[Duncan]

Hi
First put your location on the CP - makes things a bit easier!

I built myself an EV from "scratch"

http://www.diyelectriccar.com/forum...dubious-device-44370p15.html?highlight=duncan

It's great fun! - and road legal

It's also worth looking up "Locost"

Basically the secret of building yourself a car is not to do the things that take engineer decades to design and make

A "basic" car that drives and is safe and fun is about 5% of the work in a modern comfortable car

Use the Collin Chapman (Lotus) approach - Less is More - ask yourself "Do I really need this bit?"

If I was doing it again I would use the folded composite Chassis system

http://www.autospeed.com/cms/article.html?&A=112925

And try and source a complete crashed modern EV for the parts - a Tesla if I was rich or else a Leaf

 

[MattsAwesomeStuff]

I wish there would've been people doing stuff like this when I was in school. Best of luck.

An interesting place to start would be Damien's £1000 build, including the car: http://www.diyelectriccar.com/forums/showthread.php/1-000-euro-ev-build-193217.html

I'm doing a motorbike conversion, which is about as dead-simple as it goes, and I've broken my build progress down into each stage: http://www.diyelectriccar.com/forums/showthread.php?p=993802

I presume you're not building the vehicle itself, you're doing a conversion?

Here's a bit of runthrough:

Vehicle: Pick something light, and simple. An old mid-90s civic hatchback or something like that. Ideally, something that didn't have power steering or power breaks. Mid-90s is a good era because things were made light weight, but hadn't been bogged down with sensors and electronics and other crap. You probably want it to be manual, but you can make either work.

Motor: You will be using an old electric forklift drive motor. For your budget, this is the only option. And it'll perform just fine. Bottomless torque. You'll get a 150-200lb motor that's 9-11 inches in diameter. Ignore the wattage/horsepower rating, it can handle much more than that. Don't email, don't call, just show up at a forklift salvage/service/repair place, ask if they have some time to talk to you, explain your story and ask if they have any motors, controllers, cables, fuses, contactors, etc you can have, or, if they have or will be getting some vehicles that you could use your own tools to salvage from some day. If you email it will get ignored, go in person, be kind and polite. It may take weeks or months for the right one to show up, so get started on this early. Most of your build will center around the motor. Take whatever the motor mates to as well, it will save you on machining which is the most expensive part. The motor should be free or nearly free, under $1/lb for sure. Generally not worth their time to extract but this varies based on the poverty of the city you live in. This motor will be a series-wound DC motor. It will not do regenerative braking, which doesn't matter, because it has nearly zero impact on your range anyway. Alternatively, you can try buying a junk forlift but note that you'll have to find a way to move a 5-10 ton vehicle that probalbly has dead batteries.

Transmission: Your motor can spin the driveshaft fast enough directly power the driveshaft. You don't need a transmission. Sometimes it's nice so you have a manual disconnect (neutral), sometimes it's nice for hill climbing if you're underpowered. But if you want you can get rid of one. Different challenges either way, but you probably won't ever shift gears, you'll just leave it in 3rd all the time, including starts.

Batteries: This will be your biggest expense. Don't buy lead-acids, they cost more for how long they last than the gasoline you'd be burning. Figure out your range. A small light car will use 350 watt-hours per mile at highway speeds, maybe 200 watt-hours per mile at city speeds. Decide how far you need to go, this will decide how much you have to spend or salvage. There's great deals to be had buying OEM salvaged vehicle packs, maybe, possibly within your budget. If you want to build your own lithium pack, go to every single computer recycler and laptop repair place in your city with some small rubbermaids with your name, phone number, and email address taped onto them, and a "Call when full." Offer to leave the tub there and tell them you'll take them away for a school project. Some might ask you to pay. $1/lb is a normal rate. You'll be spending at least $50 on a charge tester and months scrapping and testing cells. To get 60 miles of range will probably take you a year of saving up and testing, so, start early.

