V rollers cut on a secondhand mill-turn lathe

Oct 19, 2026 Last reply: 14 minutes ago 2 Replies
Jump to replies (2)

Getting V rollers made when you have never programmed a lathe before is a real hurdle, and that is exactly where the maker started. The rollers only needed a through hole for now, and the whole part came off a CNC lathe in one flow.



The machine itself has history: a 2008 Doosan PUMA300MS picked up at a machine auction, a dual-spindle mill-turn whose sub-spindle doubles as a bar puller. It runs a Fanuc 18i control and carries a 12 inch chuck, with two inch acetal stock as the raw material. The white cylindrical workpiece sits clamped in machined chuck jaws.



The main roller program came out of Fusion360's CAM through the basic DNS lathe post processor. The mill turn post that shipped with the machine did not suit that control, so after exporting more than 40 programs already stored on it, the maker referenced them to hand write a bar pull sub program. The sequence covers facing, OD turning, live-tool face contouring, axial drilling and grooving, with the sub-spindle sliding along Z to back the part mid-process before gripping the finished end for the cutoff. A tool presetter got its first use here as well.



A couple of things stand out to me: the maker calls it a very solid machine and loves the presetter, and reports that acetal cuts nicely with honed aluminum inserts. Not everything is sorted yet. The parts catcher throws an error because a parameter makes it think it will hit something, the spindles apparently need to be turning during transfer or the machine will not feed, and OP2 plus pull transfer on the sub-spindle remain unsolved.


I am curious how the hand written bar pull sub program came together. With the supplied post not matching the control, did the maker build it from scratch or lean heavily on the exported programs for reference? Either way, doing that on a first lathe program is impressive.

Part of that has an answer: the maker exported over 40 programs from the machine and referenced them to create the bar pull sub program. The main roller program ran through the basic DNS lathe post processor, since the supplied mill turn and sub spindle post was not written for the 18i control. How much was built from scratch has not been said.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required