Astra’s Rocket Spin

I’ve made a few observations of smallsat rocket manufacturer Astra’s antics during the past few months. Generally, the company appears to be progressing toward the most basic requirement for its business: launching rockets reliably. Its latest successful launch (conducted in March 2022) just so happened to occur without the typical company launch test hallmarks, such as launching a rocket sideways, or using monkeys for installing rocket wiring or randomly choosing fuel mix ratios.

A person could be forgiven for thinking that Astra’s management would continue finessing the most successful version of its rocket, Rocket 3.3. Why mess with the potential for success? And the company appears to be launching a few more Rocket 3.3s. But only for a short while. Then it will be launching a new rocket–4.0–which is part of a new launch system–2.0. Each, and then some, were announced during the company’s “Spacetech Day” earlier this month. The company’s presentation emphasized that the concepts of “cadence, capacity, and cost” drove Astra’s business decisions. The company is also in love with the idea of “scale,” throwing that word in as much as possible.

Are those concepts more than an alliterative mnemonic for the company?

Capacity and Cost

The new rocket is not a new idea to Astra. The founder and CEO, Chris Kemp, had mentioned it (including its higher mass lift capabilities) during an interview with Jeff Foust early last year. Rocket 4.0 will still lift at most 300 kg of mass to low Earth orbit, keeping in line with Kemp’s information provided in the Foust interview. Newly provided data would be the “baseline” price to launch Rocket 4.0 at nearly $4 million per launch. The system is designed to launch Rocket 4.0 about once a week, starting perhaps closer to the end of 2022. The rocket will still be made of aluminum, which Astra emphasizes makes it different from others (the Falcon 9 and Atlas V also use aluminum–Rocket Lab uses carbon composite).

The increased mass lift capability puts Astra’s Rocket 4.0 in direct competition with Rocket Lab’s Electron, a rocket that can also lift 300 kg to LEO. But the launch price for Rocket 4.0 ($3.95 million) is a little over half of the Electron advertised price ($7.5 million). The difference also indicates that the anticipated cost per kilogram will be lower for customers using Rocket 4.0–slightly over $13,000/kg–while the Electron will cost customers at least $25,000/kg. However, when Rocket Lab initially offered Electron, the rocket’s initial launch cost was much closer to Astra’s, at $4.5 million (although it also initially carried less mass). So it may be that Astra will also up the launch price of Rocket 4.0.

Astra’s potential lower cost per kilogram will help against Rocket Lab’s and Virgin Orbit’s rockets. However, it doesn’t get to SpaceX’s Smallsat Rideshare Program’s very low $5,000 per kilogram price. That price is where Astra doubles down on the spin (and why it feels it must do so).

During the presentation, Astra’s management acknowledged the many benefits of using a larger rocket to deploy things in orbit. One of those benefits is the lower cost for smallsat operators to hitch a ride in SpaceX’s Falcon 9. However, Astra wants customers to focus on its lowest “cost per launch” benefit. In other words, a single launch using Astra’s Rocket 4.0 will cost less than a single launch with SpaceX’s Falcon 9. 

[sarcasm] This is such a blindingly obvious comparison that it’s incredible other smallsat rocket companies haven’t attempted to do the same thing.[/sarcasm]

Of course it will cost less! However, recent history demonstrates that low launch cost wasn’t always the case. Orbital Sciences/ATK/Northrup Grumman’s Pegasus smallsat rocket has a payload capacity of around 450 kg. NASA paid over $56 million for the privilege of using Pegasus during the rocket’s last mission (a whopping per kilogram cost of ~$126,000). That launch occurred not that long ago, in October 2019. 

Considering the customer, the market segment, and the proximity in time, perhaps it’s good for Astra to be so obvious in launch cost comparisons. But the comparison still seems born of desperation, a condition that has inspired other rocket companies to talk up the value of what they offer instead of customer-friendly pricing. 

The materials used for manufacturing Astra’s rockets aren’t anything special, either. Although, based on SpaceX’s selection of steel for Starship, which is cheaper than aluminum (and easier to work with), it seems as if Astra might have benefited from being a little less–legacy–in its material selection. It might have been a worthwhile consideration since its rockets don’t have to be near as robust as Starship. However, it’s not clear whether the company even considered alternatives to composites and aluminum. Using steel might have put more truth behind the company’s “cost” and “cadence” commitments.

