The Ergosphere
Tuesday, August 02, 2005
 

It's about time

NASA is finally going to quit messing around with the Model T of space vehicles and go back to what works; the Shuttle's replacement will be two vehicles, one heavy-lift cargo hauler and one much smaller people-carrier.  Both will use Shuttle SRB's, and the heavy-lifter will use SSME's as well.

NY Times coverage.

(Hat tip:  Slashdot.) 
Comments:
Wonderful. Now we'll never get back to the Moon.
 
Never?  This re-establishes a Saturn-class lifter, which is a prerequisite for manned missions beyond LEO (the Shuttle is quite incapable of such travel).  Abandoning Shuttle is the first step on the road back.
 
Two problems.
1. The simplest problems are that this Saturn-class booster is not a) man-rated or b) designed to go to the Moon. It's designed to put large cargoes in LEO. This gets rid of one of the main advantages of the Shuttle--that you could put something large in orbit (like, say, space station components) and do something with it, and gets us no closer to the Moon.
2. It diverts money, time, and people away from developing an actual lunar program. It took eight years and change from the last landing on the Moon to the first Shuttle launch. Most of the engineering was done after 1975 because people were working on Skylab and ASTP. But the real problem here is that it took years to develop Shuttle. It will take years to redesign it as two separate spacecraft. It will then take years to land on the Moon again. Realistically, what is this vehicle going to do? If we abandon the Shuttle after the ISS is complete, and then want to land on the Moon, what is a heavy-lift LEO launcher going to do?

This represents a political decision. Somebody in power doesn't want to land on the Moon, be it George Bush, NASA management, or some other random turkey.
 
What we have to do first is use the Shuttle to build the rest of the ISS, then retire it and go to the Moon.

The Shuttle was a compromise NASA-USAF design which was abandoned by USAF, meaning NASA is stuck with many design features they didn't like and can't use. A small manned capsule should have been put on a simple panel-based payload bay using a reusable LOX/LH2 staged system (probably 1 1/2 stage like the Shuttle). This would have allowed large payloads to be taken into orbit, and with modifications to the payload bay, back down. If we had only known in 1972.

Why, again, does building the Shuttle (almost) right the second time get us closer to the Moon?

Two shuttle accidents are more than enough for me. Apollo 17 was enough for me. We never should have ended the Apollo program. We should have just kept upgrading and kept going. The way it happened, we wasted all the infrastructure and R&D money by letting the results rot. Most of the engineers who were involved in building Apollo died with their knowledge of the systems. Redoing it today, we'd have to start from scratch and in today's sociopolitical environment, it would take longer to get back to the Moon than it took to get there the first time.
 
Goodness, Stewart.  It's been ages since I was a true space geek and I can still pick your objections apart point by point:

"1. The simplest problems are that this Saturn-class booster is not a) man-rated or b) designed to go to the Moon. It's designed to put large cargoes in LEO."

Curious you should say that.  Every part of the stack save the new fairing, from hydrogen motors to SRB's, is man-rated.  The Saturn V booster and the second stage happened to do nothing more than put the third stage of the stack into orbit.  On top of this, the second-stage motor for the manned booster is a re-engineered J-2... the same motor used in the two hydrogen-powered Saturn stages.

In short, you could trivially man-rate the heavy booster; you could put the manned capsule and its escape system on top of a cargo fairing.  Or you could just ignore the issue and launch your crew on a second vehicle, hooking up in orbit.  It's not like we forgot how to do a rendezvous.

"2. It diverts money, time, and people away from developing an actual lunar program."

As opposed to the Shuttle program, which did even more of the same without providing a vehicle which could get people anywhere near the moon.

"But the real problem here is that it took years to develop Shuttle. It will take years to redesign it as two separate spacecraft."

Is that so?  What part?

1.  The SSME's are done.
2.  The SRB's are done.
3.  The ET would no longer require insulation to prevent icing, or fancy non-shedding foam.
4.  We no longer need wings, landing gear or most of the other Shuttle systems.
5.  Are you seriously arguing that we no longer know how to build an Apollo-class re-entry capsule?!

