如果从明天起,所有从地球飞往火星的飞船都停飞了,而且没人知道什么时候能恢复,火星上那座城市还能活下去吗?
这是马斯克在 2020 年火星学会(Mars Society)大会对谈里反复回到的一个问题。他管它叫试金石。
如果答案是「活不下去」,那就不算数。
这个标准一下子砍掉了大多数听起来宏伟的计划。派几个人上去、建一个前哨站、插几面旗子、拍一轮照片,统统不达标。他要的是一座自给自足的城市(self-sustaining city),一座在跟地球彻底断联之后仍然能运转下去的城市。
他把话讲得很直白:
如果飞船因为任何原因停止到来,火星上的城市会不会死掉?如果会死,那我们就还没到安全的位置。(马斯克)
「任何原因」包括乏味的官僚扯皮,也包括核战争。原因无所谓,结果作数。
他把火星城市卡在第三次世界大战之前
马斯克提了一个词,大过滤器(Great Filter)。
宇宙看起来应该有大量文明,可我们什么都没听见。费米悖论给出的解释之一是,每个文明都会在某个发展阶段被一道筛子拦下来,绝大多数过不去。这道筛子就是大过滤器。
按马斯克的逻辑,人类要么在筛子落下来之前把文明复制到第二颗行星上,要么就和其他沉默的文明一样,成为统计数字。
问题在于时间窗口。
他说,自给自足的火星城市如果建在第三次世界大战之后,概率很低。他不希望发生世界大战,但他把话说得很清楚:所以得赶在任何可能的第三次世界大战之前,把这件事做完。
然后他补了一句,语气不像在演讲,像在自言自语:
坦白说,文明现在看起来不算强壮,有点摇摇欲坠。(马斯克)
这句话放在 2020 年说,分量很重。他没有展开,但意思很明确:赌局无关「人类会不会永远存在」,赌局是「未来几十年会不会出岔子」。彗星撞击、超级火山、某种自我熄灭,烈度可能不是一声巨响,更像一个漫长的冬天。
他还说了一句容易被忽略的话:这座自给自足的城市,大概率不会在他有生之年建成。
这句话很关键。一个创始人公开承认自己看不到终点,说明他要谈的是一项跨越几代人的工程,产品发布会那套愿景话术在这里用不上。
救生艇得能开,停在那儿的不算
很多人把火星计划理解成「给人类买一份保险」。马斯克对这个比喻做了修正。
如果你只是搬到一个你很快也会死掉的地方,那不叫救生艇,那叫换个地方等死。
真正的救生艇,得能自己浮起来。
所以他谈火星,谈的其实是两个东西叠加:把整个文明的生存风险压到最低,同时给未来提供一个值得期待的方向。
他的原话大意是,一个太空文明、一个多行星物种的未来,比一个被永久困在地球上等待某次灭绝事件的未来,要令人兴奋得多。后者让人沮丧,而且不体面。
他还说了一句特别朴素的话:人需要一些让你早上愿意起床的事情。
这句话听着像鸡汤,放在火箭工程师嘴里就不一样了。他是认真的。他相信「成为太空文明」这件事本身,是一种集体动力源。
猎鹰 1 号:账上现金为零的那个下午
对谈里回顾了早期历史,这段值得单独讲。
猎鹰 1 号(Falcon 1)连续失败了三次。第一次、第二次、第三次,都没进轨道。
第四次成功了。马斯克的说法是「第四次是幸运的」。
他紧接着说了一句让全场安静的话:那次发射成功的时候,公司账上的现金基本是零。如果第四次也没成功,公司就完了。
这不是修辞手法。2008 年的 SpaceX 确实只剩最后一次机会,钱只够再射一次。第四次入轨之后,NASA 的合同才进来,公司才活下来。
所以他说,这一路从来不是顺风顺水,是极其艰难的旅程,很多东西都是边走边发现的。
这段放在整场对谈的最前面,作用很明确:后面他要讲的「一次性火箭是笑话」「可重复使用是唯一出路」,全是烧过钱、炸过火箭之后剩下的结论,事后包装不出来。
先回答「目标是什么」,再回答「哪个设计好」
主持人问了一个技术问题:从猎鹰 9 号到猎鹰重型,再到星舰,这条设计路线是怎么走过来的?
