Initial Tension in Extension Springs: Setting It Right
Short answer: Initial tension is the internal force a close-wound extension spring already carries before any load is applied — it is what keeps the coils pressed together. Typical values run 5–15% of the spring's rated load, though 2–25% is seen in practice depending on index and material. Set it too low and the spring rattles and loses free length; set it too high and the coils bind, the hook stress rises, and the spring may not extend at all within the working travel. Specify it as a force (N or lbf) at zero deflection, not as a percentage, then verify on a load-deflection curve where the first segment is offset from the origin.
What exactly is initial tension?
A close-wound extension spring is not a relaxed part. During coiling, the wire is fed under tension and the coils are wound tight against each other. When the spring is released from the mandrel, the wire's own elastic recovery tries to unwind the helix, but the coils are already touching — so instead of expanding, the spring stores that recovery as a residual torsional stress. The result is a spring that behaves like a rope under tension: it resists separation of its ends until the applied load exceeds the internal preload.
On a load-deflection graph, that shows up as a straight line that does not pass through the origin. Extrapolate the linear region back to zero deflection and it intersects the force axis at a positive value. That intercept is the initial tension, usually written P₁ or Fᵢ.
Two consequences matter for design:
- The spring does not extend at all below P₁. If your mechanism needs the spring to start moving at 0.5 N and your initial tension is 2 N, the spring is a rigid link until you hit 2 N.
- The rate (spring constant) is unaffected. Initial tension shifts the curve up; it does not change its slope. Rate is governed by wire diameter, coil diameter and active coils.
This is the key difference between extension springs and compression springs, which are normally wound with a gap between coils and have no equivalent preload — their force curve genuinely starts at zero.
Why does initial tension matter more than most engineers expect?
Because it is the parameter most often left off a drawing, and the one that most often causes field failures that look like something else.
| Symptom in the field | Actual root cause | What was usually missing on the drawing |
|---|---|---|
| Spring "rattles" or hooks knock at rest | Initial tension too low; coils not held together | No P₁ value, no minimum |
| Spring will not extend until load is well above spec | Initial tension too high | No maximum P₁, or a percentage that was never converted to a force |
| Free length drifts longer after a few cycles | Coils relaxed; low P₁ combined with high index | No free length tolerance tied to P₁ |
| Hook breaks at the first coil | High P₁ concentrating stress at the hook bend | No hook radius or stress limit specified |
| Spring extends but force is low across the whole range | Rate wrong, not initial tension | Rate and P₁ conflated into a single "load at length" |
The last row is the most common misdiagnosis. A buyer specifies "10 N at 30 mm extension" and receives a spring that reads 10 N at 30 mm — but with the wrong split between rate and initial tension. It passes incoming inspection and then behaves badly in the assembly, because the force at 5 mm extension is completely different from what the mechanism needs.
How is initial tension created during manufacturing?
Initial tension is not a separate operation. It is a by-product of how the spring is coiled, and it is controlled by four things on the coiler:
1. Wire feed tension. Higher feed tension during coiling produces more residual stress and higher P₁. This is the primary dial.
2. Coiling speed and pitch. Extension springs are coiled close-wound, so pitch is nominally zero; speed affects how much the wire work-hardens as it passes the feed rolls.
3. Mandrel diameter relative to coil ID. A tighter wind relative to the finished ID leaves more recovery stress locked in.
4. Stress relief / heat treatment. A low-temperature stress relief after coiling relaxes some of the residual stress. More time or temperature means lower P₁. This is why P₁ must be specified on the drawing and the heat treat condition must be fixed — otherwise the same spring design can come back at 3 N or 9 N from two different runs.
For a springmaker, hitting a target P₁ is a matter of setting feed tension and then confirming with a pull test. It is a controllable process, but only if the target is stated as a number with a tolerance. "Close wound" is not a specification.
The relationship between initial tension and spring index
Spring index (D/d, mean coil diameter divided by wire diameter) strongly influences how much initial tension is practical. Low-index springs — say D/d of 4 to 6 — are stiff and can hold high initial tension without the coils slipping. High-index springs, D/d above 12, are wound from relatively thin wire around a large diameter, and the residual stress needed to hold the coils closed is limited by the wire's yield strength.
