EPP Foam in Sports Protection: Guards, Pads, Armor

Overview

Molded EPP has quietly become one of the default cores in sports protection: inside football shin plates, knee and elbow pads, riding armour segments, ski and hockey guards, and the padding that sits between a rider and a rock garden. It is also routinely mis-specified, because the same three numbers that make a good packaging part — density, thickness, coverage — make a mediocre protector when they are copied across without re-testing. This article is about the difference. It covers what an impact protector is actually asked to do, how EPP compares with the EVA, EPS and viscoelastic foams it competes with, how the standards that govern protectors are written and what their numbers mean, and where the material should not be used at all.

Two boundaries before we start. First, the scope here is sports and recreational protectors — the guards, pads and armour segments sold to players and riders. Ballistic, stab and military protection is a different problem with different test methods and different materials; it is outside the scope of this article, and nothing below should be read as guidance for it. EPP is not a ballistic material and we do not position it as one.

Second, we have written around this subject before. The helmet side of the story — how a moulded EPP liner manages energy inside a shell — is covered in EPP Helmet Liner: Lightweight Impact Protection and in EPP Foam Helmet Pads. The material fundamentals behind all of it are in our density guide, and the test-driven design loop is set out in Drop-Test-Driven Packaging Design. What follows is specific to worn protection: parts that a person straps to their body and then falls onto.

Three jobs, three different foam answers

Buyers tend to ask for “protective foam” as if it were a single product. In practice a protector programme contains three quite different jobs, and the foam that is right for one is usually wrong for the next. Getting this mapping wrong at the drawing stage is the most expensive mistake in the category, because it is discovered only after tooling.

Table 1 — The three jobs. A part that is asked to do two of them at once is a compromise by design, and should be drawn as one.
Job	Typical parts	What the foam is doing	What it is not doing
Load spreading behind a shell	Football shin plates, ski and hockey guards, forearm guards	Turning a point load into a distributed one, and taking the sharp edge off a blow	Absorbing high energy on its own — the shell, the frame and the strap share that work
Repeated-impact joint pad	Knee, elbow, shoulder and hip pads, MTB and riding armour	Absorbing a moderate strike, then returning to shape so the next one is also absorbed	Surviving a single vehicle-level event; that is a different class of product
Comfort and retention layer	Cuffs, straps, liner pads, contact surfaces	Pressure distribution, removal of hard edges, some thermal separation from the shell	Any structural role whatsoever

The distinction matters because the failure modes are different. A load-spreading plate fails by local bottoming out: the foam compresses to solid under the strike point and the shell edge arrives at the shin. A repeated-impact pad fails by losing thickness over a season — it still looks intact, but it has nothing left to crush. A comfort layer fails by chafing, which no impact test will ever catch.

What a protector actually has to do

An impact delivers energy. What injures a player is peak force, delivered through a small area. Those three words explain almost every design rule in this article. The foam has one physical job: extend the distance and the time over which the incoming mass decelerates, and spread the resulting load over as much of the body as the anatomy allows.

The standard test that most European protectors are judged by uses a 5 kg striker delivering 50 J — roughly a 5 kg mass dropped from one metre. That striker is travelling at about 4.5 m/s when it lands. If the 50 J were absorbed evenly over a 25 mm crush, the average force would be about 2 kN; over 10 mm it would be about 5 kN. Certified protectors still record peaks between roughly 9 kN and 35 kN, because force is nothing like constant: it climbs as the foam stiffens, and it spikes the moment the striker closes on the anvil through densified material. That is the design problem in one sentence — a protector fails by bottoming out, not by being too soft.

The anvil is not a leg

Every one of these tests strikes a protector resting on a steel anvil or a defined hemispherical striker. That gives a repeatable number, and it also means the number is not an injury threshold: a body deforms, a muscle absorbs, a bone edge concentrates load differently, and the garment that holds the pad in place is not part of the test at all. Read certification as “this insert met a defined bar in a defined rig”, not as “this will prevent a fracture”. The same caveat applies to the published force limits below, which are quoted as orientation and should be verified in their current amendment with the test house.

Why EPP: the multi-hit argument

Expanded polypropylene is a closed-cell, semi-crystalline bead foam. Three of its properties matter for worn protection, and one of them is the reason it has displaced EVA in a lot of joint-pad programmes.

