This was honestly a pretty poor choice of model to demonstrate their 3D printing capability. While it looks cool, the designer clearly does not know much about the design of rocket engines or fluid dynamics in general, and worse, this model doesn't really make it clear what they're capable of in terms of say minimum wall thickness or maximum unsupported span, surface finish, processing time, etc. Nothing here seems particularly impressive from anything other than an aesthetic standpoint.
Further, it seems they're trying to make parametric CAD, the industry standard for the past 30 years, sound impressive to the layperson using buzzwords.
I'll add on that Aerospike engines tend to be popular with non-Aerospace engineers because they look cool. In reality Aerospike engines are optimizing for a design parameter space that doesn't need optimization. To draw an analogy to coding, it's like optimizing code in the non-crtical path. It doesn't really help you.
Rocket engine bells expand high pressure subsonic choked flow into supersonic low pressure flow. They are designed to do that most optimally for a given external pressure (if you try to expand the gas to a lower pressure than the atmospheric pressure then vibrations begin and can tear an engine apart because of flow backwards into the engine bell). Aerospike engines, if designed correctly, work by automatically optimizing to always expand the gas to whatever the atmospheric pressure is, maximizing total specific impulse over the flight. However they tend to have problems with cooling and so tend to be extremely heavy compared to other engine designs and you can equivalently design for this space by instead using multiple rocket stages, which all rockets do. More so, if you're trying to re-use rockets as is the current trend, having a heavy engine makes even less sense as you need to land that heavier engine.
Aerospike engine research made sense when it was the current fad to design single-stage-to-orbit rocket designs because they thought that re-using multiple stage vehicles wasn't possible. However that's no longer the case. Aerospike engines are a technological dead end.
Disclaimer: I am not an aerospace engineer, though I did take a couple of aerospace engineering courses before switching my major (turns out I wasn't great at the type of math needed) and I have a strong interest in the field and have talked to people in the field (and former classmates/labmates who are aerospace engineers).
“Published 11 May 22” — since then, is there any news of fire tests?
I’m sure its cool they’ve been able to design and 3D print a complex structure - however, the company blog appears to only link to additive manufacturing companies: I couldn’t find a subsequent link where this engine has been (even) ground tested, demonstrating basic validation of the design in test / fire conditions: till that happens, this is just (imo) a cool looking untested, unvalidated design and therefore, untested, unvalidated algorithmic design..
None of the reports covering this even mention any testing, or any anticipated testing. Is it an art piece?
This is apparently a demonstration of their geometry software, not something that will ever be fired, since they are an "organic" modeling company, not a rocket company.
What's amazing about the linked blob post is that the headline is "But Does It Work?" and the word "no" only appears once, in the phrase "there’s no looking back."
The article asks really tough questions of itself, like "Have you tested it, does it work?" and then very carefully does not say yes or no. Why ask yourself those questions if you don't want to answer them?
Imagine a politician who starts a campaign speech "People are asking me, 'Am I a Ghost Wizard?' and here's my answer: ghost wizards are valid concerns. This builds upon the latest in crytozoological knowledge of supernatural studies. The first study of such wizards was conducted in the 1960s and 1970s. As I've discussed previously with my constituents, the United Nations has previously looked into ghost wizards. In a way, it seems that..."
Algorithmic design engineering is a fairly mature field that's been around for 20 plus years.
Looking at their website, they are simply developing a tool for more integrated algorithms from the park level to system level.
With this in mind, the the component integration of the model itself is extremely impressive and a proof of concept. It doesn't matter if it actually works or not, that's not really the point.
We don't know if they sold the design and whether it was tested. The design itself is very interesting - in particular the complex heat exchange structures.
On it being optimal, or even more efficient, as long as it's cheaper to build than a conventional engine, it's game.
I think all of that is basically out of scope. I don't think they are claiming it is a good design. The impressive point is their software can make a design at all. It seems like they want to sell it to people that will use it to try to make designs that are good
I think this level of algorithmic design is impressive, and as far as I know, novel. I certainly haven't seen anything this complex.
It is like a computer displaying "hello world". The output isn't particularly exciting, I could do better with an index card.
However, the fact that the a computer could take a set of inputs and display something at all shows exciting promise. The same is true here.
For a real engine design, you would have to work extensively on the inputs, boundaries, and goals. The impressive part is that this software can handle this complex of a parameter set at all.
as far as we know, "this level of algorithmic design" could be randomly generated volumes which serve no purpose and render the device completely non-functional
indeed, since it does't need to do anything but look cool, the inputs could be "weirdness factor" and "excessive internal wall ratio" and nothing else, and they thought "this kinda looks like an aerospike"
This is exactly why I am impressed. I have some basic understanding of rocket engine and aerospike design, and can recognize most of the features and the functions they would serve in this design. I have no clue how optimal they are, but it is clear that they are taking function into consideration.
Moving from the outside of the engine in, you can see that support lattice has been added to the exterior and density adjusted on the exterior to reenforce high stress areas.
Next, the coolant jacket around the combustion chambers has has baffles (verticals s curves) that seem to adjust cooling in specific areas.
It looks like the spike cooling system (the dense horizontal lines) is trying to balance fluid resistance with how much the cryogenic oxygen.
The one part I don't understand is what the Y shaped channels in the very center are supposed to do. This is the liquid oxygen flow to the tip of the spike before it returns along the outside of the spike as a heat exchanger (mentioned above). It looks like some sort of expansion chamber but I would have to ask a professional aerospike engineer what it is for.
This is a long way of saying the structures show functional intent to converge on some solution. Some parts seems random, but most algorithmic design is random change and iterative testing; Essentially evolutionary code, where permutations are made, compared, killed off, then fed into future models. I wonder how they handle so many degrees of freedom, and how much of what we see in the output was pre-defined.
