Cunjo Carl
-
Posts
881 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Bug Reports
Posts posted by Cunjo Carl
-
-
I found a fun anecdotal talk from Mueller about the development of the Merlin engine, which I thought I'd share.
Part 1
Part 2
He talks briefly about the pintle injector he used for Merlin, so I did some research to find they're singularly great for throttlability, low-cost, and combustion stability. Apparently Sea Dragon was going to use pintle injectors, and there's (unsubstantiated) speculation that Raptor may use them as well. The engine's rapid tuneability kinda suggests it is, and the fact raptor is gas-gas suggests it isn't. It's apparently still a mystery! Here's a nice easy-read review paper on pintle injectors, and how they do what they do so well.
http://www.rocket-propulsion.info/resources/articles/TRW_PINTLE_ENGINE.pdf
The Evelyn Woods is, the fuel is sprayed radially from a single spray nozzle at the center (like the cone setting on a garden hose), and the oxidizer is sprayed straight through a cylindrical nozzle around that. The two flows meet at 90degrees to each other, so they hit hard and mix mid-flight into a cone shaped flow. A little bit of the radial fuel spray will sneak through, and cause 'film cooling' along the walls of the combustion chamber, helping prevent the engine-rich exhaust we saw in last week's Raptor test.
Probably the biggest bonus for SpaceX, the combustion can be quickly throttled right at the injector head with a single hydraulic actuator for both fuel and oxidizer.
Anyways, I hope someone finds it an interesting watch/tidbit like I did. Cheers!
-
I don't think it would be stable over astronomical timescales, but it seems to me it should be able to exist hypothetically for long enough to make a story happen. After all, such a system would require divine intervention to set up anyways right? No one says it has to last forever! Admittedly, lasting long enough for life to develop is almost definitely a no. However, it might even be a fun plot point that citizens already on these planets would need to pump water around to help keep it all stable for as long as possible. Anyways, I decided to take the idea and see how far I could run with it.
Assuming the bodies are rigid, we can get a little idea of some limits on the system, and then maybe use those to test the sanity of an (insane) example case. For a pair of orbiting bodies, the period of rotation is (Kepler's third) T = 2*pi*sqrt(a^3/GM) where 'T' is the orbital period, 'a' is the sum of the planets' semi-major axes to the mutual center of gravity, G is the gravitational parameter and M is the combined masses. If we assume 2 equal planets (rigid for starters) with radii r and mass m, this would become . T = 4*pi*sqrt(r^3/Gm) We could assume our planets have super dense cores for simplicity, but maybe let's assume they're water-density through and through just to see what happens. In that case the mass would be m = (4/3)*pi*r^3*Rho where r is the planet radius and Rho is the density of water (1000kg/m^3). Bringing it all together, we get T = sqrt(12*pi/GRho) .
For the planets to be grazing, they'll have a specific period that depends on their densities, but not their size! That's interesting, so perhaps we can have them test this up on the ISS
For water the orbital period T = 6.5hours regardless of planet radii (as long as they're the same). Sounds nice! The actual number should be a bit larger because the planets will deform towards eachother.
Of course in real life, the surface of the planets will be pulled into an oblate shape due to centrifugal forces. Also, let's assume everything's tidally locked (it should be!). We can say for any drop of water, it will feel 3 forces: gravity from the 2 planets' centers of mass, and the centrifugal force orbiting around the planets' combined center of mass. Now we can just solve for the shape of the potential well, and thank our lucky stars some people on the internet have done this for us already! Copyright Hale Bradt '09, Section 8 in https://www.cambridge.org/us/files/1913/6681/8626/7708_Tidal_distortion.pdf
Though these potential wells are for tidally locked stars, they can be stand-ins for our tidally locked planets just as easily. The potential wells in the photo on the right are just 3d interpretations of the one on the left. Here we can imagine our planets are as though we filled the potential wells up with water to the level it just bridges between the two, and makes a channel of some thickness. Notice in the picture on the right, we can fill it up to make a stable channel that won't leak out to space.
The very center of the channel won't experience any net force, but it will have a net pressure, which is a good sign. The outer surface of the water channel bridging between the worlds will feel a net inwards force towards the center of the channel due to gravity from the two planets. At the outer surface of the bridge, the gravity of the planets will cancel eachother in the back-and-forth direction, but will still pull towards the point that's directly between them in the 'radial' direction. It's the same force that allows people to orbit satellites around certain Lagrange points. As for the stability in the back-and-forth direction, the water in the very center of the channel will be supported by the water pressure of the attached worlds. If you were to draw a line between the planets' centers, the middle of the bridge would be the lowest pressure point, but could still be submerged and have a pressure if there was enough water. We can imagine the supporting pressure towards the middle of the channel as being provided by the inwards pull from all other points on the planets' surfaces.
So, if you were sailing along the surface of the channel bridging worlds and dropped a penny, it would sink 'quickly' to the inside of the channel, and then slowly towards whichever world was closer. Meanwhile neutrally buoyant things would just happily float like the rest of the water.