If you gave this job to an outfit like Von Braun's NASA or the Lockheed Skunk Works, you'd have the vehicle flying within 2 years.  They have about as much to do as the engineers charged with making a first stage to launch the Apollo-Soyuz mission.  The Saturn 1B was cobbled together from parts of other rockets (link,
link).

"It will then take years to land on the Moon again."

And how long would it take with a Shuttle which can only loft 20 tons (and cannot be uprated) and is due for retirement?

"If we abandon the Shuttle after the ISS is complete, and then want to land on the Moon, what is a heavy-lift LEO launcher going to do?"

I dunno... maybe put the lunar-injection stage, the lunar lander and the return vehicle in place for the TLI burn?

" What we have to do first is use the Shuttle to build the rest of the ISS, then retire it and go to the Moon."

It would make a lot more sense to complete the ISS by building a 100-ton module or set of modules to go up in that payload fairing, and launch the whole kit and kaboodle with one rocket.  (Did you look at the diameter of that thing?  It's bigger than the ET, which is more than twice the size of the payload bay.  You could cluster 4 ISS modules inside it and have space left over for another Hubble telescope.)  You wouldn't have to worry about anyone dying if the launch failed, either.

"We never should have ended the Apollo program."

It's ironic that you lament the decision to return to a proven model of getting to space as "a political decision", then turn around and denounce the political decision which gave us the Shuttle.
 
Every part of the stack is man-rated (almost), but the stack itself isn't. Systems integration is a huge challenge in aerospace. Furthermore, you can't stack a payload on top of the ET as it is now. It was not designed to bear loads that way. 'Re-engineered' is the key word here. The fact that the system requires massive changes would make it a lot further away from being man-rated than you think.

Let me make my views clear on the Shuttle. We need it to build ISS. We needed it to launch and maintain large payloads, like Hubble, for example. We don't need it to go to the Moon and in fact can't go to the Moon while we have it. What we really don't need is to build the Shuttle semi-right the second time. Then we're stuck with a system we shouldn't have built, can't use, and prevents us from doing anything while we have it. We need to complete the ISS and then go to the Moon using a totally different vehicle designed for that purpose. If I was not clear before I hope I am now.

The components are done, to a certain extent. The vehicle isn't. Cobbling something together that was designed to work in a specific configuration is harder than you think. If this was actually done, we would soon find that we needed to redesign many of the "done" components.

Yes, I am saying we no longer know how to build Apollo. Most of the plans were destroyed. Most of the engineers who knew the systems are dead. The technology is very different. We would have to start off from scratch.

The Saturn 1B was designed in the 1960s to test the Saturn V's engines and to work as a LEO launcher for Apollo. It was most certainly not designed between 1972 and 1975. The V-2 was a "cobbled-together" project--it was necessary because [brilliant insight] rockets don't scale up [/brilliant insight]. It took somewhere around six years. There is no time advantage to combining existing components; at most there is a cost advantage, and the only really effective COTS area is avionics.

We cannot land on the Moon with a Shuttle. Let's build a system that works for the purposes we'll be using it for.

A launcher made from Shuttle components will not work for translunar missions. There's too much propellant in too few stages, for one. A heavy-lift launcher cannot be easily converted (if at all) into a translunar booster. The missions are so fundamentally different that you have to design the boosters differently.

It would make a lot more sense to save $15 billion, use what we have for the next six years, and then build a lunar vehicle. Building something else to finish construction would only delay ISS construction by six to eight years, as well as delaying a Moon landing by an equal or greater amount of time.

The decision to go ahead with a shuttle replacement when we're supposed to be pursuing a lunar landing program would be a political decision that would be indicative of a lack of commitment to a lunar program. The Shuttle shouldn't have been designed the way it was. Now, however, is not the time to correct it when we have a completely different set of mission requirements. Of course we'll learn from it, but we shouldn't pretend it's 1972 and build another Shuttle.
 
The illustration in the times for the lifting vehicle was miss-labeled go to the SafeSimpleSoon.com web-site.

Personally, I am thrilled that people connected to a government program have figured out a way to solve problems by recycling and reusing existing components rather than just throwing wads of money at it.
 
OT, but interesting:
Shortages Stifle a Boom Time for the Solar Industry
 
What exactly has human spaceflight ever done for us peons anyway?
 
