close the 'gap' after Shuttle retires.
Ares-V then requires an additional stage too, the Earth Departure Stage (EDS) which will use very similar J-2X engines for power.
Additionally, neither Ares vehicle is planned to utilize much of the existing infrastructure elements from Shuttle - what isn't to be replaced, is to be heavily re-worked. The Ares-I requires two all-new Mobile Launcher's (ML) with an all-new launch tower on each. Additional considerable changes are then required at both of the Kennedy Space Center's launch pads and inside the cavernous VAB to support Ares-I also. After that, the Ares-V will require complete re-building of the existing three Space Shuttle Mobile Launcher Platforms (MLP), with three more massive new launch towers on each, and yet more alterations to the VAB.DIRECT Jupiter:
Compare that extensive list of changes to the single-vehicle solution which DIRECT proposes.
The Jupiter launch systems propose to retain the existing 4-segment SRB's from Shuttle completely unmodified. No new Segments, Nozzle, TVC, Separation Systems, Parachutes, Aft Skirt or any other hardware needs replacing. The existing systems are more than adequate already and have more than thirty years of life in them still.
The Core vehicle will be manufactured on precisely the same equipment used today to turn out External Tanks (ET) for Shuttle. Fundamentally, the tanking sections of Jupiter bear a great deal in common with ET, although are manufactured stronger. Extra production line facilities are needed to prepare the unique parts for the different in-line stage arrangement, but these represent approximately 25% of the total stage, compared to ET. There is sufficient in-common with ET that theoretically, many of Jupiter's Core Stage elements could be made in parallel at the same time as Shuttle ET's, using the same equipment and personnel.
Like Ares-V, the Jupiter also uses the RS-68 from Delta-IV - but does not
require the higher performance variant. It can certainly benefit from additional performance if the USAF does complete their upgrade program as planned by 2012, but Jupiter does not require
that additional performance in order to meet all of its performance targets for all ISS, Lunar and Mars missions. Should the upgrade be canceled by USAF for any reason before completion, it would not be a crippling performance blow to the Jupiter systems.
Then, only limited changes are required to the three current Shuttle MLP's to support Jupiter launches - they will not a complete re-build as Ares-V needs because the Shuttle and Jupiter systems both share a common basic 'footprint' thanks to using the identical 8.4m diameter Core Stage/ET. To support Jupiter, more than 75% of the existing support hardware inside the VAB processing Shuttle's today can be retained without change. The other 25% require only relatively minor modification compared to either Ares vehicle. The KSC pad modifications are also limited to alterations only - not complete replacement - of the existing two fixed launch towers in place currently. The Jupiter system certainly does not require two new ML's and five new launch towers.
Jupiter will need the J-2X engine for its similar EDS to Ares-V's - although this stage is optional for Jupiter and is not required at all in order to 'close the gap' after Shuttle - it is only required for the later Lunar mission phases when real heavy lift becomes a necessity. Without this EDS, the base Jupiter vehicle, called "Jupiter-120", is a 50-ton to LEO launch system. Adding the EDS and one extra main engine (All Jupiter Core's are designed to fit up to three RS-68's and engines are essentially considered 'bolt-on' items when necessary) the larger "Jupiter-232" system increases the same basic vehicle's capability to more than 100-tons. The development schedule for J-2X on Jupiter is also less taxing - it can take 2 extra years - and delivery date of the J-2X has no impact at all for 'closing the gap' after Shuttle - the engine and its EDS are simply not utilized at all on the smaller Jupiter-120 configuration used initially to service the ISS from 2012 thru 2016. The EDS is only required in 2017 to begin Lunar exploration missions - And Jupiter-232 will be ready to do that 2 years sooner than Ares-V will be.
What these differences all boil-down to is that the one-vehicle Jupiter solution can accomplish all of the same requirements as Ares-I and Ares-V together - and actually offers potentially useful additional performance capabilities for both ISS and Lunar missions at the same time - without compromising Mars performance later.
Dr. Griffin very graciously confirms that when "launching two identical vehicles... ...Non-recurring costs are lower because only one launch vehicle development is required, recurring costs are amortized over a larger number of flights of a single vehicle, and the knowledge of system reliability is enhanced by the more rapid accumulation of flight experience."
This is 100% correct. But the precise use of this argument in his remarks does not consider that the "two identical vehicles"
might possibly be designed in such a way to have the option
of flying in a different configuration using the same hardware.
Dr. Griffin states that the "two identical vehicles... ...approach carries significant liabilities when we consider the broader requirements of the policy framework discussed earlier. As with the single-launch architecture, dual-launch EOR of identical vehicles is vastly overdesigned for ISS logistics."
This is certainly very true in the majority of cases. But in the specific case of the Jupiter launcher, the large Jupiter-232 configuration vehicle used for 2-launch Lunar missions can 'become' the much smaller Jupiter-120 vehicle for ISS missions by simply removing the EDS Upper Stage element and one Core Stage main engine. Remember that this turns the 100-ton Lunar launch system into a 50-ton launch system, and that smaller and simpler system has flight costs very similar to Ares-I.
