Rocket launch prices don't scale linearly with payload mass. Launching small payloads is very expensive per unit mass. The Falcon 1 variants are a direct competitor to the Pegasus family, but at 1/3rd the price. *Assuming* they can keep their prices down (or even drop them, such as with reuse). This is always a tricky aspect; often what happens is that issues that came up in development or a market that failed to materialize increase prices. Whether that will happen with SpaceX, only time will tell.
One thing I find encouraging is, as Musk notes, how much simpler the Falcon is than the Pegasus. The Falcon is a standard two-stage liquid-fuelled rocket designed ground-up for simplicity and bulk production. Pegasus is partially made of hardware borrowed from earlier rocket programs. You're looking at five stages. The first has to be man-rated, since you drop the rocket from a plane. You have to maintain the plane. The first rocket stage is effectively a hypersonic aircraft, complete with flight surfaces. And so forth. So, I think there's a good shot of them staying cheaper than Pegasus, probably by a good margin.
Also, I just in general like the approach they've taken with the Falcon series; there are a lot of clever design choices. My favorite: the cross between balloon tanks and rigid tanks. Balloon tanks are very light but very flimsy tanks that rely on internal pressure to keep stable. This gives you a better payload fraction but makes them hard to handle on the ground without damaging them; you have to transport them inflated, for example. Rigid tanks are heavier, but easier to handle. The tanks on the Falcon are rigid enough to not have to be transported under pressure, but not to withstand the forces of launch without their internal pressure. It's a "best of both worlds" type situation.
Even better than GOTO -- if your programming language supports them, COMEFROM statements can be a godsend. If your language doesn't support them, have your program call programs written in languages that do; wise usage of INTERCAL linkages can help ensure that you're seen as irreplaceable in your position.
I strongly agree. Self-describing code is much better than comments. Comments are only useful, IMHO, when you need to describe a complex situation. If it can be expressed in one sentence or less, it should probably be part of the code itself. I.e., instead of:
std::string mkdec(std::string x)// Converts x, a string representing a hexidecimal number, to a decimal string.
The latter says the exact same thing, but is far likelier to be maintained properly. Also, if you get in the habit of coding like that, you never have to worry about forgetting to comment. Furthermore, the "comment" is effectively replicated every time the function is used. Hence,
you see: hex = "0x" + number_str; return convert_hex_string_to_decimal_string(hex);
Now, if you had a function that implements a complex algorithm that can't be summed up in short order, then sure, use a comment. But in my experience, 95% of comments in code are like the above "mkdec" comment, and would be better expressed just by using a more descriptive function or variable name. I think a lot of coders are just lazy and don't want to have to type in longer, more descriptive variable and function names.
I got it back when I was planning to reinstall a system from scratch and started randomly rm -rf'ing directories in order from my subjective judgement of what would be least important to what would be most important. That's what I started getting after wiping out/etc.;)
Yep -- Flight 2 would have been orbital if they had just installed a baffle *or* reduced the stage separation kick (they modified the design for both). Barring something completely unforseen, Flight 3 would have been orbital had they not upgraded the engine *or* had they been able to test in a vacuum (as far as I can tell, the Merlin is too big for even the largest vacuum chambers), *or* had their CFD simulations shown the residual thrust more accurately.
Unfortunately, while rocketry may sometimes look like hand grenades, "close" doesn't count. One error, even slight, can easily doom an entire mission.
Anyways, I'm looking forward to this weekend, watching the launch on the big screen with my partner. This sort of thing is a geek equivalent of the superbowl. I'm rooting for the Earthlings to win.;)
Well, given that since they resolved the corrosion problems of the first flight, they haven't had one in-flight part failure, I don't think you can yet criticize them for quality control. Flights 2 and 3 failed due to engineering issues that didn't show up in simulations or ground testing, not defective parts.
FYI, developing a new launch system from scratch is an extremely high risk endeavour (our early rocket programs experienced the same thing). Imagine if software that you wrote could only be run for testing purposes once or twice before release, and any single bug in your program likely meant that your customers would irrevocably lose all of their data. That's the sort of situation that rocket engineers have to face.
If this were serious, of course some occasional parts that are likely to be good but have a question of doubt that they may be bad are replaced. Heck, I do that when I'm repairing my automobile... for exactly the very same reason.