A practical rule of thumb for what is achievable:
| Spring index (D/d) | Typical achievable initial tension | Notes |
|---|---|---|
| 4–6 | 15–25% of rated load | Stiff, robust, hook stress is the limiting factor |
| 6–10 | 8–15% of rated load | The sweet spot for most industrial extension springs |
| 10–14 | 4–10% of rated load | Thin wire; P₁ becomes sensitive to heat treat |
| Above 14 | 2–6% of rated load | Difficult to hold; consider a different mechanism |
These are indicative ranges, not guarantees — the achievable value also depends on material and wire tensile strength. If you need a specific P₁ outside these bands, it is worth a conversation before the drawing is frozen. The underlying stress relationships are covered in more depth in our article on spring index and stress.
How do I specify initial tension on a drawing?
Specify it as a force at zero deflection, with a tolerance, and state the test method. A complete extension spring call-out includes:
- Free length (with tolerance)
- Outside or inside diameter (with tolerance)
- Wire diameter
- Active coils or total coils
- Initial tension P₁ in N or lbf, with a tolerance band
- Rate in N/mm or lbf/in
- Load at one or two reference lengths (a cross-check)
- Hook type, hook radius, and hook orientation
- Material and finish
- Heat treat / stress relief condition
A worked example makes the arithmetic clear. Suppose you need 12 N at 40 mm extension, and you want the spring to begin moving at roughly 1.5 N.
- Required rate = (12 − 1.5) / 40 = 0.2625 N/mm
- Initial tension P₁ = 1.5 N ±0.4 N
- Cross-check load at 20 mm = 1.5 + (0.2625 × 20) = 6.75 N
Now the springmaker has two independent constraints that pin down both the geometry and the coiling setup. If you only gave the 12 N at 40 mm figure, there would be an infinite family of rate/P₁ combinations that satisfy it — and most of them would be wrong for your mechanism.
Choosing the tolerance band
Tighter is more expensive, and P₁ is inherently more variable than rate. As a working guide:
| Application | Suggested P₁ tolerance | Rationale |
|---|---|---|
| Static tensioner, non-critical | ±25% | Only needs to hold parts together |
| General industrial mechanism | ±15% | Balances function and cost |
| Precision return / sensor spring | ±10% | Position repeatability matters |
| Instrumentation, low-force actuation | ±7% or tighter | Requires 100% inspection, higher unit cost |
Tolerances tighter than ±7% are achievable but should be discussed early, because they usually mean screening every piece on a load tester rather than sampling.
How is initial tension measured and verified?
The standard method is a load-deflection pull test. The spring is mounted on hooks or pins, extended slowly, and force is recorded at several lengths. Initial tension is then found by fitting a straight line to the linear portion of the curve and reading the force-axis intercept.
Practical points that determine whether the number is meaningful:
- Use the right grip. Hooks must sit on pins of the correct diameter. A hook that is squeezed or side-loaded reads low and gives a false P₁.
- Test at the specified rate of extension. Pulling fast adds a dynamic component. Standard practice is a slow, quasi-static pull.
- Measure over the linear region only. The first millimetre or two can include hook seating and coil separation; the last portion may approach yield. Fit the middle.
- Check free length on the same sample. Low P₁ and long free length usually travel together and confirm each other.
- Sample size matters. For ±15% tolerance, a 5-piece sample from the first article plus periodic in-process checks is normally adequate. For ±7%, plan on 100% testing.
If you do not have a load tester, a simple comparative check works for incoming inspection: hang a known mass from the spring and measure the extension. If the spring extends by x mm under a load that should produce y mm, the discrepancy tells you whether P₁ or rate is off — provided you take two readings at two different loads. One reading alone cannot separate the two.
When should I increase or decrease initial tension?
Adjust it deliberately, not by trial and error.
Increase initial tension when:
- The spring must hold a mechanism against vibration or rattle at rest
- You need the spring to act as a light preload or anti-backlash element
- The assembly must not move until a threshold force is reached (a detent-like behaviour)
- Free length must stay stable over many cycles
Decrease initial tension when:
- The mechanism must respond to very small forces
- Hook stress is already near the limit — high P₁ concentrates stress at the first coil after the hook
- The spring operates near its maximum extension, where the added P₁ pushes total stress toward yield
- You are seeing fatigue failures at the hook rather than in the body
The last point is worth emphasising. Initial tension adds directly to the stress in the hook and the first active coil, which are already the highest-stressed regions of an extension spring. A spring that is fine at 10% P₁ can fail at the hook at 20% P₁ with no change in the working load. If you are chasing a fatigue problem, our breakdown of spring overload and yield covers how to tell a preload problem from a genuine over-stress problem.
