Electric car: DIY!

Frank

Certified not a majority
Veteran
So, you want to drive an electric car on the cheap? And you have an old gasoline car in the back yard, that you would want to convert to run on electricity, if you knew how? Then this is for you!

I would like to see how much it would cost to do so, what the options are and what would be the hard parts. Because I am thinking about doing so myself.


The goal:
Refitting a gasoline car to all electric for the same money it takes you to drive the gasoline car for one year if you want it cheap, or 3 years if you want high performance and/or a longer endurance between refills.


The running costs of that all-electric car are almost zero (fill up your tank and drive 100 kilometer for 10-90 Euro cents, depending), while you would pay quite a bit more otherwise (that same 100 km would cost you between 5 and 50 Euro, depending).

Let's say it costs 0.5 cents/km on electric, and 22 cents/km on gasoline, so our electric car is 44 times cheaper per kilometer. The average daily commute is 30 km in Europe and 30 miles in the US, so that makes 37.5 km/day. Let's say 40 km/day, every day, including shopping and recreation. That makes about 15000 km/year.

Although the above calculation doesn't take into account, that many people who live close to their job don't use their car to get there, so in reality the people who do drive their car daily tend to use it 20,000 km/year, which fits the statistics.

So, we would have 20,000 * 0.22 - 20,000 * 0.005 = 4300 Euro to spend. For convenience, let's say 4000 Euro. Or 12,000 Euro for the high performance/long duration package.


What we need:
1. An electric motor (AC, DC or brushless, 50-200 kW), or 4 wheels with motors in the hub if you want the best.
2. Batteries. Lead-acid would do for the cheap package, but NiMH would be better, and LiIon would be best.
3. Electronics. Those go from simple for a single, straight DC motor, through reasonably complex for a single AC motor, to quite complex for 4 brushless motors in the wheels.
4. Mechanical adjustments. From battery clamps, through installing the motor, to ripping out the complete drive train and fitting hub motor wheels.
5. Controls. Fit sensors and rewire the paddles and other controls.
6. Instructions. How do you install all that?


So, does anyone knows where we can get the parts, and has a good idea of how we could fit it all?
 
At which speed do you want/need to drive? Because that's a big factor here.

Also, the lighter your car, the better, so that's an important start. Fast accelleration and higher speeds are important drains on your battery. I'm learning one or two things from my Prius. ;) (it has an Electronic Vehicle mode, that only goes up to 50km/h though, after that, the regular engine kicks in).
 
Would the car's existing computers even allow a new engine to be installed? I have no idea what kind of lock-in vendors put in.

Also, it can get quite hot inside a car so you'd want some kind of active cooling system for that battery, or a battery that can handle very high temperatures.
 
4000€ won't even get you a proper battery pack :)

I worked in hybrid development for a while and I can tell you it's really not trivial. Loads of coding needed (with rather deep knowledge about engines/transmission required), also stuff like ESP must be tuned to fit, also for start/stop you need to change lots of stuff (for example what will you do with AC when stopped, how to keep pressure on the clutches when the engine's not running and on and on). Also you'll have to emulate all the stuff the motor ECU did, including bus communication (in many cars the motor ECU is also a gateway between 2-3 CAN busses).
 
Last edited by a moderator:
Well the last battery pack I ordered was ~8000€, but it was a development part and thus way more expensive than usual. Still, 4000€ is not enough by any means.

Frank, should it be purely electric? That way you'd at least have a realistic chance to make it work, but just forget about anything hybrid-wise - you'd have to calculate in millions for the latter.
 
Actually, there seem to be quite a few people that have done this already. This seems to be a nice one, with a good online blog on everything he did (and that was many years ago already, since he has a 2002 update ;) ):

http://jerryrig.com/convert/
 
Well building the electrics is easy, but as said modern cars have loads of important bus communication which is vital for operation. So if you remove the motor ECU, you'll have to emulate everything it does. Also doable but only if you know what it is that you need to emulate - so you need the full info on that first. And last but not least, you'll have to circumvent the immobilizer which could be a huge pita.

I'd say go with some 15-20 years old model, there you'll have no such troubles since cars were pretty much bus-free back then.
 
Frank, should it be purely electric? That way you'd at least have a realistic chance to make it work, but just forget about anything hybrid-wise - you'd have to calculate in millions for the latter.
Yes, that was the idea: an old car with as less digital electronics as possible, remove the engine, fit an electric motor and see what batteries are affordable.

But you raised some good points. And I agree: the less dependence between the parts and the simpler, the better. The electronics aren't very challenging nowadays, but the programming even more so. Custom mechanics are often the hardest part.
 
Btw, unless wheel hub motors are an available alternative for the high duation/power package (which I doubt), the only difference between the two would be in the size and quality of the battery pack.

Yes, driving hard and fast acceleration are a major drain, but you could always install a switch that allows selecting between economic and sport.
 
With most battery packs heat isn't really a problem, as you need to cool the batteries.

If you need air conditioning you are screwed.

An all electric thing should be possible though. You might be able to get a much better pack than those linked as well. Hymotions pack should be out in a year.

Otherwise just get nasty lead acid batteries as they are cheap.
 
With most battery packs heat isn't really a problem, as you need to cool the batteries.
That sounds like the thing to do, but I'm very doubtful if they would produce enough heat. Looking at their efficiency, they propably won't.

Still, I should definitely make a duct to lead that heat into the compartment, with a complimentary heater added for when it's quite cold.

If you need air conditioning you are screwed.
It would definitely be a serious power drain. But it could be done. Although I don't know if reducing the endurance so much would be worth it.

An all electric thing should be possible though. You might be able to get a much better pack than those linked as well. Hymotions pack should be out in a year.

Otherwise just get nasty lead acid batteries as they are cheap.
Lead acid are cheap and easy, but very bulky. I think NiMH are the current best for price/performance.

New innovations are great, but most of the time very expensive.

Then again, if it would work out, I might start a business doing this, which would almost require wheel hub motors and such high-performance battery packs. I might even be able to make a deal with them based on that. So I am planning on trying all the leads supplied.
 
And I'm pretty sure that electric car would outperform that ramjet powered cart and have better endurance, while still being street legal.

But it's a nice idea all the same. :)
 
Totally off topic: If I had the money, I would build a flying chair. I would take a bunch of these micro-turbines, attach them to a chair, put a fuel tank underneath, and add some servos to be able to steer.

Efficiency: worst. Sheer thrill: best ever!

:D :D :D
 
Back
Top