Stability... We're imagining the planets are the same size, and in this case it all works out nicely. We should doublecheck the system won't destabilize (have the channel grow) from small perturbations though. For example, what would happen for a small perturbation like water flowing from one planet to the other? Since a planet's radius is dependent on the cuberoot of volume, the receiving planet would grow less than the losing planet shrinks, causing the channel to narrow. This is a stabilizing factor then! Also, gravitationally, the receiving planet would pull towards the combined center of mass less than the losing planet would pull away, which is also stabilizing. Over astronomical time scales, there's no doubt the system would work its way towards becoming a single oblate sphere, but over short time scales I think it could be stable as two planets with a connecting channel!
... Or I'm full of hooey, that's always a possibility, too! Thoughts?
-
7 hours ago, wumpus said:
From what I've heard, Computer players have a huge built-in advantage in StarCraft 2 in being able to click with mouse/keyboard far faster than even pro players. Are they just considering AIs that have been limited to pro (or even amature) production rates? This sounds like a real challenge, as a computer has a sufficient advantage to being able to produce a zerg rush (and force a human or other AI to deal with that) far more effectively than a human, with or without an advanced AI.
What you're talking about is totally a thing. The most recent bots vs people tournament I'm aware of had a case just like that, where the machine built and controlled a rush of zerglings perfectly to muscle out some poor human protoss player. I don't speak a word of Chinese, but you can hear the commentator's reaction to seeing the sudden flood of computer units "Aaaah.... GG?" (in other words, is that game?)
Human players can eke out victories even with these disadvantages though, often making it look easy like Stork does later in the same tournament. In Starcraft (and SC2), you can lose every battle but still win the war with advantages in economy, technology, positional control, etc. It's never enough to just have an advantage though, you need to know how and when to use it, and that's where humanity has always shined.
What's made AlphaStar interesting is seeing that it actually tries to get advantages in these other ways, and watching how it tries to leverage them.
In the final show match it lost against Mana, you can see it trying to win using an economic choke hold rather than brute force. It builds an additional resource-mining base ahead of Mana and then does its best to stunt Mana's economy by killing his workers. After that Alphastar maintains its very slight army and positional advantage by keeping back and defensive, meanwhile rapidly growing its economic advantage. As the game progresses, it constantly keeps tabs on Mana's economy to make sure he's stuck, and (this is the cool part) it forgoes army conflict to best leverage its bigger economy for an even better advantage in the bigger later battles. However, it wasn't counting on Mana happily building up an unstoppable high tech army on just 2 little bases and then marching his way to victory. Being able to feel out these complex and constantly evolving interactions of advantages is the strong point of the human mind versus programs and AI. So, humans can find a way to win (for now) even given a wild handicap in terms of speed and control.
Saying this there's definitely some moments when Alphastar is in its element and its decision making looks almost... human! That's what everyone wants to see developed ultimately, the point isn't just winning, so for this to be fair and worth-while in the long run the DeepMind team will need to find a way to impose human control abilities onto their AI agents.
-
8 hours ago, KSK said:
Interesting article. I had mixed feelings about the way the single game was won against the AI. (Attack back of base, withdraw when AI responds, watch AI go back to what it was doing, rinse-wash-repeat.)
Granted, it was a lot harder to execute but that kind of exploit spamming seemed a bit ‘same old, same old’ to me. Much effort expended in training an advanced AI - and it still gets stuck in that kind of decision-making loop.
To double down on the point that the AI was exploitable, Mana had already exploited the AI's playstyle to put himself into a commanding advantage even before that hit and run harassment started. In other words, Mana could have won the game even if the AI responded correctly to the harassment by splitting off handfuls of units to defend each base.
The real key to Mana's success was that he (and TLO) realized each AlphaStar agent only had one tech path (play style) in mind, and wouldn't deviate from it. So, once he guessed and confirmed the techpath (mid game stalker push), he built an army that countered it hard. He won with better unit composition rather than the AI's better unit control, and the harassment was just icing on the cake. Not that it wasn't super funny to watch, mind you!
AlphaStar still has a long way to go, but even being able to play toe-to-toe for a little bit with professionals is something brand new for AI.
Conveniently, Mana talks about it a bit in his stream, I'll embed the video in here.
Spoiler -
Recent news in AI advancement
After mastering tic-tac-toe (the only way to win is not to play) AI has steadily advanced to tackle Chess, Go and is now burgeoning into Starcraft 2. AI developers have considered StarCraft 2 to be the next great stepping stone for AI because it's substantially more difficult to play- it has imperfect information, real-time play, a vastly greater move space and much more complex and layered interactions. This makes it very entertaining for us viewers because it gives the game a rich strategy which allows for different players' personalities to shine through.