马斯克打断了他,说这个问题得倒过来问。
你得先说目标是什么。目标定了,各个设计方案才好拿同一把尺子去量。否则你根本说不清一个设计为什么比另一个好。
他的目标是一句话:尽快往火星运送足够多的吨位和足够多的人,让火星能够自给自足。
注意「尽快」这两个字。它是整套设计的约束条件,写进去之后,后面所有的取舍都被它管着。
然后是反推。
要让火星自给自足,需要多少东西?他给的数量级是十万吨到一百万吨,运到火星表面的净质量。
按当时的估算,把一吨有效载荷送到火星,大概需要在低地球轨道(low Earth orbit,LEO)上有五吨。这个换算率不算漂亮,但按他的说法是比较有把握的,乐观一点四吨也许够。
那么一百万吨到火星,就需要四五百万吨进入低地球轨道。
这个数字一出来,问题就变味了。
全世界的一次性火箭加起来,差了一万倍
马斯克给了一个对照组。
当时全球所有一次性火箭的入轨能力加起来,一年大概五六百吨。
一边是四五百万吨,一边是五六百吨。差了一万倍。
他做过一个极端的假设:假如地球明天就要毁灭,前提是你必须提高入轨能力,全世界拼了命去干,也许能把现有能力提高一千倍。
一千倍之后,还差一个数量级。
他的结论毫不客气:一次性火箭这条路,从数学上就走不通。问题出在量级上,效率高低还在其次。
火箭在他眼里就是交通工具,交通工具的一次性化是荒谬的。
他举了三个例子。你要卖一架只能用一次的飞机,人会把你笑出房间。你要卖一辆只能用一次的汽车,一样。你要卖一匹只能用一次的马,人会怀疑你精神状态有问题。
可偏偏有人心安理得地卖一次性火箭。
所以他反复讲一个判断:完全、快速的可重复使用,是入场券。没有这张票,后面全免谈。
可重复使用的账:4% 变成 2%
可重复使用听着简单,做起来是另一回事。
亚轨道级别的可重复使用火箭,难度不大。轨道级才是真正的门槛。
他给了一组参照数字。很多聪明人投入了大量精力,能做到把起飞质量的百分之二三送进轨道。真正顶尖的火箭能到百分之四。他不确定有没有人真的做到过百分之四。
算盘在这儿:如果你要做可重复使用,你的火箭在一次性形态下大概得具备百分之四的入轨能力。
因为可重复使用本身要吃掉一半。回收、隔热、着陆腿、剩余推进剂、结构加强,全是死重。
吃掉一半,净剩下百分之二。
这就是为什么小火箭做不了可重复使用。你连百分之四都没有,扣完就成负数了。
猎鹰 1 号当年也试过回收。第一级装了降落伞。
结果第一级再入大气层的时候,像一头撞上混凝土墙。降落伞根本来不及起作用,箭体直接解体。
马斯克说他当时挺生降落伞供应商的气,后来才意识到怪不得人家,是自己太天真。大气层在高马赫数下的阻力和热量,完全超出想象。
这个教训后来变成了猎鹰 9 号(Falcon 9)的设计前提:必须想办法让箭体柔和地进入大气层,得有气动控制面,得有点姿态控制,得有反推减速。
火箭越大越好,因为「大脑」不会跟着变大
这一段是整场对谈里技术含量最高、也最少人讲的部分。
马斯克说,规模在这件事上是有硬优势的。
第一条,航电占比。火箭小到一定程度,你送上去的基本上就是自己的电子设备。一枚一万公斤级的火箭,有效载荷可能接近于零,全被航电吃掉了。
火箭做大之后,箭体变大了,但「大脑」不用变大。同一套飞控、同一套通信、同一套导航,装在大火箭上,占总重的比例趋近于零。
第二条,他用了「gauge」这个词,指的是材料厚度。
东西做得太小,厚度就不好控制。铸造有最小壁厚,板材有最小规格,误差棒是绝对值,不会因为你做小就同比缩小。一枚小火箭的蒙皮厚度误差可能占总厚的很大比例,一枚大火箭就没这个问题。
这条听上去枯燥,但它是航天成本结构的真实来源之一。很多人以为大火箭贵,其实单位载荷的成本反而是往下走的。
他顺手举了两个类比。运货要用半挂卡车,不是一堆小货车。跨洋运输要用集装箱船,不是成千上万条带舷外机的小艇。
道理一样。规模本身就是效率。
猎鹰 9 号已经接近它的天花板
马斯克对猎鹰 9 号的评价很有意思。他没有吹,反而说这型火箭在自己的架构里已经接近局部最优了。
什么意思?
燃气发生器循环、煤油加液氧、3.6 米直径。这三样东西一旦定下来,优化空间就那么大。
直径为什么是 3.6 米?因为要能走公路运输。再大,陆运就成了噩梦,桥梁、隧道、转弯半径全是大问题。
可细长的火箭在质量效率上吃亏。箭体越长越细,弯矩越大,蒙皮就得加厚,加厚就是加死重。
然后是燃料。煤油不是对的燃料。
煤油基本上就是航空燃油的精制版本,RP-1 是比 Jet A 更严格的等级。它密度高、便宜、成熟,但比冲上不去。
比冲(specific impulse)可以粗暴理解为单位质量推进剂能产生多少推力。这个数字越高,同样的燃料能把火箭推得越快。
马斯克用最朴素的语言解释了火箭方程:
一枚火箭能飞多远,取决于两件事。气体从尾部喷出去的速度有多快,以及推进剂占总质量的比例有多高。
就这么简单。喷得快,带得多,飞得远。
煤油在这两项上都吃亏。甲烷(CH₄)就不一样。
为什么换成甲烷:喷得更快,还能在火星上现做
甲烷的优势有三层,一层比一层重要。
第一层,比冲更高。同样质量的推进剂,甲烷能喷得更快。
第二层,氧燃比更优。煤油的氧燃质量比大概 2.5 比 1,甲烷大概是 3.5 比 1。
这个比例越高越好,因为氧气密度大、便宜。地球上制氧的成本基本上就是电费。马斯克还开了个玩笑,说地球上到处是整天制氧的工厂,还有浮游生物也在制氧,从不干别的事。
推进剂里氧气占比高,意味着每吨推进剂里便宜的那部分更多,贵的燃料更少,整体成本往下走。
第三层,也是最决定性的一层:甲烷能在火星上制造。
火星大气主要成分是二氧化碳,地下有水冰。二氧化碳加氢,可以合成甲烷和水。这就是所谓的原位资源利用(in-situ resource utilization)。
这一条直接改变了整个任务的架构。返回用的燃料如果必须从地球带过去,那成本会翻好几倍。能在火星本地生产,返程燃料就不用占发射质量。
所以星舰(Starship)从设计之初就锁定液氧甲烷加猛禽发动机(Raptor)。这个选择由任务约束倒推出来,技术偏好排在后面。
星舰的时间表:八成把握,五成回收
主持人问到进度,马斯克没有回避,先声明了一句:
我们没有藏着什么秘密数据,这些都是猜测。(马斯克)
然后给了两组概率。
第一组,星舰明年(也就是 2021 年)入轨,他有八成把握。
第二组,飞船和助推器都能完整回收,他有五到六成把握。他自己用了「dicey」这个词,意思是悬。
他预计会丢几艘飞船,才能把大气层再入和着陆这两件事摸透。助推器他希望能不丢,因为那上面发动机太多,掉一架就肉疼。
早期助推器试飞不会带 28 台发动机。第一次上去的,可能就两到四台。
高频次飞行,他估计要到 2022 年,也就是两年之后。
他紧接着说的一句话更值得记住:他正在努力让创新的速率往上走,而不是往下掉。
他加了一个判断:如果创新速率不能接近指数级增长,火星就到不了。纯线性的进度,永远做不完。
然后是一句自嘲:如果只是线性推进,那他自己都活不到那一天。
这话不好笑,但很真实。按线性速度,一项跨代工程需要的不是一代人的寿命。
四年后,赶第二个火星窗口
载人之前要送无人飞船。马斯克给的时间是四年左右。
原因是物理限制。火星和地球的窗口期每 26 个月一次。2020 年有一个,两年后一个,四年后一个。
他说有机会赶上第二个窗口,用的词是 fighting chance,意思是有一拼。
这个时间表和前面那两组成败概率是配套的。入轨八成,回收五成,无人任务四年后。三个数字串起来,能看出他心里的真实节奏:算不上稳,但确实有戏。
目标定错了,创新就不够猛
对谈里有一段话,我觉得是全场最重要的。