What has human spaceflight done for us?
-MRIs
-Smoke detectors
-The modern (i.e., not teletype-swiches-lights) computer
-Much research in medicine that was done on space stations and shuttles
-Accurate weather forecasting
Just off the top of my head.
 
Mr. Gloor, I find you saying more and more absurd things with each post.

"It is not a trivial task to mount SSME's on a completely new platform."

I'm sure that it takes reasonable care, but no more than that; if it did, standard articles like test stands would have been useless in its development.

"The SSME are the highest performing rocket engines ever designed and stretch material and processes to the edge. You cannot simple bolt them onto a completely different platform and expect them to work.... To my mind it would be akin to developing it all over again..."

You're saying that:
1.  Replacing the Orbiter from the thrust-frame forward with a wingless cargo pod would be akin to developing the engines all over again.
2.  Moving the engines from a position beside the fuel tank to directly beneath it would be akin to developing the engines all over again.

To make such a bold yet highly questionable assertion you must have much relevant education or experience, like being one of the SSME designers or test engineers.  Would you care to list it?

"A far far better route would be the lower performance RD-180 used on the Atlas V. Additionally it uses LOX/RP7 so the lower stages can be more compact and easier to build. The Saturn V used LOX/RP7 for the very good reason that LH2/LOX's high specific impulse advantages are lost on lower stages."

The Atlas is typically used in conjunction with the Centaur upper stage.  The Centaur uses the RL-10 engine, which burns LOX/LH2.

RD-180:RL-10::SRB:SSME.  Not only is the RD-180 an inappropriate replacement for the SSME, using them on a Shuttle-derived HLV would mean clustering them, with the greater probabilities of failure.

"The SRBs are man-rated but only over the most strident objections of just about everyone."

So, launching the people-carrier with just one of them instead of two (half the probability of failure), and well behind the passengers instead of alongside them, is a BAD thing?

The design does come from the builder of the SRB.  If I were managing this and there were no pork requirements, I'd ask someone to look at pressure-fed "boilerplate" rockets burning cheap room-temperature liquids.  But I'm not, and there are.

"The capsule is just a total throwback. While it is cheap and 'easy' it has no cross range or divert capability."

The Apollo capsule had considerable lift, and hundreds of miles of cross-range capability.  The proposed capsule is a very flattened cone like Apollo and would have similar aerodynamics.  The Shuttle's cross-range capability was demanded by the military for a mission profile that has never been flown and never will be.

What divert capability do you need when you're trying to hit e.g. Nebraska?

"A winged re-entry vehicle has the advantage of landing like a plane."

In other words, if you can't get properly lined up on a nice, long, paved runway, your vehicle is toast and possibly cargo and crew as well.  A capsule needs an appropriate surface roughly as big as its underside; to a Soyuz, steppe is thousands of square miles of landing pad.

So it's not glamorous.  I rather like flying myself, especially those landings when I can just grease it on, but space is a different mileu and calls for different methods.

"Lastly the main reason that there are no 100t launchers at the moment is that there are no 100t payloads that need to be launched."

Not since Apollo.  In other words, you're ruling out both the Moon and Mars a priori.

Before commenting again, I'd appreciate it if you'd (a) RTFA, and (b) think about what you're saying.
 
Stephen, perhaps you'd like to dispute the conclusions of a real rocket scientist.

Namely, Robert Zubrin.

He doesn't agree with anything you've said.
 
Mister Gloor, you're not paying attention.

1.  The "more capable" variants are the ones with capacities much greater than 70 tons.  (A Shuttle is capable of landing at weights upwards of 99 tons, so a basic thrust structure and aeroshell weighing 20 tons should allow 79 tons of cargo.)  The SSME itself weighs 6990 pounds, and 9.5 tons should be plenty for a lightweight shell (what I can find about the Shuttle Orbiter thrust structure indicates that it weighs about 4500 pounds).

2.  Go back to your Hickam quote.  Read it again.  Notice that the Shuttle-derived vehicles, HLV and crew carrier both, get rid of:
    a.  The Orbiter with its fragile TPS.
    b.  People next to SRB's (the HLV carries no people, the crew capsule is on top of the people carrier).

"So it is my rocket engineer against yours I guess."