To date, many teams of AI developers have tried to tackle StarCraft 2 only to come up short. Just last week though, Google's DeepMind team showed some impressive results with their new AI called Alphastar. Pushing AI the next step forward in StarCraft 2 required significant advancements both in technology and in how they created and trained AI agents. In a broad picture, they trained their AI in the normal way by creating neural network AI 'agents', vying them against each other, promoting the victors and making subtle random changes to their offspring for the next round. However, the DeepMind team attributes a large part of their recent success to 3 key and interesting choices:
1. They were very hands-off in the training process. It would have been easy to directly program the AI in many of the precepts of good Starcraft 2 play in order to give the starting agents a solid base-line and head start, but they instead let their agents learn everything from scratch. This led to several interesting discoveries in new/unusual play styles!
2. They trained AI agents using a league system. Rather than just throwing out the less successful agents, they would instead be left to play and evolve for a while in lower leagues. Some times, especially in AI, good ideas take a while to develop, and the league system gives those ideas time to mature rather than (more typically) discarding them in favor of the quick successes. This closely mirrors how people develop strategies in StarCraft 2; many great play styles develop from non-professional players in lower tiers, before they gain traction and catch on with the pros.
3. They selected specifically for agents that were difficult to exploit, rather than simply those with a high win rate. The quintessential problem of AI is that it tends to be fantastic at solving specific problems it saw during its training, but then be hilariously terrible when it comes to extrapolating this experience into the real world. So DeepMind focused hard on making their AI as least 'brittle' as possible. They selected for AI agents that lost the least consistently against other agents' specific play styles. Agents that won and lost in a well-rounded way were promoted, while the obliquely talented ones were left behind.
The results: So far the AlphaStar AIs have beaten StarCraft 2 professional players 10:1*. The AI agents tended to fight bizarrely and fearlessly, focusing hard on early-mid game pushes with masses of a single kind of unit. ([Day9]'s Funday Monday monobattles return!!)
I'll spare too many details, but there's a lot of nuances that makes this 10:1 victory much less impressive than it might look at first glance. In each of its victories, the AI was granted perfect control, infinite speed, the impossible ability to view/command the entire map at once, the ability to change AI 'agents' each game so humans couldn't adapt to any one's play style, and they played exclusively on an old less-practiced patch, with only one race & match up (Protoss vs. Protoss) and on only one map. Also, the 2 professionals chosen were superbly talented (actually two of my favorites!), but still far from world's best (at maybe 200th, and 20th place). Needless to say, there's a lot of room for the Alphastar AI to still improve. However, it's an incredible feat that an AI has taken a game off a professional even _given_ these caveats! Starcraft 2 is hard.
Google's DeepMind team showed some amazing technology that promises to develop further in lots of interesting ways. I'm excited to see where they take it from here.
Figured I'd share!
(Embedded video starts in the middle of the presentation on a cool looking tech graphic) -
39 minutes ago, Xd the great said:
Well, MOAR flaming and reentry flame effects are always welcome.
What is the temperature required to burn carbon soot? Will we see a burning starship due to these carbon soot?
You're right! I should be more Kerbal about this.
Most organics will happily ignite somewhere between room temp and ~500C (that is to say, they'll become flammable but necessarily ignite without provocation 'autoignition')
Carbon (like carbon black, carbon nanotubes, etc) tends to oxidize but not neccesarily completely burn in the 500-650C range, and will burn shortly above that if there's a good oxidizer present. I've been assuming the sides of the Starship will be mostly ensconced in methane during reentry, so the oxidation potential would be poor, but could definitely be wrong about that.
Pyrolysis/Evaporation of carbons like these without an oxidizer happens above 3000C! It's quite special.
I've heard people say that the energy of the methane burning will be so much less than the energy of reentry that it won't make a visible difference. It might be interesting to see a smoke trail though, if perhaps the conditions are right? ... That would look really cool, actually- like in the movies!
-
2 hours ago, Rakaydos said:
Concerning methane disassociation products, someone on NSF explained that it wasn't a problem in hypersonic flight due to "frozen reactions". If I understood it right, a mach 5+ airstream is an effective "someone else's problem" field- by the time the methane can actually disassociate, it's been blown far from the spaceship.
They also mentioned that this is a major problem with Scramjet design, as your jet fuel leaves the engine before it can burn.
I couldn't find the article you're talking about, if it's handy could you provide the link?
It sure sounds nice, but to be honest I'm not quite seeing how this would work. It sounds like the idea is that the air flow is so high, that actual time it takes for the methane to dissociate will be longer than its residence time near the craft.This makes a lot of sense for a combustion reaction which requires multiple collisions of large-ish molecules (kerosene), each of which involves a multi-step reaction with lots of reaction intermediates. Totally makes sense! That can take time.
With methane though, it's a surface catalyzed decomposition we're interested in. The methane can happily stick to the surface covalently through the process. No collisions in the gas phase are strictly required, besides the methane initially hitting the surface. Certainly, hydrocarbons can pyrolyse in the gas phase (especially the big ones), but they'll happily do it attached to a surface as well. I'm sure there's a paper on it..... Here (see discussion) . It's actually pretty perfect, 1000C methane for 10min on Iron Oxide! (I realize stainless will have a lot of chrome-oxide on the surface as well, but still) The 'amorphous carbon' they talk about is also called carbon black, but for thermal purposes, the carbon nanotubes they grow are pretty equivalent, if not a bit worse. Any ways, the point is that chemistry can happen without gas collisions, and the methane is being injected directly into the boundary layer near the metal surfaces. (as a note, chemically, a medium speed reaction at 1000C will typically be a very fast one at 1200C)
The relative amounts are also worrisome. For an engine, you want near-complete combustion, but in this case we'd be worried if even 0.001% of our methane did this. (Assuming maybe 10tons of methane used across a 750m2 surface, given 12/16 weight loss during pyrolysis, a final solid density of ~2, and ~50nm of film being enough to effect optical properties.)