Seems like a description of Homer Simpson's nuclear power plant
yeah, it has a racing stripe, and fins, like some stuff which goes fast also has, but that doesn't mean that adding it to a power plant makes the power plant fast
as another engineer that's worked in aerospace, none of those functional design aspects seem to be actually performing a function here other than "we saw this on something weird looking", as in not enough evidence that there's functional intent converging on something functional
I'm skeptical there's any value being generated here then. Who are the customers? How do they know this fancy looking biological design is any good at all?
It's fine not to actually test fire it, but in that case prove it with simulations. Demonstrate that it's superior to existing designs, or at the very least that it's functional.
How do you do comprehensive QA on a 3D printed part with complex internal structures like that? Can you X-ray copper? What happens to an engine that has a slight void or poor adhesion in a layer, does it have a 1% degradation in performance, or does it fail to cool effectively and melt during the dangerous part of the launch?
The first several (10s or even 100s) will be run, most likely from test stands, then ripped apart and combed over for hot spots and cracks. Id even expect them to build some with shittier materials specifically to see how it causes them to fail. Rocket engines are a shocking amount of trial and error.
X-ray CT is standard. There’s also just borescoping it. You can build witness coupons at the same time and test their strength. No substitute for test firing the engine. Even operationally, you’ll want to acceptance fire the engine before putting it on a vehicle.
This is a huge field in additive manufacturing. One of the most promising solutions is in-situ process monitoring to identify defects that may occur, often using machine learning.
You can probably build a kind of acoustic signature of the whole at various different points in the boundary and check each build is sufficiently close to that signature.
This is pure PR fluff. if a company you dont already know says they have an aerospike engine, you should already be skeptical. a 3D printed one? Ok now you know its bunk. a small deep dive proves that.
Will an aerospike engine have more safety margin using the SpaceX fuel mix and not the hydrogen type maintank on Artemis I with all the fuel leak complications the media could not see because the "funnies" would embarass the lawyers and accounting arranging the "goofproofers" in the KSP simulation?
This is seriously one of the coolest things I have seen lately. Taking a moment to look at the overall engine itself, the nature of how you arrive at such eloquently-chaotic beauty, and the sheer uniqueness of the damn thing, wow!
Appears to me to be littered with waveguides upon waveguides, pinch points, and various irises. This is really cool. Whether it works to cool the nozzle down, I feel like the beginning simulations could easily yield do more of this, less of that, test, and vary in a iterative feedback loop of sorts. Maybe even hot gas at tight-knit control, induction loops for moving heat inward to segments, or outward? Think of just how fast you could build this just by going single-part alone if it works.
This looks like a fantastically useful technology and I am reminded of the fact that when FDM 3D printers were under patent, they cost $25,000, and ten years after the patents expired they cost $250. I believe some key metal 3D printing patents recently expired, but I suspect much of this process is still patented, making it unnecessarily expensive. I would love to see this kind of tech become ubiquitous and affordable to everyone, and this will not happen until the patents are fully expired and we see genuine competition in the space.
I guess I should preempt a couple of the most common responses I get to this view:
First, markets provide first mover advantage even without intellectual property restrictions. The idea that governments should provide monopoly protections on ideas is anti-competetive and anti free market. Libertarians at places like the Mises institute recognize this and have some good talks on the subject.
Also, we would not see investment dry up without IP restrictions. We would see the nature of investment change from fewer larger investments to more smaller investments as competitors race to get the latest incremental improvement to market first by seeking investment to upgrade production lines, etc.
Workers absolutely deserve to be compensated for their work, but we see that in most IP restriction regimes, businesses take all the winnings and pay only wages to the workers doing the inventing. Individual inventors can still have first mover advantage, and I would argue that most new invention is not motivated by profit but curiosity. Removing IP restrictions would vastly increase invention by curiosity, as there would be far more places where a curious engineer could tinker and improve something. Imagine that one person invents something and gets a patent. This will prevent 100 other people from tinkering and improving upon it. This is why I say that the sole function of a patent is to reduce innovation - because that is the one literal function of them. The supposed follow on effects are more of a cultural meme that are often disproved by things like the open source movement, which clearly demonstrates that a lot of the assumptions around IP restrictions are not strictly true.
Finally, people in foreign countries who cannot afford expensive machines like medical scanners etc still deserve to build copies of those machines for their own use, but patent harmonization laws like TRIPS prevent these sorts of things. The same goes with medicine.
Likely more of an artpiece for the tradeshow than anything else. For the RAPID tradeshow you see plenty of intricate prints like this showing off the capabilities of different AM processes. This one being hosted at the EOS booth is showing off the AMCM customizable platform. Printed aerospikes have been tested a bunch already but none of them are market ready.
That doesn't solve the problem if you can't get enough flow through those channels. Cooling channels doesn't solve cooling, it just helps it. From my non-Aerospace (though I took a couple of Aerospace undergrad classes) perspective, the hardest points to cool in an aerospike engine is the spike itself because it tends to be thin and narrow with no internal room for cooling pathways compared to the intense heat of sitting in the center of the flame.
TL;DR: We printed a really big, hard to print model aerospike engine. This was done to demonstrate how our printing could be used in the future to make possible as-yet impossible advances in rocket science. It's purely conceptual.
It's entirely possible that this aerospike engine is actually lighter because they don't need large nozzles and nozzle extensions and are more efficient
Standard nozzles do not need a cooled support structure right in the middle of the combustion chamber. I love these things, I made a linear aerospike (Never fired), but this thing just looks heavier than a standard TCA/nozzle
Further, it seems they're trying to make parametric CAD, the industry standard for the past 30 years, sound impressive to the layperson using buzzwords.