Spring Material Selection: A Decision Tree That Gets It Right the First Time
Material mistakes are the most expensive spring mistakes — they surface in the field, not in inspection. This decision tree prevents most of them.
Question 1: What touches the spring?
Dry indoor air → carbon steels stay in play. Humidity, water, cleaning agents → stainless 302/304. Salt, chlorides, chemicals → 316 or beyond. Body contact/implant → 316LVM or titanium.
Question 2: How hot?
- < 120°C: any material
- 120–245°C: chrome silicon, stainless
- 245–340°C: 17-7PH
- 340–550°C: Inconel X-750
Question 3: How many cycles?
- Static or < 10⁴ cycles: whatever passed Questions 1–2, cheapest first
- 10⁴–10⁶: music wire or Cr-Si with margin; consider shot peening
- > 10⁶: Cr-Si valve-spring grade or 17-7PH, shot peened, with fatigue design review
Question 4: Any special physics?
Conductivity → copper alloys. Non-magnetic → phosphor bronze, BeCu, titanium (details). Weight → titanium. None → stop upgrading; take the cheapest survivor.
The meta-rule
Upgrade one dimension at a time, and only the dimension that failed. "Stainless everywhere, just in case" doubles cost across a product for corrosion resistance most parts never use; "music wire everywhere" puts rust in your bathroom products. State environment, temperature, and cycles on every drawing — then the RFQ review can confirm the cheapest material that genuinely survives.
Frequently Asked Questions
What material should I default to if I know nothing else?
For dry indoor mechanisms: music wire (plated if visible). For anything that sees water or people: stainless 304. These two defaults cover the large majority of industrial springs.
How much does material choice change spring price?
Typical multipliers versus music wire: stainless ~1.5–2×, Cr-Si ~1.3–1.8×, 17-7PH ~3–4×, BeCu ~5–8×, Inconel/titanium 10–40×. Material is usually 30–60% of spring cost, so multipliers matter.
Can I switch material without changing the drawing?
Rarely. Different moduli and strengths change rates and safe stress. Treat material changes as engineering changes, and let the manufacturer re-run the numbers.