Finally, I'd heard the metal surface was set to be ~1200C, so the methane within the pores is bound to have portions being plenty hot enough for all this to happen as well. I'm not even sure how we'd avoid clogging the pores, to be honest! It's kinda too bad to be honest. Otherwise dissociation reactions like this are almost always endothermic (heat absorbing) and can really improve a gasses heat capacity!
I'd still be happy to be wrong, I'd like to give your article a read!
-
So, I'm still on the new heat shield. It's interesting from a technical perspective, so I can't really help it!
The more I look at methane, the more it looks... completely intractable as a reentry coolant.
The problem isn't just that it decomposes when hot, but what it pyrolyzes into at 1000C: Carbon Black (or a relative like graphite or carbon nanotubes). For some reason, I was originally just worried about this happening inside the rocket, but now I think about it even outside in the wind, the decomposition reaction would be a huge issue. Funnily enough, Carbon Black is actually _the_ black body coating of choice in science! They make spray cans of the stuff, and it works really well! However, the whole idea of this steel starship is that it needs to remain a nice white body and reflect away most of the radiant heat. Unfortunately, only a few dozen nm of Carbon is enough to completely change the emissivity of a surface. Carbon black from pyrolysis happily forms directly on many metal (oxide) surfaces and unfortunately typically sticks really really well as a thin film (I learned from a neighboring lab trying to make carbon nanotubes). Unfortunately, once it forms it can be really hard to remove completely. Carbon black is also nigh unkillable from a chemical stand point. I remember trying to remove a very similar coating by soaking it in Caro's acid (an explosively good oxidizer) and it just hung out like nothing was wrong. My boss at the time told me to try soaking it in boiling 'SulphoNitric' (a 200C version of the oxidizer used to make TNT), and that _plus_ scrubbing did the trick. The oxidizing potential of reentry atmosphere probably won't touch it. It also unfortunately doesn't evaporate at pedestrian temperatures at all. Its close cousin, Vitreous Carbon, is actually what we use as a crucible for evaporating _Tungsten_! Perhaps reentry ablation can help keep it from forming too quickly, but I remember Carbon's "sputtering efficiency" being terrible as well, and from a physical standpoint I suspect that means it'll ablate slowly too. I should note that 'thick' films (1s of microns) are typically powdery and the outer layers can be easily wiped away with a cloth, but the inner layers can stick really well.
Moral of the story, carbon black (or a relative) grows quickly on surfaces touching 1000C+ methane, which is a huge problem for white-body reentry, and once it grows (that I've experienced) it can't be convinced to move for anything.So.... I have a great new idea for a heat shield I'd like to patent. Simply inject methane through pores machined in a traditional insulating heat shield and the shield will regrow lost thickness with carbon black during reentry! Hey, engineering's all about rolling with the punches right?
Nah, I think the water version will wind up being the easy-development choice for SpaceX, but it's sadly 50-80% heavier for the same cooling, will require an extra tank (with a heater), and is actually mildly corrosive to stainless steel in its Deionized form (which we'd need to use). The vapor, hot dry steam, isn't considered corrosive to stainless fortunately.
Then again, maybe they have a trick up their sleeve for the methane. I'd be happy to be wrong!
-
19 hours ago, tater said:
I just redid my calc, I thought 5000 was what I used, but it was in fact 1500. I did 2Pirh (no need for the end caps), then checked 50 vs 55m, and rounded down to ~1500.
1500*116 (the apollo tps density/m^2)= 174,000kg (which I rounded up to 200 tonnes (fins, etc)) as an upper bound.
So I was off by a factor of 2, I meant to divide by 2 but forgot.
So upper limit on the order of 100 tonnes (Apollo TPS), and realistically PICA-X for this application is some fraction of that. 20% would be 20 tonnes.
PS--like your EV Nova avatar image.
Thanks! The game's pretty simple by today's standards, but the universe it builds is super fun to explore.
Don't worry about the nose cone everyone, they're just getting the RUDs out of their system early!
After that long post I made, I finally went and read what Musk had been saying, and I think I was a bit off base in one aspect. I still think the coolant will be pushed out in a single pass and not circulated. However, Musk was apparently mentioning that they're only intending to reduce the skin temperature by 300C using this system.
