AstroForge

I do very little long-form writing, but there are a lot of questions about orbital data centers that I don't see being answered. It's easy to dismiss the idea as absurd or to call it the next revolution. Very few people have actually dug into the math. Here's the reality: Leaders of the big AI labs keep painting a future in which humans are marginalized. The result is serious regulatory pushback to terrestrial data centers, and if that trend continues, the only way to keep growing may be to move AI infrastructure off-world. The math might not close in a static world, but the job of a company is to predict the future. Read it here: mattgialich.com/posts/busine…

Solar panels removed for new mounting. We took them off the shipping plate and moved them onto the device that allows us to rotate the spacecraft. This is how it works: constant assembly and disassembly as we enter each test campaign. The next upcoming major test isn’t even with this spacecraft...

Finally got some of our hot-fire images released. This shot was taken during long-duration testing at The Aerospace Corporation. We put over 20 hours of firing on this thruster to make sure it worked with the avionics system. People always show the glow of a hot-fire test, but understanding the setup in detail is important. This was run in a flight-like configuration. That means flight avionics, a flight Power Processing Unit, and a flight Fuel Management System. The only major piece that wasn’t flight-like was the fuel tank, because that would have been pointless. Testing all of this together is really important. On our first test, our PPU caught on fire. It was a useful reminder that engineering hardware and flight hardware are not the same thing. So we rebuilt, re-tested, and tested more. We ran full-power burn-in, low-power testing, high-power testing, starting and stopping. We really ran this through its paces. These thrusters are now strapped onto the spacecraft and ready to fly.

The difficultly of building an interplanetary spacecraft: Here are all six flight solar panels for DeepSpace-2. These are the panels that will give us just over 2 kW of power to the craft. As always in spacecraft manufacturing, there have been some bumps along the way. You might notice the silver electrical leads hanging over the side. Eventually, those will be soldered to the copper rails, or bus bars, which collect and transport electricity to the spacecraft. We make the rails in-house from a copper sheet and apply an adhesive to the back of them to bond them to the panel. The issue is that it is almost impossible to get those copper bus bars perfectly straight. Once they were applied to the panel, we noticed some lifting. More than likely, this would be fine for flight, but because we identified it early, we can easily fix it. Over the last 48 hours, we have tested a few repair options. The winner, epoxying them down in spots before covering them with Kapton, a space-rated adhesive film. We are running a few more thermal cycles to confirm everything is good before the fix is implemented on the flight panels. Then we'll be able to solder the silver leads to the bus bars.

The amount of detail needed to test an entire avionics system is never-ending. Think about it this way: all of our avionics are placed in a small chamber that creates a vacuum, then heats and cools the avionics so we can test them over a temperature range. For deep space, this changes a bit. Normally, in a thermal vacuum test, you run a bunch of cycles to simulate the spacecraft orbiting a planet. For us, the soaking is still important, but the number of cycles is less so. In order to do this testing, we need to simulate both the inputs and the loads. For example, we need to accurately simulate the power coming from the solar panels. This isn’t as simple as just hooking up a power supply, because we have to make sure we test as close as possible to what we expect to happen on our way to the asteroid. Then we need to create the loads. The avionics power a bunch of heaters. We could install a bunch of flight-rated heaters on a table, or we could put a bunch of resistors in a box; we chose the latter. This applies to every powered device: the reaction wheels, the IMU, and the thrusters. So what you end up with is a really complicated but really important test. This is all flight-like hardware, and most of it has already been tested at the individual level. This is when it gets hard. Fully integrated testing is the most important and most time-consuming part of what we do.

The level of detail required to fly any computer into deep space is insane. This is the card that controls the solar panels. What you can’t see are all the chips underneath the heatsink, each of which requires thermal pads to make a solid connection to the piece of aluminum on top. That aluminum is what lets us pull heat out of the card and keep everything operating as intended. Those connections? Those are some big-ass wires that help us handle the current flowing into this beast. The small wire hanging off the board is simply an added thermocouple as we get ready to put this thing through its paces in qualification testing. This is one of six cards that make up an avionics box - the brain of the spaceship. The only difference between a spaceship and a human is the spacecraft has two brains, and the average human only has one. We will never hide the details of this work from anyone. Deep space is the final frontier of exploration, and I hope that, if nothing else, we can help write the book on the future of space exploration. More to come.

OHHHH that says Flight!

The greatest speaker of all time is talking tomorrow. I will also be there! Who else is going?

The race to mine the universe is about to heat up.

Burn underway we’re going to moon!

Thanks for having me on @Jason!

We are the next step after DART. Commercial space is the way to take the learning from some of the greatest space missions ever built, and use it to defense the only inhabitable planet we know about.

57 years ago today, Apollo 9 splashed down after testing the Lunar Module in Earth orbit. Test, learn, *then* land. That’s NASA’s playbook, and it’s @AstroForge’s playbook too.

Three years in a row. 692,000 ounces short. Platinum is facing a real supply problem. platinuminvestment.com/suppl…