Testing standard
Helical Compression Springs — Part 1 Design and Calculations for Springs Made from Circular Section Wire and Bar
Written and technically reviewed by Dak System Inc. engineeringLast reviewed
IS 7906 is the Indian multi-part standard for helical compression springs. Part 1 carries the design calculations, Part 2 the specification for cold coiled springs from wire up to 17 mm, and Part 5 the specification for hot coiled springs from bar of 8 to 60 mm. Acceptance rests on a static load test, in which the spring is first compressed three times to block length and then loaded gradually to a prescribed length, and on a spring rate taken from the characteristic curve between 0.3 and 0.7 of the force at the minimum permissible test length.
IS 7906 is a design and specification series with its test procedures written into the product parts. Part 1 carries the calculations and nothing else; Parts 2 and 5 carry the tolerances, the acceptance criteria and the way a spring is loaded to demonstrate them. Two measurements decide a spring: the static load test reads the axial force at a prescribed length, and the characteristic curve test finds the spring rate from the slope of force against deflection.
Which part applies depends on how the spring was made. Part 2 covers cold coiled springs from wire up to 17 mm diameter, mean coil diameter to 200 mm and unloaded length to 630 mm. Part 5 covers hot coiled springs from bar of 8 to 60 mm, hardened and tempered after coiling, outside diameter under 460 mm, unloaded length under 800 mm and a coil ratio of 3 to 12. Above a lot size of 5000, dimensional tolerances come from Part 7.
The spring rate is measured over a defined band and the band is the whole point. Part 5 requires that where the rate is checked by finding the characteristic of the spring, it be done over the range 0.3 to 0.7 of the force at the minimum permissible test length. Part 2 says the same in its 1975 wording.
The standard gives its own reason. The theoretical force-deflection diagram is a straight line, but in practice the start and finish of the characteristic depart from linearity — end coils bedding in at one end, coils approaching one another at the other. Taking the rate from the middle forty percent covers the linearity with certainty. A rate measured across the full stroke is a chord through two curved regions, and it is not the spring rate.
Tolerances are applied only where they matter. The spring rate is toleranced only if it has a decisive influence on functional behaviour, and then exactly one additional spring force is toleranced alongside it.
The specimen is the finished spring. What the standard controls is its own geometry rather than the machine it is tested in.
IS 7906 specifies no test speed anywhere. Parts 1, 2 and 5 were all searched for one. What the standard gives instead is a sequence and a tolerance on the instrument.
R = (F₂ − F₁) / (L₁ − L₂) = ΔF / Δs
Taken between 0.3 Fₙ and 0.7 Fₙ. Outside that band the real characteristic is not straight, so the slope is not the spring rate.
ΔR = ±0.065 (D/d + 1) R
Part 5, clause 7.6. Applied only where the spring rate has a decisive influence on functional behaviour, and then exactly one additional spring force is toleranced alongside it.
ΔR = ±0.045 (D/d + 1) R
The tighter coefficient reflects the better dimensional control of a ground bar.
Lc ≤ (nₜ − 0.3) d
From ground bars, Lc ≤ (nₜ − 0.4) d_max. Other end forms have their own expressions in Part 5, clause 7.7.
±0.015 for rolled-surface rods, ±0.012 for machined-surface rods
The two coefficients and the rolled-versus-machined split are confirmed; the exact algebraic grouping was garbled in the source text layer and is not reproduced.
Do not measure the spring rate across the full stroke. The characteristic departs from linearity at both ends — end coils bedding in at one, coils approaching one another at the other — and a slope taken through those regions is not the spring rate. The 0.3 to 0.7 band exists to cover the linearity with certainty, and the standard says so in those words.
Direct compression platens with the spring standing vertically. The standard sets no platen flatness figure, controlling the spring's end squareness and parallelism instead, so what matters on the machine is that the platens stay parallel and the spring stays on axis.
Specifications
Hardened parallel anvils for smaller cold coiled springs. Any spring with a slenderness ratio above five needs a guide or cage, since the standard itself notes that closer squareness and parallelism are only achievable below that.
SpecificationsIS 7906 specifies no test speed. There is no crosshead rate and no loading rate anywhere in Parts 1, 2 or 5. What it gives instead is a procedure and one instrument requirement: gradually approach the prescribed load length and read the corresponding axial load, with an instrument error of ±1 percent in the load indication allowed. Any millimetre-per-minute figure attributed to IS 7906 was invented somewhere along the way.
The standard likewise sets no platen flatness tolerance, controlling the spring's end geometry instead of the machine's platens.
Force capacity follows from the spring. Hot coiled bar up to 60 mm in automotive and rail suspension service reaches high loads, so the machine is chosen from the springs in the programme. A compression cage or guide is needed for any spring slender enough to buckle, since Part 2 notes closer squareness and parallelism are only achievable where the slenderness ratio is under five.
The spring rate band is the error that matters, and one version circulates widely: that the rate is measured between twenty and eighty percent of the working range. It is not. Both Part 2 and Part 5 say 0.3 to 0.7 of the force at the minimum permissible test length. A wider band pulls the non-linear ends into the slope and reports a rate the spring does not have.
Skipping the pre-scragging is the second. Three compressions to block length are part of the method, and omitting them means measuring a spring that has not yet settled.
The third is checking solid height as routine. The standard says it may not be specified as a rule and in the normal case is not to be checked — it becomes a requirement only when the application demands it, and then only as a maximum.
| Part | Covers | Applies to |
|---|---|---|
| Part 1 : 1997 | Design and calculations | Any helical compression spring from circular wire or bar, cold or hot coiled, at room temperature |
| Part 2 : 1975 | Cold coiled springs — specification | Wire up to 17 mm, mean coil diameter to 200 mm, unloaded length to 630 mm |
| Part 5 : 2004 | Hot coiled springs — specification | Bar 8 to 60 mm, hardened and tempered after coiling, outside diameter under 460 mm, coil ratio 3 to 12 |
| Part 7 | Dimensional tolerances | Lots exceeding 5000 |
Part 1 supplies the calculations that Parts 2 and 5 assume, so a specification part is not usable on its own. The committee also changed between them — Part 1 sits with the Springs Sectional Committee and Part 5 with the Automotive Springs and Suspension Sectional Committee.