This makes a lot of sense, because having a skin temp of (apparently) 1200C will make even shiny stainless glow bright red hot and release a lot of heat by radiation, which means we need to waste less fuel to transpiration. Meanwhile, I think my setup (being done with liquid methane) would wind up cooling the skin much more than 300C. To see if my idea is consistent (it isn't), we can estimate a lower bound on the Methane evaporation rate if there was 1200C on the outside of the skin, and the ~ -160C of liquid methane on the inside. Let's consider just the conduction across the steel skin as a very low-ball estimate- a 4mm steel skin with the 1200C -> -160C temp drop and constant (estimated) properties gives us: 1360K * 20W/m.K * 750m2 /.004m = 5GW . Despite Stainless being a very poor heat conductor for a metal, it's still a ton of cooling this way! How much methane would this translate to? We add the heat of vaporization to the heat of heating it up ( 8200J/mol + 35.8J/molK * 1360K ) / (.016kg/mol) = 3.5MJ/kg = 3.5GJ/ton . Dividing through, my setup would evaporate 1.5+ ton/sec of methane if somehow the outer skin temp were maintained at 1200C! Ha! We could fiddle with the assumed values, but it still way too high.
So what instead?
I think the general idea is fine, but given this high skin conduction problem it seems smart to put an insulating thin film on the inner surface of the steel skin (like Calcium Silicate, enamel or a spray-on sinterable insulator). This will prevent heat conduction through the skin from being problematic, and will let us have the option to use many fewer pores (and less fuel) as a result. Given this extra layer though, we'd probably choose to evaporate the liquid methane before it enters the space between the steel layers. It sounds like a bad idea to allow some liquid drop to accidentally seep into the insulator, then boil inside and delaminate it off. It wouldn't be a loss-of-hull problem, because this would actually make a cool spot on the skin! It'd just a major nuisance. So, let's guess some numbers to see if they can be pushed into sounding halfway reasonable... Let's say we can boil the methane inside the rocket using heat that conducts in during the gentler early parts of reentry by running the liquid methane through narrow tubes around the inside like a regenerative cooling nozzle, and bring it up to room-ish temperature. Now, including the insulator (let's say something very non-space-age like 2mm at 0.2W/mK) the total heat flux through the skin being actively cooled would be 1200K * 750m2 * ( 1 / ((.004m / 20W/m.K)+(.002m / 0.2W/m.K)) ) = 88MW . Meanwhile, the cooling of methane would come to ( 35.8J/molK * 1200K ) / (.016kg/mol) = 2.7MJ/kg = 2700MJ/ton . Dividing through, it comes to 2tons/min. Sounds nice! In this case, the steel skin would be almost entirely the same temperature throughout its thickness at 1200C, and the temperature drop would be almost entirely across the insulator. In this case, the steel would thermally just act as a coating, meanwhile the insulator would prevent the temperature from conducting into the rocket!
So it sounds nice at first blush but leads to tricky questions.
Does this ~100MW of cooling sound like a reasonable amount of heat that will need dissipating using methane during reentry? I based it on a random and out-of-context number I saw, but it's a pretty important part of the equation that I don't remotely have a handle on~!
Also, does anyone know how long reentry would last for Starship? I'd been assuming about 10 minutes like the shuttle, but don't really know.
Also, methane starts to pyrolize at ~1000C, forming H* and H2. Both of these are very damaging to steel due to Hydrogen embrittlement. I wonder if/how they plan to handle this? It's actually fairly standard to coat the inside of steel pores using various chemical and electrophoretic processes, but can any of these coatings handle the temp?
Now we're considering a different heat flow regime, what's the new guess for a number of pores? In effect, we're now on the opposite end of the biot number, so we can probably do an order or two less! 3-30M Maybe? A very non-zero issue will also be the pore diameter, which now we'd probably choose in order to get a nice pressure drop across the pores, maybe a couple psi? Too high of a pressure drop is annoying structurally for the steel, but too low is annoying to get uniform flow rates out the pores!
Well, they're all fun things to mull over.
-
2 minutes ago, tater said:
PS--I guestimated the surface area at 5000 square meters as well in my TPS calc just up thread.
I am not an "Occupy Mars!" person in the least, but SpaceX certainly likes to think ahead to this goal, so I bet that idea would make total sense to them (both for South TX as you suggest, or Mars).
Huh, good to know. It's pi*D*H for surface area of a cylinder, and I'd wound up squaring the D by accident in my first calc! Oops.
My latest looks like: ~500m^2 = ((54m/1.5)*3.14*9m)/2
where the 1/1.5 factor on the length accounts for the pointed shape, and the 1/2 factor on the whole thing counts for the pores on just one side... But you know what, my estimation is ignoring the fins which would totally need cooling! Now I'm getting ~750m^2, but still no where close to the 5000. I'm pretty shot though. I'll re-edit my post for the 750m^2 value. It doesn't change any of the bottom lines, but it'll make me feel better.
I'd gotten into space flight as a kid when my grandpa gave me Zubrin's book on going to mars... I won't pretend it's smart or practical, but dang if it's not cool! To be honest though, for me, actually going there is an after thought. I just like the tech!
-
1 minute ago, tater said:
Sounds pretty reasonable, as does the positive pressure to mitigate corrosion at a launch site. That could presumably work for dust as well (Mars, for example, looking down the line).
