the clanker is making good progress. DC IR drop: thermal and current density, exports a file with instructions for an LLM to fill in json that pulls nominal and peak loads from every part number on your BOM.

openEMS FDTD for the cases that need three dimensions with the field animated through the cycle

infers the power tree from the netlist: rails, regulators, loads, what feeds what.

physics based DRC (copper structures + stackup) that walks every routed net and scores where its reference copper is, how far its impedance strays, and what its vias and reference changes cost. each trace is scored on color (green is good, red is bad)

still kind of buggy tbh, in about a week i want to make the repo public and let other people get their clankers to clank on it too!

okay the clanker is cooking. Bold move naming it HyperLynx though, that's literally the Siemens tool it's trying to replace lol

nah i'm calling it Aether right now, just saying it's gonna be similar function but not $10k/yr

Aether works. keep the clanker clanking, if it actually matches HyperLynx results that's the real flex lol

Ok, this look so cool! Can't wait to try it out

so so sick

Holy cow that’s amazing I was trying to make it work

this IR drop analysis looks pretty slick, killer feature for any power electronics workflows

Hmmm super interesting

Would be next level to put it in a boundary box with environmental temperature... Super cool project

非常有趣的项目，不过怎么保证系统的完备性和准确性，是面临的一个大的调整

Isn’t it a bit impractical to try recreating EDA tools from scratch ?

maybe? this is more of an experiment. hoping others pick it up and contribute. there are open source solvers (openEMS, Emerge, Elmer) that are useful enough to make a tool that at least draws your attention to what could be problematic areas in a complex layout.

@sawaiz That’s nuts. How automated is that? I’ve always done spreadsheets calculating tja + tjc along with expected wattage.