Constant-Force Retention
Constant-Force Retention in bolted joints
By ISOKLAMP Engineering, Inc. Editorial Team · Updated
What Constant-Force Retention means
Every bolted joint loses clamp force. The industry has spent seventy years building devices that stop the nut from turning, and they work. What none of them does is put back the clamp force a joint loses when the nut never turns at all.
Constant-Force Retention is the principle that closes that gap. A CFR element does two distinct jobs at once:
- Arrest rotation. A wedge geometry whose cam ramp angle exceeds the thread helix angle makes it mechanically impossible for the fastener to back off without lifting the entire clamp load.
- Recover length. A self-locking helical take-up, biased by a constant-torque spring, advances whenever the stack shortens, converting rotation of a drive ring into axial extension that refills the lost clamp length.
The second job is the one that is new. It is why CFR is a principle rather than a product feature.
Reference joint used throughout
Every figure in this article refers to the same reference joint, so numbers are comparable across articles and against your own calculations.
| Parameter | Value |
|---|---|
| Bolt | M16 × 2,0, property class 10.9, to ISO 898-1 |
| Assembly preload F_V | 70,0 kN |
| Clamp length | 48 mm, steel on steel |
| Bolt stiffness k_S | 1,04 × 10⁹ N/m |
| Member stiffness k_P | 5,71 × 10⁹ N/m |
| Load factor Φ | 0,154 |
| Transverse test | DIN 25201-4:2010-03 Annex B, 2 000 cycles, ±0,45 mm slip |
Stiffnesses are calculated to VDI 2230 Sheet 1 using the standard cone-of-compression method.
The governing conditions
Three conditions define a working CFR element. All three must hold simultaneously.
Anti-rotation. The cam ramp angle α_c must exceed the thread helix angle β:
α_c > β
For an M16 × 2,0 thread, β = 1,47°. The ISK-16 uses α_c = 4,5°, a margin of just over 3:1.
Self-locking. The take-up ramp must not back-drive under load. This requires:
tan α_c < μ_r
With a ramp friction coefficient μ_r of 0,14 for the passivated stainless pairing, tan 4,5° = 0,079. The margin is 1,8:1, which holds across the full specified friction range.
Self-loosening torque. The off-torque a preloaded joint generates is:
M_self = F_V · P / 2π
At F_V = 70,0 kN and P = 2,0 mm, M_self = 22,3 N·m. The CFR wedge must resist this continuously, not intermittently.
Why constant force rather than spring force
A Belleville washer is the obvious comparison, and the comparison is instructive. A disc spring stores energy on a load-deflection curve. As it relaxes to recover joint length, the force it applies falls along that curve. Recover 0,2 mm and you have given up a large fraction of the force you were holding.
A CFR element runs on a quasi-zero-stiffness plateau. The constant-torque spring is wound so that its output torque is flat across the working travel, which means the axial force the element delivers is effectively independent of how much take-up it has already consumed.
| Belleville stack | CFR element | |
|---|---|---|
| Force across travel | Falls along the spring curve | Flat across the plateau |
| Travel available | 0,15 to 0,4 mm typical | 0,50 mm |
| Force at end of travel | Substantially reduced | Unchanged |
| Arrests rotation | No | Yes |
| Reserve is observable | No | Yes, via the indicator window |
Comparative figures for disc springs are reconstructed from published manufacturer load-deflection data. See the test data page for the source list.
What the analysis shows
Finite-element analysis of the ISK-16 geometry, combined with closed-form VDI 2230 Sheet 1 joint mechanics, gives the following for the reference joint under DIN 25201-4 Annex B transverse loading:
| Securing method | Residual clamp force at 2 000 cycles |
|---|---|
| ISOKLAMP CFR (ISK-16) | 99,4 % |
| Wedge-lock washer pair | 93,1 % |
| Ribbed flange nut | 88,4 % |
| Prevailing-torque (nyloc) | 57,2 % |
| Conical spring washer | 44,6 % |
| Plain washer, no securing | 12,8 % |
ISOKLAMP figures are predicted by finite-element analysis of the ISK-16 geometry. Baseline figures are calibrated against published test data, sourced individually on the test data page.