Awh, jeeze. Good call! The statically charged and super-fine martian dust would have a field day with those pores.
-
I'd been hum-drumming about a post about this over the weekend, and now I'm late to the party! Oh well, maybe some of this will still be interesting to consider. Please keep in mind, these are just the thoughts floating in the head of some engineer! No claims for gospel or magic here.
How to pump the fluid? First thing's first, a big advantage of cooling by perspiration rather than just circulating coolant along the inner wall is that the pumping will be very much simplified. For cooling with circulation you'd need to keep very high flow rates of coolant across the inner surface to keep the heat transfer rate high. High circulation rates means constant pumping, which is very difficult when your fluid has evaporating gas bubbles in it. Turbopumps can't handle it, so you'd either need to separate out the gas in a vapor-liquid separator (heavy, and generally hate shaking / changing gravity) or pump the fluid with a positive displacement pump (really heavy). However, by instead 'perspiring' the coolant through microporous channels, we make sure all the pesky bubbles are leaving the rocket right as they form! The coolant could just be pumped slowly without circulation, like water irrigating a field. This lets us use a nice, 'small', 'low-power' turbo pump.
Would the coolant evaporate within the pores, or be sprayed out to evaporate 'in the wind'? I'd guess we'd evaporate in the pores, hopefully reasonably close to the surface. I'm thinking since most of the heat is apparently absorbed from radiative emissions (rather than conductively/convectively from the air) increasing the thickness of the boundary layer will be helpful but not the primary effect. I could be convinced otherwise though, it's interesting to consider!
What size of pore and how distantly spaced? Let's just ballpark some numbers. I was thinking an outer steel skin of 2-20mm would be sane. Assuming the temperature drop is still mostly across this outer skin (since it's coming apparently predominantly from radiative heat) We'd want a pore aspect ratio (length/diameter) of maybe 3:1 - 20:1 with closely packed pores in order to prevent there from being hot spots in the steel where the heat sneaks through and causes bubbles to evaporate inside the vessel. So, perhaps 250-750um pores? Just taking some easy values, a 4mm skin with 10:1 aspect ratio pores would have 400um diameter pores, hexagonally arranged at maybe a 1.6mm center-to-center spacing. This makes the total material about 80% dense (20% air gaps in the steel), and is arranged so any point on the surface is atleast 5 times closer to a pore than it is to the inner surface. This should keep most of the original strength and also have minimal conduction to the interior. Nothing's magic in these numbers, they just sound like nice guesses for an initial picture. Perhaps for practical reasons, the coolant may evaporate .5-1mm into the pore rather than right at the surface. In this case, my example would have any point on the surface being 4 times closer to the evaporation in a pore than it is to the inner surface. This was the number I hunted for to get the 1.6mm spacing.
How many pores on the craft? Starship is planned to be 9m Diameter x 54m length, and only half of it (I assume) will have pores. Given it has a tapered shape and fins, I'm seeing about a 750m^2 surface area needing pores. Dividing through I'm getting a little over 300 million pores. Yeah, that sounds about right!
How do we make 300 million tiny pores? There's a dizzying array of options, but laser ablation is the runner to beat. 400um holes with a 10:1 aspect ratio made by the billion in stainless steel is right up its ally. We could talk a lot about the particulars for a case like this, but I think it's by far the simplest option.
My main issue with everything is corrosion. Stainless steel is corrosion _resistant_ not corrosion proof in sea-spray conditions, like after a barge landing. These super-high-surface-area pores will easily be able to catch and hold water by surface tension. It's kinda asking for trouble, so maybe.... Just Maybe... They'll prevent this by applying a positive pressure of clean dry air to the inner jacket, to constantly blow air out the pores.
And just maybe... If they do... I'll get to play air hockey on it!
Looking forward to hearing what people think.
-
If spaceX's new super alloy SX500 can handle 12000psi when superheated, can we finally use it to make a drilling machine to plunge through the Earth's crust?
-
I think I'm missing context, but any loss of life is terrible, and 66 is tragic. If you were affected by the event, I'm very sorry for your loss.
-
7 hours ago, Green Baron said:
Yeah, that really does not make that much sense science wise. They'll never achieve the energies even of the LHC. Better join LHC/FCC !
In a lot of ways I want to agree, especially because CERN pays half my bills! But myself and many of my workmates really like the ILC proposal as well. I can understand why Japan doesn't want to foot the bill solo, of course.
Anyways, cheerful scientific facts! ILC would work using an electron/positron rather than a pair of protons like LHC, which gives it two advantages. The first advantage is its reactions would be a heck of a lot cleaner! It's funny to think about it, but since protons are made up of quarks, those quarks that didn't directly collide apparently wind up getting flung around, sometimes making jets, and generally making a mess of trying to observe the actual things we're interested in. One of my jobs is to help cram electronics into the LHC detector upgrades specifically to help with this problem. As it was explained to me, the ILC has an advantage in that it can make a pinpoint little glob of pure energy because the positron and electron annihilate on contact, making for generally cleaner reactions.
