I shortened the title a bit from "LEAP 71 hot-fires 3D-printed liquid-fuel rocket engine designed through Noyron Computational Model."
From the article:
- First rocket engine built entirely through a computational model without human intervention
- Likely the shortest time from spec to manufacturing for a new rocket engine (2 weeks, usually this process takes many months in manual engineering using CAD)
- First liquid fueled rocket engine developed in the United Arab Emirates
> Likely the shortest time from spec to manufacturing for a new rocket engine (2 weeks, usually this process takes many months in manual engineering using CAD)
Does anyone in the field of rocketry specifically know if this alleviates some previously annoying constraint?
My uninformed gut suspects that these rocket spend an overwhelming amount of time in the post-design stage (I mean rocket engines seem to stick around for a long, long time, right?). But I’m a programmer I don’t know anything about this stuff.
"We've built a great new way to design physical structures."
"So what? The existing ways work just fine."
"We designed and built a rocket engine in two days."
However, even in the rocket field, there's a "design, simulate, build, test" cycle. They can do two of those steps in effectively 0 time and with significantly lower cost.
Moreover, it looks like the design has incremental feedback from something akin to simulate.
One might argue that CAD is "computer-aided human design" whereas this engine was designed using "human-aided computer design". That is, the computer isn't aiding in the design, the computer is the designer (and I assume humans are just providing some basic constraints). The difference between the two is subjective and perhaps meaningless, but it does poetically describe the technological advancements that are being made in physical design.
That being said, I think stuff like this is governed by homeostasis: bleeding-edge technological advancements eventually get turned into regular features. In this case: I'm sure we'll continue to see CAD software to build more complex structures with less human intervention; and maybe eventually designers will expect their CAD software to generate whatever rocket engine their product requires.
I was going to say that this is nothing Hyperganic hasn't done....and then looked up Lin and Joesefine who were previously at....Hyperganic. I wonder what the story is over there. Open sourcing their geometry kernel is a very confident move.
Interested to see what happens between Lab71, Hyperganic and nTopology - traditional CAD/CAM packages are integrating topology optimisation / generative design but are simply not voxel-first. Perhaps there's a middle-ground to be found (though possibly requires more developed use cases first).
>The engine was designed autonomously without human intervention
Hmmm. My software compiles itself 'without human intervention' when I click the compile button (ignoring the thousands of hours of work I put into writing the code and the even larger amount of work that went into creating the compiler).
This appears to be a pressure-fed rather than pumped engine, so limited real-world utility. Nonetheless, it’s incredibly impressive especially given that it seems to have been successful on the first try.
I wonder how practical it might be to integrate turbo machinery into an automated design system like this?
Oh, and it really is beautiful with copper construction and that fascinating swirl.
> This appears to be a pressure-fed rather than pumped engine, so limited real-world utility
This is addressed in the article:
> This is a relatively compact engine, which would be suitable for a final kick stage of an orbital rocket.
It has lots of real world application, just not currently as part of a lift stage since you're right it's a pressure based one as opposed to a pumped engine.
All are pressure fed. A pump generates pressure. It's common to test engine components without pumps using high pressure vessels in lieu of pumps. The E Complex at Stennis Space Center specializes in this approach.
Pressure fed is a fixed term when applied to rocket engines and means “fed only by the pressure in the tank (which is most of the time generated by having a pressurization system fed by another high pressure helium tank) and not by a pump”.
“ The engine uses thin cooling channels that swirl around the chamber jacket, with a variable cross sections as thin as 0.8mm. The Kerosene is pressed through the channels to cool the engine and prevent it from melting.”
Kerosine/LOX burns very bright, compared to the Methane/LOX tests you may have seen in other tests, which produce almost invisible exhaust and have bright shock diamonds. Unfortunately the cameras were calibrated for the less-bright tests they usually do on this test stand, so the footage is overexposed. We tried to stop down the cameras, for the long duration run, but it was still too bright. So no pretty shock diamonds in the footage. Exhaust is supersonic after engine throat, reaching around 2300m/s at nozzle exit.