No, Hickam and Zubrin are in broad agreement; they both disagree with you.  The proposal is exactly the sort of "return to drawing board" that both of them have called for.
 
I sort of agree with Stephen Gloor here. The SSME's should be avoided. And the design for the heavy lifter seems rather ambitious for the little we currently have planned. Also, I really don't trust Zubrin's judgement in this matter.

Having said that, I don't see a real problem with much of the designs. If they can find or make substantial launch volume for these rockets, then the costs will probably be reasonable.

However, it strikes me that Mr. Gloor is right in another area. We also have an opportunity to encourage private industry to enter in this range. All NASA has to do is promise enough business (IMHO).

At some point NASA will need to exit the launch industry. It's not their job and they'll be taking business away from US launch companies.
 
"The SSME's should be avoided."

Why?  They've never been associated with a mission failure; though a premature shutdown once did cause one mission to wind up in a lower orbit, it was a sensor malfunction and not an actual mechanical failure.  They're also the only hydrogen engines big enough to do the job, and they happen to be in production.

I suppose we could engineer something based on the aerospike engine of the VentureStar, which would probably have nice performance if it was expanded to cover the entire aft end of an ET with virtual nozzle.  But that would cost money to develop; we can buy SSME's off the line.

"And the design for the heavy lifter seems rather ambitious for the little we currently have planned."

That's another way of saying we shouldn't plan anything bigger, especially not Moon missions, Mars missions or e.g. new space stations lofted in a single shot.

The little lifter with the crew capsule would be much cheaper than a Shuttle.

But much as I'd love to see NASA get out of the launch business, I don't see that happening as long as Washington is footing the bill; the only way they're going to write the checks is if they can get pork for the folks at home, and the winner of a competitive bid is all too likely to be in the wrong district.  We'll see NASA get out when the business of space hotels or something like that gets going.
 
E-P, IMHO the SSME just isn't suitable as an engine for several reasons.

The SSME haven't overtly been linked to a mission failure. However, I suspect they are a leading cause of mission cost and delays for two key reasons. First, they are extremely complex. That means increased costs of building one and of inspecting and refurbishing one after each mission. Second, they leak. I suspect hydrogen leaks are probably the leading cause of delay for a shuttle mission.

Well, here's some supporting evidence. Doesn't quite align with my suspicions, but agrees in principle.

"The SSME is the single largest contributor to Shuttle delays & escalating costs. The DDT&E cost accounted for 25 per cent of the total Shuttle development cost, and various technical problems pushed back the STS-1 launch date by two years. Nothing about the SSME is ordinary; it costs $40 million per copy and Rocketdyne needs four years to produce each new engine."

[something is wrong with the italics html tag]
 
Hmmm, the same author indicates that the SSME's have improved substantially in reliability in the last ten or so years. However, there's another issue here.

The role of the engine is being shifted from reusable to disposable. This is another good reason to avoid the SSME's. They're designed to be reusable. That drives up the cost. Given the complexity of the system, redesigning the engines for the reduced reliability needed for an expendable, may mean effectively a new engine.
 
Mr. Gloor:  You're right about the RS-68.  It came along after I quit paying close attention to the field, so you've got me there.

However, your cite for the RS-68 shows a relatively low T/W (51.2) and vacuum Isp (420 sec) compared to the SSME (73.1 and 453 sec).  The lower Isp is going to cut the mass fraction to orbit, and the greater weight is going to take a chunk out of the payload.

The RS-68's thrust is about 1.5x that of the SSME, not 2x.  A pair of RS-68's is a rough replacement for a trio of SSME's; you'd have less delta-V offset by slightly lower gravity losses.

As for the SSME's DDT&E, that is a sunk cost.  Scrapping it is not going to get any of that money back.

"SSME should be avoided ... because they are just not suitable engines by design for 1st stages."

The SSME cannot be started or re-started in flight.  The mission for which it is designed, and the only one for which it is suitable, is the sustainer of a parallel-staged booster.

" I just think we already have the launcher [Delta IV common core] - we just do not realise this."

The Delta-IV Heavy has a maximum payload to orbit of a mere 10,843 kg, less than a third of the typical Shuttle-derived booster.  This is not a vehicle for Mars missions.  Even the heaviest Atlas variant is only capable of 20 metric tons to LEO, far below the 35 tons of the minimal Shuttle-C.

