#11
|
|||
|
|||
Rand Simberg wrote:
On Mon, 09 May 2005 04:49:24 GMT, in a place far, far away, Scott Lowther made the phosphor on my monitor glow in such a way as to indicate that: Shuttle is currently about $400M per launch. That's not a meaningful number. Marginal cost is much less, and average cost is much more, at expected flight rates. Shuttle-C or similar would not be more. It would be if the launch rate is lower. Right. Take the number of people employed building, training, stacking, launching, recovering, and refurbishing a launch system, multiply that by $100K, then multiply that result by about 2-3 or so, and you get a rough estimate of the total costs of running the launch program for one year. At low launch rates - which is what NASA human launch systems are always going to operate at, this cost won't vary much with the number of launches. The best way to cut costs is to reduce the number of workers. The current shuttle budget is about $4.5 billion per year, which translates to $750 million per launch if there are six launches in a year, or $1.125 billion per if there are four launches, or $4.5 billion per launch if there is one flight, etc. But note that this is the combined launch and mission costs, which includes feed and care of astronauts, cleaning the shuttle toilet, processing the payloads, etc.. It's not that much when you consider that Titan IV Milstar missions (for an unmanned communications satellite - no astronauts to care for) cost more than $1 billion each. So to get the costs for a shuttle-derived system down to something reasonable - say $400 million per flight at four launches per year - NASA would have to establish a program that only employed around 6,500 (total, for everything from factory to control center to parking lot security, etc). This is only about one-third of the total currently involved in the space shuttle program. (Or it could employ the same number of workers at 1/3rd the pay.) - Ed Kyle |
#12
|
|||
|
|||
Ed Kyle wrote:
So to get the costs for a shuttle-derived system down to something reasonable - say $400 million per flight at four launches per year - NASA would have to establish a program that only employed around 6,500 (total, for everything from factory to control center to parking lot security, etc). This is only about one-third of the total currently involved in the space shuttle program. (Or it could employ the same number of workers at 1/3rd the pay.) Or NASA could do something remarkable and design CEV to fly atop existing EELV assets. If four launches were to occur each year, then the EELV production rate would be increased from 8 to 12 annually and the per-flight launch services cost would drop from about $120 million to $80 million (a bit of guessing). Then, if NASA could consolidate centers, shut down Complex 39, and run the vehicle and mission part of the program with only 2,000 workers or so (complementing the 3,800 or so who build and fly the EELVs), the entire CEV program could be run for maybe $1 billion per year ($250 million per flight). It won't happen. - Ed Kyle |
#13
|
|||
|
|||
Ed Kyle wrote: Or NASA could do something remarkable and design CEV to fly atop existing EELV assets. .... It won't happen. The astronaut corp apparently desperately hopes not. The hydrogen fire on Delta IV scared the bejeesus out of everyone who saw it; DIV cannot fulfill the CEV mission without solid-rocket strapons (up to 7, if memory serves), thus negating the already dubious safety superiority to solids; and the DIV has to fly an odd trajectory (due to structural concerns) that means that there are points in the ascent when abort is *not* survivable. Atlas V heavy is a bit of an unknown, but apprently has a number of the same performance issues. |
#14
|
|||
|
|||
Ed Kyle wrote: So to get the costs for a shuttle-derived system down to something reasonable - say $400 million per flight at four launches per year - NASA would have to establish a program that only employed around 6,500 (total, for everything from factory to control center to parking lot security, etc). This is only about one-third of the total currently involved in the space shuttle program. (Or it could employ the same number of workers at 1/3rd the pay.) Is there any breakdown regarding how many of those are related to orbiter inspection, TPS repair, and SSME (I assume that the cargo version wouldn't use SSMEs, but rather expendable RS-68s for the sake of economy and system simplicity) postflight checkout and maintenance? Also if we go over to a unmanned system, we don't have to meet the stringent requirements for SRB safety, so we can probably reuse more SRB segments. Pat |
#15
|
|||
|
|||
