Put the lid back on the can until I get a chance to rethink logistics.
Still, after an unexpectedly intense year it feels good to make any progress on this at all.
Clean room assembly, meet my dusty shed.
Plan for today was to remove the bus bars and seal it up again, so next time I don't need as much PPE. But (unlike the last time I did this) the bus bars aren't readily accessible...
New post in my Hyundai Kona Electric reverse engineering series: introducing the Fakon project
https://www.projectgus.com/2024/12/fakon/
In all seriousness I am amazed at the absolute crimes you can commit against modern communication protocols like USB and CAN and they just keep on going.
Pretty decent engineering.
(This was me splitting out the CAN junction that joins all the individual CAN buses of the Kona's motor stack together. By bridging them in one at a time I was able to identify which modules send a few of the remaining unknown CAN bus IDs.)
Spent a bunch of time today refactoring the Fake Kona firmware to use dynamic dispatch in #Rust. It now has ~25% smaller binary (less monomorphisation) and is clearer structured.
However, was just reading the new glue code and immediately realised I'm passing around *way* too much mutable state. No wonder this was so fiddly to do!
Old habits die hard, and I've accidentally gone and recreated the original Python architecture of this program in Rust. With Rust quietly pushing back at every step! đ¤Śââď¸
New blog post with latest update of my Hyundai Kona Electric reverse engineering odyssey...
Back in the shed wrestling the Bench Kona this afternoon. Hyundai's Vehicle Control Unit (VCU) is now satisfied that it's seeing a (non-existent) brake controller. đ
Only one stray fault code left, as it still complains that it can't see the main vehicle gateway.
This has been a case of taking the long route on the long route to what already would have been a huge project anyway, but maybe starting to get somewhere!
Spent the afternoon in the shed gradually pulling the unused wires out of the Bench Kona's rats nest to get something tractable.
Getting there, moving one wire at a time to the discard pile... and it all still seems to be working.
New post in Kona EV reverse engineering series. Nothing super huge to report this time, although there are some short videos with graphs of underdamped motor controller behaviour...
Got to do some "bush mechanics" style Kona EV investigation yesterday.
The runaway motor speed behaviour happens even in a real car with everything operating normally apart from the front wheels off the ground!
Even if you put it in Neutral, it will still spin up to max rpm until you apply the brakes. It's a very small amount of torque, but no rpm limit at all!
Best guess, this is intentional to pre-load the drivetrain while stationary and avoid wear.
Although it's kind of wild that Hyundai engineers don't seem to be worried about the case that a mechanic puts it in Drive on a hoist to check something, or a driveshaft breaks...
Gus