Mr. Hallowell:  The Shuttle launch scrub data indicates that the SSME is now responsible for perhaps 20% of the launch scrubs, not the 1/3 it was in 1993 (one would expect engines of any type to require much more time to fix than e.g. the weather).  The RS-68 might be more reliable, and probably could be made more so, but an HLV clustering 2 or 3 RS-68's would present greater risk than the same using 3 SSME's (which have already flown over 100 missions with no vehicle or mission losses due to the engines).

Building a vehicle for manned lunar and Mars missions could be done with Delta V parts, but it would take paired RS-68's in combination with at least two SRB's to lift a Shuttle-C equivalent payload to LEO.  It looks like no such configuration has ever been tested.  The risk appears high compared to a slightly reconfigured ET feeding 3 SSME's with the same two SRB's, and while I have no objections in principle to working the cost down I would worry about both technical (proving a never-tested cluster configuration) and political (shifting lots of money from established interests, placing total support in jeopardy) risks.
 
E-P, you wrote:

"The Shuttle launch scrub data indicates that the SSME is now responsible for perhaps 20% of the launch scrubs, not the 1/3 it was in 1993 (one would expect engines of any type to require much more time to fix than e.g. the weather). The RS-68 might be more reliable, and probably could be made more so, but an HLV clustering 2 or 3 RS-68's would present greater risk than the same using 3 SSME's (which have already flown over 100 missions with no vehicle or mission losses due to the engines)."

I think it would be better to accept the higher risk here. the RS-68's can be with experience made more reliable just as the SSME's were. Hmmm, in this news story there is a short discussion comparing the SSME with the RS-68:

"Compared to the SSME, development time for the RS-68 was cut in half, the number of parts was reduced by 80 percent, the hand-touched labor reduced by 92 percent and non-recurring costs were cut by a factor of five.

"We've been able to get a lot of hand-work out of the RS-68 and replaced it with numerically-controlled machines. So instead of having a thrust chamber built up of a lot of tubes, we've machined this thing out of a solid piece of metal, which increases reliability," Collins said."

When you toss in the claimed decline in costs and simplification of the design and conbstruction, this rocket appears to be a signficant improvement over the SSME. And worth sacrificing some reliability over.

E-P, you wrote:

"The risk appears high compared to a slightly reconfigured ET feeding 3 SSME's with the same two SRB's, and while I have no objections in principle to working the cost down I would worry about both technical (proving a never-tested cluster configuration) and political (shifting lots of money from established interests, placing total support in jeopardy) risks."

The cluster hasn't been tested either way though the SSME are I allow less risky in the near future. Second, we have already determined (see the history of the ISS and the numerous experimental programs to study replacements for the shuttle) that the established interests are harmful to effective NASA operations. IMHO, going with the better designs is more likely to extend NASA's useful lifespan (such as it is) than continuing to feed the established interests.
 
That's an impressive list of Delta-variant vehicles, but I have to note several things:

1.  You're flat wrong about the Shuttle cargo variants using SSME's as first-stage motors.  They use SRB's; the SSME's are burning from the pad to orbit, but the SRB's provide a majority of the total vehicle thrust for the first 2 minutes.

2.  None of the Delta IV configurations which have Shuttle-C class capability have ever been tested.

3.  All of the Delta vehicles require an upper stage.  Shuttle and its variants are 1.5 stages to orbit; all engines are burning as the vehicle leaves the pad, so there is no way for an ignition failure to result in failure to make orbit.  This is not true of the Deltas.

4.  While the basic Shuttle-C would have a capacity of 70 tons and upgrades could bring it to nearly twice that, the Delta IV Heavy can only orbit 28 tons and the top of the upgrade path (which requires 4 solid boosters) is only 85 tons.  This is insufficient for manned missions to the Moon or Mars.

