The wings are not pressurized, but will leak at a fairly constant rate. Let's call it R. Now, a wing with a bloody great hole in it will also leak at a constant, but different, rate. Let's call it R'. It does not matter what R' is, only that it is strictly greater than R.
So, you have a simple pressure guague and measure the difference between observed pressure drop and your expected value of R. If the total discrepency exceeds some critical threshold, then there is a problem with the wing that would create a serious problem.
Maybe you should have studied harder when you were with the NASA people you worked with.
The Shuttle wing sections are unpressurized (true) and are faced with solid metal sheets that form the wing structural skin. The thermal protection system goes over those metal sheets. You can completely remove all the tiles and not change the structural integrity of the underlying metal skin.
None of the prior damage in tiled areas punctured the skins.
The leading edge areas with the reinforced carbon-carbon segments are outside the main wing skin. They aren't sealed either, and generally retain very little pressure differential with the outside. It's possible that you'd have seen an abnormally fast equalization of pressure in the Columbia leading edge. But all that tells you is if you have a hole so bad that you're going to lose the shuttle, with no chance of repair.
Most damage will be less severe than that, and such a pressure test won't catch that more minor damage. To find that you have to look at the surface up close to check for dings and cracks.
If you have to visually inspect to check for the more minor but still potentially lethal damage, why even bother with the huge hole pressure check? It's not like the visual inspection can possibly miss an 8" hole in the leading edge with a camera two meters away.
Keep one thing in mind about adding more drives.. Mean Time Between Failure falls as a function of how many drives you are using.
Correct.
Unless you need high performance (Database spindle count, for example), using as few drives as possible saves on expensive RAID controller or RAID unit chassis slots, and improves MTBF of the set of disks.
Home users won't typically care, they want minimalist solutions, only talking about a few drives.
Enterprise sysadmin / data managers care; when you get up into tens or hundreds of terabytes, even if your storage uses SATA drives underneath, you're talking about dozens or hundreds of drives. Then you get into 10 vs 01, hot spare plans, whether SW RAID on top and mirroring across RAID units makes sense, etc etc.
Rule of thumb I have seen and heard repeated around the industry is to assume a 2% or higher failure rate for drives, per year. Even good, enterprise grade SCSI / FC drives. Probably should double or triple that for ATA / SATA components in large enterprise units (or small business or home units).
RAID 0+1 vs RAID 1+0
on
Basics of RAID
·
· Score: 2, Informative
This is a non-trivial question:
When you're setting up a RAID set using both striping and mirroring, do you want to set up two stripes and then mirror between the stripes (0+1), or do you want to set up mirrored pairs and then stripe those mirrored pairs (1+0)?
This is a quiz, and your data will grade you.
What you want, by far, is RAID 10 (1+0).
When you set up two stripes and then mirror across them, if you lose two disks, any disk in the first stripe and any disk in the second stripe, you lose all the data.
If you stripe across mirrored pairs, then the only way to lose data is to lose both drives in one of the mirrored pairs. You can lose any other disk than the second drive in a pair, or even many more disks, as long as they aren't both in the same mirrored pairs.
This doesn't make a difference with 4 drives. At 6 drives and up, use 10. Your data and users will thank you for it.
I'm not going to mod you down despite having a fistfull of modpoints at the moment, but I just went and read your blogs, and I have to question your judgement and impartiality.
What you're writing is sorta neat, but it's not necessarily edgy in a new and good manner.
What I see in Joss' stuff is good characters with depth, that develop over time, and humor that is mostly achived by putting realistic characters in absurd (but in context for the assumed macguffins) situations and watching the results.
You don't like it? Fine. But over 95% of the sci fi book writers out there can't do that well with their characters, situations, etc. Joss is good at what he does by any measure.
Baen is a huge exception to that rule; they accept electronic submissions, and have released free copies of large chunks of (and quite a few whole books) in various formats.
Eric Flint really started pushing it for them, his experiments (and sales increases) were what started the whole trend off.
In fact, even the most modern thermonuclear devices have an efficiency ratio of "only" 20% or so.
That's not true. Or rather, is such an oversimplification as to be grossly inaccurate.
It's possible to build boosted fission primaries with fission efficiency up to about 50%. Such have been built and weaponized. Modern US devices have less efficiency (around 15%, in rough terms) because they are designed to use as little fissile material as possible and to be one-point safe, and also to have limited overall fission yield. Those requirements lead to less efficient weapons than are possible and were used in the past.
Second, fusion, stages can be both highly efficient (50% or more of the possible fusion energy content) and have very high multiplication ratios of input to output energy (factor of 25 is possible, with factors of 8-15 in deployed US weapons), even before you double it again with a fissionable tamper third stage.
That then ignites a small pellet of boron / potassium nitrate.
Which ignites a small rocket motor which is about 4 inches long.
Which ignites a medium sized rocket motor about three feet long.
Which fires a jet of flame for about a tenth of a second, all the way down the whole inside length of the solid boosters, which ignites the whole inside at the same time.
You do realize that *BSD systems don't use init.d, right? They use the much more sensible (albeit less "flexible") rc.d system. The concept is simple:
This is the master startup script. After it runs, it loads subscripts from this directory. Life is good.
My own feels have always been that init.d's flexibility comes at far too high of a price in maintainence and ease of use. I'm sure there are others who would disagree, though. (No, I don't want to hear about it.)
