PDA

View Full Version : Part Schematics Library?!?!?


brickbatbae
03-15-2009, 02:28 AM
As you may have seen in an older thread, I'm engineering student bent on having my own performance engineering company. My long term goal is to have my own production car, as many performance companies end up creating, from concept to the conveyor belt.

I've noticed most, if not all specialty automobiles share parts with a larger company: from suspension parts and engines to a/c vents and weather strips.

It seems like all the guys at Ariel did were design a tube chassis and a carbon fiber tub then plugged in parts from a hundred different companies: engine, springs, etc.

How do they know what parts they should use? Do they buy thousands of different springs then plug, play, and test? Of course not, that isn't cost efficient. So how do they do it?

I remember going to the Mitsu dealership to get a few parts for my burden of a VR-4, and the service technician would type in a few things and bring the part up from a library of parts. The part will show the size, the schematics and the basic dimensions of the part, which is essential. But how about the part attributes? How do companies know what spring gives the required tension, which steering arm gives the required articulation, and so on and so forth?

Simply put...

How the hell do car companies figure out which part to use?

Is there a such thing as a part library that companies hold?
If so, could a lowly civilian like myself get a hold of this library?

Aenimal
03-15-2009, 02:57 AM
Engineers. Years of design, modeling and testing go into choosing a part. A car company will design a platform or purchase a design from another company who specializes in it. A suspension design is built into the platform through modeling/testing/design work (such as strut tower height compared to ride height) and a platform is designed to work with a specific part, or a part is put into production for it.

It's not an easy job at all. Your best bet is try and start with something prefabricated. Try your hand at building a locost or similar kit car and modifying the frame/suspension to your liking.

It's not easy, and almost impossible for a single person to start up a car company from scratch, and the design work to come up with a platform alone would take a single person years to do. Even TVR, amazing as they were, had SO many reliability issues with their cars because their designs didn't have the years of in-field experience of a major car company platform.

http://en.wikipedia.org/wiki/Locost

Nereth
03-15-2009, 03:07 AM
You said you're an engineering student, shouldn't you be able to figure it out?

Engineering decides what to get based on what their work tells them is best. In the ideal world.

brickbatbae
03-15-2009, 11:47 PM
I like the Locost idea, considering that it's a fairly good place to start if I'm really going to be serious with this car stuff. Plus, it would be hella fun.

Being a full-time student, I make about $160 a week :(.

I live with with my fam-bam, and being a fairly tight-knit family, I'm sure that they'll support me strongly, maybe even throw me a bit of extra cash. Maybe.

Where would I start? I have nowhere near the knowledge needed to make something like this, considering the engine work and the tranny would pose as the most difficult part. But let's just put a pin in that and go to the first few steps.

What tools do I need. I own basic construction tools, miter saws, sawzalls, grinders, vices, drills. I don't own any large machines like table/band saws nor lathes or mills, but I do have access to a machine shop on campus that offers these things, so small work wouldn't be too difficult, considering machining out brackets and simple stuff like that.

Since the chassis would be the first major step: I'm thinking I'll need an arc welder first, I can buy one used for fairly cheap, yes?

How would I know the dimensions I would need for the drivetrain? Would I obtain the engine/tranny/steering items first, measure them, then draft the chassis accordingly?

Edit: This is so not going to happen :P

Aenimal
03-16-2009, 01:36 AM
Get yourself a welder first. Then get yourself some steel tubes. Build this:

http://mcsorley.net/locost/

Also, try to use rounded DOM (drawn on mandrel) tubes that don't have a seam on the inside for added strength compared to square tubes.

For an engine, get a cheap JDM engine and rebuild it. Good one would be a toyota 4AG. Other than that, the rest of the info should be available easily online, and in that book that's called 'build your own supercar for under $250' or something like that. Just look up lotus seven clone or Locost cars to find out.

http://www.japanracingengines.com/engineview.asp?id=139

Thar ya go, and make sure to post pictures of the whole process.

Aenimal
03-16-2009, 01:42 AM
And also, Caterham (some british company) makes a production version of the Locost called the Caterham seven. Your fist one will probably look like crap, but if you build a frame or two (not that much cash, you can get it to start looking something like this:

http://racingaccidents.files.wordpress.com/2008/07/caterham_007.jpg

brickbatbae
03-16-2009, 01:59 AM
Thanks a lot for the information, Aenimal.

A few more things.

Where could I purchase a cheap donor car? And is the 4AG engine absolutely necessary?

Aenimal
03-16-2009, 05:13 AM
A 4ag is not necessary, but it'd be a cool cheap engine to put in there. Find a toyota corolla GTS. That's where the engine comes from. You could also use any other engine that you can find for cheap that's light.

As for parts you won't really need too many other than an engine from a car. You can get an drivetrain from the corolla, or get one that'll fit the chassis well and get a driveshaft made (around $150-$350). It'll all be covered in that book, and be covered better by people on the locost and kit car forums. Read up a bunch before doing anything, and just go for it, because you can sell one of these pretty easily after and not loose money as long as it's in decent shape.

bornkiller
03-16-2009, 09:25 AM
Thanks a lot for the information, Aenimal.

A few more things.

Where could I purchase a cheap donor car? And is the 4AG engine absolutely necessary?
I recently swapped one of these http://upload.wikimedia.org/wikipedia/en/d/d3/First_Generation_4A-GE.jpg for 2 x 12 packs of beer. :p

Kwinnie Bogan
03-16-2009, 11:44 PM
D'oh. That was worth at least a 30 block. In other news, what the hell kinda beer comes in cases of 12? Boutique? If so, maybe it wasn't such a one way transaction :).

bornkiller
03-17-2009, 01:33 AM
D'oh. That was worth at least a 30 block. In other news, what the hell kinda beer comes in cases of 12? Boutique? If so, maybe it wasn't such a one way transaction :).
http://www.atlantarenegades.com/Images/sponsor_stein.jpg
6, 12, 15, 24 packs...He was supposed to give me one 15 pack but he got it wrong and I got it right.:thumbsup:

ratfrink
03-19-2009, 01:43 PM
The Locost idea is pretty good to get you started.

As for stealing a drivetrain, I'd go for a Miata, personally. Classic FR layout, enough power for a light vehicle, and the suspension is set up for sportscar handling from the factory, and there are loads of upgraded parts available. Also, there are plenty of them about and they are cheap as chips. I'd look in a scrapyard, but make sure all the mechanicals are fine.

If you'd like a bit more of a challenge, you could go for an MR layout car using a Toyota MR2, or for even more of a challenge, take the engine and gearbox out of a FWD car and mount it in the rear of yours (although they basically did that with the Pontiac Fiero and the handling wasn't anywhere near as good as the MR2) .

As for parts libraries, yes, they do exist. I have worked in a couple of parts departments which list part dimensions and physical features etc, but you'd have to go to the engineering department of a car manufacturer to get details like spring rates. If you are looking to make a car from parts of other cars, the easiest way would be to chat with people who have done it before (must be plenty of forums out there), plus plenty of trial and error.