How does material choice affect achievable initial tension?
Material sets the ceiling. Initial tension is stored elastic energy, so it scales with the material's yield strength and shear modulus.
| Material | Relative achievable P₁ | Notes |
|---|---|---|
| Music wire (ASTM A228) | Highest | High tensile, best for small, stiff springs |
| Oil-tempered MB / hard-drawn | High | General-purpose industrial extension springs |
| Chrome-silicon | High | Good fatigue life; often used for higher-stress designs |
| 302 / 304 stainless | Moderate | Lower tensile than music wire; corrosion resistance trade-off |
| 316 stainless | Lower | Best corrosion resistance, lowest achievable P₁ |
| Phosphor bronze / beryllium copper | Low to moderate | Non-magnetic or conductive applications |
If you switch from music wire to 304 stainless for corrosion reasons, expect the same geometry to deliver noticeably lower initial tension — and plan to re-tune the coiling setup. Wire diameter selection interacts with all of this; our guide to spring wire diameter selection walks through the trade-offs.
How BQUQ handles extension spring initial tension
BQUQ runs four production lines in one Dongguan factory — CNC machining to ±0.005 mm, metal stamping, custom springs, and heat sink production — under ISO9001. Extension springs are coiled, stress-relieved, and pull-tested in-house, so the P₁ you specify is the P₁ you receive.
What that means in practice for a custom extension spring order:
- Send the drawing with P₁ stated as a force and a tolerance; if you only have "load at length", we will help you split it into rate and initial tension
- Flexible MOQ — prototype and short-run quantities are welcome, so you can validate P₁ in the real assembly before committing to volume
- Quotes in 12 working hours, including a manufacturability note if your target P₁ falls outside what your chosen index and material can realistically deliver
- First-article load-deflection data supplied on request, so you can verify the intercept yourself
If you are also sourcing other spring types for the same assembly, we produce torsion springs and compression springs on the same floor, which keeps tolerances and material batches consistent across the bill of materials.
Frequently Asked Questions
Q: What is a typical initial tension value for an extension spring?
A: Most industrial extension springs are designed with initial tension between 5% and 15% of the rated load at maximum extension, with 8–12% being the most common band. Low-index springs (D/d of 4–6) can hold 15–25%, while high-index springs above D/d 14 are usually limited to 2–6%. Treat these as indicative starting points, then confirm against your actual mechanism.
Q: Can initial tension be zero in an extension spring?
A: Yes, but it requires a deliberately open-wound extension spring with a gap between coils. Such a spring has no preload and its load-deflection curve passes through the origin. The trade-off is that the coils can shift and the free length is less stable, so open-wound extension springs are used mainly where very low actuation force matters more than stability.
Q: Does initial tension change the spring rate?
A: No. Initial tension shifts the load-deflection line up or down along the force axis but does not change its slope. Spring rate depends on wire diameter, mean coil diameter, active coils and the material's shear modulus. This is why you must specify both values independently — a single "load at length" figure cannot distinguish between them.
Q: Why did my extension spring come back with the wrong initial tension?
A: Usually because the drawing specified only a load at a given extension, leaving the rate and initial tension split undefined. Other causes include an unspecified heat-treat condition, a material substitution for corrosion resistance, or a hook geometry that changes the effective number of active coils. State P₁ as a force with a tolerance and fix the material and stress relief.
Q: How do I check initial tension without a load tester?
A: Take two extension readings at two known loads on the linear part of the curve. The slope between them gives the rate, and extrapolating that line back to zero extension gives the initial tension intercept. One reading alone cannot separate rate from initial tension. For ±7% tolerances, though, a calibrated pull tester is the only reliable route.
Related Resources
- About BQUQ and our four production lines in Dongguan (/about/)
- Custom extension springs, compression springs and torsion springs (/extension-custom-springs/)
- Industry trends in spring and precision component sourcing (/industry-dynamics/)
- Technical articles on spring design and manufacturing (/bquq-blog/)
- Frequently asked questions on ordering custom springs (/faq/)
- Case studies from BQUQ customers (/case/)
- Contact the engineering team for a 12-hour quote (/contact/)
Authored by the BQUQ Engineering Team. BQUQ (Dongguan) runs CNC machining (±0.005 mm), metal stamping, custom springs, and heat sink production in one ISO9001 factory. Source-direct from Dongguan, China — quote in 12 hours: sc@bquq.com | WhatsApp +86 13713157787 | www.bquq.com