This is a particular circumstance where the assumptions made in order to create the ESAS recommendations have a significant and critical hole in them.
The Jupiter launcher is designed
from the start to be a 100-ton heavy lift launch vehicle suitable for accomplishing the Lunar and Mars objectives of the new program - it is effectively a slightly smaller and considerably cheaper 'Ares-V'. But instead of also developing Ares-I to handle the separate mission requirement of a pure Crew Launch Vehicle, the same Jupiter launcher can also fly in a much smaller configuration and perform all of the duties of Ares-I - and actually offers 25-tons of additional payload capacity which can be utilized in the future as NASA needs.
This unique approach was simply never assessed by ESAS. It simply slipped through the cracks. This approach does, however, offer some major advantages over the two-vehicle approach in place right now so needs to be re-examined as soon as possible to ensure NASA is not missing a valuable opportunity to gain ground. Let me reiterate some of the more significant points:-
1) For Lunar missions, 2 x Jupiter-232 performance is 120% of 1 x Ares-I + 1 x Ares-V.
2) Lunar missions can begin 2 years earlier (dependent entirely on LSAM schedule) with 2 x Jupiter-232 than with 1 x Ares-I + 1 x Ares-V.
3) For ISS missions, 1 x Jupiter-120 ISS performance is >200% greater than 1 x Ares-I.
4) Jupiter-120 can be operational in mid-2012 because of its less demanding technical development requirements. Ares-I is currently slated to miss the Presidential dictate of 2014, with a first crew flight currently due March 2015.
5) Currently, with Ares-I's official Loss Of Crew (LOC) safety assessment down to 1 in 1256, Jupiter-120's analysis indicates it is 12.5% higher safety at 1 in 1413 LOC.
6) Ares-V does not achieve the minimum crew safety requirements of 1 in 1000 LOC as defined by the ESAS Report. Jupiter-232 exceeds this minimum requirement by 12%, so can be flown with humans as a Crew Launch Vehicle.
7) "Full Wrap" Non-recurring Development cost for Ares-I + Ares-V is $30-35bn. Jupiter-120 + 232 is $12-16bn.
Near-term Non-recurring Jupiter-120 Development costs to 'close the gap' are considerably reduced by deleting the requirement for any new SRB's and associated motors, nor requiring the new J-2X engine and an all-new Upper Stage with no heritage at all. The existing RS-68 only requires human-rating, and the development cost of a Jupiter Core Stage is actually less expensive than the Ares-I Upper Stage because of such direct commonality with Shuttle's External Tank. Some recent estimates suggest Jupiter-120 could be operational in 2012 for a DDT&E cost just 55% that of Ares-I due to fly in 2015.
9) A nominal >500-ton Mars mission will require the same number of launches with either system:-
5 launches = 4 x Ares-V + 1 x Ares-I, or
5 launches = 5 x Jupiter-232 (one w/ crew)
We therefore believe that ESAS missed a 'diamond in the rough' which needs to be re-considered afresh.
We believe this alternative approach offers advantages in performance, safety, schedule, workforce retention and both non-recurring and recurring cost.
We are convinced this concept, which has its roots already firmly within NASA's archives (National Launch System and ESAS LV-24/25) would allow NASA a much wider range of opportunities by reducing LV costs and allowing that money to be better spent on missions and additional infrastructure elements to the benefit of the greater VSE.
Towards the end of his remarks Dr. Griffin makes a point of saying "we are resolved to listen carefully and respectfully to any technical concern or suggestion which is respectfully expressed, and to evaluate on their merits any new ideas brought to us. We are doing that, every day. We will continue to do it."
In that same spirit, I deliberately leave all of my vitriol checked at the door. I respectfully bring a concept to NASA's table and I request you and your team take some time to examine the underlying ideas
, the principles
behind DIRECT's Jupiter launchers. I don't want you to take our word for anything. I am not after a mere assessment of the presented numbers - but a reasoned look at the core
of the approach instead - and from there allow your team to consider some of the many variations upon this theme.
The key guiding light, I feel, is this: 'One vehicle designed primarily for optimum Lunar performance, but based as closely as possible on the existing Shuttle Stack in order to keep development costs down, with an optional Upper Stage which can be removed to allow some variety of configurations and performance from the exact same hardware'.
I request you allow your team of talented engineers an opportunity to see what the basic idea
is actually capable of when placed in their hands and fully investigated. I have been personally amazed at the versatility of the system and would hope you and your team will allow the idea the chance I believe it deserves. I firmly believe that you will be able to come up with far better results than our small team has so far with this outline.
I place the concept
in NASA's hands and look forward to reviewing what I hope will be an open assessment of its merits.
Ross B Tierney