The number that matters is not the headline. It is the shape of the curve. Every other method on that table decays monotonically. The CFR curve decays for roughly the first 120 cycles, then recovers and flattens, because the take-up is refilling clamp length faster than the joint is losing it. That signature is the whole point of the principle.
Where CFR earns its place
CFR is worth the part cost wherever the joint loses length rather than turns:
- Soft or creeping interfaces. PTFE and elastomer gaskets, polymer spacers, composite flanges.
- Coated faces. Zinc flake, PTFE-impregnated, and electroless nickel all creep under contact stress.
- Thermally cycled joints. CIP and SIP washdown, busbar duty, anything ratcheting between aluminium and steel. See CIP and SIP thermal cycling and busbar joints.
- Joints where retorquing is expensive or impossible. Sealed enclosures, subsea, potted assemblies, anything behind a guard that requires a lockout to reach.
Where a joint is steel on steel, dry, at constant temperature and lightly loaded, a wedge-lock washer remains the right answer and costs less. We will tell you so.
Terminology
Constant-Force Retention (CFR) is the principle. ISOKLAMP CFR is the implementation. Vibration-Actuated Take-up (VAT) is the specific mechanism by which the drive ring advances. Preload reserve is the remaining unused take-up travel, readable through the indicator window without instruments.
Read the mechanism in detail on how it works, or the derivation in the technical report.
Frequently asked questions
What does Constant-Force Retention mean in a bolted joint?
Constant-Force Retention is a bolted-joint principle in which a constant-torque spring element continuously biases a self-locking helical take-up. When the clamped stack shortens through embedment, coating creep or gasket relaxation, the take-up advances and restores the lost clamp length. The force applied stays effectively constant across the full 0,50 mm of travel rather than falling along a spring curve.
How is CFR different from a Belleville or disc spring washer?
A disc spring stores energy on a load-deflection curve, so the force it applies falls as it extends to recover joint length. A CFR element operates on a quasi-zero-stiffness plateau, so force is independent of how much travel has been consumed. A CFR element also arrests rotation, which a disc spring does not do at all.
What is the difference between CFR and a wedge-locking washer?
A wedge-locking washer arrests rotational self-loosening and does that job well. It has no mechanism for recovering clamp force lost when the nut does not rotate, which accounts for 10 to 40 percent of preload in the first 200 load cycles of a typical joint. CFR does both jobs: it arrests rotation using the same wedge principle and adds a helical take-up that recovers lost clamp length.
What are the governing equations for a CFR element?
Three conditions must hold. Anti-rotation requires the cam ramp angle to exceed the thread helix angle, α_c > β. Self-locking requires tan α_c < μ_r, the ramp friction coefficient. The element must continuously resist the self-loosening torque M_self = F_V·P/2π, which is 22,3 N·m for an M16 × 2,0 joint at 70 kN preload.
Which standards apply to CFR testing?
DIN 25201-4:2010-03 Annex B defines the transverse vibration (Junker) test and its 80 percent residual preload criterion at 2 000 cycles. ISO 16130:2015 covers aerospace dynamic testing. VDI 2230 Sheet 1 governs the joint stiffness and load factor calculations. DIN EN ISO 16047 covers torque-clamp force testing.
Take it further
- The ISOKLAMP technical report covers the full derivation, the geometry, and every dataset behind these figures.
- Full residual clamp force dataset gives the residual clamp force numbers for ten securing methods, with sources.
- Contact sales puts you in touch with the engineering team. Bring us a joint that keeps failing and we will run a VDI 2230 Sheet 1 analysis on it.
Engineering questions go to engineering@isoklamp.com. An engineer answers, not a form.