# Lathe or Mill First? Work Backward From the Jobs

> Sticker price barely separates them. What separates them is which jobs you can bid, and how much repair evidence exists when something breaks.

Updated: 2026-08-21. Source: CNC Ledger — https://cncledger.com/guides/lathe-vs-mill-first/ (claims cited to primary sources; see /methodology/).

A shop with one machine is always a compromise. The lathe-or-mill question usually gets answered with opinion. It can be answered with numbers instead, and the numbers say something a little uncomfortable: the purchase price is the least interesting part of the decision.

## Start with what the used market actually asks

Dated Machinio value-by-year data, pulled 2026-08-19, gives real bands rather than vibes:

- [Haas ST-10](/machines/haas-st-10/) turning center: $32,950 to $33,523 for 2011-2013, $35,500 to $42,500 for 2014-2017, $39,900 to $49,900 for 2018-2020, and a 2021 unit at $52,900.

- [Haas VF-2](/machines/haas-vf-2/) mill: $8,900 to $12,500 for 1994-2008, $34,500 to $39,950 for 2010-2015, $41,900 to $62,900 for 2016-2022.

- Haas Mini Mill: $14,950 to $29,950 across 2000-2018 units.

Compare like years and the two families sit close together. A 2014-era ST-10 and a 2014-era VF-2 are both mid-to-high thirties. The mill only gets dramatically cheaper if you drop back to the 1990s and early 2000s band, and a sub-$12,500 VF-2 is a twenty-year-old machine, which is a different risk conversation entirely.

So price does not decide this. Something else has to.

## What each machine actually earns

A lathe turns. Diameters, threads, tapers, bores. The part spins, the tool moves, and the work is mostly serial: one piece cycles through, then the next. Lathes are fast at their job and slow when the job changes.

A mill cuts pockets, slots, flats, and hole patterns. Mill work is geometric. You shape several surfaces in one setup. Mills are slower per feature and far more flexible about what feature you ask for.

That is the whole functional split, and it maps directly onto your invoices. If what you sell is shaft-turned parts, buy the lathe. If what you sell is housings and plates that need finishing, buy the mill. If you cannot name what you sell, that is the actual problem, and no machine fixes it.

## Here is where the repair evidence stops being symmetrical

This is the part most buying guides skip, and it is the honest reason to lean one way.

For Haas VF-series milling spindles there is a published, specialist-sourced cost band: rebuild $4,800 to $7,200, full replacement cartridge $16,000 to $24,000, with the rule of thumb that a rebuild should not eat more than about 20% of the machine's resale value. Lead times run 3 to 6 weeks for a rebuild against 2 to 16 weeks for a replacement.

For Haas ST-series lathe spindles, no equivalent published band exists. We looked. The only cost source we could locate is framed explicitly around VF milling spindles.

Plenty of sites would quietly reuse the mill number and call it a lathe number. We are not going to. A turning spindle carries load differently, sits in a different housing, and has no reason to price identically. Treating the VF band as a lathe band would be a guess wearing a decimal point.

What that means practically: buy a mill and you can price the worst-case repair before you sign. Buy a lathe and your worst-case repair is a phone call you have not made yet. That is not a reason to avoid lathes. It is a reason to get a spindle condition check written into the deal.

## What actually goes wrong on each

On the lathe side, the on-model evidence we have is specific and reassuring. Haas publishes its own service video on tailstock sticking for the ST-10 and ST-15, and it is framed as a routine, low-cost fix. A sticking tailstock on an ST-10 is an inspection item, not a walk-away signal.

The heavier lathe evidence comes from a full ST-30 rebuild filmed by an independent shop. Different machine, same ST-series family, so treat it as family-level rather than ST-10 fact: trashed backing plates on both chucks, rail covers and rails needing replacement, a sub-spindle drive belt chewed up by a chip jam or a flung part, X-axis ball-screw bearing wear, and enough bed twist that correcting it needed a laser level and machinist jacks.

Read that list again. Almost none of it is the spindle. It is chucks, covers, screws, and geometry. Lathe money tends to leak through the workholding and the ways.

On the mill side, the documented failure that costs real money is the spindle bearing. Older VF-series spindles lubricate their bearings with an air-oil mist system, and that system is fed by shop compressed air. Run it off a compressor with no air dryer and you are pushing water into the bearing path. A VF-3 rebuild video documents exactly that: rusty bearing collar, badly worn bearings, root cause pinned on moisture in the shop air supply. That machine's rebuilt bearings had lasted about a year.

A shop air dryer costs a rounding error against a $4,800 rebuild. That is the single highest-leverage thing on this page.

## The decision, run as arithmetic

Sketch the next five jobs you expect to quote. Then answer three questions:

- How many are turned work: shafts, spools, bushings, tapers?

- How many need several milled surfaces held in one setup?

- How many could survive a facing operation plus a manual finish?

If turned work clears half, the lathe is cheaper per part. If geometry work clears half, the mill is. If it splits evenly, you eventually need both, so start with whichever covers the majority of your billable hours and send the rest to a trade partner until the second machine pays for itself.

One more filter, because it is the one people regret ignoring. Whichever way you lean, the first machine sets the ceiling on what you can bid for the next two years. Buy toward the work you actually want to be doing, not the work you happen to have this quarter.