Ed Kyle wrote: Ed Kyle wrote: So to get the costs for a shuttle-derived system down to something reasonable - say $400 million per flight at four launches per year - NASA would have to establish a program that only employed around 6,500 (total, for everything from factory to control center to parking lot security, etc). This is only about one-third of the total currently involved in the space shuttle program. (Or it could employ the same number of workers at 1/3rd the pay.) Or NASA could do something remarkable and design CEV to fly atop existing EELV assets. If four launches were to occur each year, then the EELV production rate would be increased from 8 to 12 annually and the per-flight launch services cost would drop from about $120 million to $80 million (a bit of guessing). The problem here is that you need a minimum of four EELV launches to land a man on the Moon if the LockMart design is chosen: Launch one carries the CEV, launch two carries the TEI stage and living area/docking module (as shown in the LockMart "Lunar Train" drawing), launch three carries the lunar descent/ascent module, and launch four carries the TLI/Lunar orbital braking module. If these are using cryogenic propellants, you've got to get them all into LEO in fairly short order to prevent excessive propellant boil-off. If there is a launch failure of any of the components, or a failure to assemble them in LEO, then the whole mission is off. I can see assembling two components in LEO- but _four_ (and that's a minimum; it could end up being five or six)?! Compared to Apollo's single Saturn V launch, this sucks. With the unmanned shuttle cargo variant, you could send up everything except the CEV in one launch, and then launch it separately with the crew. Since you would be using two separate pads, this would save the money on building more pads for the EELV only option's need for four or more rapid turnaround launches to accomplish this mission. Pat |
#16
|
|||
|
|||
|
#17
|
|||
|
|||
|
#18
|
|||
|
|||
Rand Simberg wrote: Or subcool the propellants and insulate the tanks properly... Still, how long would one have to get it all stuck together? Days or weeks? The Soviet's did a stage (Block D) that used an insulation sunshade for it LOX/Kerosene propellant on Proton-Zond and N-1: http://www.myspacemuseum.com/l1s_2.jpg But do we have any experience with this sort of thing? The closest we came was the canceled Shuttle boosted Centaur stage. Pat |
#19
|
|||
|
|||
Pat Flannery wrote:
Is there any breakdown regarding how many of those are related to orbiter inspection, TPS repair, and SSME (I assume that the cargo version wouldn't use SSMEs, but rather expendable RS-68s for the sake of economy and system simplicity) postflight checkout and maintenance? Assuming that the shuttle goes away in 2010, there shouldn't be any objection to using the remaining inventory of SSMEs as expendables for a heavy lift vehicle. Leaving aside the six that have been lost in flight, there seem to be about 40 engines in the inventory (2005 through 2109), most of which have fewer than 10 flights. Even omitting the workhorse engines (i.e., 2012 with 22 flights), sufficient engines remain for several flights of a shuttle-derived heavy lifter. (I get my numbers from the available vehicle descriptions in press kits -- STS-108 omitted that info for some reason). Of course, the engines would be a mix of base, Block I, Block II, etc -- but if overall performance is an issue, either mix similar engines in the same cluster or design the flights around the nominal thrust levels for the least-capable engine (i.e., use 100% if the flight includes a base engine, 104% if the worst engine is a Block I). They're bought & paid for, and their only other use is as expensive paperweights in a few museums. It would actually be *more* expensive to replace them with RS-68s, since they'd have to be purchased (assuming that the SSMEs have been stored in reasonably controlled conditions). -- Reed Snellenberger GPG KeyID: 5A978843 rsnellenberger-at-houston.rr.com |
#20
|
|||
|
|||
Rand Simberg wrote: So we don't care if we lose a billion-dollar payload? Or the price of relaunching? This notion of reliability being of no relevance for unmanned systems gets tiresome. Not to the same degree... for a manned launch you want around 99+% reliability if at all possible; for unmanned you can settle for 95%-97% (like most operational expendable rockets have) and realize that the loss of a couple in 100 launches will be more than offset by the money you save in not having to design and build to quite the high standards required to get to 99+%. It's where those last few percentage points start coming into play that you run into lots of added dollars- and extra equipment weight to overbuild things to make critical things redundant. Which cuts into your payload weight, and therefore ups your launch price per pound for large numbers of launches. Pat |
Thread Tools | |
Display Modes | |
|
|