马斯克说,如果目标只是把几颗卫星送进轨道,那这件事根本不需要激进创新。你会用一条对数曲线去推进,一年十次、十二次发射,对手一年六次,慢慢磨,也就赢了。
可如果目标是让生命成为多行星的,而且时间紧迫,那激进创新就成了唯一路径。
他用了「强制函数」(forcing function)这个词。一个足够难的目标,会逼着组织去做出本来不会做的创新。
他还顺便吐槽了传统航天巨头。
洛马和波音其实没在真搞可重复使用。他觉得这很奇怪,因为这两家公司造的飞机本来就是可重复使用的。
他的类比很损:你要是跑去跟客户说,这架 737 只能用一次,人家会觉得你有病。可卖一次性火箭,他们觉得天经地义。
他还顺手补了一句玩笑,说 737 MAX 后来还真有点像「一次性使用」的飞机,指的是那两起事故之后的停飞。
这个玩笑不厚道,但它把双标点破了。
瞄准月亮和瞄准火星,是两种公司
马斯克说,如果 SpaceX 的目标只是「在常规卫星发射市场上做到第一」,那它跟对手不会有本质差距。大家比的是谁多发几发、谁价格低一点。
但目标是火星,路上这些竞争就变成了小事情。
他补了一句:除非有别人也在瞄准火星,否则他们在那件「发几颗卫星」的小事上,不可能有竞争力。
这句话的逻辑可以掰开看看。它讲的与市场份额无关,讲的是目标函数的差异。当你的目标高出对手一个量级,你在低量级战场上反而会形成碾压,因为你被迫练出来的能力,对手根本没有。
可重复使用、轨道加注、规模化制造、快速迭代,这些能力是为火星练的,但顺手就把近地轨道市场拿下了。
除了完全快速可重复使用,他还点了两个必须完成的环节。
一个是轨道加注(orbital refilling)。飞船装满货之后没多少余量飞到火星,得在近地轨道上多次加注推进剂,才能凑够逃逸能量。这一步没有替代方案。
另一个是火星就地生产推进剂。前面讲甲烷那一段已经说了,这是架构层面的必需。
火星城市成立的条件:两个集合的交集达到一百万人
主持人最后问,火星学会这样的组织能帮上什么忙。
马斯克的回答很数学。他说需要两个集合的交集。
第一个集合,是那些真的想去火星的人。
第二个集合,是那些能负担得起去火星的人。负担的方式可以是自己出钱,可以是找到赞助,可以是政府支持,可以是贷款,方式不重要。
真正稀缺的,是同时落在两个集合里的人。
他给了一个数量:当这个交集达到大约一百万人,火星城市就会出现。
一百万人,不是一百万个座位,是一百万个既有意愿又有支付能力的个体。这个标准把问题从火箭工程转到了社会经济。
造出便宜的船,只是把门槛的价格降下来。真正决定城市能不能建成的,是有多少人愿意掏钱、并且掏得起。
这也解释了他在整场对谈里为什么反复强调成本和可重复使用。船票便宜一个数量级,交集就大一个数量级。
这期对谈里,马斯克其实在讲一套方法论
把整场内容拉通看,技术细节只是表层。他真正在演示的是一套顺序。
先定义终点状态。终点必须可验证,所以他把「自给自足」具体化成一个可执行的测试:断联之后会不会死。
再从终点反推数量级。一百万吨到火星,五百万吨到近地轨道,全球现有能力五六百吨。数字一摆出来,方案的对错就不需要争论了。
然后让数量级去筛方案。一次性火箭筛掉,小火箭筛掉,煤油筛掉,没有轨道加注的架构筛掉。剩下的那几条路,无论多难都得走。
最后,用这个目标去逼组织保持高创新速率。他自己承认,如果目标低一点,公司会活得很舒服,但也就到不了火星。
这套顺序的价值在于,它能反过来用在很多地方。你先想清楚三年后必须成立的那个状态是什么,再问现在的能力差多少个数量级,答案往往会让人不太舒服,但至少方向是清楚的。
可以立刻用起来的几件事
- 把你的目标改写成一句可验证的话。别写「提升效率」,写「如果团队全部休假两周,业务会不会停」。马斯克用「断联之后火星会不会死」来定义自给自足,你也可以给自己的项目找一个这样的试金石。
- 做一次数量级核算。把目标拆成终端需要的量,再算出现在的能力,两者相除。如果差三个数量级以上,说明路径本身要换,优化现有流程没有意义。
- 检查你手上有没有「一次性火箭」。那些只用一次就丢掉的流程、工具、组织结构,往往被当成理所当然。列出三样,问一句:飞机汽车都不是一次性的,这个为什么是?
- 给团队设一个明显超出当前能力的截止时间,然后公开承认它是猜的。马斯克给的是八成把握入轨、五成把握回收,并且明说没有秘密数据。这种坦率反而让团队敢下注。
- 算清楚你的两个集合。想做这件事的人,和能为此付钱的人,交集有多大。如果交集太小,先降低门槛,别急着扩大宣传。
- 把「规模」当成一个设计变量。系统做大之后,固定成本占比会下降,精度会变好,容错空间会变宽。很多小团队失败的原因在于做得太小,执行层面反而挑不出毛病。
0:13so let's see well on the slasher point we've gone through many iterations starting from not really knowing how to build rockets at all with the falcon one and having core failures actually in reaching orbit through yeah sorry three failures and then the fourth one got to open so fourth one is a charm so really barely survived i was at zero cash basically when we got that fourth one to orbit and if that fourth one hadn't worked we would have been curtains so it's definitely not been smooth sailing it's been a very difficult ride
0:52with just a lot that has been discovered along the way i'm just just trying to figure out what questions to ask about the design was was quite difficult i think it's helpful to have as the objective the creation of a self-sustaining city on mars i think this is this has to be the objective not simply a few people or a base but a self-sustaining city