And this is a rocket going into outer freaking space. It's not like your car will blow up if something goes wrong. Unlike your car, an orbital rocket will.
Okay, well... I'm assuming that you don't drive an early Ford Pinto.
Ramuglia said Sunday that the IP address he found in the proxy service logs didn't "look consistent" with reports identifying Kernell. By today, however, he had changed his mind.
"It became clear that the ISP, in addition to serving Illinois, also serves Tennessee, which means that the IP address could actually be consistent with the news reports," Ramuglia said today.
Representative mechanical property measurements on multiwalled carbon nanotubes
Multiwalled != Single-walled. Space elevators are proposed to be built out of SWNTs because they have lower density; the issue is the density versus the tensile strength, not the tensile strength alone. SWNTs are hollow tubes. MWNTs are filled like onion stalks, and are thus more dense.
And besides, even if you did have an SWNT with a tensile strength of of 150GPa, that doesn't mean that your *bulk fiber* will be 150GPa. It doesn't even mean that a single fiber will be 150GPa. Or even that the VdW-bonded bundles of tubes will be 150GPa. Each scaling up you do, the lower your tensile strength.
Did you actually read the diff? Here's a better one:
Q) What's the difference between the old tech page and the new one? A) The new one is half the length.
Yeah, they cut a ton of stuff out. And the net result is that it's half the size. It still talks about net neutrality, just not for nearly as long. So? They were making the page shorter. They entirely removed the sections on Immigration Reform and Promote American Business Abroad. You think that means that Obama now no longer supports immigration reform or supporting American business abroad?
For people who can't bother to read TFA, here is the revised section's *number one* (by order of listing) tech policy platform plank:
"Protect the Openness of the Internet: A key reason the Internet has been such a success is because it is the most open network in history. It needs to stay that way. Barack Obama strongly supports the principle of network neutrality to preserve the benefits of open competition on the Internet."
Cry me a river that it doesn't spend half a page elaborating.
The problem is that even the *simplest* form is way beyond what we can produce in the present day, and you're wanting to do a form that's far harder.
In a space elevator, the tether has to be long. Very, very, very long. So much that even if you could build a cable with the density of graphite and a tensile strength of 100GPa, it'd still have to taper severalfold as it reaches toward the earth. With the taper requirement, pulleys are simply right out (can't have the pulley's cable change shape as it goes, now can you?), as is *anything* that can increase the weight of the fiber. You need elevator "climbers", powered by beamed power transmission.
The problem remains the cable. 100GPa with the density of graphite is just so far beyond anything that we can achieve today it's really just a sci-fi concept that people like to dream about. The last I checked, the strongest *individual single-walled carbon nanotubes* that people had directly measured the strength of broke at just over 60GPa. This is for single tubes, let alone bundles of tubes, let alone a bulk fiber, let alone an entire tapered cable. Tubes theoretically can be stronger, but I haven't seen any measurements confirming such extreme theoretical strengths. The strongest SWNT bulk fiber I've read about was planar sheets that were about 10GPa.
Yes, you can build a space elevator with a tensile strength of less than 100GPa. But your taper factor for the elevator rises *very fast* with decreasing tensile strength or increasing density, which means that its mass increases *very fast*, which rapidly puts it outside the realm of possibility. Honestly, something more like 120GPa would be much easier to build, but that's even further from what we can achieve today. I'm not even sure it's physically possible to achieve. SWNTs are pure graphene SP2 structures; how can you get stronger than that? The only thing I can think of that could help us best today's best strengths are complete perfection, every atom of the fiber all the way up, and I'm not sure that would do it.
Meanwhile, Gabriel Ramuglia, webmaster of Ctunnel, the proxy service used to attack the Yahoo account, has identified the IP user of the perpetrator but he doesn't think it points to Kernell, Computerworld reports.
"Because I'm not in contact with the Internet service provider, I'm not 100% sure of where the IP is based," he said. "But from what I can tell, the IP address doesn't look consistent with the media reports."
The FBI will be able to close the loop, though, with the records of the ISP to which that IP address is assigned. Ramuglia said it is a small, residential ISP.
Now, it's always possible that he compromised someone else's box or drove a long distance away to someone else's computer.
No freaking kidding. The next time there's a widely publicized hack, is someone going to make a 4chan post taking credit for it under the name "president@whitehouse.gov".