Well, you're going to hear about it.
How can it be hard to maintain or understand the Solaris init.d/rc.d setup?
I mean, really. You are going to runlevel 3. The init script goes into/etc/rc3.d, which contains a bunch of K## and S## scripts, many of which are links to a master script in/etc/init.d just so that you don't have to modify five copies of a Sendmail startup script if you play with it.
It runs all the scripts with K## with "scriptname stop" and all the scripts with S## with "scriptname start".
You're up.
Dependencies are implicit in the numbering order for the various scripts, not explicit anywhere, so that's somewhat of a drag. But other than that, what's not to understand?
Sure, you could put it all in one long shell script for each run level. That works too. Do you code your hundred thousand line C programs in one.c source file too?
This is all about generating massive shockwaves to examine the properties of matter in extreme conditions (without having to heat it up to enormous temperatures).
...using strong magnetic fields to accelerate aluminum plates to incredibly high speeds in a contraption that has a strong resemblance to a rail gun.
The Z machine isn't a railgun at all.
It doesn't accellerate the flyer plate linearly. A Z-pinch machine, which the Z-machine at Sandia is an example of, implodes a thin hollow cylinder of material, a fraction of a mm thick and about 1-2 cm in diameter and roughly 1-2 cm long. The Z-pinch effect causes the cylinder walls to collapse inwards at high speed, striking a target along the axis of the cylinder.
High electrical current flows through the cylinder, which causes a magnetic field whose lines of force are circular around the wire. Basic electrical physics. Also basic physics, you get a force ExB (electrical field cross product with magnetic field). The sign of the force, from the direct electrical field and its induced magnetic field, is inwards towards the center of the cylinder.
So at high enough currents, the cylinder implodes.
There's no external magnetic field necessary. If you add one, then the implosion process is more even and stable, but that isn't necessary at all for the Z-pinch effect to work.
You can even do fairly safe home experiments in Z-pinching. Take a bunch of thin wires and a couple of nonconductive disks. Put the disks on a pole, then string the wires from disk to disk so that they form a cylindrical array. Solder all the ends at the top to one electrial lead and all the ends at the bottom together. Connect up to a power source. Watch the wires move inwards.
Z-pinch is just taking that effect, and putting the equivalent of the whole world's electrical power output through it instead of a small regulated power supply. A bit more force, eh?
Calling the Z-machine a railgun is like calling a F-16 fighter a really cool drag racer. Just because they're both fast and burn hydrocarbons doesn't make them at all the same thing.
Yes you are. In your haste to glorify Everything Google, you're leaving out the fact that Henry Spencer did not own the tapes. Spencer stole the tapes. [...]
I am not attempting to glorify everything Google in any way. I respect the good things they have done. I certainly don't believe they have never done anything wrong at all.
I asked Henry about your ridiculous accusation, and he responded, and I quote with his explicit permission:
The Usenet archive tapes were *loaned* to Dave Wiseman, of U of Western Ontario, because he was interested in copying the contents onto more modern media. We don't seem to have kept an organized record of exactly when that happened -- it wasn't considered an important event -- but indirect evidence places it no later than March 1992, and almost certainly in late 1991.
At the time, I was still full-time U of T staff and head of Zoology Computer Systems, so this was my decision to make. My superiors were barely aware that the archive tapes existed, and had no interest in being consulted about minor operational decisions. (The intrinsic value of the tapes was already approaching zero, due to their age and the transition to 8mm tape for backups etc.)
I went part-time and stepped down as the head of the facility at the end of June 1992, but remained U of T staff (although with lines of authority less clear) until summer 1993.
Last I checked, Dave Wiseman still has the tapes. They remain officially the property of U of T Zoology, but I've checked with my successors several times over the years, and they have consistently denied any interest in having the tapes back -- Zoology no longer has a drive that can read them or a rack to store them properly.
Dejas only went back to 1995 though didn't they? Google's innovation was to extend the archive back to 1982. Granted Henry Spencer's tapes were used, but no one else has put them online complete back to 1982 as far as I know.
Basically, Google helped right at the end of a roughly decade-long process to get the tapes available online...
In summary, Google only really started encouraging the tape restore project about six months before groups.google.com kicked off. The idea of restoring Henry's tapes had been widely thought of in the 1990s, and Wiseman had picked them up to start the project, but it took some years to accomplish, along with help from various people and some equipment from
Brewster Kahle.
And I'm leaving out a bunch of stuff. I won't try and credit everyone involved in the process here, but it was lots of people. Good on all of them.
Google groups basically archiving all of usenet back to 1982 or something.
Google groups is now hosting an evolving archive that started as DejaNews, plus Henry Spencer's tape collection from the 1980s, plus other independent archives and some gap-filling.
While those of us in the Usenet community widely applaud Google for their adopting and continued hosting of the archives, it wasn't their idea to start with. Permanent (nonsearchable) archives started with Henry's tapes, over 25 years ago. DejaNews made it searchable. Google figured out how to make enough money doing it to justify having done so and the continuing operations.
Google's doing a lot of good in the world, and making a lot of money off doing it. But there's no reason to give them credit for everything good that's out there. Doing that tends to slight historical contributions and effort of others.
But good on Google for all sorts of other things they have done!
