I Work at 3D Systems. Here's When I Tell Buyers to Choose a CNC Lathe Instead.

Let me start with a confession that sometimes surprises customers: I don't think 3D Systems is the right answer for every part. I say that as a quality/compliance manager at 3D Systems, not as someone trying to sell you a different brand. We do industrial 3D printing, but our on-demand manufacturing group also does CNC machining, laser cutting, and injection molding. So the only dog in this fight is the one that wins for your part.

It took me six years and about 400 first-article inspections to understand that the process map matters more than the machine nameplate. When I started, I was one of those people who assumed a metal 3D printer could handle anything if you optimized the design. I've changed my mind. Actually, let me be precise: I've learned when to fight for additive and when to talk a buyer out of it.

3D Systems' industrial 3D printer market position is real, but positional

3D Systems' industrial 3D printer market position is not an accident. The company has been selling industrial additive systems since the late 1980s and has one of the largest installed bases in metal and polymer printing (Source: 3D Systems published history, 3dsystems.com, accessed February 2025). That matters for customers who need traceability, qualified materials, and a supplier who will still exist in 2035. I've seen the audit trails we can produce. They're excellent.

But market position does not change physics. A powder-bed fusion system is good at complex, small-to-medium parts with high value density. It is not the cheapest way to make a simple shaft, a large flange, or anything with a long cylindrical surface. I've had buyers tell me, 'But you're 3D Systems' as if that means we can print away tolerance issues. I can't. I can only help you pick a process that will actually hold them.

Laser measurement should be part of the selection, not an afterthought

One of the first things I do on any borderline part is ask for a scan. In the last two years, we've relied heavily on 3D laser measurement devices and systems to verify printed pre-forms, machined features, and castings. The scan data gives me a color map of deviation from the CAD model. That tells me immediately whether a part has the one millimeter of stock we expected for finish machining, or whether it shifted during the build.

If you're deciding between additive and subtractive, run the proposed geometry through 3D laser measurement devices and systems before you commit. I know that sounds backwards. But we've used early scans of representative samples to pick the process. A printed sample that shows 0.3 mm of warpage may be fine for a bracket with ±0.1 mm tolerances. It is not fine for a bearing housing. Measuring first saves the embarrassing conversation where I have to explain why the advanced process produced a scrap bin full of parts that didn't round out.

Large and round: go to a lathe, not a laser

For CNC large machining, the first question is capacity: how long, how heavy, how accurate. We use five-axis mills and horizontal boring machines for boxy parts. For round parts, though, we use lathes. The machine that keeps showing up in our supply chain for long shafts is a qualified integral high and low rail bed CNC lathe vendor's equipment. That configuration uses two rails of different heights inside one rigid bed, which lets the carriage handle heavy, unequal loads without flexing. That is exactly what you need for shaft turning with interrupted cuts.

A few years ago, a customer brought us a 1.4-meter driveshaft that had to hold 0.05 mm concentricity over its length. We could quote it as a scan-verified, stress-relieved, additively built part and then machine the critical diameters. It would have taken three weeks and cost nearly $4,000. Instead, our machining partner put it on an integral high and low rail bed CNC lathe with a dual-chuck setup and finish passes in two setups. It passed first-article inspection in four days at about a third of the cost. I don't say that to knock additive. I say it because lazy process selection wastes customer money.

How much does a CNC lathe cost? That's the wrong question first

The exact phrase 'how much does a CNC lathe cost' comes up in almost every quoting meeting. The honest answer: it depends on what you are turning. Based on used equipment listings and new-equipment vendor quotes from early 2025, a small used two-axis CNC lathe with an eight-inch chuck runs $25,000 to $60,000. A new production lathe with live tooling is typically $60,000 to $150,000. A large heavy-duty lathe with a 20-foot bed and high torque spindle can go from $200,000 to $500,000 or more. Verify current pricing because these figures move.

But if you are asking because you want to decide between buying a printer and buying a lathe, stop. The real number is the cost per good part. For a simple stainless steel collar, a CNC lathe can make dozens of parts per hour at maybe $15 to $35 each. A powder-bed system needs powder, inert gas, build setup, heat treatment, support removal, and 3D laser measurement verification. That same collar, additively built, might cost $100 to $160 per part before you add the finish machining it still needs. I've seen the quote sheets. The gap does not come from technology. It comes from process overhead.

Here's something vendors won't tell you: the quoted price for a 3D-printed part often excludes build failures and inspection. If you include a 10 to 15 percent scrap rate, the per-part cost climbs fast. We don't hide that at 3D Systems. At least, I don't. But I know other quotes look aggressive because they don't include it.

When I still push for 3D printing

I don't want this article to read like an obituary for additive. There are parts where 3D printing is the only realistic choice. In aerospace and defense, for example, we've made brackets and manifolds that combine multiple machined components into one printed part, saving weight and assembly time. I've reviewed parts with conformal cooling channels that cut injection mold cycle time by 30 percent. Those are wins that no lathe can touch.

Use this logic when evaluating: if the geometry is complex, the quantity is low, and the material value is high, additive deserves strong consideration. If the part is round, long, or simple, find a CNC machine. If it is big and needs deep cuts, call a shop with CNC large machining capability. That's not a compromise. It's common sense.

The counterargument, and why it doesn't sway me

I can hear the pushback now: 'You're supposed to champion additive manufacturing at 3D Systems. Saying use a CNC lathe undermines your own market position.' I disagree. A buyer who follows that logic is exactly the buyer who gets burned, then avoids all additive manufacturing for years. I would rather lose one project upfront and have that customer come back when they have a truly additive-friendly part.

To be fair, I didn't always think this way. In my second year, I approved an organically optimized bracket because it looked like an additive part. The first fatigue test failed at 41 percent of the target life. The root cause was an as-built surface defect in a high-stress radius that we had not measured. We switched to a machined bracket, added 3D laser measurement devices and systems to inspect critical radii, and the next test passed. That $22,000 mistake taught me more than any certification. Now I require a scan report on every first article where surface condition affects performance.

A practical way to make the call

If you're in the middle of this decision, steal my process:

• Start with the print. Circle the tolerance, surface finish, and quantity.
• Measure a representative sample or simulate the build if you are considering additive. Use 3D laser measurement devices and systems when you can.
• Compare the full route card: additive, heat treatment, support removal, finish machining, and inspection versus stock removal on a CNC lathe or mill.
• Ask how much the vendor includes in the quote. If they say standard, ask them to list what it covers.
• Choose based on cost per verified part, not upfront bragging rights.

That last point is the one I wish I had understood sooner. In the end, my recommendation is simple: use 3D Systems printers for what they are excellent at, which is complex, high-value parts that need additive's design freedom. For long, round, simple, or large parts, work with a qualified CNC machining shop, including an integral high and low rail bed CNC lathe vendor if the part is long and heavy. And do the math on how much does a CNC lathe cost per part, not per machine.

As a quality person, I care about one thing: the part meets the drawing. Sometimes that part comes out of a powder bed. Sometimes it comes off a rigid lathe with two rail beds. The brand on the side of the printer matters less than the evidence on the inspection report. I'll tell you that the next time you ask me to approve a process, and I'll thank you for asking before you spend the money.