Speed/Motor Controller: Hopefully you could get something out of the largest forklift you could find. Buying used is at least $300, if you want a great one you can build yourself, probably $500-$1000 for a Paul & Sabrina Controller, but if you hunt Ebay for deals on the expensive parts you can get it for less.

...

That's the stuff you gotta start hunting for ASAP, because it shapes the rest of your build. The more time you give yourself to find deals, the more will appear. The less time you have, the more you will have to spend.

Best of luck.

 

[brian_]

I want to build a road-legal electric car from scratch (save the batteries and motor etc.).

I think you should clarify what you mean by "from scratch".

At the very least, as you mentioned, you would not build a motor from iron plates, copper wire, and so on, just as you would not make cells from sheets of electrode and pouch material then pack the into a custom battery case. Similarly, a builder of a custom engine-driven car is unlikely to make their own engine from aluminum and steel, or cut their own gears to make a transmission.

But what about the rest of the car? Converting a base engine-driven vehicle doesn't seem like "from scratch" to me, but perhaps you meant just not starting with an EV or a conversion kit. Most custom car builders use a lot of components either from production vehicles or aftermarket suppliers, such as Duncan's suspension (from a Subaru), plus steering and braking systems. As Duncan explained, this makes a lot of sense.

In an EV, you can actually build the controller or controller/inverter (depending on whether your motor is DC or AC). Most builders buy this component; a scratch-built version won't typically be better, but it can be good, and it can be less expensive and it can allow a design (combination of other components) which might otherwise not be practical (particularly if using a production EV motor). In mechanical / automotive engineering, my guess is that you would not want to design this component, but perhaps you might build one of someone else's design just to save money.

If you understand the purpose of starting "from scratch" (such as demonstrating fundamental knowledge of automotive design, or learning about mechanical or electronic fabrication), you might be able pin down to what extent you want to design and build components yourself... then forum members can provide more specific guidance.

 

[brian_]

I want to build a road-legal electric car...

You might also want to define "road legal". In some places low-speed vehicles have less demanding regulations, and so can be simpler; the low speed is less demanding of power and energy as well. A road-legal car in this category could be easier and less expensive; it could also be closer to scratch-built.

 

[LottaFreetime]

Hi guys, thank you for all the feedback. To clarify, I meant building from scratch within reason, i.e. if it is reasonable for me to make it instead of buying, I will do so. I was thinking that my finished product will resemble a buggy of sorts -- a chassis made out of metal tubing (unless I can get Tesla batteries as chassis or something). I'm located in the midwest (US) and will learn the regulations. Thank you for the replies, I'll be checking in frequently to post any updates, etc. Anyone have any educational material on this stuff I can read into? I'm still in the very early steps of conceptualizing my build, so please, feel free to give me some input. Thank you!

 

[Emyr]

Since this project would be academically assessed and you don't necessarily need to finish the vehicle in order to succeed academically, you should focus initially on the R&D (a waterfall-esque approach).

Your first project item should be to find the relevant regulations for NEVs, light low-volume-production automobiles etc, and write one report summarizing the key differences between each vehicle class and another report listing each restriction.

Component choices need to be left until later so you can formally compare different potential solutions in the context of the regulations, your budget and your requirements/aspirations for the vehicle's functionality/performance.

 

[MattsAwesomeStuff]

Two big issues:

1 - Legal requirements for homebuilt vehicles. There will be hundreds of pages of regulations specifying required sizes, parts, spacing of safety features, etc. Hundreds of things you've probably not considered. The frame will probably have to be sent off to be non-destructively tested, etc. The testing and certification alone will be expensive and lengthy.

2 - Insurance requirements. Most insurers will refuse to insure it because they don't have a large enough sample size to know what you're likely to cost them. Home-built will almost assuredly bump you into a "high risk/unknown" category and be quite expensive.

...

Also, I'm not sure how you're thinking of this, but if you're thinking "I'll build it myself so it will be cheaper!"... you can buy a driveable car for less than the raw material for the frame would cost you. Heck you can buy a driveable car for less than a new set of tires would cost you.

If you want to do this on a small budget like you are, I think building from the ground up is not going to be possible.