The second advantage is that if the total energy of the electron-positron annihilation (mostly relativistic mass) is ~240GeV, you apparently get these beautiful clean reactions where you 'commonly' get a Higgs and a Zboson, which itself acts as a nice marker. 240GeV is pretty darn low energetically speaking, but it's the optimum energy for this awesome reaction. Apparently the ILC could spit out Higgs like a firehose. Would have loved seen it... If only I was a multi-billionaire!
I'm happy to hear about the FCC as next steps though for CERN. In the meantime, LHC for the win.
-
1 hour ago, DDE said:
Thing is, there are options beyond 9 mm most of these tests used.
As there are options beyond ordinary kinetic FMJs.
Very true, I kinda took it on assumption your question was meant for a common handgun or .556, or similar. If you decide to move the firepower northwards, there's really no sensible limit to how massive people have made things that shoot so I'm sure there's something that'd make a bang! I think we'd have to figure a 12Ga slug followed by some of this stuff would make for fireworks.... Oh, to have a backyard that large. For the record, I'm saying this on assumption, I've never worked with it.
SpoilerBy the way, stay tuned until after the 3 minute mark to get a science lesson! I'm a fan of the channel to say the least.
-
This seemed a little too appropriate to not add in to the thread...
-
19 hours ago, Green Baron said:
Thanks, it's nice to finally hear some hopeful news after the latest dying gasp from ILC
-
3 hours ago, DDE said:
Let’s revisit the old “shoot gas tank for car explosion” trope. Justs how impossible is it?
And what happens if we use an actual gas (propane)?
Ooh, 'practical' general chemistry.
I was always taught that cars make for good concealment and reasonable cover in a shootout (better than nothing), and that the most common issue was actually a fuel line (not the gas tank) being hit and leading to rapid conflagration, but not technically an explosion. Fortunately, never had to put anything into practice!
Propane tanks are under positive pressure, 5-10 bar gauge (75-150psig) temperature pending. That's enough pressure to make an 'explosion' (a loud bang), and fling highly volatile liquid everywhere even if nothing ignites! Igniting liquid butane isn't particularly hard, and I have to figure propane is easier. Getting a proper explosion from the gas is quite a bit trickier though (thankfully!) . A little bit like spaceX, I like to differentiate 'static fire anomalies' from real explosions
. It's fairly easy to get an idea of how easy something is to explode from checking its explosive limits (though not necessarily how well it burns). https://www.engineeringtoolbox.com/explosive-concentration-limits-d_423.html . Propane looks pretty standard for a flammable gas at 2-10%. Even though only the upper and lower explosive limits are shown, there's often non-explosive concentrations within the region. Anyways, bottom line is I'd expect lots of trouble occurring, but almost definitely not a proper explosion from a single round. Hmm, that is unless you used one of the few remaining Gyrojet bullets! I think I'd pay to see that experiment! (Edit: Joking, just to be clear)
I'd be surprised if someone hasn't tried a practical experiment of these and posted it to youtube. This seems like clear mythbusters territory to me, especially given their Hollywood origins.
-
15 minutes ago, GDJ said:
A Schottky Diode has a very small forward voltage loss. Would that work for you?
It's a good idea, and I should have specified. Unfortunately, the ~.25V drop of a schottky will be too much. I'm hoping for a forward drop in the .01V range, (equivalent to an RDSon of 40mOhm). I realize this may be impossible of course!
-
Does anyone know of an electronic device (ideally an IC or a combination of discrete components of some sort) which can block reverse current without imposing a forward voltage drop? Relays and reed switches are a bit too slow for my application, unfortunately. I'll probably be working in the 10s-100s of us time scale. I'm above signal level, but ultimately still looking at low-ish currents (250mA) and voltages (20V) for starters though. Also, I'll unfortunately specifically need to block reverse current not bypass it like in a crowbar circuit.
I've heard of symmetric JFETs does anyone have experience with those maybe? Not sure if they do what I'm thinking of.
Thanks in advance!
-
I'm probably the internet's slowest surfer at the moment, so it was by some wild serendipity that I think I found your planes! They seem to be from an upcoming Bandai-Namco game called Kotobuki (full name: Kouya no Kotobuki-Hikoutai or Kotobuki Squadron from the Wastelands). Here's the links I found associated (web pages in Japanese):
https://kotobuki-game.bn-ent.net/aircraft/
https://kotobuki-anime.com/aircraft/
The planes in your pictures seem to be:
Photo 1: Ki-43 "Hayabusa" ('Oscar' or 'Army Zero')
Photo 2: N1K1-J "Shiden" ('George')
Photo 3: A6M3 "Zero" ('Hap', 'Zero' or 'Zeke 32')
Photo 4: A6M2b "Zero" ('Zero', or 'Zeke')Plane 5: Ki-44 "Shouki" ('Tojo')
There was a 5th plane shown on the web page as well, so I went ahead and added it to the list. Within my abilities, I'm happy to translate anything on the pages. The tag line for the game seems to be "So many skies, so many of our stories to tell." and it has a planned release in winter 2019. Neat!