1:22the acid test really is if these if the ships from earth stop coming for any reason does mars die out for any reason it could be from but it could be banal or it could be nucleon again it doesn't matter if the if the ship stop coming for any reason does the city on mars die out if it does we have not we're not in a secure place so i mean i think this really might come down to you know on the great filter front is this are we going to create a self-sustaining city on loss before or after world war three i think the probability of it being created after world war iii hopefully that hopefully there's never a world war three but of after is low so we should try to create let's make
2:14this city self-sustaining before any possible level three this is just a risk this is not you know i think sometimes people have difficulty dealing with with probability to see that this way or that way but it's really we just face a series of probabilities and there's some chance that we will have a giant war or through a volcano or you know a comet might hit the earth or we might just self-extinguish in some it might be more of a more of a winter than a bang yeah and frankly right now civilization's not looking super strong you know this is looking a little rickety right now frank i mean less modest mars is made self-sustaining which will probably not happen in my lifetime it is
3:02certainly not it's mean it's meaningless to have an escape you know lifeboat or or or escape hatch or something if you will you're simply moving to another place where you will soon die out that doesn't count so much of a lifeboat really so this is really about to minimizing existential risk for civilization as a whole and and then having an exciting future that you can look forward to and a future where we are a space-barring civilization and multi-planet species is far more exciting than one where we are not i mean that's an exciting future and being forever confined to earth until some eventual extinction event is depressing and not gone and we need things that make you want
3:56to get out of bed in the morning and be excited about the future and i think being a space faring civilization is one of those things that everyone can get excited about but can you maybe just kind of leave people on the path that led you from the design from falcon 9 to the falcon heavy but now starship is is rather different than falcon heavy yeah the engineering absolutely absolutely i didn't quite answer your original question you first have to say what what is the goal and once you have what is the goal you can then measure the various designs against that goal if otherwise you're saying how are you evaluating why is one design better than another
4:39what's your goal it's gotta be a goal so the goal is get enough tonnage to mars to and enough people to make mars self-sustaining as quickly as possible so then you say okay let's back back out the math on this we're gonna need we're gonna need a lot of tonnage maybe i don't know a hundred thousand tons maybe a million tons so you can't be faffing around with these expandable rockets they're a joke they're absurd even saturn five is tiny potatoes we we need it because if you want to get like let's say first order approximation a million tons the source of mars inclusive of people you know that that means probably something around four or five million useful tons of