Actually, I came up with this design (3d cell with towers to repeatedly reflect non-converted the light, esp. low-angle light) back in 8th grade or so. Seriously. I could probably dig through my old papers where I used to sketch up my ideas to dig out a copy. Back then, though, pretty much all solar was crystalline and I couldn't think of an effective way you could produce crystalline cells in the required towers. Nowadays, with thin film deposition processes, it's probably achievable.
Irregardless of clearance, body-on-frame construction generally raises CG, period. Increasing ground clearance *also* increases CG, but it's not the only factor. There are unibody offroad vehicles. In fact, Jeep has been making them since '84, and the Jeep brand is virtually synonomous with offroading. Most SUV lines nowadays seem to be headed back in the direction of unibody construction. Even the iconic Ford Explorer is going to unibody. It's unlikely now that GM is going to introduce another body-on-frame SUV in the forseable future.
Barium titanate has far, far better permittivity than conventional materials. Hyperpure aluminum oxide is a far, far better dielectric than conventional materials. There is no electrolyte in an EEStor capacitor, and no, electrolytes are not charge carriers in this case. You're dealing with the buildup of a charge charge on the opposing plates; the strength of the differential is limited by the permittivity of the dielectric (i.e., how well the electrons on either side tug at each other) and the dielectric constant (if you get voltage breakdown, that's it, it's over). Conventional supercapacitors have breakdown voltages at around 2-3 volts. EEStor's is designed to run at voltages *three orders of magnitude higher*. If you short circuit a capacitor, you short circuit it, period. It's just not an option; it's not even close to an option. It's like suggesting that since guinea pigs walk, you should make tires out of live guinea pigs to make cars go faster.
EEStor's EESU is not about surface area. It's about a material with both incredibly high permittivity (ability to permit an electric field) and an incredibly high dielectric constant (ability to resist electric currents). You'd simply blow both the dielectric constant and the permittivity by trying to force graphene into the mix.
The Scion example was just an example of how radically different of a drag coefficient you can get without a fundamental change in shape. But as for heavy duty applications, body on frame gets worse grip on cornering and leaves you with a higher CG, so for a vehicle that advertises itself as for offroad use and uses commercials of people racing around through corners on bumpy dirt roads, that's hardly desirable properties.
There's not really a point to it. EEStor's caps are a layered solid of composition modified barium titanate, aluminum oxide, and PET that has to be engineered to a ridiculous level of precision. You can't start sticking graphite in there without messing it up, and why would you want to take up space of something that's far more energy dense with something that's far less energy dense?
It's not always as hard as you make it out to be, either. Small changes in shape that don't really even impact style or cost can make a huge difference. For example, SUVs are made with body-on-frame construction, not unibody. This makes it easier to churn out a couple new models every year, but makes them heavier and less safe. For another example, the Hummer H2 and Scion xB are both boxy vehicles, but the Hummer has a drag coefficient of 0.57 while the Scion has a drag coefficient of 0.35 (and I'm talking about drag *coefficient*, not drag area; this is *before* you consider changes to the cross-sectional area). It's almost a willful disregard for efficiency. And we haven't even gotten into things that have a price point but pay off rather quickly, such as more efficient drivetrains (higher efficiency engine layout, IMA or other stop/start, diesel, HCCI, etc), aluminum in places where steel isn't needed for structural integrity, higher efficiency accessories, and so on, or more radical streamlining.
That said, consumers probably are mostly to blame, namely for insisting that their vehicles look like a brick and drive like armored tanks, complete with the high weight, low visibility and lack of maneuverability that entails. But automakers are not blameless.
Anyways, my primary hope is that the fuel crunch (which seems to be going away fast, IMHO) will help change consumer style preferences to more aerodynamic shapes and lighter bodies, as well as increasing the awareness that a higher upfront cost can pay off down the road.
Slashdot Mirror
Do remember that Musk is also wrapped up in solar power (SolarCity) and electric cars (Tesla Motors).
Rocket launch prices don't scale linearly with payload mass. Launching small payloads is very expensive per unit mass. The Falcon 1 variants are a direct competitor to the Pegasus family, but at 1/3rd the price. *Assuming* they can keep their prices down (or even drop them, such as with reuse). This is always a tricky aspect; often what happens is that issues that came up in development or a market that failed to materialize increase prices. Whether that will happen with SpaceX, only time will tell.
