ABS CAN Bus Simulation for Automotive Systems

Hello can tool,

First of all, I'm sorry for my ignorance; I'm new here and I have a problem:

I want to build an electric car (yes, I know there are different opinions about that). Unfortunately, the ABS system will no longer function because it receives data from the engine!
There are supposed to be additional control units that can simulate this, but I haven't found one yet!

My question:

'Does anyone know about these control units?'

Can you build something like that yourself?

Please help me because without a functioning ABS (Anti-lock Braking System), we won't pass the technical inspection (TÜV) --> and that would mean the project is dead!

Best regards, Gerd.

gerd4711 wrote:

Unfortunately, the ABS will no longer function because it receives data from the engine. gets!

Sure?
I would be concerned that if the engine control unit fails during a drive, the ABS system would also stop working.
What kind of vehicle platform are we talking about?

Quote:

There are supposed to be additional control units that simulate this, but I haven't found one yet!

"Plug and Play?" I have no idea; it's probably a rather small market as well.

Quote:

Can one build something like that oneself?

In principle, you can build anything yourself; it's just a matter of the effort involved (and whether or not you can obtain the necessary information).

Quote:

Please help me, because without a functioning ABS, we won't pass the TÜV inspection --> and that would mean the project is dead!

It's also important to consider the initial registration. If the vehicle already has an EC type approval, you run into the problem that anything that looks electrical, including your auxiliary control unit, needs an "e" marking with a corresponding test report; otherwise, the operating permit is invalid.

Best regards,
Ingo

hello Ingo,

First of all, thank you for your reply.

If the engine control unit (ECU) fails during a trip, I don't think it's a major problem.
The problem is that when the ignition is turned on, the ABS module somehow receives a 'wake-up' signal from the engine, as if the engine is saying, 'I'm running, so you should turn on too!' This communication happens via CAN bus.

Here's the translation:

'This is about an Audi A2 manufactured in 2001.'

Regarding the 'e-mark,' no problem, as an electromagnetic compatibility (EMC) test is already required for conversions on electric vehicles! Therefore, it's not a problem for self-built devices.
Half of the car is a self-built project without any 'e' characters .

gerd4711 wrote:

hello Ingo, First of all, thank you for your reply. If the engine control unit (ECU) fails during a trip, I don't think it's a major problem. The problem occurs when starting the engine: the ABS module somehow receives a "wake-up" signal from the motor, as if the motor is saying, "I'm running, so you should activate too!" This communication happens via CAN bus. Here's the translation: "This is about an Audi A2 manufactured in 2001." Regarding the "e-mark," no problem, as an electromagnetic compatibility (EMC) test is already required for conversions on electric vehicles! Therefore, it's not a problem for self-built devices. Half of the car is homemade without any "e" characters

.

That sounds very interesting.
I sent you a private message.

Best regards,
Ingo

Yes, they are definitely interested!

Electric car, hold on!

Unfortunately, I don't know enough about CAN (Controller Area Network) systems!
Is it possible to capture the CAN signal from the engine control unit (ECU) and then loop it back into the system?

It is certainly possible to record a sequence and then replay it on the drive's CAN bus.
I don't believe that the engine control unit (ECU) and the anti-lock braking system (ABS) control unit communicate with each other solely through a fixed sequence.
Instead, the ABS control unit will respond to specific messages from the engine control unit with certain reactions. And this is what needs to be determined: the mechanism of when which message must be sent and what the reaction should look like in response to a reply from the ABS control unit.
Without detailed information, it's certainly not a trivial matter; I can imagine that. However, I must admit that I don't know which information these two control units exchange or what the ABS control unit absolutely requires. It's hard for me to believe that it's just a wake-up signal. It probably also needs speed data and other information.

Greetings.
Jens.

Hello Jens,

First of all, thank you for your reply.

So, some kind of signal must be exchanged!

Definitely a wake-up impulse/signal! It gets its speed information from the transmission.

What tools can I use to record CAN signals?

I just ordered a few books about Can from Amazon.
Let's see, you can probably simulate something with a microcontroller like the AVR ATmega8!

--> The books cost a total of €100! I hope they provide a solution...

gerd4711 wrote:

I just ordered some books about Can from Amazon. Let's see, you can probably simulate something with a microcontroller like the AVR ATmega8! --> The books together cost €100! I hope they provide a solution...

Send the books back and give me €50, then we both benefit from .
Okay, putting the jokes aside, of course I don't know your background or current knowledge, but I think those books might not be very helpful to you.
For a basic overview of CAN technology, the internet will be sufficient.
Take a look at the Audi self-study programs (SSPs) with numbers 186, 238, and 269; that should give you a good overview. Just do a Google search, and you'll find something about them. Otherwise, feel free to contact me again.

The reason why the books aren't helping you is that CAN only describes the first two layers in the ISO/OSI layered model. In simple terms, this means that only the voltage levels on the line and the structure of a CAN message are specified.
However, each manufacturer decides for itself which values are included in the message and what meaning those values have, and Audi won't simply reveal that information to you.
That means you need to acquire a CAN logger, tap into the bus, and somehow try to interpret the data traffic and derive a pattern from it. A tool for this is the CAN-Hacker program mentioned in this forum. Someone also recently offered some CAN loggers in the "For Sale" section.

Depending on the types of messages you need to replicate on the drive's CAN bus, I think it might be challenging to do with just the ATMega8.

Greetings.
Jens.

Quote:

Depending on the types of messages you need to replicate on the drive's CAN bus, I imagine that it might be challenging with the ATMega8.

I'm not entirely sure why, but it feels uncomfortable – after about 15 years of using 68HC11 microcontrollers with 82C200/SJAA1000 interfaces (for professional purposes, without any cars involved), I find the AT90CAN32/64/128 microcontrollers (for personal projects) quite luxurious, especially because you can debug the CAN registers using a JTAG-ICE.


Best regards,
Ingo

Hmm, okay, so the books are probably junk. No problem, I'll return them then!

I found something that might work. From what I'm reading, it seems like I can log the data and extract specific parts from it.

Send this as a message on a specific schedule.


'Should it work like this, or what do you think about it?'

Well, I don't know how much that thing costs, but I wouldn't buy it.
While it can be used extensively for recording, it seems too inflexible for sending data.
4 zyklische Nachrichten? Hmm.
The problem is that you first need to know what data you have to send and whether you might also need to respond to specific CAN messages in a defined way and at the right timing. And before I spend a lot of money on something like that, I would first think about it carefully.

Greetings.

To ensure that all these components work together seamlessly, you will likely also need to deal with network management.
Simply sending a few telegrams to the bus is probably not enough.
I would advise against using IXXAT.
As mentioned before, it's inflexible.
It's better to get an interface with the LAWICEL protocol and work with the CAN-Hacker.

Regards,
Maxx.

Hello, thank you all for your responses.

That LAWICEL thing sounds like a good starting point!

Regarding the part I posted earlier, I just thought it might be possible to run it independently, meaning install it in the car and have it operate on its own.
(to the team: please send this message repeatedly in a loop)

That would be the best thing!

'With the forum's interface, I could simulate the signals of a motor control unit, at least under laboratory conditions. Am I understanding that correctly?'

Sorry for my silly questions, but I'm new to this topic.

Yep, that should work.