Home › Blog › EV & Battery › Compression & Cushioning › CFD plateau, set & relaxation
The CFD Plateau, Compression Set, and Stress Relaxation: Why Compression Pads Hold or Lose Their Force
A compression pad is a spring with a memory. The compression-force-deflection (CFD) curve says what force it pushes back with today; compression set says how much of its thickness it will permanently surrender after years of being held compressed; stress relaxation says how much force it will quietly let go of even when the thickness survives. A pad that is specified on only the first of those three is specified for the day the module is built, not for the decade the pack has to live.
This article explains the three behaviors, the test methods behind each (ASTM D3574 / ISO 3386 for CFD, ASTM D395 and ASTM D3574 for set, ASTM D6147 / ISO 3384 for relaxation), and the design rules that follow. The companion material guide covers which foam family delivers them; here the subject is the behavior itself.
Request a quote for your project →
The duty: two thickness clocks running at once
A lithium-ion cell changes thickness in two ways, on two time scales. Reversible breathing is the small, repeating change each charge-discharge cycle as lithium moves into and out of the graphite anode: slightly thicker on charge, recovered on discharge. Irreversible swelling is the slow, permanent growth over life from gas generation and solid-electrolyte-interphase growth; it never comes back, and it commonly leaves cells about 10 to 20% thicker at end of life.
The pad between cells has to serve both clocks at once. It flexes a little with every cycle, and it absorbs the one-way growth over years, all while keeping its push-back force inside the window the cell maker recommends. Pouch cells, which have no rigid housing, depend on that external force most heavily; prismatic cells are stiffer but still benefit from controlled stack pressure. That dual demand is exactly why the three behaviors in this article, plateau shape, set, and relaxation, are the specification, and a single firmness number is not.
Behavior one: the CFD curve and its plateau
Compression force deflection, measured per ASTM D3574 Test C (the related ISO method is ISO 3386), reports the stress a foam pushes back with at a given percent compression. For microcellular foams the curve has a characteristic three-region shape: a roughly linear-elastic rise at low compression, a wide plateau through the working zone where force barely changes with deflection, and a steep densification region where the cell walls have collapsed and force climbs fast.
The Three Regions of a Microcellular CFD Curve
Qualitative shape only, drawn from the ASTM D3574 Test C three-region description; no force or deflection values are represented. Confirm the measured curve for your grade and gauge on the manufacturer's current technical data sheet.
The design rule that falls out of the shape: choose grade and thickness so the deflection at the beginning-of-life gap and the deflection after end-of-life swelling both land on the plateau. Run off the left edge and the force falls below the window early; run into the densification knee and the force spikes as the cell grows.
A slightly thicker pad on a flatter grade often holds the window across the whole life better than a thin pad that exhausts its plateau. This is also why PORON microcellular polyurethane and the PORON EVExtend battery grades are specified by curve rather than by durometer: the families are engineered for plateau flatness, and grade selection is the act of placing your window on it (the PORON EVExtend product page lists the current battery-grade options).
Behavior two: compression set, the thickness you never get back
Compression set is the permanent thickness a foam loses after being compressed, held, and released: the unrecovered fraction of the deflection, expressed as a percentage. Lower is better. A pad that takes set thins under sustained load, pushes back with less force at the same stack dimension, and can eventually go slack, letting the cells move and electrode layers lose contact.
Three method details matter when you read a data sheet:
- The method family. Rubber and silicone foams commonly cite ASTM D395 (Method B, constant deflection, is the most specified); flexible polyurethane foams cite ASTM D3574.
- The reporting basis. ASTM D395 and D3574 report set against the original height, while ASTM D1056 reports against the compressed height. Values from the two bases are not directly comparable, so a smaller number under one method is not automatically the better foam.
- The conditions. Set is reported at a stated temperature, deflection, and duration. A value earned at room temperature says little about a pad that lives warm; match the conditions to your duty before comparing.
Behavior three: stress relaxation, the force that leaks away
Stress relaxation is the companion problem: hold a foam at a constant deflection and the force it exerts decays over time, even before any permanent thickness loss shows up. It is measured as force decay under constant deflection per ASTM D6147 and ISO 3384. Heat accelerates it, and so does vibration, both of which a traction pack supplies in abundance.
The distinction from set is worth keeping sharp, because the two are often blurred into one "aging" bucket:
| Behavior | What is held constant | What decays | Governing methods | Design consequence |
|---|---|---|---|---|
| Compression set | Deflection, then released | Recovered thickness | ASTM D395 / ASTM D3574 (basis differs in ASTM D1056) | Pad thins; force window drops over years |
| Stress relaxation | Deflection, continuously | Push-back force | ASTM D6147 / ISO 3384 | Force decays even at full thickness; preload drifts down |
| CFD plateau placement | (Design choice) | Nothing, if sized right | ASTM D3574 Test C / ISO 3386 | Working window stays on the flat region BOL to EOL |
A material can be respectable on one axis and weak on another, which is why the parent application page treats set and relaxation as a twin-track watch list rather than a single number. When the two are reviewed together with the CFD curve, the question "will this pad still be pushing correctly in year ten" becomes answerable from the data sheet plus your duty cycle, rather than a hope.