If your blowtorch produces 5kN, all the power to you. Even small rocket engines are surprisingly strong.
Lin (co-founder of LEAP 71 — and "I was there, when we heard the rocket rumble through 3m of concrete bunker walls")
Yeah, no pretty looking shock diamonds in that exhaust. Which makes me thing the exhaust velocity is pretty low, which I'm not too surprised by since the throat of that engine looks pretty large. And the specific impulse (efficiency) of a rocket engine is directly tied to the effective exhaust velocity [0].
Still amazingly cool, but to the other questions on this thread I'm sure the performance is not comparable to an existing rocket engine design.
An incredible achievement and in to my eyes a thing of beauty. This is not the first time I've seen computational geometry (played with it myself) but this output seems something else.
Readit News
From the article:
- First rocket engine built entirely through a computational model without human intervention
- Likely the shortest time from spec to manufacturing for a new rocket engine (2 weeks, usually this process takes many months in manual engineering using CAD)
- First liquid fueled rocket engine developed in the United Arab Emirates
- Engine worked on the first attempt
- No CAD was used in the design
Does anyone in the field of rocketry specifically know if this alleviates some previously annoying constraint?
My uninformed gut suspects that these rocket spend an overwhelming amount of time in the post-design stage (I mean rocket engines seem to stick around for a long, long time, right?). But I’m a programmer I don’t know anything about this stuff.
"We've built a great new way to design physical structures."
"So what? The existing ways work just fine."
"We designed and built a rocket engine in two days."
However, even in the rocket field, there's a "design, simulate, build, test" cycle. They can do two of those steps in effectively 0 time and with significantly lower cost.
Moreover, it looks like the design has incremental feedback from something akin to simulate.
> No CAD was used in the design
This is amusing -- while I understand they mean "CAD tools" like 3d modeling software, the entire engine was literally "computer-aided design", no?
That being said, I think stuff like this is governed by homeostasis: bleeding-edge technological advancements eventually get turned into regular features. In this case: I'm sure we'll continue to see CAD software to build more complex structures with less human intervention; and maybe eventually designers will expect their CAD software to generate whatever rocket engine their product requires.
Deleted Comment
Interested to see what happens between Lab71, Hyperganic and nTopology - traditional CAD/CAM packages are integrating topology optimisation / generative design but are simply not voxel-first. Perhaps there's a middle-ground to be found (though possibly requires more developed use cases first).
Hyperganic, Lab71 or nTopology?
Do you have a link?
Interestingly the author of it is Lin Kayser, former CEO of Hyperganic.
Hmmm. My software compiles itself 'without human intervention' when I click the compile button (ignoring the thousands of hours of work I put into writing the code and the even larger amount of work that went into creating the compiler).
- Your code specifies design constraints--you need it to do X and Y and Z.
- The compiler takes your instructions and optimizes them for the processor instruction set. It seems like both represent huge productivity leaps from laboriously making things in the original low-level languages.Computer: generates natural language output in response to arbitrary prompts
Programmer: it’s just a computer program, it doesn’t require intelligence to do that. It’s not doing rocket science.
Computer: does literal rocket science
Programmer: sure, but it’s still just running a computer program
I wonder how practical it might be to integrate turbo machinery into an automated design system like this?
Oh, and it really is beautiful with copper construction and that fascinating swirl.
This is addressed in the article:
> This is a relatively compact engine, which would be suitable for a final kick stage of an orbital rocket.
It has lots of real world application, just not currently as part of a lift stage since you're right it's a pressure based one as opposed to a pumped engine.
If your blowtorch produces 5kN, all the power to you. Even small rocket engines are surprisingly strong.
Lin (co-founder of LEAP 71 — and "I was there, when we heard the rocket rumble through 3m of concrete bunker walls")
Still amazingly cool, but to the other questions on this thread I'm sure the performance is not comparable to an existing rocket engine design.
[0] https://en.wikipedia.org/wiki/Specific_impulse#Specific_impu...