Joni Pelham 🇬🇧🇱🇹 ✈️ 🚀

I’ve looked at alternatives. I won funding back when I was at NASA to study using dust instead of liquid for heat transfer. The idea was that dust has gigantic surface area to mass ratio so it is efficient at heat exchange, and that in zero g it should be easier to disperse and convey the dust pneumatically. I called the project “dust as a working fluid.” We built a prototype and got a patent out of it. But I think fluid is still the best way to go. Dust has lots of new challenges including internal wear of components and electrostatic arcing. We got shocked really painfully a few times when we stood too close to the pipes 😅 It might be possible to do heat transfer using dust without any pipes, just using magnetic steering rings, covering long distances with very little mass apart from the dust itself, but it would take years of work to mature that concept.

Your daily reminder that the fastest Dallas-to-Houston trains did the whole 280-mile run in 4 hours flat, often going faster than 100 mph. They managed to do this in the 1930s, with trainsets powered by V8s only capable of 600 horsepower but that still weighed 12,000 pounds!

1/4 LLMs solve research grade math problems but struggle with basic calculations. We bridge this gap by turning them to computers. We built a computer INSIDE a transformer that can run programs for millions of steps in seconds solving even the hardest Sudokus with 100% accuracy

1/4 LLMs solve research grade math problems but struggle with basic calculations. We bridge this gap by turning them to computers. We built a computer INSIDE a transformer that can run programs for millions of steps in seconds solving even the hardest Sudokus with 100% accuracy

1/4 LLMs solve research grade math problems but struggle with basic calculations. We bridge this gap by turning them to computers. We built a computer INSIDE a transformer that can run programs for millions of steps in seconds solving even the hardest Sudokus with 100% accuracy

America's Concorde. The Lockheed L-2000 was America's Mach 3 supersonic jet project from the 60s, built to carry 250 passengers across the Atlantic at Mach 3 in 2 hours. The initial design called for a beefed-up turbofan derivative of the P&W J58 engines that had already been proven on the A-12 and subsequently on the SR-71—That would change in future evolutions. Unfortunately, It lost to Boeing's design, and sadly both got axed in '71 over costs and eco issues.👀

a mutual here on x asked me how professionals design wings for aircraft and drones in cad, so I put together a quick, high-level overview of how this process is generally done, and showed how the 3d model is constructed if anyone wants a high-level or technical breakdown of anything aircraft or jet engine-related, just shoot me a dm. if I’m knowledgeable on it, I’d be happy to make a video explaining it at the level of complexity you’re comfortable with. same goes for any powder bed metal 3d printing processes, specifically dmls and electron beam sintering