Putting the three behaviors into a specification
- Set the force window first. Take the preload window from your cell supplier's guidance; the published research record (from Cannarella and Arnold's 2014 stack-pressure work through the 2021 and 2025 Journal of Power Sources module-compression studies, surveyed alongside SAE 2024-01-2430) supports light-to-moderate preload helping life and over-compression accelerating aging, with wide variation across cell types.
- Establish the deflection range. Beginning-of-life gap, expected breathing percentage, and end-of-life swelling allowance, from the cell data and your stack tolerance budget.
- Choose the family for the environment. Microcellular polyurethane across normal module temperatures; BISCO closed-cell silicone foam where temperature or UL 94 flame class governs; a V-0 silicone sponge where the vertical-burn listing leads the brief.
- Place the window on the plateau. Pick grade and thickness so BOL and EOL deflections both sit on the flat region of the published CFD curve.
- Check the aging pair. Confirm compression set (right method, right basis, right conditions) and stress relaxation for the grade at your temperature, against the service life on the requirement.
How H-O Products helps
H-O Products is a precision die-cut converter. We do not write your preload window, and we do not produce the foam; we convert the engineered families this article describes into the die-cut, kiss-cut, laser-cut, or waterjet-cut pads your drawing calls for, laminated to thickness where a single gauge cannot reach it, with adhesive, liners, and pull-tabs for line installation.
What we add at the specification stage is the drawing review: flagging a window that runs off the plateau, a set value quoted on the wrong basis, or a gap-fill foam sitting in a compression slot, before tooling.
Documentation available on request includes the material TDS, certificate of conformance, lot traceability, and first-article inspection, under our ISO 9001:2015 certified quality management system.
Frequently asked questions
What is the difference between reversible breathing and irreversible swelling?
Breathing is the small, repeating thickness change each charge-discharge cycle as lithium moves into and out of the anode; it recovers on discharge. Irreversible swelling is the permanent growth over life from gas generation and solid-electrolyte-interphase growth; it never recovers and commonly leaves cells about 10 to 20% thicker at end of life. The pad must flex with the first and absorb the second while keeping its force inside the window, which is why plateau shape, set, and relaxation together form the real specification.
Why can't I compare compression-set numbers from two data sheets directly?
Because the reporting basis and conditions may differ. ASTM D395 and ASTM D3574 report the unrecovered loss against the original height, while ASTM D1056 reports against the compressed height, so the same physical behavior produces different percentages. Temperature, deflection, and hold duration also change the value. Compare like-for-like methods and conditions, and put the method on the drawing alongside the limit.
If a foam has low compression set, do I still need to worry about stress relaxation?
Yes. Set measures thickness that fails to recover after release; relaxation measures force decaying while the pad is still held at constant deflection. A pad can keep most of its thickness and still let a meaningful share of its force leak away, especially warm and under vibration. The two are measured by different methods (set per ASTM D395 / D3574, relaxation per ASTM D6147 / ISO 3384) and reviewed as a pair.
Where should my working deflection sit on the CFD curve?
On the plateau, at both ends of life. The beginning-of-life gap sets the initial deflection; end-of-life swelling adds to it. If the initial point sits too low on the elastic rise, the force starts below the window; if the end-of-life point reaches the densification knee, the force spikes as the cell grows. Grade and thickness are chosen together so the whole life stays on the flat region, using the manufacturer's published curve per ASTM D3574 Test C / ISO 3386.
Have your pad spec reviewed against all three behaviors
Send the stack drawing, the gap and swell numbers, and the target window, and engineering will check plateau placement, set basis, and relaxation exposure before anything is tooled.
Samples typically ship in 3 to 5 business days for common die-cut configurations; standard production runs about two weeks; special orders are quoted to the job.
Family-owned in Winsted, Connecticut since 1971 · ISO 9001:2015 certified · OEMs worldwide send us a drawing — and get back their part.
Related reading
Request a quote for your EV cell compression & cushioning part
Send H-O your drawing or describe the part—engineering reviews every RFQ and converts samples on your geometry.
Last updated · Reviewed by H-O Products engineering (ISO 9001:2015 certified converter, Winsted, CT).
References
- ASTM D3574 (flexible cellular polyurethane test methods; Test C compression force deflection) and ISO 3386 (CFD / CLD).
- ASTM D395 Method B (rubber compression set, constant deflection); ASTM D1056 (sponge and cellular rubber; compressed-height reporting basis).
- ASTM D6147 and ISO 3384 (stress relaxation: force decay under constant deflection).
- Cannarella & Arnold (2014), stack pressure and capacity fade; module compression and aging, Journal of Power Sources (2021); localized compression and lithium plating, Journal of Power Sources (2025); SAE Technical Paper 2024-01-2430.
- Manufacturer technical data sheets: PORON industrial 4701/4790 and PORON EVExtend grades; BISCO HT/BF closed-cell silicone foam; V-0 silicone sponge grades.
This article explains test-method concepts (ASTM D3574, ISO 3386, ASTM D395, ASTM D1056, ASTM D6147, ISO 3384) and design reasoning for EV compression pads. The figure is qualitative and carries no measured values. Grade-level curves, set, and relaxation data come from the manufacturer's current technical data sheet, and pack-level test outcomes belong to the tested pack.
All parts are custom manufactured to order; minimum production run quantities vary by material and part. H-O is a die-cut converter, not a raw-material producer.