Hacker News new | ask | show | jobs
by scrooched_moose 1709 days ago
Modeling a tractor in CAD isn't the issue. Manufacturing it is.

A 25 year old tractor design (probably out to 75+ year old tractors even) still has hundreds of millions of dollars in infrastructure and tooling behind it. Molds or dies for a single part can easily reach into 6 figures, without taking into account the $10 million injection molding machine.

The simplifications you could do would be limited, as every design decision affects the placement and access to a dozen other parts.

Take a part like this, which is for a Deere 750 (small tractor) released in 1989: https://www.compactractorparts.com/tractor-parts/john-deere/...

I could model that in CAD in under an hour. Lining up manufacturing is a minimum 12 month process with preexisting relationships, and would take upwards of $250k to get in the door. This also doesn't take into account the "societal collapse" angle - post-apocalypse there's no way to make that yourself, and manufacturers in China would presumably no longer be an option.

4 comments

Then I’m unsure what problem the project is trying to solve.

We have a real problem today with affordability and right to repair. My wife’s family are rice farmers from rural Thailand. My father in law had tears in his eyes when I bought him a $25k basic Kubota tractor; his individual productivity went through the roof while the amount of backbreaking work he had to do fell. Small time agriculture does not pay enough for such folks to make such a purchase on their own, and a major repair could still be out of reach. I thought this project was trying to address these problems. I don’t see such people as having to tooling or time to build their own tractor.

As for “post apocalyptic”- I will band with others with guns and tractors and use our guns to protect our tractors; I’m not interested in building a road warrior tractor; I’m interested in easy to repair.

I think you've nailed the reason they have gravitated towards bolted tube frame design, despite the quite glaring mechanical downsides to that style of assembly. They are driving down the number of custom operations and shrinking the BOM as much as possible. Bolting on more struts is preferable to welding custom angle brackets (which would be more skookum) when that's yet another part you have to jig and fabricate yourself.
> I could model that in CAD in under an hour.

That's very impressive.

An open source foundry might be an interesting project.