Second, if you look at the diagram you'll see that it is initiated a gun-type trigger, something that is impossible for Pu. This makes the diagram look like the work of someone that doesn't know what they are doing.
You have to have sufficient plutonium to measure its various isotopes' spontaneous fission and decay rates , thus its neutron emission rate, before you can know that it's not suitable for gun type bombs.
The Nazi bomb scientists didn't have the opportunity to make and separate much if any, so they couldn't have known what the SF rates were.
The US bomb programs looked at Plutonium gun type bombs until we actually made Plutonium and figured out that we couldn't, then shifted to Uranium gun type and Plutonium implosion weapon designs.
That is just plain wrong. The shape needed for a plutonium bomb to go off needs some very high tolerances, not achievable by average guys in a garage shop.
Spherical implosion is moderately tricky.
There are other, simpler, but still very functionally workable geometries and core assembly methods.
Cylindrical implosion, linear implosion, etc.
Rei, these shapes are less stable at lower speeds (high supersonic, transonic particularly) than at hypersonic speeds.
That's why you use a supersonic drogue chute, followed by the main chute(s), ala Apollo.
You can also use a ballute to increase your effective reentry diameter, or large extendable drag plates, or all sorts of other things.
Not having to deploy anything to remain stable is a very good design characteristic. Things that are mission critical (such as, vehicle will spin out of control if you don't release it between Mach 4 and Mach 3.5...) need to be very reliable, and get very expensive and heavy as a result. Moving parts that are mission critical and require (for example) pyro deployment are always a weakspot.
If you can do a design which doesn't require any such components, because it's just inherently stable, you win. Your engineering is easier, you don't have to qualify the component to 99.999% including corner operational envelope cases, etc. You don't suffer from potential lingering maintenance or engineering design error issues.
If it's not there, it can't fail.
Good hypersonic L/D ratio with moderate angle of attack, and a geometry which doesn't make angle of attack heating changes hard on the capsule's "aft body"
Apollo had a hypersonic L/D of 0.75, according to Astronautix; are you claiming better? Cite for that and heating (comparative), please?
As for the geometry and heating issues... if you happen to own
Human Spaceflight Mission Analysis and Design
then you just need to look at the table of comparative L/D at Alpha for Apollo style and sphere-cone style capsules. I don't have it in front of me to give you the page number, but you can see that you can get good L/D at very mild alpha (25-35 degrees gives you 0.25-0.3 L/D).
That's a very simplified table, for one particular sphere-cone, of course. But it shows you the general case.
Geometry is conducive to CG layouts which let you attain enough angle of attack to be useful
As opposed to losing your orbital energy higher up from having a larger blunt base in addition to a good L/D?
The overall drag will be similar for a high alpha sphere-cone and a basic blunt capsule. The blunt capsule has higher drag, but incrementally more rather than a large factor.
By the time you dial in enough alpha with either one to get a good enough L/D to get low enough peak entry Gs for easy human tolerance, the differences are very minor. You always want to see what the conditions are for a 3-4 G reentry, with the required lift (and thus L/D, etc). That's what matters.
If you use passive energy absorbtion for landing on land, such as shock absorbers, crush structure, or space for seats to move downwards on shock absorbing mounts, the narrow part of the capsule being down at impact means that the least fraction of total usable capsule volume is taken up by the space needed for the impact attenuator.
At the same time, it gives you the least amount of space for the impact attenuator.
Space doesn't really matter. Length of attenuation material does... the G loading depends directly on how much time and space it takes to decellerate you. But the volume is just a question of what impact attenuation material you use. If you have a low energy dissipation density than you use lower density or lighter attenuators or crush structure... light foam, thin sheet metal constructs which are lightly connected, etc. Higher energy density per unit volume just requires denser energy absorbtion, such as heavier foams, thicker structures, denser aluminum foam
Ok, the terminology here is somewhat ambiguous with capsules that tend to launch in one direction and reenter in the other. I meant, take the end away from the heatshield on an Apollo like design, which normally comes to a point, and cut it off flat halfway up the cone section.
Which is what Stardust did, some of my prior capsule designs have done, etc. For different reasons than what you think, though.
You make a big deal out of reducing beta, but in most practical designs, the capsule outer diameter is limited by some other factor such as fitting under a launch vehicle shroud, ground transport concerns, maximum feasible hammerhead on the launch vehicle, etc. If diameter were never an issue, capsules would be a lot larger; to first order approximation, TPS mass doesn't change if you decrease beta/increase surface area, but the reentry happens higher and more gently. Systems mass doesn't increase. The structural mass doesn't necessarily have to increase, though for pressurized volumes with people inside it will to some degree (less than you think, though; much of that mass is dealing with point loads such as hatches and the like, which don't increase at all if the overall volume goes up).
Fluffy capsules are a great thing. If you can actually build, transport, and fly them. Which, is the problem...
For example, the Stardust capsule. You get more space for a given amount of mass (as the shape is a closer approximation to a sphere), and you have a lower beta from the (proportionally) larger blunt area so that you need a slightly less massive TPS. It's also more stable at hypersonic speeds (due to shifting the center of gravity toward the nose, and makes for easy parachute deployment.
Rei, these shapes are less stable at lower speeds (high supersonic, transonic particularly) than at hypersonic speeds. If it's hypersonically marginally stable, it's going to be uncontrollably unstable as it slows down... generally a bad thing.