5:29payload in the into a low earth orbit you know for every time you get to live with overwatch you're gonna get four or five tons hopefully closer to five it's it's you know this math is it's you really saw squeezing like tiny percentages but let's say it comes confidently if you got five tons to low earth orbit you can get one ton to mars that's that's confident maybe you can get i mean maybe you only need four anyway the point is you need five million tons into earth orbit to get one million tons to mars now let's put this into perspective total global capacity to orbit of all expandable rockets is around around five for 600 tons i think and if you said okay the world's gonna
6:18end if you do not increase your capacity perhaps they could do a thousand times okay so that's one five five thousand one five thousandth of what's needed this is ridiculous you know it's it's not even you know point one percent would be you know one thousandth so way less than point one percent we have way less than point point one percent of the capability needed to create a if everyone went full tilt with expandable rockets external rockets are the absolute are just utterly stupid in my opinion utterly stupid they're a complete waste of time people should stop wasting their time if you try to sell an expendable plane people would laugh you out of the room
7:12if you're trying to sell an expendable car they would laugh you out of the room you're trying to access an expandable horse they would like you out of the room and think there's something wrong with you mentally so all these things are reusable it's essential to be reusable now creating a reusable rocket orbital rocket is very difficult but doing a sub-orbital reusable rocket is easy doing a reusable oval rocket is hard even when a lot of smart people have put quite a bit of effort into it they might get two or three percent of the electric mass to lower the orbit and a really epic rocket would get four i'm not sure every i think anyone's ever gotten four
7:54so but you basically need to have something that in expendable form would probably get about four percent of his payload order such that you can spend about half of that four percent on reusability and and still net out to around two percent of payloads so you have to make both the booster and the upper stage and the varying everything reusable with falcon 9 so so with falcon 1 we did actually attempt to do this so we had a parachute in the first stage but really did not appreciate that that first stage was going to hit the atmosphere like a concrete wall so at first i got pretty mad at the parachute supplier until i realized yeah it's not their fault yeah we were just
8:35being cool that that thing was exploding as soon as it hit the atmosphere and you know so you really got to do something to ease the transition into the atmosphere at high mach number it's just very hot and there's a lot of force and a lot of heat so then with falcon 9 we made a bigger rocket and our scale matters here because you do get basically economies of scale you can't have a tiny rocket there's a tiny rocket you basically just end up carrying your electronics to orbit so you're avionics you know so in a little rocket you if you're small enough just your your avionics loan ends up being a significant percentage of your payload and then you
