John Deere is Injection Molding some HUGE parts these days. They have gone to big double screw injectors for big tractor parts. They had an article about it a few months ago in a trade magazine (Cant remember which one).
There's no theoretical reason why it couldn't be done ... if you could get a mold blank that was big enough.
A related process called "Resin Infusion Molding" or some such flackery has taken over production boatbuilding. In it, dry reinforcement is pre-placed, the hull mold cavity is closed, and thermoset resin is pumped in. The pressure is much lower than is common in thermoplastic molding, but the idea is pretty much the same. Its big virtue is that the smell, and the "organics" are contained, and not released for the EPA to sample.
That would require at least 60,000 tonnes clamp force to mould any thermoplastics, even with thick section and lots of injection points, hot runners, and a very hot mould.
Unless it had numerous screws and barrels, melt residence and injection time would be 10s of minutes.
Thick section foam moulding reduces these numbers a bit.
Last I heard, a company called RELN in Sydney Australia has the largest injection moulding machine in the world. They do rainwater tanks, I would estimate from memory at about 12 or 13 foot diameter.
I can get the actual numbers for shot size, clamp force and maximum projected area moulded to date if you like.
Regards
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
According to my calculations, you would need 153,000 tons of clamp pressure (2.2 tons/sq inch) to conventionally injection mold the part. Then 31,000 square inches of parting line bearing area to keep from crushing your mold which would make your mold about 27 feet square. Not to mention trying to mold a 2" thick wall thickness and the problems with that.
With Reaction Injection Molding (RIM) we currently mold body panels in the 7' x 11' range. I've seen 1500 lb parts molded with this process. As far as I know, there is no theoretical limit with RIM.