This episode of Real Engineering is brought to you by Brilliant, a problem solving website that teaches you think like an engineer. If you have been on the internet in the past month, you have probably seen a picture of Elon Musk’s latest project. A rocket that looks like the brainchild of a H. G Well’s fever dream of the future. It doesn’t look like any current generation rocket by any shape or measure. It’s shorter and fatter than your typical Space X rocket, and most strange of all, it’s made of stainless steel. A material that has largely fallen out of use for propellant tanks since the 60s. Steel is strong, but it’s pretty heavy. Making it unsuitable for flight structures. Reducing the weight of the launch vehicle is an art form in rocket science. Every kilogram matters, and engineers have come up with some innovative ways to reduce weight. WD-40, was originally developed to displace water, which is where its name comes from, to protect the metal tanks of the Atlas rockets from rusting, because they weren’t painted to save weight.  And those Atlas rockets were made of stainless steel. In those days aluminium alloying material science hadn’t quite developed far enough, and the engineers of the Atlas rockets instead opted to use extremely thin stainless steel for their propellant tanks, varying from 2.5 millimetres to about 10 millimetres. These were essentially metal balloons.
As they were structurally unstable when unpressurised. In one infamous case on May 11th 1963, an Atlas Agena D lost pressurisation on the launch pad, allowing the weight of the upper stage to buckle the thin steel. Pressurisation adds strength to pressure vessels as the pressure provides a restoring force for small deformations, so if the metal attempts to bend inwards the internal pressure pushes it back out. This strengthens all rocket tanks allowing their thickness to be minimised, but this application took it to the extreme to make up for steels density. Our choice of material for aviation and aerospace applications has evolved with our mastery of material science. Specifically with the materials available to us that have the highest strength to weight ratios.
We can visualise these strength to weight ratios on graph like this. Plotting the strength of the material against its density. Looking at this it’s pretty clear that steel adds a significant amount of weight, while not adding a proportional amount of strength. Steel is typically 2.5 times heavier than aluminium, but it is not 2.5 times stronger. So why use stainless steel? Well, strength to weight ratios are not the only factor engineers have to consider. Something you may not consider are things like thermal conductivity. Aluminium has a much higher thermal conductivity than steel, and thus can conduct heat from its surroundings into the cryogenic fuel much faster. This can vaporise the fuel, which requires boil-off valves to vent the vaporised fuel. To min
imise this problem, rocket fuel tanks are often sprayed with foam insulation, that’s what gave the external tank of the space shuttle it’s distinctive orange colour, but this adds a substantial amount of mass itself, which in turn decreases the weight saving benefits aluminium provides. 
However, the Falcon-9 fuel tanks are not insulated. To prevent major boil off of the fuel, the fuel is loaded as late as possible. This reduces the amount of fuel that will be vaporised, but also makes the job of getting the Falcon 9 certified for human payloads a bit of a nightmare. NASA did not want Space X to fuel the rocket with passengers on board, because as we saw earlier things can go wrong during this phase. In August 2018, they finally approved the Falcon 9 for this “load and go” style of fueling for human flight.  The aluminium-lithium alloys used in the Falcon-9 were not developed until the late 50s and early 60s, which increased their strength to weight ratios, allowing the introduction to aerospace applications. 
The stainless steel balloon tanks of the Atlas rockets were eventually made with this aluminium alloy metal, and their strength to weight ratio were boosted by using a unique stringer pattern called an isogrid, which boosted the aluminiums ability to resist buckling, like that of the Atlas Agena D. NASA performed these huge compressive buckling tests on the aluminium lithium tanks of the SLS rocket. Typically you use little strain gauges, whos electrical resistance change as you stretch them forcing the electrons along a longer path to keep track of the strain in the material, but for something this big they would have needed thousands. Instead they painted dots all over the structure to allow computer imaging software to keep track of the strain. That isogrid structure is excellent for maximising strength while minimising the material needed. It is essentially an inter woven pattern of I beams that increase the stiffness of the overall structure. You will see this pattern everywhere in aerospace. From these sixties era rockets to Space X’s new dragon 2 capsule. Space X, to date, has used aluminium-lithium alloys in their propellant tanks. But they opted not to use this isogrid structure, even though it provides fantastic strength to weight performance, it is absurdly expensive to manufacture. To manufacture isogrids you start off with a thicker piece of aluminium and machine it down using a CNC machine. This results in about 95% of the material going to waste. Instead Spacex opted for a thin skin of aluminium-lithium alloy and then stir welded strengthening stringers in place. We are constantly balancing a huge number of factors. Here the cost of manufacturing the rocket influenced it’s design. Typically the cost of launching an extra kilogram of material to space far outweighs the cost of material, but in cases like this the waste in the manufacturing process can influence our material choice. For example Musk attributed the cost of carbon fibre composites as one of the primary reasons he abandoned it as a material for the Starhopper. Carbon fibre composites cost about 135 dollars per kilogram, and a significant amount of it is thrown away in the lay-up process. The manufacturing process for carbon fibre composites is extraordinarily expensive and difficult. As explained in my carbon fibre video. Carbon fibre composites gain all of their strength from the long and thin carbon fibres inside the plastic resin that holds them together. This means that their strength is not the same in all directions, and in order to ensure the material can be strong in all directions you have to layer your carbon fibre composite in a very specific way. You then have to cure it in a pressurised oven. This was one of the major flaws I pointed out in predicting the failure of the early prototypes of the BFR carbon composite tanks, which were made in two parts presumably because they couldn’t find tooling and an autoclave big enough to cure a full sized tank. Being perfectly honest this is the only subject area where I have enough expertise to make comments on other peoples designs, and I was surprised Space X were pursuing the material at all, for the reasons stated above, and as it’s unsuitable for a vehicle that not only has to withstand the freezing temperatures from the cryogenic fuel on assent, but the scorching temperatures of re-entry. Not once, but twice. As this will be the first vehicle in history expected to visit Mars AND return. Here we really start to see where stainless steel shines, and why Musk is opting for a stainless steel vehicle. Let’s plot another graph, this time plotting strength against maximum operating temperature. Here we can see that stainless steel outperforms both aluminium alloys and carbon fibre composites by a significant margin.  The Falcon 9 first stage rocket serves only to boost the second stage to about 65 to 75 km in altitude and between 6,000 to 8,300 km/h, before flipping over and performing re-entry burns to slow down before entering the thicker atmosphere at relatively slow speeds. Even then, the engine nozzles, which are designed to tolerate massive temperatures take the brunt of the re-entry heating, allowing the aluminium tanks to avoid any major reentry heat. This is not how the Starhopper is intended to work, because it is being built as an interplanetary vehicle. The starhopper can expect to enter into the Martian atmosphere at speeds of up to 21,000 km/h and experience temperatures up to 1,700 degrees. Well above the maximum service temperature of both aluminium and stainless steel, but we have ways of leaching some of that heat away before it can heat the metal. The curiosity rover utilised a phenolic impregnated carbon ablator, which is extremely extremely light, has a low thermal conductivity, and can resist extreme temperatures of up to 1,930 degrees.  But nothing this heavy has ever entered the Martian atmosphere before, and it’s not going to be any easy task for it to slow it down. It’s going to have to enter the martian atmosphere at an extremely high angle of attack to allow the thin martian atmosphere to sap away speed through drag for an extended period, but drag comes with heat. Stainless steel may be heavy, but it will require significantly less heat shielding that an aluminium or carbon fibre composites. Once again closing that weight advantage gap of these alternate materials . In fact Musk has stated that the rear side of the Starhopper will require no heat shielding at all, and he plans to use a strange technique to cool the wind facing side of the vehicle. Using the same method humans use to cool down, by sweating. Musk plans to pump liquid methane between two steel panels on the windward facing side of the Space X rocket, where it will gain heat, vaporise and evaporate through small holes in the rockets surface. This is pretty weird way of cooling a ship, and I wondered why you would not just opt to use the tried and true method of ablative tiles. Then I remembered that this ship needs to make a return journey, and the entry into the Martian atmosphere will damage the tiles and require maintenance. There is no oil on mars to manufacture new phenolic resin or the carbon needed for ablatives. So, using methane, the fuel the new Raptor engines that Space X will use for the Starhopper, makes a lot sense. It reduces the equipment the rocket will need to carry to Mars, making the rocket significantly lighter. They can just use the equipment they already needed for refueling, making it double purpose. They just need to mine water and extract carbon dioxide from the atmosphere, and then do some fancy chemistry to produce methane and oxygen. The prototype they are building at the moment is likely just to test the manufacturing techniques needed to build it, and test it’s flight capabilities. This ship does not need to be space worthy, it just needs to have the same weight, centre of gravity and shape to allow space x to test it. On the surface though the whole operation looks like a bit of a shitshow, and I really try to be positive about engineering advancements, but the thing literally fell over in the wind last week. I’m really curious on how this whole thing is going to unfold. Sometimes you just need to make mistakes to learn, which is why you should sign up to Brilliant. Brilliant recently introduced a new feature, called “Daily Problems”, which will present with you with interesting scientific problems to test your brain. Like this one that will teach you how solar sails allow spacecraft to gain speed without rocket fuel. If you answer a question wrong like I did here, it teaches you exactly why. Allowing you to learn from your mistakes. Brilliant even have an app that you can download to play these brain teasers on your morning commute. Each Daily Problem provides you with the context and framework that you need to tackle it, so that you learn the concepts by applying them. If you like the problem and want to learn more, there’s a course quiz that explores the same concept in greater detail. If you are confused and need more guidance, there’s a community of thousands of learners discussing the problems and writing solutions. Daily problems are thought provoking challenges that will lead you from curiosity to mastery one day at a time. So what are you waiting for? Go to brilliant.org/realengineering/ and finish your day a little smarter. And the first 500 of you to do so will get 20% off the annual subscription to view all problems in the archives. As always thanks for watching and thank you to all my patreon supporters. If you would like to see more from me the links to my instagram, twitter, discord server and subreddit are below.