Ok, just found out about this a few days ago and thought I'd share. The CANBus Triple is a programmable device that interfaces with an automotive CAN network. It has 3 pairs of CAN connections which allows you to not only listen and send messages out to the network, it can also isolate devices on the network as well. This lets you intercept certain messages on the network, modify them and resend a forged message, all while letting everything else filter through. Something like this should blow the doors open on some of the things that can be done with on modern computer-laden cars. A few CR-Z specific examples that I can think of:
- Speed signals going to the IMA controller may be "corrected" for a swapped final drive and the IMA can start working properly again.
- It may also be possible to fake the IMA controller all together and keep the gauge cluster happy when doing a K swap...
- Expanding on the engine swap idea, things like the battery charge indicator and IMA assist gauge may be hacked to display other things, even data from external devices provided it can interface with the CANBus Triple.
Anyways, you guys can check it out for yourselves and get some ideas going... Here's the website:
It's also in the final four days of it's Kickstarter deadline. If you want to get in on that, NOW is the time. Backers throwing in $75 or more get a CANBus Triple from the first production run. I'm going for the 2 CANBus Triple package myself... Check it out here:
Yeah, once the way is paved for the crz anyone with a little technical knowledge and motivation could do it. Check out this live dyno from an s2000, using a similar tool for interfacing the can bus:
The first problem to overcome is deciphering the data which can be time consuming. I'm going to a computer security convention in Louisville this weekend; reverse engineering is a hobby for some of these guys so I'll try to make some new friends.
From what I've read this was a hurdle for some of the early first gen Insight mods. Say you're trying to grab battery state of charge- does the car send a percentage value? Battery voltage? Number of bars on the gauge cluster? In a flood of real time data it's enormously difficult to pick out even a known value, let alone an unknown one.
That being said I'd love to see some more work done on these kinds of things. I made my own little scangauge-type device with an Arduino but I can only access the basic OBD2 identifiers. Judging by the service manual it's possible to access the fast CAN IMA line from the OBD2 port, but I haven't had any success.
i like the idea of fixing the aftermarket final drive goof. how about something like allowing more usage of the battery than current limits allow such as a little more charge or drain, and how about getting a little more juice from the battery pack making sonething like the s+ more worthwhile?
If someone actually knows how to read/code with this, I'd be more than willing to see if we can assemble a group fund to get them one of these to work on our cars.
While looking for some tips for the OBD-II / CAN-Bus project I am working on I stumbled across THIS.
I had seen a previous iteration of this by the same person but that method was a bit too complicated / annoying to use. The "revamped" version in the link above is sooo much simpler... if you don't mind the DIY aspect for the hardware.
I too am curious about this. I'd like to hack the engine computer to engage auto-stop at about 40 mph, not 17. When I'm getting off the highway, the IMA is charged to 8 after using 6th and 5th gear to engine brake down to 55, and regen from 55-40.
Better yet, I'd like to engage auto-stop without the brakes, like ITEM9's paddle-shifter hack.
The CAN signals are numerous codes from various systems all utilizing a finite number of CAN wires, read: lots of codes, so the appropriate term of CAN-hacking is closer to computer programming than basic electronics. It's not multimeter testing, splicing, soldering, relays, fuses and so on, its reading, identifying, altering or creating code, not basic electronics.
Ah. No wonder the jumper from the Big Orange Wire to the 12V didn't seem to help much. And smelled something awful, with a horrid cloud of little pluses and minuses wafting over the engine.
With the fairly-common 12V death issues, I wonder why they don't use some of the hybrid battery's juice to jump it when needed, though.