Every year, around the same time, I end up in a meeting where someone asks some version of: 'Which process is cheapest?' Usually it's about a part that needs to exist in three weeks. My answer never satisfies anyone at first: 'It depends.' But after seven years of watching orders blow budgets because the wrong process was picked, I've learned that 'it depends' is the only honest answer.

I manage procurement for a 40-person engineering company, and our prototyping and low-volume production budget is about $120,000 a year. I've negotiated with 30+ vendors, logged every invoice, and built a cost tracking spreadsheet that has embarrassed more than one supplier. This is how I actually decide between a service like Sculpteo 3D printing service, a CNC machine, or a repair method like laser welding.

Let's break it into three scenarios. There is no single right manufacturing process. There is only the right process for your specific situation.

Before we start: the three questions

In my experience, these three questions predict the right process better than any cost-per-part calculator:

  1. How many parts do you need? One batch of 3? 300? 3,000?
  2. What are the loads and tolerances? Will it be hot, stressed, visible?
  3. Is the design final or still changing? If it's still changing, 3D printing is usually the no-brainer.

If you can answer those, you'll know which of the three scenarios below you're in.

Scenario A: Low quantity, complex geometry, design not final → Sculpteo 3D printing service

I can't count how many times I've heard someone say 'we'll just machine it' when the model has internal channels, organic curves, or a part count that makes assembly illogical. That's when 3D printing wins, especially for verified functional prototypes.

'How much does 3d printer resin cost?' is usually the first question. The honest answer from my purchase orders: standard resin has landed between $50 and $80 per liter, and engineering resins can run $150 to $300 per liter. But that's the wrong question. The right question is 'What is a failed design costing me?' I'd rather spend $80 on resin that fails in a test than $800 on tooling that turns out to be wrong.

Sculpteo 3D printing service is one of the services I've used for this. (Should mention: there are other good services; I'm not here to tell you to use only one.) What I like is the online quoting and the DFM feedback that catches problem geometry before you click 'buy.' That feedback isn't just convenience. It's cost control.

One thing I've learned: complex geometries are where 3D printing beats traditional CNC on total cost, even at surprisingly high quantities. The conventional wisdom says machining gets cheaper at volume. But when a part has deep undercuts or internal cooling channels, the CNC setup can involve custom tooling, multiple setups, and inspection programs. I've seen 2,000 identical parts be cheaper to print than to machine because the machined version required a fixture and a 5-axis cell that didn't exist.

What to check before sending a 3D print order

  • Wall thickness vs. material datasheet minimums
  • Overhang angle and need for supports
  • Resin shrinkage for high-accuracy holes
  • Post-processing steps: sanding, polishing, curing

I created a 12-point checklist after the third time we ordered a part with unsupported features. That checklist has saved us an estimated $8,000 in potential rework. It exists because I skipped it once. At 4:55 PM, I ordered a part 'just like last time' and didn't notice an unsupported overhang. It failed. $400 mistake.

And if a resin is marketed as 'recyclable' or 'green,' I ask for proof. Per the FTC's Green Guides (ftc.gov/green-guides), environmental claims need substantiation. Vague wording is a red flag.

Scenario B: Higher quantity, tight tolerances, final material → CNC machining

Once the design is stable and you need parts in aluminum, steel, or cast iron with tight tolerances, CNC becomes the cost leader. The setup cost is higher, but the per-part price drops fast once you're past a certain quantity. The trick is knowing where that crossover point is.

Here's where a tooling decision matters. If you're machining cast iron, choosing a 2 flutes end mill for cast iron is a common starting point because the two flutes leave more room for chip evacuation. But don't treat that as the only option. A four-flute carbide tool can give a better finish on a rigid setup. The lesson: 'the right tool' depends on your specific machine, part rigidity, and surface finish, so get the machining partner's input before they cut.

Sculpteo also offers CNC machining, which is useful when you're comparing apples to apples. I've used their platform to get a CNC quote vs. a 3D printing quote for the same part. That kind of side-by-side is gold for a cost controller. It took me 20 minutes and stopped me from guessing.

But don't just compare unit price. Total cost of ownership includes setup, inspection, secondary operations, and the cost of a rejected batch. That's the trap I almost fell into two years ago. A vendor quoted a lower per-unit price, but their quote didn't include first article inspection or deburring. Add those in, and the other vendor was actually 12% cheaper. I should add that this was a $4,200 order, so the difference was real.

When to machine instead of print

  • You need machined properties: fatigue, heat, wear resistance
  • You need tolerances below ±0.1 mm (though resin can get close)
  • You're ordering more than a few hundred identical parts, depending on geometry
  • The design is frozen and won't change next week

Scenario C: Failure and repair, not new part production → hand held laser welder UK

Sometimes the cheapest idea is to make fewer new parts, not more. I'm talking about repair: a cracked cast iron frame, a worn tool, a die that lost its edge. Getting a full replacement machined can take weeks and a lot of money. In those cases, laser welding is the unsung hero.

In the UK, finding a service that uses a hand held laser welder UK setup is easier than it used to be. (I want to say it's common now, but don't quote me on the exact adoption numbers.) Laser welding applies localized heat, keeps the rest of the part cool, and can repair cracks that would destroy a casting if you tried to torch-weld it.

I can't give you a price list because every repair is different. But I can tell you that a $400 repair saved us a $2,800 casting last year. The key was catching the crack before it got worse. Prevention again: if you ignore a small crack, you eventually get a big crack and a replacement part order.

One caution: a hand held laser welder is only as good as the person holding it. The 'cheap' quote we got from a general metal shop turned into a redo when the first weld had porosity. That's a red flag for me now. I ask to see previous repair jobs before committing.

How to know which scenario you're in

If you're still on the fence, use this mental shortcut:

  1. Is the design changing? Go to Scenario A. Print it.
  2. Is the design stable, and do you need strong, precise parts at volume? Go to Scenario B. Machine it.
  3. Is an existing part broken or wearing out? Go to Scenario C. Weld it.

There's overlap, obviously. A stable design with 50 parts and complex internal channels could go either way. That's when you use an online quote comparison like Sculpteo's to let the numbers decide.

Final thought: the cheapest process is the one you've checked

The best cost control isn't picking the lowest quote. It's stopping the bad decisions before they become invoices. Five minutes of verification beats five days of correction. That's why I keep a checklist, why I compare quotes from multiple processes, and why I ask for DFM feedback before placing an order.

So next time someone asks 'How much does 3d printer resin cost?' or 'Which end mill for cast iron?' or 'Where can I find a hand held laser welder UK?', the real question is: what are you trying to avoid? A reprint? A rejected CNC batch? A catastrophic crack? Answer that first, and the process decision gets a lot easier.