And you still haven't answered the question I posed:
Why choose this design, if their concern wasn't a higher beta? I mean, what is the point then - to shift the center of gravity *back* to make it less stable during reentry? I'm asking a serious question: *why*?
Well...
More available volume for same outer diameter (similar to Soyuz shape, but more stable)
More protection for parachutes and docking hatch area during reentry
Can be designed to be stable "nose first" only, throughout whole speed regime from subsonic through transonic through supersonic through hypersonic through newtonian flow
Good hypersonic L/D ratio with moderate angle of attack, and a geometry which doesn't make angle of attack heating changes hard on the capsule's "aft body"
Geometry is conducive to CG layouts which let you attain enough angle of attack to be useful
If you use passive energy absorbtion for landing on land, such as shock absorbers, crush structure, or space for seats to move downwards on shock absorbing mounts, the narrow part of the capsule being down at impact means that the least fraction of total usable capsule volume is taken up by the space needed for the impact attenuator.
Like I said... design a few, or look at the designs, in detail. It clarifies tradeoff issues magnificently.
Heatshields are among the cheapest components of a spacecraft on a per pound basis.
Just ignoring that extra heatshield mass involves a lot of other mass and complexity to get it up there, that's simply not true - especially with high beta heatshields. Have you ever looked at the thermal properties of materials at reentry temperatures? It's amazing how quickly most materials become easily pliable.
Yes, we're talking about a small craft here (which helps), but even still, you're going to need a good heat shield. Heat shields are costly (what on Earth made you think that they're cheap? Have you ever checked out RCC prices, for example? Even ablative shields cost a fortune) and are common failure points.
As I pointed out earlier, your assumption about a high beta reentry is incorrect, or at least unfounded.
RCC is not on my menu; I have priced ablative TPS before, yes, and have talked to the ablative heatshield manufacturers within the last few months about a current project.
The engineering on ablative heatshields is nontrivial, but the actual materials are quite affordable.
Existing capsules were designed to be expendable, not reusable. You don't generally reuse things you designed to throw away.
They've been expendible because of what reentry does to a capsule. The capsule itself corrodes, and the heat (which gets channeled to every part of the skin, usually by copper, to make the nose heat bearable) damages essentially everything that isn't occupied by people. There's risk of warping, weld damage, damaged electronics, and numerous other problems. Unlike a craft like the shuttle where you radiate your heat away before it becomes a problem, a capsule doesn't have that luxury, especially a high beta capsule.
I'm sorry, Rei, but that's a bunch of baloney.
The capsule doesn't corrode; the ablative heatshield material is heated, expands, and ablates away as it is supposed to do.
The metal structure gets heated somewhat, but there's no necessary reason for it to get heated to the point that it suffers any structural damage. The inner capsule with people in it is protected just fine, and the whole rest of the capsule structure other than the ablative skin can be protected equally well if you care about it. And if you're reusing it then you care.
Despite it being an expendable capsule, a Gemini was flown a second time in an unmanned test flight.
Unmanned for a reason. Why do you think it was that we never made a reusable capsule series, given how many capsules we flew? Russia started one (Zarya), but not until the mid 1980s, even with all of their capsule experience, and it got cancelled. Now we have some newcomers who claim that they're going to do what even superpowers with decades of experience couldn't, on their very first try? Please.
Zarya had two technical defects: one, the landing rocket motors were so loud they would damage the crew's hearing, and two, there was no backup landing system if those rockets failed to fire.
It was cancelled because they didn't have missions for it and didn't have money for it.
Prior capsules were made expendable because it is, arugably, just as cheap to do it expendable and it takes less up front research and development cost.
The Gemini flew unmanned the second time not because it was too damaged to fly with people in it, but because they were testing out another, truly risky at the time idea... having a hatch in the middle of the ablative heatshield. It passed the test just fine, fortunately.
Reading your reply in total again, I can't think but that you really don't understand what you know and don't know about spacecraft design. Most of your detailed objections about capsules are incorrect. You're making assumptions about what was hard and what wasn't hard which don't stand up to close inspection.
To be a good skeptic, you have to have good information to start with. You need to ask more questions first.
You just admitted that you know that they are saying they'll launch cargo on another vehicle. And yet you go right on in there and do a cost per pound of cargo comparison.
Per kilogram. And that's per kilogram of payload - in this case, the payload is humans. The shuttle carries both human payloads and cargo payloads; would you rather I choose a different craft for comparison?
I would rather you not use cargo costs for judging the cost of orbiting people, as they're not the same figure of merit. The Shuttle is the only vehicle which carries both in quantity, and nobody is proposing followons which do that.
Use reasonable methodology.
The capsule shape is called the sphere-cone geometry... Use its proper name.
A) What percentage of slashdotters would recognize the term "sphere-cone geometry"?
Without an introduction? Probably vanishingly few. But you can introduce it and explain it, as I did, and then educate people some about the technical field as you go.
B) What percentage of slashdotters would recognize the term "bell shaped" (which is a reasonably accurate description - in fact, the Mercury capsules were named "Liberty Bell X", where X was a number).
Be realistic here:
1) I'm posting on Slashdot, not alt.space.
2) "sphere-cone" isn't even a widely used term - it isn't even mentioned on places like astronautix.com.
