I've been building an electric motorbike for the past couple of years, and have been involved in the Canberra Electric Vehicles group for that time. In the process of construction I've learnt a fair amount; I've seen numerous people's descriptions of individual builds but not a lot of general principles. So here I've tried to distil what I've learnt. This is not a guide to building a particular electric vehicle - every vehicle is different - but hopefully it'll be useful somehow.
The basic question to start with is: what are you wanting to build? This can be divided down into:
At the moment, building your own electric vehicle costs about as much as a regular fuelled vehicle. Aim to spend somewhere between $10,000 and $20,000 on a motorbike, $20,000 to $60,000 on a car, or $3,000 to $5,000 on a go-kart. You can adjust your total budget to match your other expectations, but electric vehicles are not cheap. They pay off in the long run though.
Ah, yes, range anxiety. First, remember that this figure should really be what is the absolute minimum distance that the vehicle has to go to be useful. The usual measure for this is a daily commute to work with a side trip. For most people that's somewhere between 50Km and 100Km.
This equates to how much power and torque you need. Power in electrical kilowatts is roughly equivalent to power in motive kilowatts, for the purpose of comparing petrol vehicles to electric. But most electric vehicles are continuous drive - in other words, there's no gear box or automatic transmission to switch between various gear ratios while driving, so the electric motor is always at the same speed ratio to the wheels. Electric motors can do this because they don't need to idle (so they don't need to be disconnected from the drive wheels when the vehicle is stopped) and because their power output range is very wide compared to a standard internal combustion motor (that has a narrow band of speeds in which it is most powerful and that actually delivers less power at lower speeds).
Electrical kilowatts can (again, roughly) be calculated as voltage multiplied by amperage. That gives us a couple of parameters for the whole vehicle: the battery supplies a narrow range of voltages and up to a certain maximum (burst or continuous) amperage. As an example, in my electric motorbike the battery supplies a bit over 120 volts at up to 180 amps. So the battery would be capable of delivering (120 * 180 = ) 21.6 kilowatts.
This is where torque really applies. The torque of the motor is the strength of its twist - it is measured (in the sane world) in Newton-metres. The maximum torque that a motor delivers is entirely dependent on how it's wired. But how much torque actually gets applied is (basically) proportional to how many amps are supplied to the motor by the controller. The graph of torque versus speed in an electric motor is usually a flat line up to a particular point, where (for complicated electrical reasons) the power stored in the motor's internal windings pushes against the current coming in from the controller; then the line basically descends at a constant slope.
While power is (obviously) important to actually push the vehicle forward, torque affects both the acceleration and the top speed (because of having to overcome wind resistance). The top speed that any motor can go is governed by the voltage you can supply to it.
Okay, that's a bit of the theoretical background. The factors that really determine what vehicle you build are:
Obviously, this is the most important factor. It's what determines whether you can convert a vehicle at all, and what its parameters will be.
The good news is that it's not an amount that has to be paid in one go. You can save up, purchase the controller, save up, purchase the motor, save up more, purchase the battery, and so on. None of these parts loses its efficacy over time - even the battery doesn't really stop working unless you leave it for decades. It takes longer to assemble a vehicle this way, but it does mean that you don't have to take out a loan.
It's also the reason I reckon converting a motorbike to electric drive is the most feasible vehicle to attempt: it's just cheaper.
From my budget, this is roughly how much I spent on the conversion:
Obviously, if you've got your heart set on converting a Rolls Royce Siver Phantom or a Volvo 1800, then finding one is going to be a challenge and may be the prime expense of the build. Likewise, if you've already got an old car body that you want to convert, then you've eliminated a major part of the cost.
Likewise, choosing certain parts dictates other choices. There's a distinct jump up from the high end Kelly controllers, only able to deal with 156V maximum, to the more powerful controllers such as Rinehart Motion or EVnetics that deal with 350V and above. A build with limited space might require more expensive batteries with a higher energy density. Some models of car are easy to find spare parts for; some are more difficult or require custom manufacturing.
The other factors to consider here are aesthetics and registration. If you're prepared to just weld a bunch of angle iron to the frame and have a bike that you can throw around the back yard, then the build will be cheaper. If you're going for a particular look, then that adds to the cost. If you register the vehicle, then you not only have to pay for an engineering certificate to say that the vehicle complies with the Australian National Code of Practice - Vehicle Standards Bulletin 14 (phew!), you are going to have to spend money to make sure the build meets the standards for safety, compliance and reliability. The more you have to modify the vehicle to fit the electrical equipment, and the more you change it from its original configuration, the more you (generally) pay.
The process of building a vehicle takes time, and certain actions (such as welding or electrical wiring) require knowledge, equipment and skill. Some will require specialised equipment - for example, my bike has parts machined on metal lathes, something I don't have. You may need to buy or pay people to make parts for the build. Depending on the vehicle you're converting, you may have to spend money on it just to repair previous damage or to get it roadworthy.
Dave Rowe did a conversion of a Charade in a week, but this is probably the exception to the rule. It's easier when you've already done it - and yet most people doing conversions are doing it first time. It's why the people I've met who have converted vehicles to electric drive are all unfailingly happy to donate their time and knowledge to anyone willing to take on a build. But every build is different, and technology is constantly changing - there's no point in slavishly copying Dave's build (for instance), because battery technology and motor technology have improved during that time.