It recovers. EPP deforms elastically over a large strain range and comes back close to its original thickness after a strike. EVA of the same density takes a permanent set: the first impact leaves a compressed zone that behaves like a harder, thinner material next time. For a knee pad that will be hit twenty times in a season, that difference is the whole product. It is stable. Density and stiffness change relatively little between -10 °C and +40 °C compared with EVA, which stiffens in the cold and softens in a hot car. It does not hold water. Closed cells mean sweat and rain sit on the surface rather than inside the foam, and the part dries without the musty weight gain that open-cell PU develops.

Table 2 — An honest comparison. The right answer is often a laminate of two of these, not a winner.
Material	Real strength	Real weakness	Where it wins
Moulded EPP	Multi-hit recovery, temperature stability, low water uptake, mono-material PP	Firm at low density; needs thickness to work; limited drape and softness	Joint pads, guards, armour segments, anything hit more than once
EVA	Soft, cheap, easy to cut and laminate, good hand feel	Takes a permanent set, soaks up sweat in open grades, loses thickness with use	Comfort layers, low-volume prototypes, parts that must feel soft
EPS	Very cheap, very light, high energy absorption per gram in a single event	Friable: crushes once and sheds beads; no recovery at all	Single-event sacrificial packaging; not a worn pad
Viscoelastic PU (D3O-class)	Best performance per millimetre at thin profiles; soft until loaded	Cost, temperature sensitivity, limited thickness scalability, harder to recycle	Low-profile armour where 8–10 mm is the entire budget
ETPU	Very high resilience and rebound, soft and springy feel	Lower energy dissipation per unit thickness than EPP in a crash pulse; higher cost	Footbeds, comfort and energy-return layers, not the primary impact core

We should be explicit about the trade we are recommending. If the brief is “maximum protection in 8 mm of profile”, a viscoelastic PU will beat EPP and pretending otherwise wastes everybody’s prototype budget. EPP wins where there is room for 12–25 mm, where the part must survive repeated hits, and where cost per unit at volume and a mono-material PP recycling story both matter. A large share of production knee and elbow pads land in exactly that zone.

The standards buyers actually meet

Sports protection is not governed by one document. The vocabulary most European buyers use comes from the motorcycle clothing family, which has become the de facto reference for sports protectors generally; football shin guards, roller sports and garment-level protection each have their own standard. Knowing which one applies to your product — and which one your supplier is actually quoting — is the first line of any technical conversation.

Table 3 — Orientation only. Force limits are quoted as published; standards are amended, and the applicable version and level must be confirmed with your test house or notified body before any label is printed.
Standard	Scope	What the pass limits look like
EN 1621-1:2012	Limb joint impact protectors (shoulder, elbow, knee, hip), worn or inserted in clothing	Level 1: mean transmitted force ≤ 35 kN, no single strike ≤ 50 kN. Level 2: mean ≤ 20 kN. 5 kg striker, 50 J, five impact points; optional high (+40 °C) and low (-10 °C) tests
EN 1621-2:2014	Back and lumbar protectors, in full-back, central-back and lower-back types	Level 1: mean ≤ 18 kN, single ≤ 24 kN. Level 2: mean ≤ 9 kN, single ≤ 12 kN. Different anvil and striker geometry from part 1
EN 1621-3:2018	Chest protectors, divided and full designs	Two levels; level 2 additionally addresses force distribution, not only transmission
EN 13061	Shin guards for association football players	Coverage areas plus impact attenuation of the guard assembly
EN 14120	Wrist, palm, knee and elbow protectors for roller sports users	Impact and abrasion performance of the protector as worn
EN 17092	Motorcyclists’ protective garments, classes AAA to C	Abrasion and impact of the garment; a protector inside it is certified separately

Certification attaches to the protector, not to the foam

A CE mark on a foam core is meaningless on its own. The test is performed on the finished protector — foam plus any shell, plus the textile envelope that holds it, at the thickness and coverage actually shipped. When a supplier says “our foam passes EN 1621”, the correct follow-up question is: which part number, which level, which test house, and can I see the report? A change of foam density, thickness, colour batch or shell material is normally a re-test, not a paperwork update. We supply cores and moulded protectors; the certificate belongs to whoever puts the finished protector on the market, and we will not imply otherwise.