Astronautix is a great encyclopedia for the public, but that doesn't make it authoritative in the industry. Everyone at NASA I talk to about capsules, at Bigelow Aerospace, at the FAA AST, the DOD, etc.... everyone knows and uses sphere-cone routinely. Even in presentations at conferences where most of the audience is nontechnical, it tends to be used as the designator, along with a description of what the word means.
3) You completely misread my post: "Bell shaped" was in reference to later, Mercury/Apollo-style capsules. The key difference in the Corona-style capsules is that the gently sloping sides continue outwards away from the nose, instead of inwards - that is, the nose end is smaller than the aft end.
Then you really have me confused, as Apollo/Gemini/Mercury blunt leading edge capsules look (to me) even less like a bell than sphere-cone geometry capsules.
A little history: The first reentry designs proposed were long needlelike designs, designed to be low drag. These hardly worked at all; they burned up. For the next generation, they widened the capsule, still having a taper, but blunting the nose, thus driving the heat away from the craft (most ICMBs use shapes like this).
US and Russian ICBMs haven't used that shape in 40 years, actually. Using much better thermal protection systems (reinforced carbon-carbon mostly) the current types of ICBM Reentry Vehicle (RV) are very pointy cones, with a roundoff at the nose of less than 10% of the base diameter typically.
Later capsules optimized this further by having the capsule itself withdraw inwards further away from the shockwave. I've seen some more recent designs that call for further optimization with a "loaf" shape, in which the top of the cone isn't rounded but truncated; it provides better mass and space ratios.
I've designed several capsules which were generically Apollo shaped but truncated the cone lower down; it has its uses, but it's not really better from a mass or space ratio point of view. You can stick a larger hatch in the nose that way, you can stick cargo outside the pressurized capsule that way; I put pressurized Mars rovers riding outside some big (40 plus foot diameter) capsules of that geometry for one mission design.
Rei, look, you have been a skeptic on slashdot about alt.space for as long as it's been an active thread that I recall. Healthy skepticism is fine. But a lot of what you keep doing is just ridiculous.
The problem is, this bid isn't cheap. 20 million dollars per launch, 4-6 people, no cargo (their proposal is to have all cargo launched on unmanned systems) seems to imply a cargo capacity of around 1200 kg at 16,700$/kg. These are Space Shuttle prices.
You just admitted that you know that they are saying they'll launch cargo on another vehicle. And yet you go right on in there and do a cost per pound of cargo comparison.
That's a complete nonsequiteur.
They're using an outmodded reentry design (the bell-shaped reentry design wasn't chosen by the US, Russia, and China for no particular reason - they did extensive testing, and it proved to be the most efficient, most reliable shape)
The capsule shape is called the
sphere-cone geometry, for the rather obvious reason that it has a nose that's a section of a sphere, and then a conical mid and aft body.
It's not bell shaped. Use its proper name.
They are far from the first use of the sphere cone design. As a matter of fact, more sphere-cone reentry vehicles (400 plus) have been used in history than any other shape. More than all other shapes put together, as a matter of fact. The Discoverer film capsules provide plenty of flight history on the geometry.
Nor is t/Space the first company to propose using them for modern manned capsules.
Nor is modern use the first proposed use: the same geometry was proposed for space lifeboat capsules in the 1960s and 1970s. It has always been felt to be a very easy to design, robust, and simple concept.
Sphere-cone requires slightly more thermal protection system mass than a conventional blunt cone shape like Apollo or Gemini, or even the Soyuz or Shenzhou. On the other hand, it has
superior characteristics during the touchdown and landing phase of the spacecrafts flight, and is simpler and lower risk in other areas as well. You are trading off ablative heatshield mass for other technical risks being lessened.
Heatshields are among the cheapest components of a spacecraft on a per pound basis. If your objective is to minimize risk and cost, then more heatshield mass is a good thing if it simplifies other problems.
...and they plan to make reusable capsules out of it when capsules have seldom proven realistic to refurbish for a second flight in the past. Furthermore, they plan to do this on their very first space attempt.
Existing capsules were designed to be expendable, not reusable. You don't generally reuse things you designed to throw away.
Despite it being an expendable capsule, a Gemini was flown a second time in an unmanned test flight. So it has been done before.
Slashdot Mirror
The Shuttle wing sections are unpressurized (true) and are faced with solid metal sheets that form the wing structural skin. The thermal protection system goes over those metal sheets. You can completely remove all the tiles and not change the structural integrity of the underlying metal skin.
None of the prior damage in tiled areas punctured the skins.
The leading edge areas with the reinforced carbon-carbon segments are outside the main wing skin. They aren't sealed either, and generally retain very little pressure differential with the outside. It's possible that you'd have seen an abnormally fast equalization of pressure in the Columbia leading edge. But all that tells you is if you have a hole so bad that you're going to lose the shuttle, with no chance of repair.
Most damage will be less severe than that, and such a pressure test won't catch that more minor damage. To find that you have to look at the surface up close to check for dings and cracks.
If you have to visually inspect to check for the more minor but still potentially lethal damage, why even bother with the huge hole pressure check? It's not like the visual inspection can possibly miss an 8" hole in the leading edge with a camera two meters away.
Unless you need high performance (Database spindle count, for example), using as few drives as possible saves on expensive RAID controller or RAID unit chassis slots, and improves MTBF of the set of disks.
