Electric Conversion of a Modern Car

Buying new parts plus the car will cost far more than just buying an EV from a manufacturer in the first place. That would be if you can even find OEM quality parts for sale to the public. There are many aftermarket EV parts directed towards hobbyists but it is unlikely they are anywhere are robust as what is supplied to the manufacturers. Some of these crate motor plus inverter combos make the same amount of power as an entry level EV while costing 10x the used parts. Batteries can get extremely expensive if you get top quality Li-ion new.

As for the CAN bus issue, it isn't an impossible task if you have the computer skills to solve it. Each device is only sending out a minimal range of data, but the quantity and rate will be large (that is handled by computer). A CAN bus view tool is a must, and then you would have to make an substitute for the device you are trying to replace. You could also fake the sensors connected to the device to replicate the actual sensors, then let the device act as if it was normal. There are some projects out there that have done it but it can be a complex extra step which could be why so many people choose to use an older car.

yertiz said:

Been finding it difficult to find any examples where a modern car which was new in 2020 or thereabouts was converted to electric. All conversions happily talk about converting old cars, just with modern cars there are so many electrical modules, sensors etc, if I rip out the engine the ECU etc will light up like a Christmas tree let alone electronic systems not working rendering the car just mechanically useable bar basic electrics. Would be great to see examples. If your wondering why you'd convert a nearly new car its because currently electric cars are so expensive would it make sense to do a DIY conversion on a nearly new car? Lots of questions but no answers...help!

Click to expand...

I have a 2008 B6 Passat and have recently been wondering the very same. I have electronics and computing experience and some CanBus experience but I am not in the aurtomotive trade. I figured that everything would revolve around the ECU so the biggest issue would be if you threw away the ECU you'd probably end up with a rather expensive brick that lit up the dashboard like a christmas tree ... or not at all !
My thoughts were to flip it on its head and fool the existing ECU into thinking it was dealing with an engine...sending the appropriate signals to it to think the engine was there and running. This approach also might be fraught with problems in emulating what the missing sensors might be doing. Simple things like the fuel gauge you could emulate, but getting things like airflow sensor to return the "right" value for the currently emulated "load" might be tricky.....unless that car would happily run around at speed when the emulated conditions were that the engine was at idle all the time....

This is something I've pondered for a while! How possible is it to convert a 'modern' car?

I've read several dicussions about whether to try & simulate the sensors on the engine to fool the ECU or to replace the ECU altogether - with no clear winner between the two. Both seem difficult.

I decided to go for a third option! Most critical vehicle systems will work OK without the engine actually running - ABS, SRS, EPAS, Body controls etc. This is important in case your engine stalls on the highway - you still need steering & braking to work properly.

My solution is to insert a CAN Bridge between the original ECU and the rest of the vehicle. Filter out the messages for speed, rpm, engine temperature which may be read by other systems to change their action. PAS usually provides more assistance at lower speeds. ABS only functions at higher speeds for example.

The filtered messages can be replaced with values derived from the EV controllers and injected into the bus. I've done some preliminary work on this on a 2021 Polaris General Although this is relatively simple compared to most vehicles (only one CANBUS, 3 ECU's). My 2005 Range Rover Sport has 3 CAN networks & 16 ECU's. I've reverse engineered (hacked) a lot of that too!

I think a lot of the problem is fear of the unknown, rather than it being technically difficult. It is, though, much easier to convert a stone-age vehicle & not have to worry about all the computer stuff. Reverse engineering takes time, even if you know what you're doing

I probably wouldn't embark on this if it were my first EV conversion. Even though, this is still a challenge. Luckily, I like challenges!