Learn basic mechanical skills first. Then buy a project car or two. Once that's done, find a classic car that's very common in your country already, that way it'll be easier to find parts, advice and contacts. Beetles for example are pretty much universal no matter where you are in the world. Join an Owner's Club forum by the time you get your machine running, they'll help you keep it running. Then after you've got used to driving and maintaining your first restoration, try another of the same model. Then another. Gradually increase the level of degradation of the cars you find. Then if you feel proficient try a new car, something rarer. Also, here's what you want to avoid as a beginner: >Chassis damage >Rust that has literally eaten away at the bodywork >Pillar damage (a single cut through any of the six pillars reduces your car to immediate scrap) >Interior damage outside of seats/belts and steering wheels (if the dashboard's been ruined/torched/hacked at with a machete just leave it) Those four things don't necessarily make a restoration impossible, but they will add significant chunks of time and money to any project.

Buy an old motorcycle in poor shape but popular enough to have cheap parts avalible, it's better if it's in working condition or hasn't been fired up for a couple of years. You'll get it running and start encountering problems that you'll be working on, once you get it in good enough shape sell it for profit or keep it ( doesn't take much space and can be fun on weekends ), you'll learn some basics from that and later move on to cars. Protip: never buy asian reproduction parts they're SHIT and a waste of money. >>17050 summed it up pretty good. It's also good if you know someone who knows something about mechanics and is willing to help you out. What kinda car would you like to fix up OP ?

>>17051 >What kinda car would you like to fix up OP ? Nothing specific, it's just that I'm in my 20's and I'm already seeing cars I like disappear from the roads. Over time I wanna help cars not being replaced by self-driving electric crossovers, that's all.

>>17053 Then open up your equivalent of Craigslist. Select every manufacturer. Then sort by price. Find something popular for about 300 euros/dolars, and fix up that. I still do not regret any cent i have spent on my 1988 Audi 80. No garage, so get to work on it very damn rarely. Don't buy russian though, parts are cheap but unreliable.

>>17049 How much time and money are you willing to spend? You can repair or restore almost anything given enough time and money. I've seen utter scrapheaps restored to mint condition because somebody was willing to pour in the cash to save a unique model. At a certain point it's not practical to do it yourself, but there are shops that specialize in restoring or repairing pretty much any kind of damage, corrosion, or wear you can think of.

Oh right, forgot to ask. Is there any recommended general service literature you could read up if you wanna learn how to repair shit. There are always forums and repair manuals for specific cars but I was thinking more of a compiled font of experience and know-how that could be applied and extrapolated. And another question I forgot to ask. Since even now a lot of parts are hard or impossible to find, especially for old cars that aren't popular, what kinda stuff can you repair on your own without having to replace it?

>>17065 CAD and 3D printing. No, really. That shit will revolutionize the restoration market by offering 99% accurate parts at a cheap price and immediate/ready availability. I've already seen small parts such as hubcaps and rear nameplates casted brand new for when the cars in question got too rare to source a replacement for, and when the technology improves we'll probably see more significant parts.

>>17066 I doubt the technology will evolve beyond the practical point that it is in right now, and there is a small amount of parts you can replace with 3D printed ones. Hubcaps, nameplates, light housings maybe. Maybe even grills for some cars. But anything that's part of the motor? Maybe individual components.

>>17066 You can't 3d print a forged piston.

>>17066 3D printing a shit for anything other than prototyping. Terrible tensile strength compared to ABS plastic or fiberglass reinforced plastic.

>>17049 Mechanic here (industrial tho). Most stuff you can learn from books and videos and trying for yourself. Biggest hurdle is imho the tools. Access to good tools and machinery is the key and that stuff is expensive.