I don't know if it's applicable on these, but on the older chrysler ECUs we used to desolder the eprom and replace it with a socket to make changing them easier. The original eprom usually ended up destroyed but it wasn't a big deal as the stock calibration files for different vehicles were pretty easy to find. Of course that's all minor compared to finding a programming adapter.
-
Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.
AW-4 gurus, need advice.
- Thread starter tonyxj2k
- Start date
I don't know if it's applicable on these, but on the older chrysler ECUs we used to desolder the eprom and replace it with a socket to make changing them easier. The original eprom usually ended up destroyed but it wasn't a big deal as the stock calibration files for different vehicles were pretty easy to find. Of course that's all minor compared to finding a programming adapter.
- Location
- Rainy side of Washington
I wish that would work. Sadly on these, the ROM is built into the microcontroller itself. Rip the ROM off and you lose the brains too...
Eventually I'll figure it out, I don't give up that easily, but it hasn't happened yet.
Eventually I'll figure it out, I don't give up that easily, but it hasn't happened yet.
I like the idea of your pushbutton controller combined with some sort of steering wheel mounted buttons. That with a switch to change back and forth between the stock TCM seems like an ideal setup to me. Don't get me wrong the rail and baja shifters look great but I really like the idea of being able to manually shift without having to take my hands off the wheel.
I'll definitely be looking into something like this in the future. Right now though I'm waiting for work to start back up, so buying a few components to attempt this overdrive hack is about the limit. That should change sometime around the middle of this month.
Right, all I was getting at was using a socket for the microcontroller so you don't have to desolder/solder every time you want to remove it. I'm doing good not to mess up a board trying to solder to it once. I can imagine needing a pile of spare TCMs if I had to desolder every time I wanted to change something.
- Location
- NJ, U.S.A.
How about replacing the microcontroller with an FPGA running an emulation of the original part? With a reprogrammable FPGA and the right JTAG hardware, you could have exactly what you need. There seems to be a fair number of IP cores around to allow FPGAs to replicate microprocessors and microcontrollers, maybe this one's available.
You know, this sounds like the beginnings of a MegaSquirt system for Transmissions - is anyone using the trademark MegaShift?
Edit: Is it that you can't FIND that programmer, or you can't get anyone who has one to give theirs up? Any chance there's a general-purpose adapter that can emulate the one you need, or that Fujitsu might take pity on you and give you schematics?
- Location
- Rainy side of Washington
Could do that, but run into packaging concerns. It's either a 64 or 100 lead PQFP package, by the time you get done adapting an FPGA's pinout to the existing board footprint you could have just bought a pallet of the original MCUs. There is a small EEPROM (a 24C02, 2kbit i2c bus EEPROM) off to the side that is probably connected to the microcontroller, after reverse engineering the stock firmware it shouldn't be hard to figure out what sections of the EEPROM are not used and could be utilized by hobbyist code.
I looked into it and though it's something I'd love to do eventually, I just don't have the time right now. And by the time you get FPGAs and footprint adapters involved, it is pretty expensive to build, too, and well out of the reach of anyone who wants to follow any kind of write-up I'd be doing.
I looked into it and though it's something I'd love to do eventually, I just don't have the time right now. And by the time you get FPGAs and footprint adapters involved, it is pretty expensive to build, too, and well out of the reach of anyone who wants to follow any kind of write-up I'd be doing.
- Location
- NJ, U.S.A.
Good point. Maybe it'd be easier just to reverse-engineer how the controller does what it does, and then build your own circuitry (possibly FPGA-based) to replace the whole TCU. In essence, a true MegaSquirt for the AW4.
Any way you slice it, a fair bit of work.
Any way you slice it, a fair bit of work.
Similar threads