Legal hurdles and certification would be the make-or-break issue, so do your research on that first.

 

[LottaFreetime]

Well it doesn't necessarily have to be road legal, nothing is set in stone yet. Also I believe this build is very possible within three years' time. Again, I won't be building the entire thing from ground up. Some parts will be bought. I'm planning on this project because I like to work on cars and this seemed like a fun and time-worthy project. I would be extremely happy with a car that seems street legal but hasn't been cleared.

 

[brian_]

I was thinking that my finished product will resemble a buggy of sorts -- a chassis made out of metal tubing (unless I can get Tesla batteries as chassis or something).

Although Tesla people like to show the key components of the powertrain and suspension together and call it a "skateboard", the big flat battery pack in the middle is not structural. If you think that these parts are connected into something that you could stand on like a skateboard, then the marketing people have fooled you, as they have many others. The pack is not a floor that supports the car, with the wheels and motors stick under each end; in fact, the battery hangs under the floor of a conventional car body structure; even the motors and gearboxes and subframes at each end are hung from the body structure.

This is true of essentially every current production EV. The structure of the vehicle is the body, and the battery pack is under the floor, supported by the body. The pack case encloses and supports the modules of cells, but does not support the rest of the vehicle. In most conversions which use the battery of a production EV, the pack case is opened and not used; the production modules are re-arranged into a configuration which suits the vehicle, and enclosed in some sort of custom-made box.

 

[brian_]

Well it doesn't necessarily have to be road legal, nothing is set in stone yet.
...
I would be extremely happy with a car that seems street legal but hasn't been cleared.

This opens up a lot of possibilities. A design must have a purpose (and your project will demonstrate engineering to meet that purpose). If your purpose is an non-road use (such as recreational off-road use, or some sort of short-duration competition such as autoslalom, autocross, or drag racing), with requirement for eventual road legality, you could build something interesting and successfully complete the project and use the car regardless of regulatory and licensing issues.

 

[LottaFreetime]

I don't know why I thought that the battery acted as the chassis, that sounds very wrong now that you've pointed it out, thank you. In regards to the drivetrain, what are the pros/cons of a motor with a differential vs multiple motors?

 

[brian_]

I don't know why I thought that the battery acted as the chassis, that sounds very wrong now that you've pointed it out, thank you.

I suspect that you thought that because it's exactly what Tesla marketing people wanted, and what technically ignorant online authors have told you.

Some Tesla displays have included some of the vehicle's structure, so it is closer to a rolling chassis, but even then it is structurally incomplete because the top half of the body (which is important structure). Note that in the photos in that article, the rear suspension struts (which hold up the entire back half of the car) are attached to nothing.

GM's "skateboard", in contrast, really was a structural chassis which held up the body:
Meet The Father Of The Auto 'Skateboard' Chassis Used By Tesla: Chris Borroni-Bird
AUTOnomy
2002 GM Hy-Wire Concept

Since GM's skateboard predates the Tesla Model S (and the original Tesla Roadster was structurally a Lotus Elise and not the battery-under-floor style), I think Tesla Motors copied the term without actually using the design (despite the title of the Forbes article above).

 

[Duncan]

Multiple motors is a bad idea

Motor torque is roughly proportional to weight - so everybody uses a reduction gear to allow a lighter motor to produce enough torque for the wheel

Having two motors and two reduction gears will be heavier than one motor and a diff

PLUS - you can get a diff and reduction gear from any scrapyard

 

[brian_]

In regards to the drivetrain, what are the pros/cons of a motor with a differential vs multiple motors?

We can debate that subject in detail - and we have - but I have a suggestion: decide what the vehicle needs to do, and through the design stage of the project determine from that what characteristics it needs and what design features could deliver them. That will lead to a discussion of the number of motors (and the number of driven wheels) at a point when there is some basis to assess the alternatives and make a decision.

At this point I think you're mostly going to get set opinions from people who will give the same direction completely without regard to what you want the car to do. You'll also get contradictory information. I realize that you're asking for pros and cons - not a decision - but even that is hard to do with so little of the rest of the car's design known, such as power level, motor technology, gearing requirements, component layout, etc.