One thing I find encouraging is, as Musk notes, how much simpler the Falcon is than the Pegasus. The Falcon is a standard two-stage liquid-fuelled rocket designed ground-up for simplicity and bulk production. Pegasus is partially made of hardware borrowed from earlier rocket programs. You're looking at five stages. The first has to be man-rated, since you drop the rocket from a plane. You have to maintain the plane. The first rocket stage is effectively a hypersonic aircraft, complete with flight surfaces. And so forth. So, I think there's a good shot of them staying cheaper than Pegasus, probably by a good margin.
Also, I just in general like the approach they've taken with the Falcon series; there are a lot of clever design choices. My favorite: the cross between balloon tanks and rigid tanks. Balloon tanks are very light but very flimsy tanks that rely on internal pressure to keep stable. This gives you a better payload fraction but makes them hard to handle on the ground without damaging them; you have to transport them inflated, for example. Rigid tanks are heavier, but easier to handle. The tanks on the Falcon are rigid enough to not have to be transported under pressure, but not to withstand the forces of launch without their internal pressure. It's a "best of both worlds" type situation.
Should they have consulted all the sages they could find in Yellow Pages first?
Even better than GOTO -- if your programming language supports them, COMEFROM statements can be a godsend. If your language doesn't support them, have your program call programs written in languages that do; wise usage of INTERCAL linkages can help ensure that you're seen as irreplaceable in your position.
I strongly agree. Self-describing code is much better than comments. Comments are only useful, IMHO, when you need to describe a complex situation. If it can be expressed in one sentence or less, it should probably be part of the code itself. I.e., instead of:
std::string mkdec(std::string x) // Converts x, a string representing a hexidecimal number, to a decimal string.
std::string convert_hex_string_to_decimal_string(std::string hex_string)
The latter says the exact same thing, but is far likelier to be maintained properly. Also, if you get in the habit of coding like that, you never have to worry about forgetting to comment. Furthermore, the "comment" is effectively replicated every time the function is used. Hence,
instead of:
hex = "0x" + number_str;
return mkdec(hex);
you see:
hex = "0x" + number_str;
return convert_hex_string_to_decimal_string(hex);
Now, if you had a function that implements a complex algorithm that can't be summed up in short order, then sure, use a comment. But in my experience, 95% of comments in code are like the above "mkdec" comment, and would be better expressed just by using a more descriptive function or variable name. I think a lot of coders are just lazy and don't want to have to type in longer, more descriptive variable and function names.
I got it back when I was planning to reinstall a system from scratch and started randomly rm -rf'ing directories in order from my subjective judgement of what would be least important to what would be most important. That's what I started getting after wiping out /etc. ;)
My favorites that I've gotten:
"This wizard will complete the installation of:
AQP AA002! P O a @ P @1 Ae IoD'i"
And:
"You don't exist. Go away."
Yep -- Flight 2 would have been orbital if they had just installed a baffle *or* reduced the stage separation kick (they modified the design for both). Barring something completely unforseen, Flight 3 would have been orbital had they not upgraded the engine *or* had they been able to test in a vacuum (as far as I can tell, the Merlin is too big for even the largest vacuum chambers), *or* had their CFD simulations shown the residual thrust more accurately.
Unfortunately, while rocketry may sometimes look like hand grenades, "close" doesn't count. One error, even slight, can easily doom an entire mission.
Anyways, I'm looking forward to this weekend, watching the launch on the big screen with my partner. This sort of thing is a geek equivalent of the superbowl. I'm rooting for the Earthlings to win. ;)
Well, given that since they resolved the corrosion problems of the first flight, they haven't had one in-flight part failure, I don't think you can yet criticize them for quality control. Flights 2 and 3 failed due to engineering issues that didn't show up in simulations or ground testing, not defective parts.
FYI, developing a new launch system from scratch is an extremely high risk endeavour (our early rocket programs experienced the same thing). Imagine if software that you wrote could only be run for testing purposes once or twice before release, and any single bug in your program likely meant that your customers would irrevocably lose all of their data. That's the sort of situation that rocket engineers have to face.
