Ups and Downs In Brownsville, Texas

If you had a chance to watch SpaceX’s test of a Starship prototype (dubbed SN5) last night, then you know you saw the company making history -- again. During the trial, the rather silo-like body rose from its launch pad, sitting on an exhaust plume. It was supposed to fly as high as 150 meters, then come back, balanced on the plume, and land.

It did that.

One Step of Many, One Stage of Two

There are many YouTube channels with the footage of this feat available. If you haven’t seen it, just choose a channel and watch the event (actually, SpaceX ,,released official footage--it’s much nicer). It didn’t take long. And there was no drama. Ars Technica’s Eric Berger summed up the company’s activities perfectly for the successful Starship test ,,on Twitter last night:

It is the third time SpaceX has flown a prototype for its Starship/Super Heavy program (the first two were completed using a water-towerlike prototype called Starhopper). It took slightly over a year from the Starhopper’s last test to the full-scale prototype test using SN5 last night. While the rocket appears to be doing things SpaceX is renowned for with its Falcon 9/Falcon Heavy rockets (reliable launch and landing), the materials making up the SN5 are quite different. The engine that powered it to 150 meters is different. The scale of the prototype itself is different.

Keep in mind that Starship is supposed to be the “upper stage” of a much larger launch vehicle. Mounted on top of Super Heavy, the entire vehicle will be 120 meters tall--about 10 meters longer than an American football field. Starship itself looks to be about one-third the length of the system (about 50 meters long). When finished, Starship will be just 20 meters shorter than a Falcon 9’s 70-meter length. When Starship and Super Heavy are combined, they will nearly be twice as tall as the Falcon 9. But the launch vehicle will lift over four times as much mass to low Earth orbit (LEO) --100 metric tons.

What SpaceX flew yesterday evening was slightly shorter than Starship’s ultimate length--around 30 meters (about the length of two semi-truck trailers lined up end to end).

The rocket engine that powered and guided SN5 from launch to landing was a Raptor. It is a beast of an engine when it is compared with SpaceX’s Falcon 9 engine of choice, the Merlin. For details as to why the Raptor is so different, please read Everyday Astronaut’s “,,Is SpaceX’s Raptor engine the king of rocket engines?” It’s a long article, with some useful comparisons with other rocket engines--including the Merlin.

The upshot is the Raptor appeared to work during the test. It’s supposed to be relatively inexpensive to manufacture (~$2 million) and reliably reusable for at least 50 launches. That makes the engine economical to operate over time (which are all critical traits considering the combined Starship/Super Heavy will have 37 Raptor engines).

Building Up Production, but Who Will Use It?

While this is only a step, it’s far from the usual raft of vaporware ideas thrown around by other space startups. We don’t see in this test all the hard work SpaceX’s team brings to the production aspect of the rocket. SpaceX’s Chief Engineer, Elon Musk, is focusing his team not just on building Starship prototypes but also on building up Starship production. Berger interviewed Musk earlier in the year, and made this observation:

Yet Musk has not been spending so much of his time in South Texas just to build a Starship. Rather, he’s trying to build a production line for Starships. He wants to build a lot of them. And fast, always fast.

In the same article, Musk notes he wants his factories to manufacture two Starships per week--and that they’d only cost $5 million each to produce. Even if the Starships were double that in manufacturing costs, they’d still be a bargain to use.

Usually, with orbital launch vehicles that can lift large masses to orbit, such as 100 metric tons, there’d be limited use-cases for such a capability. The limited market possibilities are because traditionally, large payloads would cost too much for any person or organization to launch something that large. Applying the “traditional” label to the U.S. Space Launch System (SLS), the costs to launch 130 metric tons (only 30 tons more than Starship/Super Heavy) could be as much as ,,$2 billion per launch.

When I worked at Quilty, we came up with ,,a launch cost estimate for Starship/Super Heavy of $20 million per launch. Such a low projected launch cost means, potentially, that for the price of a single SLS launch, NASA could instead have 100 Starship/Super Heavy launches for its purposes. That’s 10,000 tons NASA could put into orbit with Starship, instead of 130 tons with SLS.

Such a scenario is why it is difficult to say what the market is for Starship/Super Heavy--no other launch system in the world potentially offers such a low price per kilogram--none. Such low pricing is obviously attractive to NASA, and likely the U.S. military--so the government will definitely be a SpaceX customer.

But such a potentially affordable launch cost may attract more customers who would’ve never considered space to help with their businesses/projects. It would be the most affordable way to put up various space-based infrastructures, such as broadband communications satellites and possibly “guardian observer” satellites for general space situational awareness purposes. But I suspect that would be just the tip of the iceberg.

All the speculation above depends on how quickly and earnestly SpaceX moves with Starship. The test the company conducted yesterday makes it appear that SpaceX is moving in the right direction.