This. A friend buys and restores old Porsche 911s. He will pay anything for custom made parts that are rare as hens teeth. As an example there is a housing around the main cooling fan that seems to be titanium or aluminum or something very exotic. [1]
A company does sell a replacement, which he paid a fortune for, which had garbage tolerances and we spent weeks modifying it until it fit more or less well enough.
I suspect if you spent time on "classic car" forums you'd find owners of very specific models of cars searching for things you could make and sell for any markup, and every owner would buy one.
I am into older Maserati's. Own a couple of them. I do a lot of manufacturing myself. 3d printing, cncing. The problem is that whenever I design a replacement panel, wishbones or various other stuff and try to sell them commercialy I will get sued by the original Maserati company. On a low scale it's fine, but you will never get a living out of it.
The only parts which are in a gray area are rims, exhausts, filters, springs and couple of other parts.
What do they sue you for? If you are upfront about what you are selling, and you didn't steal any IP, it isn't fraud or theft of trade secrets. Maybe they have patents, and claim you are violating them? I can't see how it is illegal to reverse engineer a part and sell a similar version.
Similarly, a German-based company started producing replica Mercedes 300 SL bodies but they were shutdown and the bodies had to be crushed. Although, this isn't quite the same, because the bodies weren't used as replacement parts to keep existing cars on the road.
If we had a right to repair law, presumably it would allow third parties to make replacements when the original manufacturer lo linger was.
Even without that, I'm kind of surprised they sue. Are they patenting each part? Is there case law that covers this? Or is it just a case of bringing a suit to scare people I sto stopping? I wonder.
I would actually like to hear a little more about why the tolerances were poor. Were the major diameters the issue here? I could see a lot of adjustment happening on a car-to-car basis to get concentricity between the alternator fan and the shroud just right. I have very little experience with Porsches, but am a coordinate metrologist for a major aerospace parts manufacturer. I don't see something like this being a huge issue, at least compared to what I deal with on a day-to-day basis.
Edit: I actually looked into it a little more- it looks like the part is probably machined from casting given that there appear to be four stators which also locate the center mounting point. That's where the concentricity would be set; not by the casting, but by the machining of the center flange. So assuming that the fan's mounting points are within spec, I would say that your friend's part could just be chalked up to machining process that leaves something to be desired. This is actually a trickier part if machined from casting (without the appropriate tooling), but actually pretty simple (and wildly expensive) if machined from billet.
Probably some entrepreneur ordering them in small quantities from China from a place that makes frying pans one day, car parts the next.
I used to order a lot of custom things from China. Took a lot of trial and error to find good suppliers. They only accept bank wire as payment, so they risk little other than repeat business if they choose to be sloppy.
You're right, it was a cast part that had been machined, and (from memory) the machining was not precise enough to locate the alternator (and thus the fan) in the exact center. It kind of seemed liked a backyard job, but the price was anything but.
I think from memory the inside of the circle (which was cast) was not perfectly round. From memory the original Porsche part was titanium, and this replacement was aluminum. But I might have that backwards.
It doesn't surprise me in the least that circularity on the casting would be messed up, you would just think that the person on the mill would have the presence of mind to take a look at it before sending it off to the end-user. We reject castings pretty regularly, and that ID could be turned true without a great deal of difficulty provided the deviation in roundness didn't exceed something like 1/2 material thickness.
What doesn't sit right with me is the fact that for a small run, the price of even an aluminum casting would absolutely motivate me to get it right the first time. At the level of volume that the guy would have to be operating at, I would be test-fitting every part to a car before sending it out the door.
I feel like the relative scarcity of air-cooled 911s coupled with the difficulty of machining the part correctly as a third party is why the part itself is so hard to come by. If a person were motivated, though, there is definitely another way to make a suitable replacement more cheaply and easily with a modified design. That's provided someone is okay with not having a completely factory 911.
A 5 axis mill? Once you had the model you could get one offs in many contract shops. It is by no means trivial but if it is an expensive part you could make a very workable copy. Given the required characteristics of the air flow there are some programs that calculate blade profiles.
Maybe the part would end up even more expensive but that sounds like an interesting and worthwhile project.
I do this for old (15yrs+, not old-old) JDM cars. There's a massive untapped potential in this for any car-nuts who have 3D printers.
Just take a quick look through your favourite forum for your car of choice and there's endless threads begging for some random old plastic part that can no longer be found.
Interior trim clips, centre console fascias, exterior body plugs (where tow hook goes, where roof racks go, etc), stereo surrounds, door handle surrounds, etc etc etc.
Could even be something silly, like a plastic fuse extractor tool that people will pay $50 for if it looks OEM.