Density and thickness: the first two knobs

Two numbers set most of the behaviour, and both are usually fixed too early. Density sets how much load the foam carries before it densifies; thickness sets how much distance it has to give before it bottoms out. They are not independent: a thin high-density part and a thick low-density part can both be “firm”, but they fail differently under a point load.

Table 4 — Indicative starting points only, for orientation during concept work. Density alone does not determine a performance level; the level is whatever the finished protector demonstrates in the test.
Part type	Indicative density	Indicative core thickness	Note
Shin and forearm plates	50–80 kg/m³	6–12 mm	Usually bonded behind a hard shell; the shell does the spreading, the foam does the sting removal
Knee and elbow pads	70–110 kg/m³	12–25 mm	Thicker and firmer because it is the primary energy absorber and must recover repeatedly
Riding armour segments	60–90 kg/m³	10–20 mm	Segmented and articulated; overlaps matter more than any single segment
Hip, tail and shoulder inserts	60–90 kg/m³	8–15 mm	Thin sections inside garments; conformance to the body is the constraint
Comfort and liner layers	30–50 kg/m³	3–8 mm	Not an impact part; chosen for hand feel and pressure distribution

The counter-intuitive part: more density is not more protection. Raising density stiffens the part, which raises peak transmitted force for the same geometry and makes it conform to the body less well. It buys durability and load spreading, and it costs weight and comfort. That is why a protector programme should start from a target level and a thickness budget, and then find the lowest density that achieves it — not from the densest grade the supplier stocks.

Geometry decides whether the lab result survives Saturday

A protector that passes on an anvil can still fail on a player, and the reason is almost never the foam grade. It is the geometry, and specifically these seven details:

Coverage beyond the pad. A foam edge concentrates load. Extending the part 5–10 mm past the hard shell, and chamfering the edge, moves the pressure peak off the shell rim and onto supported tissue.
Edge radius. A square moulded edge behaves like a blade under load. A 2–3 mm radius on every body-facing edge is cheap in tooling and expensive to retrofit.
Articulation gaps. A single rigid plate over a knee bridges the joint and lifts off when the knee flexes. Segmentation with 3–5 mm gaps keeps coverage where the joint actually is.
No undercut traps. Deep undercuts that hold the pad in place also hold grit, and they tear at the moulding seam. Retention belongs to the textile pocket, not to a snap feature in foam.
Ventilation channels. Moulded ribs or a channel pattern give air a path and reduce the contact area. They also stiffen the part directionally — useful if the channels run across the impact axis, unhelpful if they run along it.
Drainage and drying. A blind recess collects sweat and salt. Any recess that faces the body should drain, or it becomes the place the part fails first.
Strap and shell interfaces. Anchor points must bear on the shell or on a densified boss, never on a thin foam web. Foam creeps under sustained strap tension, and a pad that has crept 4 mm sits in the wrong place.

The most common field failure in the category is not a crushed pad. It is a pad that has rotated or migrated, so that the impact lands on the edge. A perfectly certified insert in the wrong position performs like no insert at all — which is why we treat retention as part of the protector design, not as the garment designer’s problem.

Multi-hit, set and recovery in the real world

“Multi-hit” is a relative claim, not an absolute one. EPP recovers far better than EVA, but every strike costs something: cell walls fatigue, the part loses a little thickness, and the load it can carry before densification falls with it. A pad that measured 18 mm in the lab may measure 16 mm after a season, which is exactly the margin that made it pass.

Two practical consequences follow. First, ask for a compression-set figure after cyclic loading at service temperature, not only a single-impact result; a useful rule of thumb in the field is that a protector should be retired when it has lost roughly 10–20 % of its original thickness or shows visible cracking at the flex lines, and that any protector involved in a hard crash is replaced regardless of appearance. Second, design the wear indicator in: a moulded witness step or a nominal-thickness callout on the drawing makes the retirement decision objective rather than a judgement call.

This is also where segmentation earns its keep. Ten articulated segments that each recover are more durable in service than one large plate that creases along the same line every time the joint flexes. The crease is where the cell structure fatigues first, and it is always invisible from outside.