 

[dain254]

I'm also located in the midwest and during my years in college also completed a couple semesters of "independent study", the project goal for the initial semester was to study the feasibility, vehicle layout, potential driveline, seating configuration, styling etc. of a vehicle that could achieve 100mpgE. The next semester I completed work to determine what it would be like to manufacture the vehicle and as a senior design project I designed and fabricated a CVT transmission for the vehicle (since the powerplant was a diesel engine). I never built the actual vehicle but did complete an electric variant of it some years later that is on this forum called the CozE.

As Brian_ started to mention, what really needs to be determined is what goals you have for this "electric vehicle". Does it need to carry multiple people or just 1, what kind of range and performance goals do you want to achieve? What kind of budget are you working ($3000)? building a practical EV from scratch is a highly multidisciplinary undertaking which many will attempt but few will accomplish. You have successfully stumbled upon this forum, which is a world of knowledge - my advice is to start spending time reading threads of successful projects!

If your goal is road legality, the first step is getting ahold of your local DOT, all states are different on what they let fly or not. In Iowa, we can basically build anything and legalize is so long as we follow the rules and it passes a safety inspection which is somewhat simple and easy to pass.

 

[galderdi]

I agree with everything that has already been said. Duncan's build would be a great point of reference for a road legal build from scratch.

I also would stick to a single motor as it really does reduce complexity.


My build was never intended to be road legal. But I did take regular pics as the build progressed. And my build shows a different alternative for chassis layout.

Edit: I used a full Front Wheel Drive gearbox but placed it behind the seats to achieve a pseudo mid engined layout. Advantages of this approach are the inclusion of the diff in the gearbox, it swings the motor to become east/west, you get to choose the ratio depending on circumstances and you effectively get the reduction gear as part of the deal. Not to mention it surprises a lot of people when you tell them it has a full gearbox.

http://www.diyelectriccar.com/forums/showthread.php/aussie-ev-autocross-special-ii-185897.html

 

[potvinguy]

I have a VW bug EV. It is very easy to convert, just unbolt the ICE from the transaxle and bolt in the electric motor. If you don't want to use a whole VW, the transaxle from one would be a good part for a from-scratch car.

Deep cycle lead-acids are fine for short range, say 25-50 miles. I use 10 but 12 is another popular configuration. I paid $130/each for Interstate brand. Find a dealer in your area, because you don't want to pay shipping on these. If you take care of them, they can last up to 5 years.

I put a big analog ammeter in front of the speedo; it's your fuel flow gauge, so keep the reading low.

I use 2nd and 3rd gear. 2nd is good up to about 30mph, then I shift to 3rd when I get on the freeway. No clutching required!

 

[cheechthechi]

Just an idea since you are an engineering student... does your school have a formula hybrid or formula electric team? If not, maybe consider starting one. You'll probably get a lot more funding from external companies and it'll be one of the best experiences you can have in college. Automotive companies always look to hire students with SAE project experience (formula, baja, etc.) so it really gives you a leg up on all the other students out there.

https://formula-hybrid.org/

https://www.sae.org/attend/student-events/formula-sae-electric/

 

[Pop-boy racer103]

I have a VW bug EV. It is very easy to convert, just unbolt the ICE from the transaxle and bolt in the electric motor. If you don't want to use a whole VW, the transaxle from one would be a good part for a from-scratch car.

Deep cycle lead-acids are fine for short range, say 25-50 miles. I use 10 but 12 is another popular configuration. I paid $130/each for Interstate brand. Find a dealer in your area, because you don't want to pay shipping on these. If you take care of them, they can last up to 5 years.

I put a big analog ammeter in front of the speedo; it's your fuel flow gauge, so keep the reading low.

I use 2nd and 3rd gear. 2nd is good up to about 30mph, then I shift to 3rd when I get on the freeway. No clutching required!

I would be very interested in more details of your set up. I have access to trade prices on lead acid batteries at work and 10-12 batteries Wouldnt cost much money.