Normally, once the Kick Stage does its job, Rocket Lab de-orbits it to burn up in the atmosphere. However, this time it sent a command that switched it into Photon satellite mode to continue on a standalone mission called “First Light.” Intended as a demonstration, it’s equipped with solar panels and a camera that can snap images of itself and the Earth.

Eventually, customers will be able to choose a “launch-plus-spacecraft” mission with the Electron Rocket and Photon satellite, which “eliminate[s] the complexity, risk and delays associated with having to build their own satellite hardware and procure a separate launch,” said Rocket Lab CEO Peter Beck in a statement.

During a press conference, Beck said that the company launched Photon in secret to “make sure it’s all good and it works before announcing it.” Rocket Lab said that a high-energy version of Photon will eventually fly “lunar and interplanetary missions,” including NASA’s Capstone mission in early 2021. In that mission, Photon will fly as a “pathfinder” that will help the Artemis program’s Gateway spacecraft safely approach the Moon.

Rocket Lab is on a roll this week. After successfully returning to launch following a failure in July, the company announced that it has received FAA clearance to launch small satellites from its Electron rockets in the US. So far, all its missions have flown from the Launch Complex 1 (LC-1) facility in New Zealand, but it can now officially operate from its LC-2 pad at Wallops Island, Virginia.

Rocket Lab unveiled the Wallops site late in 2018 and had the opening ceremony towards the end of last year. The company also built a third launch complex in New Zealand, all to serve its goal of eventually launching up to 130 Electron rockets per year. It hopes a lot of those will be for US government customers, so the FAA permit is a big step to “streamline the path to orbit and enable responsive space access from US soil,” the company wrote in a press release.

The Electron booster’s payload was the “Sequoia” satellite from San Francisco-based Capella Space. The mission was called “I Can’t Believe It’s Not Optical,” because the satellite uses synthetic aperture radar (SAR) and not optical lenses to provide detailed Earth images. “Capella’s space-based radar can detect sub-0.5 meter changes on the surface of the Earth, providing insights and data that can be used for security, agricultural and infrastructure monitoring, as well as disaster response and recovery,” Rocket Lab wrote.

The mission is the 14th successful launch for Rocket Lab. Prior to the July 4th failure, the company had made 11 consecutive successful launches.

If all goes according to plan, Rocket Lab will resume flights less than two months after an “anomaly” led to an Electron rocket failing shortly after launch, destroying its payload in the process. The company eventually pinpointed an electrical connection as the cause.

A fast turnaround might have been necessary. While Rocket Lab had a good track record for missions up until July, a significant delay might have cast doubt on the firm’s ability to identify and recover from setbacks. It’s not as experienced as companies like SpaceX. A quick (but not too quick) return to flight suggests it has the issue under control, and that customers won’t have to wait several months or more to get their cargo into orbit.

In its announcement, Rocket Lab said that it was able to gather the data it needs, because the vehicle was unharmed and was able to continue sending information to its ground team. It also explained that it wasn’t able to detect the issue before the flight, because the electrical connection remained secure throughout testing. However, its now knows that the issue can be avoided through additional tests. Rocket Lab founder and CEO Peter Beck said in a statement:

“The issue occurred under incredibly specific and unique circumstances, causing the connection to fail in a way that we wouldn’t detect with standard testing. Our team has now reliably replicated the issue in test and identified that it can be mitigated through additional testing and procedures.”

The next Electron launch doesn’t have an exact date yet, but it will take off from Rocket Lab’s Launch Complex 1 in New Zealand.

The rocket had been carrying payloads for three customers, most notably Canon. It was supposed to be demonstrating an Earth imaging camera system ahead of plans for mass production. Planet was expanding its giant constellation of Earth observation satellites, while In-Space was launching its first cubesat mission as part of an initiative to help researchers and startups get payloads into orbit.

The loss comes after 11 consecutive successful Electron launches, including some with a relatively rapid turnaround. This isn’t the end, then, but it suggests that Rocket Lab may have work to do before it’s a completely reliable option for getting machines into space.

A brief statement about today’s mission from our founder and CEO, Peter Beck. pic.twitter.com/QUShtzp7J0

— Rocket Lab (@RocketLab) July 5, 2020

Public fireworks shows may be in short supply this weekend due to the pandemic, but Rocket Lab is ready to fill the gap. It’s planning to launch its 13th mission, “Pics Or It Didn’t Happen,” from a complex on New Zealand’s Mahia Peninsula at 5:19PM Eastern with a livestream (below) available 15 minutes before takeoff. The launch window is open until 6:03PM Eastern. This is only the third Rocket Lab launch this year, due in no small part to the pandemic, and it includes some significant payloads.

The main payload is a Canon satellite, CE-SAT-IB, that will demo high-res and wide angle cameras for Earth imaging — if all goes well, this will pave the way for mass production. Planet, the company behind the largest network of Earth observation satellites, is also including five of its newest-generation SuperDove satellites. It’s also an important mission for the UK with In-Space launching its first Faraday cubesat mission to help startups and researchers get payloads into orbit.

Next, Rocket Lab will attempt to recover a full Electron first stage following a launch. It won’t pull that from the air but will retrieve the rocket stage after it lands in the ocean. A parachute will help slow its descent, and like previous versions, it will include instrumentation to “inform future recovery efforts.” That mission is planned for late 2020.

Of course, catching a rocket stage after an actual launch is a lot different than catching one that’s dropped neatly by a helicopter. But the feat is a key milestone, as Rocket Lab’s plans to reuse the rockets depend on this recovery method. If it’s successful, Rocket Lab will be able to lower costs, and in theory, that may lead to more launches.

The tiny satellite (it weighs just 55lbs) will first conduct a navigation demo with the Lunar Reconnaissance Orbiter, but it’s ultimately meant to be a “pathfinder” that will help the Artemis program’s Gateway spacecraft safely approach the Moon.

This deal isn’t surprising given the current NASA administration’s privatization efforts. At the same time, it appears to be a logical fit. Rocket Lab’s speciality is launching rockets quickly and on a relatively low budget, and it appear to be delivering — it’ll blast off roughly a year from this writing with a cost just under $10 million, or less than the $13.7 million NASA is paying to create and run CAPSTONE. If everything goes well, this could lead to more high-profile team-ups between NASA and Rocket Lab.

Photon can handle missions lasting as long as five years in low Earth orbit, with payloads up to 375lbs depending on the configuration and its flight path. Its communications aren’t exactly blazing fast at 512Kbps, but it can hold up to 1TB of data.

You won’t have to wait too long to see Photon in service. The first launch of any kind is slated for the fourth quarter of 2019, while customers are currently aiming for 2020 flights.

This represents a relatively uncommon strategy. SpaceX and other private companies tend to build only the rockets for satellite missions. Rocket Lab, meanwhile, is betting that it can thrive by taking care of virtually everything. There’s a degree of risk involved — it’s not clear that customers will want to pay someone else to handle the satellite versus designing something themselves, especially if there are highly specialized tasks involved. However, this might be appealing to companies and researchers that can’t justify developing their own satellite platforms.