Slashdot Mirror
Carmack's Throatless Rocket Engine
Baldrson writes "John Carmack is working a potentially disruptive technology: A throatless rocket engine. Its made from plain aluminum pipes with few machined fittings. Carmack says: "The great thing about these engines is that it only takes me two nights to machine the parts, so we can test two engines a week if necessary." It scales too: "If this line of tube engine development works out, we can make a 5,000 lbf engine with very little more effort than the test engine." This is what makes disruptive technology development work: Cheap, fast turnaround on on redesign producing technologies that scale. If this works, the NASCAR guys may really start entering space competitions like the X-Cup."
Not that I'm one to criticize (large liquid-prop rockets built by Skyshadow: 0), but everytime they get an engine together and start encountering difficulties it seems like they scrap it and just go to another design. Assuming that rockets are anything like the mechanical things that I understand (cars), this just isn't how you can go about these things -- you've got to settle on a promising, well thought-out design and then dedicate your efforts towards ironing out the kinks or you'll perpetually be just past "go".
Anyhow, just the impression I get from reading the updates.
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
since I'm not a rocket scientist, I fail to understand the importance of what John is doing (or has discovered? surely throatless engines aren't an entirely new concept are they?).
I understand that this *might* impact manufactoring costs, but exactly how is this revolutionary, or going to affect us? Are we going to sport some pocket engines in the future? Are they more environmental friendly? Do they scale well? Will it run Linux?
Seriously, after reading the story and the article a few times I haven't yet understood half of it.
Hack your mind out of its sandbox.
No, we are not "stubborn". And yes, the standard system SUCKS. The reason we haven't moved to metric is that we have too much momentum built up in society for anyone to switch over. What needs to happen in America is a migration. This is slowly being done, but it will take many more generations beyond me.
I was tought the metric system in grade school. However, I only use it personally when working on cars. Most of the time GM will have a mix of standard and metric bolts these days.
Life is not for the lazy.
"If you would have read through armadillo's website you would see that he has been putting a serious effort in. As an aerospace engineer who has been keeping tabs on John for several years I can assure you he's got his design well thought out." - by everphilski (877346) on Sunday August 07, @11:07AM
:)
That is good to hear, that another pro from that SPECIFIC field is seconding his designs, and as far as thinking things out (when it comes to this guy, Mr. Carmack)?
I believe you.
E.G.=> Ever heard of "Carmack's Reverse"? If not, look it up, & check it out... this showed me this guy can THINK, & "outside of/above & beyond the box"...
Not every computer scientist has things like that to his credit!
IMO, these types of things (engines/algorithms) to overcome limitations ARE the mark of what others called him here:
A genius.
IMO, he is.
After all, his wares from IDSoftware & contributions in the OpenSource world show it. There's little arguing with that & little arguing with success.
*
APK
P.S.=> I wish him the best of luck to be quite blunt & honest about it...
Like he himself has said (not a direct quote):
Once you've accomplished a decent amount in 1 particular field? It's good to take a "hard right turn" into another one that interests you and make a career of it!
In his case it would NOT surprise me one iota if he does not end up becoming a breakthru maker in it as he did in computer science...
People like these, just give them time & experience: They're the types that always come thru!
Yes, this is "intellectual/technical hero worship" & he is one of the few out there I give that much credit to!
He, along with Mr. Anders Hejlsberg (of Borland (TurboPascal/Delphi designer) & Microsoft (C# & Visual Studio improvements) fame)... they're 2 of the 'outstanding' individuals in THIS field (computers) out there now, truly outstanding ones.
There IS a diff. between pretty good, good, great, & absolutely outstanding/in a league of their own.
Never underestimate folks like those, they often surprise... apk
Now "disruptive technology" will become the new buzzword. Suddenly AJAX will be called disruptive technology. Linux will be called disruptive technology. The next P2P file sharing protocol will be called disruptive technology. Fuck!
I know it sounds crazy, but sometimes it is much easier to evolve a nice solution to an engineering problem than to calculate one. For example calculating robustness and reliability is very difficult. It is much easier to simply measure it.
Change the unreliable bits to something else, and see if it breaks. If you kill off the most unreliable part of your design each time, I guess you can work out what happens.
I'm not saying that you shouldn't use simulations or calculate anything at all. Use them as guiding hands, but don't let them alter the process. The problem is many people think that it is easy to create some sort of `breakthrough' design. It isn't; if it was, then everyone would have them.
but he doesn't work for NASA
Neither does Burt Rutan.
oh?
is that so?
keep in mind that the US had to "borrow" german scientists like and Werner von Braun from the third reich nazi regime in order to develop the technology.
no need to tank me for reminding...
You've got your judgement of Carmack's mentality exactly backwards. From the Armadillo site:
"A chamber with no contraction ratio at all will lose 20% of its thrust due to pressure losses from accelerating gasses in the straight section, but the Isp loss is only 1.5%."
How much *design* is actually going into these if they are expecting to build 2-3 a week???
What does it matter? Rapid turnaround means he can develop a lot of them, and pick the best performers.
Maybe John, as brilliant as he is, should go to school for awhile to learn a bit about fluid dynamics and thermal dynamics and the equations that govern those sciences.