Home users won't typically care, they want minimalist solutions, only talking about a few drives.
Enterprise sysadmin / data managers care; when you get up into tens or hundreds of terabytes, even if your storage uses SATA drives underneath, you're talking about dozens or hundreds of drives. Then you get into 10 vs 01, hot spare plans, whether SW RAID on top and mirroring across RAID units makes sense, etc etc.
Rule of thumb I have seen and heard repeated around the industry is to assume a 2% or higher failure rate for drives, per year. Even good, enterprise grade SCSI / FC drives. Probably should double or triple that for ATA / SATA components in large enterprise units (or small business or home units).
When you're setting up a RAID set using both striping and mirroring, do you want to set up two stripes and then mirror between the stripes (0+1), or do you want to set up mirrored pairs and then stripe those mirrored pairs (1+0)?
This is a quiz, and your data will grade you.
What you want, by far, is RAID 10 (1+0).
When you set up two stripes and then mirror across them, if you lose two disks, any disk in the first stripe and any disk in the second stripe, you lose all the data.
If you stripe across mirrored pairs, then the only way to lose data is to lose both drives in one of the mirrored pairs. You can lose any other disk than the second drive in a pair, or even many more disks, as long as they aren't both in the same mirrored pairs.
This doesn't make a difference with 4 drives. At 6 drives and up, use 10. Your data and users will thank you for it.
What you're writing is sorta neat, but it's not necessarily edgy in a new and good manner.
What I see in Joss' stuff is good characters with depth, that develop over time, and humor that is mostly achived by putting realistic characters in absurd (but in context for the assumed macguffins) situations and watching the results.
You don't like it? Fine. But over 95% of the sci fi book writers out there can't do that well with their characters, situations, etc. Joss is good at what he does by any measure.
Eric Flint really started pushing it for them, his experiments (and sales increases) were what started the whole trend off.
Once you build a Farnsworth Fusor, you have a tabletop nuclear fusion device. Which part of "Fusor" was unclear?
If all you want is neutrons and to violate your local Nuclear-Free Zone, just stop at that point...
It's possible to build boosted fission primaries with fission efficiency up to about 50%. Such have been built and weaponized. Modern US devices have less efficiency (around 15%, in rough terms) because they are designed to use as little fissile material as possible and to be one-point safe, and also to have limited overall fission yield. Those requirements lead to less efficient weapons than are possible and were used in the past.
Second, fusion, stages can be both highly efficient (50% or more of the possible fusion energy content) and have very high multiplication ratios of input to output energy (factor of 25 is possible, with factors of 8-15 in deployed US weapons), even before you double it again with a fissionable tamper third stage.
Look at references like the Nuclear Weapons FAQ at http://nuclearweaponarchive.org/
The solid boosters ignition starts with a small Nasa Standard Initiator (NSI) http://www.hstc.com/pdf/nsi.pdf.
That then ignites a small pellet of boron / potassium nitrate.
Which ignites a small rocket motor which is about 4 inches long.
Which ignites a medium sized rocket motor about three feet long.
Which fires a jet of flame for about a tenth of a second, all the way down the whole inside length of the solid boosters, which ignites the whole inside at the same time.
How can it be hard to maintain or understand the Solaris init.d/rc.d setup?
I mean, really. You are going to runlevel 3. The init script goes into /etc/rc3.d, which contains a bunch of K## and S## scripts, many of which are links to a master script in /etc/init.d just so that you don't have to modify five copies of a Sendmail startup script if you play with it.
It runs all the scripts with K## with "scriptname stop" and all the scripts with S## with "scriptname start".
You're up.
Dependencies are implicit in the numbering order for the various scripts, not explicit anywhere, so that's somewhat of a drag. But other than that, what's not to understand?
Sure, you could put it all in one long shell script for each run level. That works too. Do you code your hundred thousand line C programs in one .c source file too?
It doesn't accellerate the flyer plate linearly. A Z-pinch machine, which the Z-machine at Sandia is an example of, implodes a thin hollow cylinder of material, a fraction of a mm thick and about 1-2 cm in diameter and roughly 1-2 cm long. The Z-pinch effect causes the cylinder walls to collapse inwards at high speed, striking a target along the axis of the cylinder.
See for example http://plasma-gate.weizmann.ac.il/ZP/
High electrical current flows through the cylinder, which causes a magnetic field whose lines of force are circular around the wire. Basic electrical physics. Also basic physics, you get a force ExB (electrical field cross product with magnetic field). The sign of the force, from the direct electrical field and its induced magnetic field, is inwards towards the center of the cylinder.
So at high enough currents, the cylinder implodes.
There's no external magnetic field necessary. If you add one, then the implosion process is more even and stable, but that isn't necessary at all for the Z-pinch effect to work.
You can even do fairly safe home experiments in Z-pinching. Take a bunch of thin wires and a couple of nonconductive disks. Put the disks on a pole, then string the wires from disk to disk so that they form a cylindrical array. Solder all the ends at the top to one electrial lead and all the ends at the bottom together. Connect up to a power source. Watch the wires move inwards.
Z-pinch is just taking that effect, and putting the equivalent of the whole world's electrical power output through it instead of a small regulated power supply. A bit more force, eh?
Calling the Z-machine a railgun is like calling a F-16 fighter a really cool drag racer. Just because they're both fast and burn hydrocarbons doesn't make them at all the same thing.