Part 5, Table 4. Ground ends buy tighter geometry, and the tolerances say by how much.
| Forged or rolled flattened ends | Ground ends | |
|---|---|---|
| Squareness e₁ | 0.03 L₀ (about 1.7°) | 0.01 L₀ (about 0.57°) |
| Parallelism e₂ | 0.025 Dₑ (about 1.5°) | 0.015 Dₑ (about 0.9°) |
| Rate tolerance coefficient | ±0.065 (D/d + 1) R | ±0.045 (D/d + 1) R |
| Total turns tolerance | ±0.015 nₜ | ±0.012 nₜ |
A spring out of square loads its platen on one side, which both biases the force reading and encourages buckling. That is why the standard tolerances the spring's ends and says nothing at all about platen flatness.
It is the Indian Standard series for helical compression springs. Part 1 covers design and calculations for springs made from circular section wire and bar. Part 2 specifies cold coiled springs and Part 5 hot coiled springs, each with their own scope, tolerances and test procedures. Part 7 carries dimensional tolerances for large lots.
Between 0.3 and 0.7 of the force at the minimum permissible test length — the middle forty percent of the working range. Both Part 5 clause 10.2 and Part 2 clause 7.2 say so, and Part 5's characteristic curve figure is labelled 0.3 Fₙ and 0.7 Fₙ on the force axis and 0.3 sₙ and 0.7 sₙ on the deflection axis.
No. That figure circulates but appears nowhere in IS 7906. The standard says 0.3 to 0.7 of the force at the minimum permissible test length, in two separate parts and in two separate decades of wording. Measuring over a wider band drags the non-linear ends of the characteristic into the slope and reports a rate the spring does not have.
Because the real characteristic is not straight at either end. The standard explains it directly: the theoretical force-deflection diagram of a cylindrical helical compression spring is a straight line, but in practice the start and finish depart from linearity, as end coils bed in at one end and coils approach one another at the other. The 0.3 to 0.7 band is chosen so that the linearity is covered with certainty.
None. There is no crosshead rate and no loading rate in Part 1, Part 2 or Part 5. What the standard prescribes instead is a sequence: compress the spring three times in quick succession to block length, or to the length corresponding to the maximum permissible static stress if that is greater, then gradually approach the prescribed load length and read off the axial load. An instrument error of ±1 percent in the load indication is allowed. Any millimetre-per-minute figure attributed to IS 7906 has been invented.
Because an unscragged spring has not yet taken its permanent set. The first few compressions to block length settle it, and the standard asks that after them further scragging produces no further change in height. A force read before that settling has happened describes a spring that will not exist after the first few working cycles in service.
No, and it is worth knowing that it does not, because the question comes up constantly. The standard controls the geometry of the spring rather than of the machine — squareness of the ends to 0.01 of the unloaded length for a ground-ended hot coiled spring, parallelism to 0.015 of the outside diameter, and bow to half the out-of-squareness tolerance within the middle third of the spring. Do not quote a platen flatness figure for IS 7906.
Normally no. Part 5 states that the solid height may not be specified as a rule on the spring drawing, and that in the normal case it is not to be checked. It becomes a requirement only where the application depends on it, and then it is specified as a maximum value.
They occupy the same ground in different jurisdictions. EN 13906-1 is the European design standard for helical compression springs, and DIN 2095 and DIN 2096 the German specification pair for cold and hot coiled springs respectively — a split that mirrors the Part 2 and Part 5 division in IS 7906. ASTM A125 covers heat-treated helical steel springs but ASTM now records it as historical rather than active. The detailed technical differences were not verified and none are asserted here.
Part 1 supplies the calculations Parts 2 and 5 assume, so a specification part is not usable without it. Part 7 carries dimensional tolerances for lots above 5000. EN 13906-1 is the European design standard for helical compression springs, and DIN 2095 and DIN 2096 the German pair for cold and hot coiled springs. ASTM A125 covers heat-treated helical steel springs, though ASTM records it as historical. IS 4454 specifies the steel wire cold coiled springs are made from.
Dak verifies against whichever standard the method names, and where a class applies our frames sit a class tighter than it asks.
| The method asks for | Dak supplies | |
|---|---|---|
| Capacity | Follows from the spring. Hot coiled bar up to 60 mm in automotive and rail suspension service reaches high loads, so the machine is chosen from the springs in the programme rather than from the standard. | Load cells from 1 kg to 60 ton on the Series 7200, and 0.5 to 100 kN on the Series 9000 |
| Force accuracy | none stated as a class, but an instrument error of plus or minus 1 percent in the load indication is expressly allowed for the static load test | ISO 7500-1 Class 0.5, verified to ASTM E4, DIN 51221 and BS 1610 |
| Gripping | Parallel compression platens with the spring standing vertically, plus a guide or cage for any spring slender enough to buckle. The standard tolerances the spring end geometry rather than the platens and sets no platen flatness figure. | Our compression anvils, built to the specimen |
| Environment | Ordinary room temperature. Part 1 states that for operation at considerably higher or lower temperature the spring manufacturer should be consulted. | 3009 series chambers, −150 °C to +400 °C — temperature only |
This page describes the method as practised. The governing text is the current edition from the issuing body. Tell us what you are testing and we will answer with the machine, the fixture and a quotation.