Why would you assume that he doesn't already know a great deal about these subjects?
I really don't get the knee-jerk reaction around here. Whenever someone does something interesting and potentially significant, there's always this chorus of people looking for something to bitch about to try to look cleverer than the guy who's acually doing something.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
Plus it gives people something to bitch about and blame pretty much everything that goes wrong on. They always appreciate that.
Switch back to Slashdot's D1 system.
The Russian approach to rocket science was along the lines of "let's see how it blows up and make sure it blows up in a different way next time".
That's a complete crock. The Russians, just like the USA, started from the V2s they captured, and applied the best engineering practices that were known at the time. They blew up a lot of rockets, and so did anyone else who every tried to build a rocket.
The fact that they got Sputnik into orbit first, and got Gagarin into space before anyone else is a testament to their skill.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
This idiot should stick to games.
Well, thank you for that well-reasoned critique of his efforts.
I have to wonder, what is it about achievement, or even effort, that brings people like you out of the woodwork? Why are you so jealous of him?
'Disruptive' is one of those buzzwords that business school types throw around when they are trying to deceive investors.
It's also a term that describes any number of scientific and technological advances that came from private effort like John's. In any case, he's doing this with his own money, so what's it to you?
What is the advantage of this engine design? What are the reactants, ISP?
Hey, here's a wild idea: why not RTFA and find out?
Meaningful details like that don't get you posted on slashdot I guess.
The same could be said for meaningful critiques. Better luck next time.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
Yes! For the slight inconvenience of having to work with four different length units (inch, foot, yard, mile) with easily remembered conversion factors, we can finally express lengths of 1/3 or 1/6 inch as umm... 10.667/32 in, or... 0.3333 mill. Better example: Highway exit signs. The Englisch system is obviously superior, since we can easily say 1/4 mile, where you only need to divide 1760 by 4 to get yards. Easy! Converting something like 1/4 kilometer to meters would involve dividing 1000 by 4: Totally ARBITRARY!
You win, I'm switching over to Imperial today.
chl
PS: Is 543 yards plus 1563 yards longer than 1-1/4 mile? I lost my calculator.
PPS: Sarcasm is not funny. Mod this insightful.
So space flight is all about weight reduction.
That's the NASA myth, and why they build crappy spacecraft.
Certainly weight reduction is important, but if you don't keep looking at the overall system, and the trade-offs, you'll fuck up the design. The focus on Isp is a symptom of that. The Shuttle is a prime example of it.
Sure, high Isp means you need less weight of propellant per unit of impulse (thrust * time), which sounds great. So the propulsion guys start focusing on Isp and design exotic high-pressure engines that use fuels like LH2 which is really light and gives you an engine with a 400-range Isp. Never mind that you have to practically rebuild the thing after every flight. (It also gives you a relatively low thrust to weight ratio, so you end up having to augment it at launch with something with high thrust even if crappy Isp, like SRBs)
The propellant guys figure LH2 is cool too -- highest energy per unit weight, and all that.
And the airframe guys just build the lightest structure they can to hold what the propulsion and propulsion guys give them.
All of which leads to a suboptimal design.
LH2 is about the lowest density liquid around. High density urethane foam would sink in the stuff, it's about the same density as lighter weight foams. Which means you need a bloody big tank to put the stuff in. The Shuttle uses 8 times, by weight, as much LOX as LH2, but the LH2 tank is about three times bigger than the LOX tank.
Tankage is heavy. The portion of airframe weight devoted to tankage scales as the volume of the propellant, not the weight.
Replace LH2 with something like liquid methane and your tankage -- and its weight -- becomes much less, which ends up improving your fuel fraction, even with the slightly lower Isp of methane-LOX.
The original Atlas launch vehicle, which used LOX and kerosene and stainless steel tankage, could reach orbit without shedding any stages. (Although it did shed two of its three engines.) It also couldn't carry much payload that way, but we're talking late 1950s technology.
NASA technology is fragile and unreliable because they're more interested in engineering projects for the sake of engineering projects, and then have to hack it back because of budget limitations. Anyone in software development who has seen a project get designed with all sorts of bells and whistles and the designers' favorite new technlogies, only to be turned into some ugly hacks toward the end when deadlines are looming and the budget has run out, will understand this.
-- Alastair
Didn't Mr Edison make over 100 prototype of lightbulbs before finding the successful version?
With Carmack is trying many different technologies, I presume that he is learning from each problem/difficulty with each model he works with. This shall eventually result in him having a broad and deep knowledge of the technologies involved, and give him the ability to achieve his goal, given enough time.
See my art -> http://herbevore.deviantart.com
"Look at it this way - the amount of fuel it takes to get to orbit will get you from the US to Australia in a 747. The reason it's cheaper to go to Australia is they don't throw away the plane when you get there (expendables) or take it apart and rebuild it (the shuttle) before the next flight."
I do love this old chestnut: it's perfectly true, and yet so misleading.
Helpful Hint #1: How long does it take to fly the 747 to Australia? What would happen if you took that long getting to orbit?
Helpful Hint #2: What is the difference between energy and power? If two machines release the same total energy, but machine #2 must release it much more quickly with equal or greater precision, which is likely to be more complex and expensive?