If this were serious, of course some occasional parts that are likely to be good but have a question of doubt that they may be bad are replaced. Heck, I do that when I'm repairing my automobile... for exactly the very same reason.
And this is a rocket going into outer freaking space. It's not like your car will blow up if something goes wrong. Unlike your car, an orbital rocket will.
Okay, well... I'm assuming that you don't drive an early Ford Pinto.
Link
Ramuglia said Sunday that the IP address he found in the proxy service logs didn't "look consistent" with reports identifying Kernell. By today, however, he had changed his mind.
"It became clear that the ISP, in addition to serving Illinois, also serves Tennessee, which means that the IP address could actually be consistent with the news reports," Ramuglia said today.
You are mistaken. From the article:
Representative mechanical property measurements on multiwalled carbon nanotubes
Multiwalled != Single-walled. Space elevators are proposed to be built out of SWNTs because they have lower density; the issue is the density versus the tensile strength, not the tensile strength alone. SWNTs are hollow tubes. MWNTs are filled like onion stalks, and are thus more dense.
And besides, even if you did have an SWNT with a tensile strength of of 150GPa, that doesn't mean that your *bulk fiber* will be 150GPa. It doesn't even mean that a single fiber will be 150GPa. Or even that the VdW-bonded bundles of tubes will be 150GPa. Each scaling up you do, the lower your tensile strength.
Did you actually read the diff? Here's a better one:
Q) What's the difference between the old tech page and the new one?
A) The new one is half the length.
Yeah, they cut a ton of stuff out. And the net result is that it's half the size. It still talks about net neutrality, just not for nearly as long. So? They were making the page shorter. They entirely removed the sections on Immigration Reform and Promote American Business Abroad. You think that means that Obama now no longer supports immigration reform or supporting American business abroad?
For people who can't bother to read TFA, here is the revised section's *number one* (by order of listing) tech policy platform plank:
"Protect the Openness of the Internet: A key reason the Internet has been such a success is because it is the most open network in history. It needs to stay that way. Barack Obama strongly supports the principle of network neutrality to preserve the benefits of open competition on the Internet."
Cry me a river that it doesn't spend half a page elaborating.
The problem is that even the *simplest* form is way beyond what we can produce in the present day, and you're wanting to do a form that's far harder.
In a space elevator, the tether has to be long. Very, very, very long. So much that even if you could build a cable with the density of graphite and a tensile strength of 100GPa, it'd still have to taper severalfold as it reaches toward the earth. With the taper requirement, pulleys are simply right out (can't have the pulley's cable change shape as it goes, now can you?), as is *anything* that can increase the weight of the fiber. You need elevator "climbers", powered by beamed power transmission.
The problem remains the cable. 100GPa with the density of graphite is just so far beyond anything that we can achieve today it's really just a sci-fi concept that people like to dream about. The last I checked, the strongest *individual single-walled carbon nanotubes* that people had directly measured the strength of broke at just over 60GPa. This is for single tubes, let alone bundles of tubes, let alone a bulk fiber, let alone an entire tapered cable. Tubes theoretically can be stronger, but I haven't seen any measurements confirming such extreme theoretical strengths. The strongest SWNT bulk fiber I've read about was planar sheets that were about 10GPa.
Yes, you can build a space elevator with a tensile strength of less than 100GPa. But your taper factor for the elevator rises *very fast* with decreasing tensile strength or increasing density, which means that its mass increases *very fast*, which rapidly puts it outside the realm of possibility. Honestly, something more like 120GPa would be much easier to build, but that's even further from what we can achieve today. I'm not even sure it's physically possible to achieve. SWNTs are pure graphene SP2 structures; how can you get stronger than that? The only thing I can think of that could help us best today's best strengths are complete perfection, every atom of the fiber all the way up, and I'm not sure that would do it.
Link
Meanwhile, Gabriel Ramuglia, webmaster of Ctunnel, the proxy service used to attack the Yahoo account, has identified the IP user of the perpetrator but he doesn't think it points to Kernell, Computerworld reports.
"Because I'm not in contact with the Internet service provider, I'm not 100% sure of where the IP is based," he said. "But from what I can tell, the IP address doesn't look consistent with the media reports."
The FBI will be able to close the loop, though, with the records of the ISP to which that IP address is assigned. Ramuglia said it is a small, residential ISP.