I asked Henry about your ridiculous accusation, and he responded, and I quote with his explicit permission:
See for example David Wiseman's history of the recovery or the Salon.com overview article.
In summary, Google only really started encouraging the tape restore project about six months before groups.google.com kicked off. The idea of restoring Henry's tapes had been widely thought of in the 1990s, and Wiseman had picked them up to start the project, but it took some years to accomplish, along with help from various people and some equipment from Brewster Kahle.
And I'm leaving out a bunch of stuff. I won't try and credit everyone involved in the process here, but it was lots of people. Good on all of them.
While those of us in the Usenet community widely applaud Google for their adopting and continued hosting of the archives, it wasn't their idea to start with. Permanent (nonsearchable) archives started with Henry's tapes, over 25 years ago. DejaNews made it searchable. Google figured out how to make enough money doing it to justify having done so and the continuing operations.
Google's doing a lot of good in the world, and making a lot of money off doing it. But there's no reason to give them credit for everything good that's out there. Doing that tends to slight historical contributions and effort of others.
But good on Google for all sorts of other things they have done!
Blue Team
The Nazi bomb scientists didn't have the opportunity to make and separate much if any, so they couldn't have known what the SF rates were.
The US bomb programs looked at Plutonium gun type bombs until we actually made Plutonium and figured out that we couldn't, then shifted to Uranium gun type and Plutonium implosion weapon designs.
There are other, simpler, but still very functionally workable geometries and core assembly methods. Cylindrical implosion, linear implosion, etc.
You can also use a ballute to increase your effective reentry diameter, or large extendable drag plates, or all sorts of other things.
Not having to deploy anything to remain stable is a very good design characteristic. Things that are mission critical (such as, vehicle will spin out of control if you don't release it between Mach 4 and Mach 3.5...) need to be very reliable, and get very expensive and heavy as a result. Moving parts that are mission critical and require (for example) pyro deployment are always a weakspot.
If you can do a design which doesn't require any such components, because it's just inherently stable, you win. Your engineering is easier, you don't have to qualify the component to 99.999% including corner operational envelope cases, etc. You don't suffer from potential lingering maintenance or engineering design error issues.
If it's not there, it can't fail.
Actually, it had a hypersonic L/D of 0.25 to 0.3, as the Astronautix entry for the Apollo CSM Command Module section indicates.
As for the geometry and heating issues... if you happen to own Human Spaceflight Mission Analysis and Design then you just need to look at the table of comparative L/D at Alpha for Apollo style and sphere-cone style capsules. I don't have it in front of me to give you the page number, but you can see that you can get good L/D at very mild alpha (25-35 degrees gives you 0.25-0.3 L/D).
That's a very simplified table, for one particular sphere-cone, of course. But it shows you the general case.
The overall drag will be similar for a high alpha sphere-cone and a basic blunt capsule. The blunt capsule has higher drag, but incrementally more rather than a large factor.
By the time you dial in enough alpha with either one to get a good enough L/D to get low enough peak entry Gs for easy human tolerance, the differences are very minor. You always want to see what the conditions are for a 3-4 G reentry, with the required lift (and thus L/D, etc). That's what matters.
Space doesn't really matter. Length of attenuation material does... the G loading depends directly on how much time and space it takes to decellerate you. But the volume is just a question of what impact attenuation material you use. If you have a low energy dissipation density than you use lower density or lighter attenuators or crush structure... light foam, thin sheet metal constructs which are lightly connected, etc. Higher energy density per unit volume just requires denser energy absorbtion, such as heavier foams, thicker structures, denser aluminum foam
Ok, the terminology here is somewhat ambiguous with capsules that tend to launch in one direction and reenter in the other. I meant, take the end away from the heatshield on an Apollo like design, which normally comes to a point, and cut it off flat halfway up the cone section.
Which is what Stardust did, some of my prior capsule designs have done, etc. For different reasons than what you think, though.
You make a big deal out of reducing beta, but in most practical designs, the capsule outer diameter is limited by some other factor such as fitting under a launch vehicle shroud, ground transport concerns, maximum feasible hammerhead on the launch vehicle, etc. If diameter were never an issue, capsules would be a lot larger; to first order approximation, TPS mass doesn't change if you decrease beta/increase surface area, but the reentry happens higher and more gently. Systems mass doesn't increase. The structural mass doesn't necessarily have to increase, though for pressurized volumes with people inside it will to some degree (less than you think, though; much of that mass is dealing with point loads such as hatches and the like, which don't increase at all if the overall volume goes up).
Fluffy capsules are a great thing. If you can actually build, transport, and fly them. Which, is the problem...
Rei, these shapes are less stable at lower speeds (high supersonic, transonic particularly) than at hypersonic speeds. If it's hypersonically marginally stable, it's going to be uncontrollably unstable as it slows down... generally a bad thing. Well...- More available volume for same outer diameter (similar to Soyuz shape, but more stable)
- More protection for parachutes and docking hatch area during reentry
- Can be designed to be stable "nose first" only, throughout whole speed regime from subsonic through transonic through supersonic through hypersonic through newtonian flow
- Good hypersonic L/D ratio with moderate angle of attack, and a geometry which doesn't make angle of attack heating changes hard on the capsule's "aft body"
- Geometry is conducive to CG layouts which let you attain enough angle of attack to be useful
- If you use passive energy absorbtion for landing on land, such as shock absorbers, crush structure, or space for seats to move downwards on shock absorbing mounts, the narrow part of the capsule being down at impact means that the least fraction of total usable capsule volume is taken up by the space needed for the impact attenuator.