Cadence

Astra also plays up a much better reason to use a smallsat rocket–the ability to insert a satellite directly into a customer’s desired orbit. But, of course, that doesn’t differentiate Astra’s offerings from those of its competitors. Rocket Lab’s Peter Beck noted the same characteristic during an interview over three years ago:

“It’s control of destination and timeline. That’s the number one thing. “I’m going to this orbit on this day” because that’s one thing that [traditional] ride share can never offer. Because you just don’t know where you are going to go, when you are going to go. So that’s pretty much the order of conversation when someone comes and sees you is “I have got this spacecraft that requires to get to this orbit on this day, is that good?” Then other discussions occur later, like the environment of the vehicle and price, and all of that kind of stuff. But really that’s the number one thing that people need.”

Beck’s quote also encompasses the third concept Astra considers–launch cadence (or frequency). In another 2018 interview, Beck alluded to the goal of launching an Electron once a week. That is also the goal Kemp set for Astra’s Rocket 4.0 during the presentation (he previously set a goal for daily launches). Like the ability to select a specific customer orbit, Rocket Lab’s CEO mentioning the importance of increased launch frequency means that the concept is likely not something unique to Astra. In the five years of operating the Electron, Rocket Lab has not come close to launching at a once-a-month frequency.

Would Astra, with its much more confused approach to business strategy implementation and goals, be able to beat Rocket Lab’s cadence out of the gate? First, Astra must build and launch Rocket 4.0. Maybe that’s more likely now that it has had a few successes with Rocket 3.3 (although two successful launches out of six are poor odds). 

Astra also highlighted the mobility of Launch System 2.0 and elaborated the system could launch Rocket 4.0 from countries that are unable to build rockets. In addition, Kemp made it sound as if it would not violate ITAR because Astra’s personnel would maintain constant control of the system. But that mobility isn’t that unique, as Virgin Orbit can launch its rocket from under a 747’s wing…which can soar from many runways worldwide.

“If at first you don’t succeed…redefine success.”

None of the company’s concepts to form its business decisions are novel, and the cost comparisons it is using seem downright desperate, even though some customers might find the rocket’s launch cost to be cheap. Its desired launch cadence requires a “wait-and-see” approach as an established competitor, Rocket Lab, is having difficulty achieving a similar goal. While the company touts its manufacturing process, including automation, as a way for the company to manufacture its rockets quickly–well, it hasn’t demonstrated that ability yet. 

It’s not just manufacturing but reliability, too, which must be integral to the company’s manufacturing process. If Astra’s launch reliability is no better than two out of six (maybe aim for one out of ten or better) with its new rocket, that pushes customers away. So Kemp attempted to change what he perceived as customers’ expectations for perfection from launch vehicles: 

“The expectation I think that a lot of people have is every launch has to be perfect,” he told TechCrunch. “I think what Astra has to do, really, is we have to have so many launches nobody thinks about it anymore.”

He doubled down on that by saying customers’ definitions for mission success need some redefinition:

“If two out of the three [launches] are successful, it’s not mission failure,” he said. “It’s just a lower refresh rate for the constellation.”

That is an interesting company perspective about satellites that don’t belong to it, making Astra unique compared to its competitors. Kemp made that comment during the segment about NASA’s TROPICS mission, which Astra will (hopefully) launch successfully.

There’s nothing wrong with using common materials and concepts for rockets and launches. Astra, after all, is already following the footsteps of many other smallsat launch companies. Some of those companies are successfully launching rockets. Astra’s success in launching two rockets is something to be proud of–few startups get to that stage. But during Spacetech, Astra tries too much to excite investors by presenting its activities as unique in the industry. The company’s hand-waving regarding launch system reliability demonstrates that it is still not satisfied with its position in reality. I’ve already provided a few reasons why its various attempts seem to cross the line between truth and fiction to gain investor interest. In attempting to raise funds, companies may initially try to court the three “Fs:” Family, Friends, and Fools

Astra’s pitches make it pretty clear which F it’s focusing on.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s