Now, it's always possible that he compromised someone else's box or drove a long distance away to someone else's computer.
Yeah. He confirmed that his son is being widely named on blogs and chatrooms. No freaking kidding.
No freaking kidding. The next time there's a widely publicized hack, is someone going to make a 4chan post taking credit for it under the name "president@whitehouse.gov".
much better than the Seinfeld ones, anyway.
So is stepping into a bathtub full of agitated electric eels, but I wouldn't recommend Microsoft use that as a marketting strategy either.
Actually, I came up with this design (3d cell with towers to repeatedly reflect non-converted the light, esp. low-angle light) back in 8th grade or so. Seriously. I could probably dig through my old papers where I used to sketch up my ideas to dig out a copy. Back then, though, pretty much all solar was crystalline and I couldn't think of an effective way you could produce crystalline cells in the required towers. Nowadays, with thin film deposition processes, it's probably achievable.
Irregardless of clearance, body-on-frame construction generally raises CG, period. Increasing ground clearance *also* increases CG, but it's not the only factor. There are unibody offroad vehicles. In fact, Jeep has been making them since '84, and the Jeep brand is virtually synonomous with offroading. Most SUV lines nowadays seem to be headed back in the direction of unibody construction. Even the iconic Ford Explorer is going to unibody. It's unlikely now that GM is going to introduce another body-on-frame SUV in the forseable future.
Barium titanate has far, far better permittivity than conventional materials. Hyperpure aluminum oxide is a far, far better dielectric than conventional materials. There is no electrolyte in an EEStor capacitor, and no, electrolytes are not charge carriers in this case. You're dealing with the buildup of a charge charge on the opposing plates; the strength of the differential is limited by the permittivity of the dielectric (i.e., how well the electrons on either side tug at each other) and the dielectric constant (if you get voltage breakdown, that's it, it's over). Conventional supercapacitors have breakdown voltages at around 2-3 volts. EEStor's is designed to run at voltages *three orders of magnitude higher*. If you short circuit a capacitor, you short circuit it, period. It's just not an option; it's not even close to an option. It's like suggesting that since guinea pigs walk, you should make tires out of live guinea pigs to make cars go faster.
EEStor's EESU is not about surface area. It's about a material with both incredibly high permittivity (ability to permit an electric field) and an incredibly high dielectric constant (ability to resist electric currents). You'd simply blow both the dielectric constant and the permittivity by trying to force graphene into the mix.
The Scion example was just an example of how radically different of a drag coefficient you can get without a fundamental change in shape. But as for heavy duty applications, body on frame gets worse grip on cornering and leaves you with a higher CG, so for a vehicle that advertises itself as for offroad use and uses commercials of people racing around through corners on bumpy dirt roads, that's hardly desirable properties.
There's not really a point to it. EEStor's caps are a layered solid of composition modified barium titanate, aluminum oxide, and PET that has to be engineered to a ridiculous level of precision. You can't start sticking graphite in there without messing it up, and why would you want to take up space of something that's far more energy dense with something that's far less energy dense?
It's not always as hard as you make it out to be, either. Small changes in shape that don't really even impact style or cost can make a huge difference. For example, SUVs are made with body-on-frame construction, not unibody. This makes it easier to churn out a couple new models every year, but makes them heavier and less safe. For another example, the Hummer H2 and Scion xB are both boxy vehicles, but the Hummer has a drag coefficient of 0.57 while the Scion has a drag coefficient of 0.35 (and I'm talking about drag *coefficient*, not drag area; this is *before* you consider changes to the cross-sectional area). It's almost a willful disregard for efficiency. And we haven't even gotten into things that have a price point but pay off rather quickly, such as more efficient drivetrains (higher efficiency engine layout, IMA or other stop/start, diesel, HCCI, etc), aluminum in places where steel isn't needed for structural integrity, higher efficiency accessories, and so on, or more radical streamlining.
That said, consumers probably are mostly to blame, namely for insisting that their vehicles look like a brick and drive like armored tanks, complete with the high weight, low visibility and lack of maneuverability that entails. But automakers are not blameless.
Anyways, my primary hope is that the fuel crunch (which seems to be going away fast, IMHO) will help change consumer style preferences to more aerodynamic shapes and lighter bodies, as well as increasing the awareness that a higher upfront cost can pay off down the road.