Like I said... design a few, or look at the designs, in detail. It clarifies tradeoff issues magnificently.RCC is not on my menu; I have priced ablative TPS before, yes, and have talked to the ablative heatshield manufacturers within the last few months about a current project.
The engineering on ablative heatshields is nontrivial, but the actual materials are quite affordable.
I'm sorry, Rei, but that's a bunch of baloney.The capsule doesn't corrode; the ablative heatshield material is heated, expands, and ablates away as it is supposed to do.
The metal structure gets heated somewhat, but there's no necessary reason for it to get heated to the point that it suffers any structural damage. The inner capsule with people in it is protected just fine, and the whole rest of the capsule structure other than the ablative skin can be protected equally well if you care about it. And if you're reusing it then you care.
Zarya had two technical defects: one, the landing rocket motors were so loud they would damage the crew's hearing, and two, there was no backup landing system if those rockets failed to fire.It was cancelled because they didn't have missions for it and didn't have money for it.
Prior capsules were made expendable because it is, arugably, just as cheap to do it expendable and it takes less up front research and development cost.
The Gemini flew unmanned the second time not because it was too damaged to fly with people in it, but because they were testing out another, truly risky at the time idea... having a hatch in the middle of the ablative heatshield. It passed the test just fine, fortunately.
Reading your reply in total again, I can't think but that you really don't understand what you know and don't know about spacecraft design. Most of your detailed objections about capsules are incorrect. You're making assumptions about what was hard and what wasn't hard which don't stand up to close inspection.
To be a good skeptic, you have to have good information to start with. You need to ask more questions first.
I would rather you not use cargo costs for judging the cost of orbiting people, as they're not the same figure of merit. The Shuttle is the only vehicle which carries both in quantity, and nobody is proposing followons which do that.
Use reasonable methodology.
Without an introduction? Probably vanishingly few. But you can introduce it and explain it, as I did, and then educate people some about the technical field as you go.
Astronautix is a great encyclopedia for the public, but that doesn't make it authoritative in the industry. Everyone at NASA I talk to about capsules, at Bigelow Aerospace, at the FAA AST, the DOD, etc. ... everyone knows and uses sphere-cone routinely. Even in presentations at conferences where most of the audience is nontechnical, it tends to be used as the designator, along with a description of what the word means.
Then you really have me confused, as Apollo/Gemini/Mercury blunt leading edge capsules look (to me) even less like a bell than sphere-cone geometry capsules.
US and Russian ICBMs haven't used that shape in 40 years, actually. Using much better thermal protection systems (reinforced carbon-carbon mostly) the current types of ICBM Reentry Vehicle (RV) are very pointy cones, with a roundoff at the nose of less than 10% of the base diameter typically.
See for example: The Nuclear Weapon Archive W-78 warhead / Mk-12A RV webpage.
I've designed several capsules which were generically Apollo shaped but truncated the cone lower down; it has its uses, but it's not really better from a mass or space ratio point of view. You can stick a larger hatch in the nose that way, you can stick cargo outside the pressurized capsule that way; I put pressurized Mars rovers riding outside some big (40 plus foot diameter) capsules of that geometry for one mission design.
Masswise, it end
That's a complete nonsequiteur.
The capsule shape is called the sphere-cone geometry, for the rather obvious reason that it has a nose that's a section of a sphere, and then a conical mid and aft body. It's not bell shaped. Use its proper name.They are far from the first use of the sphere cone design. As a matter of fact, more sphere-cone reentry vehicles (400 plus) have been used in history than any other shape. More than all other shapes put together, as a matter of fact. The Discoverer film capsules provide plenty of flight history on the geometry.
Nor is t/Space the first company to propose using them for modern manned capsules.
Nor is modern use the first proposed use: the same geometry was proposed for space lifeboat capsules in the 1960s and 1970s. It has always been felt to be a very easy to design, robust, and simple concept.
Sphere-cone requires slightly more thermal protection system mass than a conventional blunt cone shape like Apollo or Gemini, or even the Soyuz or Shenzhou. On the other hand, it has superior characteristics during the touchdown and landing phase of the spacecrafts flight, and is simpler and lower risk in other areas as well. You are trading off ablative heatshield mass for other technical risks being lessened.
Heatshields are among the cheapest components of a spacecraft on a per pound basis. If your objective is to minimize risk and cost, then more heatshield mass is a good thing if it simplifies other problems.
Existing capsules were designed to be expendable, not reusable. You don't generally reuse things you designed to throw away.Despite it being an expendable capsule, a Gemini was flown a second time in an unmanned test flight. So it has been done before.
List price of a SpaceX Falcon V launch vehicle, with 5.5 ton payload to International Space Station orbit:
$16 million
Estimated / planning price for a Delta IV Heavy launch, with payload of 22 tons or so to space station orbit...
$280 million
4 x $16 million = $64 million
1 x $280 million = $280 million
$64 million is a little bit less than $280 million
The only caveat here is that Falcon V hasn't flown yet and Delta IV Heavy has flown once, with a moderate non-catastrophic failure.