9:27you know if you got a rocket that was i don't know they're trying to get a 10 000 pound rocket for example or even a 10 000 kilogram rocket to orbit i think you would basically get zero pillar now as you get bigger the rocket gets bigger but the brain doesn't get bigger the brain can stay the same size so your avionics for example become almost not almost zero percent of the weight at a per big rocket then for big rockers you also get gauge advantages so this is we're really in the nuances of rocket design and manufacturing here if things are very small it's difficult to get your gauge accurate so but basically how thick is the material like you want to do castings for
10:11example there's a minimum gauge or thickness for a casting there's a minimum kind of error bar on the you know on this the material skins even for a composite rocket you've got you start getting granularity issues as you get bigger you know you're no longer gauge limited and you can get your percentage accuracy on the thickness of walls and castings can be can be very good these are nuances that i think almost
10:45no one appreciates but spice to say that there are advantages to size and you can certainly see this many walks of life where if you've got a truck that's carrying cargo you it's more efficient to have a big semi truck not a bunch of little trucks for ships it would be pretty silly to see container ships or containers going across the ocean one at a time with little outboard motors that would be silly you put them on a container ship you have big ships not little little tiny ships so anyway so size matters it really does and full reusability matters so with with falcon 9 after demands effort we were able to achieve reusability of the booster
11:32and we're mostly achieving reusability at this point with the fairing as well but this is a this is a monumental effort and i think within its architecture falcon 9 is close to a local maximum if you say your gas generator cycle kerosene oxygen vehicle of this particular size with a 12 foot or 3.6 meter diameter
12:09which is which is that size for because of road transport limitations so go bigger than that you can transport it over the road and your logistics class become extreme so but having a long thin rocket is not very mass efficient you're you're not having to have thicker skins to take out the bending moment so and then having
12:36like kerosene is not the right fuel methane is a much better fuel you can get higher isp a specific impulse basically efficiency i mean but for those who i think probably a lot of us are listening know what the rocket equation is but in simple terms it's actually very simple it's like a rocket is going to go further yeah if if the gas if it shoots the gas out of the end faster and if a bigger percentage of its mass is propellant it is obvious so that's what that's what the rocket equation says so shoot gas out faster in the right direction and increase the propellant the percentage profile that's going to get you go to go further with methane you can shoot
