What is this all about? Well, it's simple. I want to (re-)design the electrical system the way it would be designed today, replacing the current system that was designed over 20 years ago. I picked the name 2003 because that is 20 years after the year the first Fiero was sold to the public, and it also marks the year that I start this project. And if GM doesn't make a 20th anniversary Fiero, I will.
The project (which is a huge undertaking, no doubt) will consist of a variety of electronic modules, controlling most of the functions which are controlled by individual relays and wiring in the original Fiero. All of these modules are interconnected with various networks, and can work together as a whole. The ability of all modules to exchange information allows things like e.g. speed dependent wiper control, and the possibilities are endless.
Though possibilities are endless, there are some things I can not do for legal reasons (i.e. to be able to drive the car on the road legally). I will not mess with safety relevant circuits, so I will not be adding antilock brakes or airbags to the car. I will also not mess with the engine control system and emissions. But I can mess with convenience functions like power doors, interior lights, seat controls, instrument panel, etc.
Apart from the apparent advantages in day to day operation, there are also other advantages: the wiring will be dramatically reduced, imagine only a power, ground and data line going into each door. New functionality can be easily added: just add a module to the bus that implements that function. No need for extra control cables. With less wiring, reliability will be improved automatically. And when a wire no longer has to carry power, it can be thinner and lighter. And finally fault finding is a lot easier
Before you get too excited: my car will be an experimental car, my hobby, and unique. So I will generally not sell any module or code or schematics. But maybe someone is interested in working on this project as well, then I will consider sharing information provided that person is actually working on this project as well. I've had enough "you give me something, and I'll give you something in return", because from experience this does not work because the incentive is gone once that person has received what he/she wanted.
Why am I doing this in the first place if I don't want to sell it? Simple. Reason #1, it's fun to do, #2 I use my car every day, and I like the features found on "modern" cars. And last but not least: I don't have to design the most cost effective approach, squeezing every last cent from the design, and then still having to struggle with people who find it too expensive. This way I design for me and myself alone.
Oh, and maybe this will also serve as a reference to any potential future employer...
This should be a simple choice, or is it? Well, it depends. No doubt CAN is the faster and more powerful system, but I have a lot of experience with J1850, and I also have a lot of J1850 chips in the parts bin. Also, CAN is a European/Asian approach, with J1850 being uniquely American. And since the Fiero is an American car, I am going to go the GM J1850 route. If GM had planned to make a 20th anniversary Fiero, it would have J1850 too. Changing the design to CAN will not be all that difficult, but again, not having to build modules for the public but for myself alone does make choices like this a lot easier. Most modules will be based on the Motorola 912B32 which is available in the pin-compatible CAN version 912BC32, so a new transceiver is all that's necessary to switch to CAN. But it will give me the option to change to CAN if I find it useful for some reason without redesigning everything.
But to avoid excess traffic on the bus, I am going to use the LIN-subbus where appropriate. I can use cheaper CPUs, for instance as slave-only units with very little memory, for stuff like seat control (heater, and possibly in the future lumbar control). This way there are only very few nodes on the J1850 bus for major functions, with a second level of LIN subbus units connected to them.
With this being a huge project, where do I start? What modules do I design and in which order? My design is going to consist of the following subprojects (if you can click on them, they are being worked on or even finished):
A few words of explanation for each of these subprojects. The first one is the J1850 Scantool. I built this one some time ago, and it was actually finished long before I even thought about starting this project.
Next, what good is a protocol without a protocol sniffer? So I built this protocol converter that allows me to hook up my PC or laptop to the J1850 bus. The device features a RS232 and 10 Base 2 Ethernet port, and the software I wrote for the PC allows me to monitor and decode messages, and also transmit messsages to the bus with a keystroke. The configuration file lets me customize the program for different jobs, and it's been fun to write.
Now the real fun began. And since the mirror and window switches are on the center console, there is a center convenience module to query those switches (and some more) and communicate with the DCM via a LIN subbus. The two DCMs and a CCM can work together to control the power doors and power mirrors without interaction with any other module, so this was a good starting point.
Next in line is a Body Control Module controlling interior lights, and other non critical body functions like wipers (the "high speed" mode is still controlled directly, providing a manual override should the BCM fail) or parking brake switch.
The final module will be the instrument panel cluster module (IPC), which will be based on the Fiero instrument panel, but will have a more modern look, complete with driver information system and scantool. I am thinking of using the LCD panel and maybe other parts from a VW instrument cluster. The instrument cluster itself will also be redesigned, remaining a modular unit fitting the current dash panel, but shaped differently to accomodate the new IPC module.
In the course of this the aux. gauge pod will be thrown out and replaced by a 5" TFT screen for use with navigation systems, entertainment or watching DVDs when the vehicle is stopped and the engine is off. But that's really down the road... (although I already have the automotive TFT monitor running here on my desk ;-)
I will define the protocol messages as the project progresses, in accordance with J2178 parts 1-4.
Keep an eye on this page for progress and changes, especially links popping up in the module/subproject list above.