Rush Orders for Springs, Stamping, and Machined Parts: 8 Questions We Answer Weekly

We answer these by phone almost every week, so here they are written down instead. These are the questions that come up when someone needs springs, stamped parts, or machined components on a timeline that doesn't leave room for a second attempt.

The short version of what's below: what a closed coil spring actually is, how fast custom precision metal stamping can realistically move, what drives torsion spring cost, when high speed CNC machining beats a die, and why CNC spare parts orders go sideways.

What is a closed coil spring, and when do you actually spec one?

A closed coil spring is a compression spring where the end coils sit tight against each other (meaning the last coil or two are wound flat, with no gap). You'll usually see it written as "closed ends," or "closed and ground" if those ends are also flattened so the spring sits square.

It's tempting to think this is just about winding the spring tighter. It isn't. The pitch along the working length can be identical to an open-end spring. What changes is how the spring bears load.

You spec closed ends when the spring has to sit flat in a pocket without cocking, when solid height matters (closed and ground ends give you a more predictable stacked height), or when the spring sees side load and you want to reduce buckling.

Where people get burned: they design a pocket around catalog dimensions, then discover the closed-end version has a different solid height than the open-end one they modeled. Not a catastrophe. Just a redesign at the worst possible moment.

How fast can custom precision metal stamping actually move?

Depends almost entirely on whether tooling already exists. Two very different answers:

Existing tool, material on the shelf: days. Sometimes 24–48 hours if the press is open and the coil is already there.

New tool: weeks, not days. Design, build, tryout, first article. You can compress that, but the risk of a second tryout never fully goes away.

In March 2024 a client called at 4:40 on a Wednesday needing 2,000 stamped brackets for a Friday line start. Existing tool, but the print called out 304 stainless and we had 301 on the floor. We pulled 304 from a local service center Thursday morning at a premium over their quoted price, ran the parts Thursday night, shipped Friday at 6 a.m. The client's alternative was pushing a product launch.

The bottleneck in custom precision metal stamping is almost never the press. It's certified material, inspection capacity, and whether the print is actually manufacturable as drawn.

What actually drives torsion spring cost?

Six things, roughly in order of impact:

• Wire diameter and material — chrome silicon and 17-7 PH cost meaningfully more than music wire, and not just per pound.
• Leg geometry — straight legs are cheap. Legs with multiple bends, hooks, or bends in two planes mean secondary operations, and secondary operations are where the money goes.
• Tolerance — a ±5° leg angle is a different job than ±1°.
• Number of coils — affects cycle time and wire consumption.
• Volume — setup and tooling amortized across the run.
• Finish — passivation, plating, or coating adds handling and lead time.

Granted, unit price is the number people compare. But on a 500-piece order, the line item that actually moves torsion spring cost is usually setup and secondary operations, not the wire itself.

Practical advice: ask for a breakdown that separates material, setup, run, and secondary operations. A single number tells you nothing about where there's room to negotiate.

What separates precision metal parts from "metal parts"?

Three things: the tolerance band, the evidence, and how stable the process is.

Tolerance band — ±0.005 in. on a 0.75 in. feature is a different world from ±0.030 in. Both can be correct. They are not the same job.

Evidence — first article inspection reports, CMM data, material certs, traceability. If a shop can't tell you what they measure and how often, you're buying hope.

Stability — can they hold the same tolerance on piece 1 and piece 50,000? Process control, tool-wear monitoring, and SPC on critical features are what make precision metal parts repeatable rather than lucky.

The standard "just match the drawing" advice ignores that drawings frequently don't state what actually matters. I've seen prints with tight GD&T on a cosmetic surface and nothing at all on a bore that carries load.

When does high speed CNC machining beat stamping?

Short answer: when volume is low enough that tooling won't amortize, or when the geometry is going to change.

Prototypes. Bridge production. Parts that will be revised. In that order.

High speed CNC machining (meaning high spindle speeds with light radial engagement, not simply a fast feed rate) earns its keep on thin walls, deep pockets, and harder materials, because heat leaves with the chip instead of soaking into the part. That's usually what keeps a tight-tolerance feature from drifting after the part cools.

Where it loses: any volume where a progressive die makes sense. Past a few thousand pieces a year the per-part gap gets ugly, and no amount of process optimization closes it.

Why do CNC spare parts orders show up late — or wrong?

The most frustrating part of this job: the same three or four failures, over and over. You'd think a print plus a STEP file would settle everything, but interpretation still varies wildly between shops.

Where it actually breaks:

• Model and print disagree on a critical dimension. Happens more than anyone admits.
• No datum structure, so every shop picks their own.
• Finish spec says "as machined" on a part headed into a wash-down environment.
• Material cert requirement shows up after the PO, not before.
• Nothing called out for burrs or edge break on a sealing face.

There's a pattern here. CNC spare parts are usually replacement components for something already in service, which means the print may be 15 years old and was never written for a one-off run.

Is the rush premium actually worth paying?

Yes — with one caveat about what you're buying.

What you're buying isn't speed. It's a date you can plan around. Those are different products. A shop that says "we'll try" has sold you speed. A shop that says "Thursday, 4 p.m., or the expedite is free" has sold you certainty.

Speed is a promise. A date is a commitment. They aren't the same product.

In 2023 we paid $800 in expedite fees on top of a $3,400 order of CNC spare parts. The alternative was a packaging line sitting down for a full shift. Downtime math isn't subtle — that decision paid for itself several times before lunch.

To be fair, not every order needs this. If the assembly start is three weeks out and you have float, standard lead time is the right call and the premium is pure waste.

The rule I use: pay for certainty when the cost of a missed date exceeds the premium. Most of the time it does by an order of magnitude. When it doesn't, don't.

What should I have ready before I call for a rush quote?

Five things. Having them cuts hours off the front end of the conversation.

1. Volume, and whether it's a one-time run or the start of a program.
2. The print and the model — and know which one wins if they conflict.
3. Which dimensions are functional and which are reference. This is the highest-leverage thing you can tell a shop.
4. Material, or permission to substitute.
5. Your real date, not your comfortable date. If the deadline can flex three days, say so. It changes the quote and it changes how the work gets sequenced.

That's it. Everything else we can figure out on the call.