13:22it out faster and you can make it on mars you make it on mars for sure exactly so being able to do institute profound development is production is very important so you don't have to carry your return fuel with you or return fuel and action it's like rockets are mostly oxygen or oxidizer so
13:51there's there's some other subtle advantages with a oxygen methane system in that you can go to a higher percentage a higher mass ratio of oxygen so with kerosene you'd be about roughly two and a half to one oxygen to fuel mass ratio with methane you're more like three and a half to one and you actually want that higher mass ratio because oxygen is very dense and it's inexpensive especially on earth so you're gonna you know you have all these plants just making oxygen all day long and plankton just making oxygen never do anything so the cost of oxygen is basically cost of electricity anyway so going from falcon going from you know kerosene which is basically
14:39the same as jet fuel it's like rp1 rock propellant kerosene is just a tighter grade of jet fuel you want to go from that to something which has where the gas shoots out faster and that's methane and where the institute production of propellant is easier so that's that's why they change from from kerosene to methane but i think it's just ch4 it's one carbon four hydrogens and then oxygen pairs pairs together so you have it's called o2 because oxygen pair and obviously you know all this stuff and well it's not like i i we're obviously venturing into unknown territory so it's not as though i i i have all these secret dates and i and i i'm you know just keeping them from people
15:25but so so that mike these are just guesses obviously i'm pretty i'd say i'm 80 tonight i'm confident that we will reach orbit with starship next year i think probably 50 or 60 50 confident that we'll be able to bring the ship and booster back that's that's more of a dicey situation
15:50we'll probably lose a few ships before we really get the atmospheric return and landing right we might lose hopefully we don't you lose it hopefully we don't lose any boosters because that's a lot of engines our initial booster flights will just have maybe two to four engines not twenty-eight twenty-eight engines
16:13so yeah and i think we'll probably be in doing high volume flights i think probably in 2022 took a couple years from now but i'm trying to make sure that that our rate of innovation increases it doesn't decrease this is really essential in fact if we do not see something close to an exponential improvement in our rate of innovation we will not reach mars like a pure linear doesn't get there no no well i'll be down anyway before it gets there if it's purely here it's exponential i think we could get to mars we could probably send an uncrewed mission there in maybe four years
17:06you know there's a moss conjunction every 26 months there's one this year so that means a couple years from now is another one and then four years from now is another one i i think we've got a fighting chance of of making the that second moz transfer window no or focused on like i said it's