-
On 1/11/2019 at 12:52 PM, zolotiyeruki said:
I don't know anything about StarLink--does it somehow not require directional antennas?
Given the number of simultaneous users and datarates, I'm guessing it will use a special kind of directional antenna called a 'phased array' whose direction can be controlled electronically (without anything actually moving). They're already common in AM/FM radio (whenever you see 3-4 adjacent antenna towers) and modern cell phone towers, and can work to create a narrow beam using a grid-shaped array of tiny antennas. As for the user end... who knows?
https://en.wikipedia.org/wiki/Phased_array
https://en.wikipedia.org/wiki/Beamforming
Image care of Chetvorno on wiki, CC0.
-
1 hour ago, Nich said:
Could the wolfhound be good because of economics of scale? Also if you are using the wolfhound you have an enormous payload. Perhaps you need to be more efficient with your payload. I have always wished I had versions of the terrier and poodle that were 70% smaller.
Could be, the economics of scale definitely appear in the engines, where larger versions are typically a little bit better on the whole than smaller versions. The most direct example is Poodle/Terrier or Mamoth/Mainsail, but it's definitely a trend in KSP that bigger is in general a bit better. That said, despite them sharing a niche, Wolfhound (2.5tonne) is dramatically better than Rhino (9tonne) in almost all conditions they'd normally be used in, so I don't think economics of scale quite cover it here...
(** Here using economics of scale to refer to larger objects being cheaper per mass rather than the other meaning of things being cheaper by the dozen)
As for payload efficiency, it's definitely super important. But fortunately, the whole kit'n'kaboodle up at the top is expressly to find the best possible payload efficiency, so that's all covered! It's been a long time coming.
I also wish we could tweak the scale of the engines. I'd totally want a little baby twin boar for trips to Eve!
SpaceX Discussion Thread
in Science & Spaceflight
Posted · Edited by Cunjo Carl
Oh, I figured out why they aren't worried about their 1200C steel surfaces sooting up into a black mess with carbon from the Methane!
First, I found a paper confirming that sooting occurs on metal oxide surfaces even with 0 reaction time, so the winds from reentry really aren't going to make the problem completely go away. Above about 800C, we'd slowly get soot even from methane. Grown in this way, the soot would be harder and more chemically resilient than the stainless steel, so cleaning it would be a bear.
But then I had an epiphany (well, perhaps it was more of a 'no duh' moment :D). Above 1000C even stainless steel will 'carburize'- it will actually soak up Carbon into the metal! This is great because above 1000C it will allow the surface to stay nice and shiny. However, it's also bad because if done wrong it can change the structure's properties during reentry, which doesn't sound like the greatest even in the context of SpaceX's joy for cowboy engineering.
Done correctly, carburization hardens steel surfaces by increasing Carbon content, permanently improving both scratch and corrosion resistance. However, too much carburization causes brittleness and enormous internal stresses. An additional concern, once the heat shield is carburized, keeping it between 500-1000C for longer than about 10 minutes will cause the Chrome to precipitate out (as Chrome Carbide) and result in all sorts of nastiness. Heating it back above 1000C will reset the timer though, so in the context of reentry to sea level this isn't so difficult, but it's still an added requirement.
Methane is great at carburizing steel, and can work in the 1s of minutes timescale at 1200C, causing problems for repeated use. Fortunately, the carburization reaction can also be reversed. H20 and CO2 are excellent at decarburizing steel even in moderate concentrations, effectively removing the excess carbon from the steel in the form of CO. Now, if only we had some way to make H2O and CO2 midflight (nudge nudge, wink wink @cubinator ) . So if the correct levels of CH4, CO2, and H20 are maintained, the carburization (carbon content) in the steel will also be maintained, and the shiny hot steel will remain intact for many uses.
As a result, the steel heat shield would stay at a carbon equilibrium and would be effectively be catalytic for:
CH4 + CO2 -> 2CO + 2H2 and
CH4 + H2O -> CO + 3H2
This would be a nice side bonus because these reactions should be endothermic (absorbing heat), and are also fantastic for increasing the specific heat of the resulting gas! Generally for holding heat, the more moles of gas you can split your molecules into the better they become at soaking up heat. The above reactions both double the moles of gas from their reactants, so they're great from a thermal standpoint. The Oxygen from the reentry wind will also be decarburizing, but I'm guessing it's to a lesser extent in context, and the SpaceX engineers would need to account for carburization inside the pores and heat shield as well.
This is a much nicer looking picture now, but it brings up a whole new interesting set of questions. How will they manage carburization/decarburization during reentry? Or will the just accept the heavy surface carburization for what it is and maintain/swap heat shields accordingly? This sounds like the sort of nuts-and-bolts problem SpaceX excels at solving, and I'm interested to see what they come up with.
Images care of JFP Technical and Phase-Trans respectively. Images show carburization (left) and decarburization (right) in the surface of stainless steel. SpaceX will probably want to be somewhere in between!