Sweat, sun and temperature: the ageing story

Worn protection lives a harder life than packaging. It is soaked in saline sweat, washed with detergent, left in a hot car, used at 0 °C and stored in a garage. EPP handles most of that better than the materials around it, but not indefinitely and not without rules.

Sweat and salt. The foam absorbs very little water because the cells are closed, but the textile envelope and the strap do. The realistic problem is a wet assembly held against the foam for hours, which is a hygiene and odour issue long before it is a mechanical one.
Washing. Mild detergent at 30–40 °C, no solvents, no chlorine bleach, no tumble drying. The foam tolerates water; adhesives, printed graphics and laminated comfort layers frequently do not. Where the protector is bonded into a garment, the garment care label governs — and it should be written with the foam in mind.
Heat. Polypropylene softens in the 100–110 °C range and creeps under load well below that. A closed car in summer is uncomfortable but not a problem; a drying room or a hot press is.
Cold. The material stays usable well below freezing, but it does stiffen, and a stiffer protector transmits more force. This is precisely why the standard offers optional +40 °C and -10 °C impact tests: if your product is sold into winter sport, that claim on the label is worth more than any room-temperature figure.
UV. Unpigmented or lightly pigmented PP will chalk and embrittle under sustained sunlight. Inside a garment it never sees UV; on an exposed outdoor part it needs a pigmented grade, a coating, or a cover.

Comfort is a construction problem, not a foam problem

Bare EPP against skin is a chafe generator. It is firm, it has a moulded texture, and it does not breathe through its cells. Every successful EPP protector we have seen solves this the same way — as a three-layer construction rather than as a single foam choice.

1. The core

Moulded EPP at 60–110 kg/m³, shaped to the anatomy, segmented where the joint flexes. This is the only layer doing impact work, and it is the only layer that has to be re-tested when it changes.

2. The comfort layer

3–8 mm of EVA, ETPU or a spacer textile between core and body. It removes pressure points, absorbs shear, and lets the core be as firm as the protection requires without the wearer noticing.

3. The outer layer

Abrasion-resistant textile, a hard shell, or a laminated film, depending on the sport. It also protects the foam from grit, which is what turns a smooth surface into a sanding belt over a season.

Weight is worth doing the arithmetic on, because it is the argument that decides most comfort debates. A knee-pad core of 150 × 100 × 15 mm encloses 0.000225 m³; at 80 kg/m³ that is about 18 g, and at 110 kg/m³ about 25 g. On one pad that is nothing. Across a full riding suit with eight segments, the same choice moves 50–60 g, and it moves it to the outside of the limb, where it is felt most. The illustrative conclusion: choose the lowest density that passes, and spend the saved grams on a better comfort layer.

From pattern to first article: the tooling path

There are two routes to a foam protector core, and the choice is driven by volume and shape rather than by preference. Cut foam from block or sheet needs no tooling, suits 50–500 pieces and simple 2.5D shapes, and cannot produce channels, bosses or undercuts. Moulded EPP needs a tool, gives repeatable density and geometry, and is the only route to a part with ribs, a skin, or a shape that wraps a joint.

Anatomical input: a 3D scan of the wearer population, or a physical pattern taken from the garment the pad lives in. Sizes matter — a pad that fits the 50th percentile fits nobody well.
Coverage and level target: which standard, which level, which zones. This sets thickness and area before any styling happens.
Core design: density, thickness, segmentation, channels, edge radii, and the interface to the textile pocket.
Tool and T1: mould, first shots, dimensional report after stabilisation, sectioning to check fusion and density distribution, and a fit trial on the actual garment.
Protector-level testing: the finished assembly goes to the test house. Budget a re-test cycle; a first-shot pass is the exception.
Production and change control: density tolerance, colour batch, and a written rule that any change to foam, thickness or shell triggers re-evaluation.

On cost, the structure is simple enough to model without quoting anyone’s price. Part cost is dominated by mass × material price plus a share of machine time and, at low volume, tooling amortisation. Because mass is density × enclosed volume, dropping a core from 90 to 70 kg/m³ removes roughly 22 % of the material cost — and may remove the protection with it. Tooling for a two-part aluminium protector mould typically sits in the low thousands of euros as an order of magnitude, amortised over the programme volume; treat any such figure as illustrative and non-binding and get it quoted against your actual drawing.

