Asteroid mining startup AstroForge is racing against the clock to complete its refinery demonstration on orbit after encountering a number of unexpected setbacks before and after launch, the startup said in a candid blog post published today.
This demonstration mission may prove to be a case study in the difficulties of putting hardware on orbit for the first time. Such challenges are often gestured to in the common refrain that “space is hard,” but they are rarely elaborated upon with much detail.
AstroForge is changing that. In the blog post, the company dives into the problems it encountered — some even before the mission launched. One of the most consequential issues cropped up during final integration with OrbAstro, the satellite bus provider for the mission. AstroForge learned that its refinery system generated a magnetic field that would prevent the satellite from being able to actively orient itself in space.
The issue was massive: Essentially, it meant that the spacecraft would wobble, making communications difficult. Even worse, it meant that the spacecraft would eventually settle into an orientation that would sever communication with Earth completely.
AstroForge faced a decision: delay the mission by at least nine months and eat the launch costs, or move forward with the understanding that the satellite would wobble and eventually settle into an orientation that would cut off communication with Earth.
“We made a decision to go as fast as possible,” AstroForge co-founder and CEO Matthew Gialich said in a recent interview. “We had the magnetic field problem that we knew about before we launched but we decided, hey, we’re a company that wants to go as fast as possible. We’re going to take some risk here. Let’s go for it. Let’s see what happens.”
As a stopgap, the company worked with OrbAstro to incorporate an extra antenna on the satellite, so they could at least downlink data on the health status of the spacecraft. There wasn’t enough time to also add an uplink antenna, but the mission moved ahead regardless.
A separate issue occurred immediately after the satellite separated from the launch vehicle, and AstroForge had trouble identifying it among the 50 other payloads discharged by SpaceX close to the same time. It’s an issue that’s more common than is often talked about, AstroForge co-founder and CTO Jose Acain said.
“You can think that you’re tracking one satellite, but in reality, it could be somebody else’s. So that whole time you’re spending trying to talk to this one object, it could be completely not yours,” he said. “We’re finding other companies having to deal with this as well.”
Engineers were finally able to connect with the spacecraft on May 5, 20 days after launch. The nominal health packets were a good sign, but the magnetic field issue reared its head when it came time to deploy the solar arrays. Because the satellite was unable to orient itself well, there was a decreased chance of the onboard antennas for telemetry and command aligning with ground stations.
The answer was to establish more ground station passes. Contracting and licensing with a ground station can take up to a year, but with the help of Australian ground segment venture Capricorn Space, AstroForge was able to complete it in just three months.
The company finally managed to deploy the solar arrays on September 2, and completed checkouts on November 8, around seven months after launch. AstroForge now has just three months to complete its initial baseline check of its onboard sensors and the refinery demonstration before it loses the ability to command the satellite completely.
Of course, errors are not without value: The two co-founders said the mission has added up to lessons learned that are already informing the next, considerably more ambitious, mission to deep space. Oddly enough, that mission, which is scheduled to take place next year, will benefit enormously from these first mistakes.
Some of those changes include: establishing protocols to get fast tracking and communications established with spacecraft after launch vehicle separation; adding an uplink capability; and adding sequences to command critical systems, like the solar arrays, to deploy automatically.
“This is a project that I care a lot about,” Gialich said. “This is what I want to do with my life’s work. When you do something that really you see as your life’s work, there’s two outcomes: either we are the successful company that does it, or we show other companies the kind of mistakes we made along the way that they can avoid to become the successful company.”
Acain echoed these comments: “We always said that if we fail, if this business fails, learnings that people can take away from that is just as important. And that’s part of the reason why we were as transparent as we were in that blog.”