5.  The Delta IV Heavy upgrade has twice the solid-motor risks of Shuttle-C.

There's one thing you don't appear to appreciate about specific impulse.  LEO orbital velocity is roughly 7700 m/sec, and typical solid boosters drop off somewhere around 1 km/sec; this means the liquid booster engines need to impart 6700 m/sec of delta-V.  (The vehicle is effectively in vacuum at that time, so its sea-level performance is irrelevant from separation of the solids to engine cutoff.)  If you have two engine options, of 453 seconds and 420 seconds Isp, the amount of fuel which gives you 6700 m/sec of delta-V in the first will only yield 6212 m/sec from the second.  That's half a kilometer per second you have to make up with more fuel, which could otherwise have been payload.  Extra engine mass comes straight out of payload.

Half a kilometer per second is a huge number.  IIRC, 1 m/sec of delta-V near LEO changes the altitude at the other side of your orbit by roughly a mile.  Losing 488 m/sec at MECO would put your perigee at almost 300 miles below sea level.

It might be possible to adapt the RS-68 to achieve SSME-class impulse using an aerospike nozzle; such an advance would eliminate the need for an upper stage to achieve orbit, and improve both the vehicle reliability and payload.  But this is a long-term project, and getting e.g. Mars Direct launched soon can only be done with something like Shuttle-C.
 
The original Mars Direct program called for 45 tons landed on Mars.  The Hohmann transfer orbit data I've got calls for 2.411 km/sec minimum delta-V required to go from Earth's orbit to MTO; if you're starting from a circular parking orbit at 7700 m/sec the total delta-V is about 3.2 km/sec.  Applying this kick with an H2/O2 upper stage using RL-10's at 450 sec impulse requires a mass ratio of about 2.07; with the transfer stage tankage, plane-change requirements and whatnot, call it an even 100 tons to orbit.

You can put 100 tons into LEO with 3 SSME's, 2 SRB's and a standard ET.  You don't have to modify anything about the physical configuration of the engines or their fuel plumbing, though you could hard-couple them to the ET and save the weight and expense of the disconnects.  Doing another upgrade of the Delta IV heavy is a stretch upon a stretch; perhaps worth doing, but not the slam-dunk of the Shuttle-C.
 




<< Home
Talk largely about energy and work, but also politics and other random thoughts


Mail Engineer-Poet

(If you're mailing a question, is it already in the FAQ?)

Important links

The FAQ
Glossary
The Reference Library

Blogchild of

Armed and Dangerous

Blogparent of

R-Squared




The best prospect for our energy future:
Flibe Energy

ARCHIVES
January 1990 / February 2004 / March 2004 / June 2004 / July 2004 / August 2004 / September 2004 / October 2004 / November 2004 / December 2004 / January 2005 / February 2005 / March 2005 / April 2005 / May 2005 / June 2005 / July 2005 / August 2005 / September 2005 / October 2005 / November 2005 / December 2005 / January 2006 / February 2006 / March 2006 / April 2006 / May 2006 / June 2006 / July 2006 / August 2006 / September 2006 / October 2006 / November 2006 / December 2006 / January 2007 / February 2007 / March 2007 / April 2007 / December 2007 / January 2008 / May 2008 / June 2008 / August 2008 / October 2008 / November 2008 / December 2008 / February 2009 / March 2009 / April 2009 / May 2009 / June 2009 / July 2009 / August 2009 / September 2009 / October 2009 / November 2009 / December 2009 / January 2010 / April 2010 / May 2010 / June 2010 / July 2010 / August 2010 / September 2010 / October 2010 / November 2010 / December 2010 / January 2011 / February 2011 / March 2011 / April 2011 / May 2011 / July 2011 / August 2011 / September 2011 / October 2011 / April 2013 / November 2013 / December 2013 / January 2014 / February 2014 / March 2014 / April 2014 / July 2014 / August 2014 / September 2014 / October 2014 / November 2014 / February 2015 / April 2015 / October 2015 / March 2016 / April 2016 / May 2016 / June 2016 / July 2016 / November 2016 / December 2016 / February 2017 / May 2017 / June 2017 / September 2017 / October 2017 / November 2017 / March 2018 / May 2018 / June 2018 / October 2018 / December 2018 / January 2019 / March 2019 / June 2019 / October 2019 / November 2019 / March 2020 / June 2020 / December 2020 / March 2021 / April 2021 / May 2021 / July 2021 / January 2022 / February 2022 /


Powered by Blogger

RSS feed

Visits since 2006/05/11: