Comments by "ke6gwf - Ben Blackburn" (@ke6gwf) on "" video.

Some observations from someone who enjoys working with metal. The base section is completely structural. It is using welded heavy wall stainless steel pipe for a bulletproof internal support frame and landing legs, heavy enough to withstand "slight miscalculations" on the suicide burns and punch holes in the concrete if needed. This looks like it was designed by an engineer familiar with building bridges and roller coasters and other such durable structures, and I see signs of very good workmanship in the welds and little details like the contoured gusset plates and such. It really looks like it's designed to support a bridge rather than fly. But considering that Elon probably wants something close to actual weight, overbuilding this is not a problem, and makes it more likely to survive heavy grasshopper testing without needing repairs. (landing pad is more likely to need repairs lol) The shell of the base section is heavy gauge SS as well, and was all bent in large probably CNC plate rolls making perfect curves that fit and weld together. And it is heavy enough not to distort from the welding. I assume that this will all be polished up to a mirror finish, and was built by a structural engineering team. The upper sections are built totally different, and I suspect by a different team or company. They seem to be an open framework with thin sheet metal panels fastened over it. It makes sense that it is a water tank company, because they would have experience doing such things. Now the difference is that they normally aren't trying to do compound shapes with thin sheets on water tanks, and it is nearly impossible to do it with flat sheets without lots of wrinkles and divots, just like of you put a sheet of paper over a basketball. They may have some trick to use heat to stretch the sheet metal and smooth it out after the structure is finished, it's pretty much black magic, but it is used by good sheet metal workers, so I say wait a little while and see what they are able to do. I also will not be surprised if the company that bid to do it had never done anything quite like this before, and are finding that they are fighting the metal, and it is winning lol So this may turn out to be a total failure on their part of being able to get a good looking result. Elon may be fuming about how ugly it is looking as we speak. On the other hand, as long as it is all attached properly, it is totally flyable as a grasshopper, and once the polishing is done it will be hard to see the surface shape. So it is possible that Elon decided he wanted something of the right size and shape and shiny to be in the air asap, and so is using his preferred fail fast prototyping method to knock together something that can hop, even if the appearance is wanting. If that is his goal, this is the way to do it. If you have seen the videos of that NASA guy who worked with SpaceX, and helped develop the Pica heatshield, (can't remember his name, but it was a series of short interviews), you are familiar with the culture there of being able to order stuff up from McMaster Carr, and put a test prototype together the next day to see what happens, and this seems to fit that profile. Though they probably didn't get the metal from McMaster lol So whether Elon is angry at how it looks, or this was the expectation, it is the fastest and cheapest way to get a Starship prototype hopping, and the appearance really doesn't matter for the testing, since it won't be going fast enough to have any effect. And using stainless steel is just because he wants it to look shiny like the final version lol As far as tanks, he doesn't need large or super light tanks for this first prototype. For initial testing, he can probably just use umbilicals, or small tanks. In order to make this shape with stainless steel, you basically need to use the same method that car body panels use, and hydroform or press them, so each panel is already shaped properly for the location it needs to go. This however requires extensive time and work making the dies for each section, since the sheets change shape and size as it goes up, so this is a big investment, and I can totally see him saying to just wrap it in tin foil to get it flying. They very likely are currently working on building dies and such, and may replace the upper section as soon as they are done and some hydroformed panels are available. So let's see what they are able to do to smooth out the tin, and remember that as long as it is being moved around and welded on, any straightening they do will just get messed up again, so they may be waiting until it is stacked before smoothing it. Oh, and if they were building this for anything other than an actual test prototype, they would NOT be making the top to look like a steam punk Burning Man art installation. The very fact that it is ugly and utilitarian, for something that Elon is doing, tells me that it is for hard science quickly.
908
Oh, and some types of stainless steel sheet are harder to work with than others, and I suspect that they are using a grade or hardness that is more difficult to get wrinkle free. For instance the landing leg gore with just a flat piece of sheet on it, it's all distorted, and that should be a simple flat surface, so I think this probably a hardened, crackly type that would be a bear to work with, and probably was chosen because it was the same type planned for the final version, or maybe because it polishes up nice. Shiny!
53
@Volpenowski this is a rough prototype for engine and computer testing doing hopping, so yes, the are stick welding stainless steel sheet metal over a steel frame, but it is perfectly serviceable for what it is designed for. And stainless steel over aluminum ribs like a final version would be, isn't much heavier than with an aluminum skin (such as an airplane) , because the SS can be much thinner than aluminum. And since the SS can give them thermal properties that aluminum or composite can't, replacing a thick and heavy heat shield for reentry, it gives them a net benefit, even if at a slight weight penalty.
4
@aaronpeters4394 McMaster is liked for steel by most machinists and such, but I meant because of the volume of metal they were using, so it would be by the truckload from a steel supplier lol
4
@rolsen1304 that's basically what they did on the base section, though they used panels to make shipping and assembly easier, but it also looks like they are using at least 1/4 inch thick steel plate on the bottom. It is totally possible to make giant steel tubes like that with giant industrial plate rolls, but it only works in steel thick enough to hold its shape, and the upper sections are thin sheet metal, more like tin foil at that scale. You know how you used to build model planes with a balsa wood frame and cloth or plastic stretched over it? That's basically the construction of the upper section, using what looked like a 1" square tubing frame with the sheet metal stretched over it. This is much much lighter than any plate that you could bend. The finished version would use construction like an airplane, with probably aluminum ribs and preformed pressed sheets of thin and super strong stainless steel riveted or tig welded onto it, but that takes a major investment in tooling and construction, and isn't needed for a hopper.
3
@tomhardy604 if it were a simple cone, rolling would be no problem, but it looks like it goes from a cylinder to a cone to a sphere, and so the transition areas at least would be compound curves. Even then, getting precision panel size and shape would be difficult without some sort of punch press die process it would seem to me, because if you tried to laser cut it first, then roll it, you would get distortion and uneven edges from the rolling process, and then have to find some way to trim the edges to within spec, and since this is a BIG rocket, you might be talking about a few hundred panels per unit, and so I think since Elon likes automation, and because I suspect that the cost of making dies so that each panel is created in basically a single instant step is offset by avoiding the hand fitting that other methods would require, that it would be the natural choice, at least once he thinks he has the basic shape decided on. And he may find another way to shape panels even better, like maybe hydroforming or drawing it in one giant piece, or maybe 2 clamshells or something. I am sure that he would love to find a way to make it seamless like the renderings lol One way or another I am sure that he is going to figure out how to make it big, make it fast use robots, and most importantly, make it perfect and shiny! Because, Elon
3
@jez6208 lol, you should see me when I am really into something, and have data to work with!
2
@camielkotte I think I remember hearing that they were hoping to get these down to 10 percent throttle, but whether they will be reliable that low is unknown yet. In any case, I had not thought about it that way before, but you may be right, they may actually be building this to be heavier than the expected maximum final weight, just so they can keep the engines at higher thrust levels for testing certain parameters of balance and vectoring and such with multiple engines.. On the other hand, a few tons of extra steel probably weighs less than all the life support systems and such that isn't on here, and it's going to be balanced on the thrust, so center of mass vs center of aerodynamic force probably doesn't matter much lol I mean, it's going to be spending half its life flying backwards!
2
@mal2ksc oh, I wouldn't be surprised if some small quantity metal needs still came from McMaster, but since they are dealing with several truckloads of specialty steel in large sizes, I am sure it would be cheaper and quicker to just call up the steel supplier and have them send a truckload direct. That's what I was talking about, the quantity, since I doubt that McMaster has that much in stock lol Also, this is being built in Texas.
2
@tomhardy604 yes, the current rough prototype has a stick built square steel tubing ladder frame, so it may have a stepped cone, but the renders of the finished product seems to be a compound curve to my eyes. And it looks like the nose should be the same material. It is the side of the capsule that will get the high heat after all.
2
@gerardo garnica I did! It's published by YouTube and Scott Manley! ;)
2
@spacedout5692 you know that what they mean by "they lost the technology" is that they designed amazing things and kept all the blueprints, but because it was a bunch of young guys operating in skunk works/startup mode at the time, the team actually building the engines and such were having to solve problems on the fly, and having to develop new techniques to build things, and making tweaks to the blueprint design to get it to work right, and since it was the same team that continued building them, no one thought to document and preserve the techniques and specialized skills they had developed, and so when those guys retired and died, the knowledge of how to make it work reliably went with them. Teams since have built engines based on the blueprints, and found that they differed slightly from the remaining engines they compared them to, and without knowing the exact techniques used to build the pieces, they were not able to exactly recreate it, and decided that the amount of time and money to try all the different techniques until they figured out the right one, and how to apply it, would be more expensive than designing a new engine. So it isn't so much that the technology was lost, it is that they relied on human craftsmanship rather than technology, and the craftsmanship wasn't passed down, as the programs were canceled and new ways of doing things were developed. It's just like we don't know how the Egyptians built the pyramids, but we could achieve the same results now, just using different technologies.
1
@avid0g I had not looked at the pictures for a couple of days, but I just went back to David's and Cowboy's Twitter feeds, and Shiny! It had looked like they may have tried polishing the bottom section, but may have decided it was easier to just wrap it, because that heavy mill finish plate takes a LOT of work to polish lol Anyway, thanks for the heads up.
1
@tomhardy604 , is don't see getting those compound curves through rolling, unless you are talking about hand rolling with an English wheel or something, they need a higher degree of precision than rolling could give. Pressing would be the simplest way, and cryoforming would probably be the finish step after welding to get the smooth shape, but there are many ways to do that. And stir welding would make sense, but would require extreme precision in sheet fitment.
1
@TheBustyOnion lmao, I just reread my comment, and man, he moved faster than even I was expecting! I was right about not worrying about the "nose cone" since it was scrapped not long afterwards and was for decoration only. Watching the iteration through the various methods and designs really shows the way they work, and how if you don't like something, wait a few weeks.
1