17:29important to have that have the objective be right that's why i was talking so much about the importance of making wealth mars a self-sustaining as bullying self-sustaining city on mars if that's the objective then obviously you know just putting some satellites in orbit or there's not that that's not important you have to achieve full and rapid reusability amp size full and rapid reusability is only relevant to the degree as rapid and complete and and then you also have to do orbital refilling this is essential as well and and then propellant production on mars also essential so you know and with that as the goal then
18:20you know that that means that that that creates i think a good forcing function for radical innovation because in the absence of radical innovation we have no chance of meeting that goal whereas if i go simply you know defeat lockheed and boeing or something like that that we would probably achieve
18:44that really wasn't even a thing you know i wish it was yeah like they're not really trying to do they're not even trying to do reusability which is bizarre because they make planes that are reusable so i mean if they if they talk to them like you know if they talk to one of their customers about bias or a lucky fighter jet or a boeing aircraft like hey we're going to sell you a 737 it can be used once and it's not 737 max but that was turns out that was a single user plan at times but it really
19:21it would be an absurd thing for them to sell a single-use aircraft but they feel quite comfortable selling a single-use rocket and anyway but if our goal is simply we're going to have to be the leaders in launching the conventional satellites that exist we would probably approach that in on a sort of a logarithmic basis where you know you get there and you sort of slowly make progress towards doing 10 launches a year 12 launches a year and while they do six or something like that but since the goal is hey we need to make life multi-planetary before it's too late and time really matters so we're it's like we're shooting for mars not just the moon
20:08it's not much suit for the moon shoot for mars and then and then the you know these these competitive things are are kind of small things along the way unless somebody else is shooting for mars they will not be competitive with something as pedestrian as launching a few satellites into earth orbit so how can the mars society help you
20:30well i do think there's in order to for there to be a self-sustaining city on mars there's we're going to need an intersection of sets here one set is the set of people that want to go and can either find sponsorship you think either for themselves or find government sponsorship or take out a loan whatever the case may be and but somehow you've got to have the set of people who want to go to mars and can and can come up with a fund somehow to do that and then then you know it's it's i should say there's two sets desire to go to mars and can afford to go to mars when it's when desire to go to the people who want to go to mars and who could afford to go to mars when
21:18that intersection of sets reaches a million roughly then i think we will have this city on mars
资源与出处
- 视频在 YouTube 收看
- 档案silicon-valley.tech 档案页本页视频与文稿数据来自 Elon Musk Archive
- 精读本篇精读 Markdown 版