Where EPP should not be used

Most of the value of a technical conversation is in the negative list. These are the jobs we decline or redirect:

Ballistic, stab and fragmentation protection. Out of scope entirely; EPP is not a ballistic material and has no role in those products.
The energy-management system of a certified helmet. We mould liners, but the helmet’s certification belongs to the helmet assembly and its own standard. A liner is a component inside someone else’s approval, never a substitute for it.
Parts that must drape. If protection has to conform to a wrist or an ankle in 4 mm, a film, a thin TPU or a viscoelastic sheet will do it and EPP will not.
Visible cosmetic surfaces. Foam surfaces show flow marks, weld lines and bead texture. If the part is the product’s face, either give it a shell or accept a technical look.
Structural fittings. Buckles, strap anchors, hinges and press-fit joints are fatigue parts. PP foam creeps under sustained load and cracks at flex lines; use moulded PP or PA components for those.
Sustained heat under load. Anything held above roughly 80–90 °C while carrying a load will creep permanently.
Long-term exposed outdoor parts. Without a pigmented grade and a UV strategy, an exposed EPP surface will degrade within a few seasons.
Sealing and gasketing. Foam is not an elastomer. Sealing belongs to EPDM or silicone; a foam gasket leaks as soon as it takes a set.

What to ask before approving a core

The document set is short, and each item answers a question that cannot be answered from a sample in the hand:

Table 5 — A seven-document ask. If a supplier cannot produce the second row, the conversation is about foam, not about protection.
Document	What it proves	What it does not prove
Material datasheet with grade and density tolerance	What the foam is, and how much it can vary batch to batch	That any protector made from it passes anything
Protector test report naming standard, level and test house	That this finished protector met that bar in that rig	That it fits, stays in place, or survives a season
Ageing and wash-conditioning statement	Whether the result holds after the conditioning the standard requires	Long-term field durability
Temperature claim (high and low)	That the optional +40 °C and -10 °C impacts were tested and claimed	Anything about humidity or wet use
Skin-contact compliance: REACH SVHC statement, colour fastness to water	Material innocuousness for a worn part — the standard family requires it, including colour fastness	Biocompatibility for medical claims, which is a different regime
First-article dimensional report and sectioned sample	Wall thickness, density distribution, fusion quality	Impact performance
End-of-life statement	Mono-material content and how the part routes into a PP stream	That a recycling route exists locally

The short version

EPP earns its place in sports protection because it recovers, stays stable across the temperature range a sport actually happens in, and does not hold sweat. It does not earn its place in 8 mm profiles, in ballistic work, in structural fittings, or in any part where the foam is expected to be the certified article. Design the coverage and the retention first, pick the lowest density that passes, put a comfort layer between the core and the skin, and ask for the protector report rather than the material datasheet. Everything else in this category is detail on top of those four decisions.

Can EPP simply replace the EVA in an existing shin guard?

Sometimes, and never as a drop-in. EPP at the same thickness is firmer and will recover better over a season, which is usually an improvement for the impact side. It is also less conformable, so a guard that relied on EVA to wrap the shin may now sit proud at the edges and concentrate load there. The realistic route is a hybrid: EPP as the core, a thin EVA or spacer layer against the skin, and a re-check of the coverage geometry. Then re-test the assembly, because the protector is what is certified.

Is a CE mark on the foam core enough?

No, and it is not how the system works. Certification applies to the finished protector — core plus shell plus the envelope that holds it, at the thickness and coverage shipped. A core supplier can tell you the foam grade, the density tolerance and the mechanical data; only the party placing the protector on the market can hold the certificate. Ask for the report: part number, standard and version, level, test house, and the conditioning applied.

What density should a knee pad use?

As a starting point for concept work, 70–110 kg/m³ with a 12–25 mm core covers most recreational knee pads, and 50–80 kg/m³ at 6–12 mm covers plates that sit behind a hard shell. Those are orientation numbers, not specifications. The density that matters is the one that lets your finished protector meet the level you are claiming, and that can only be established by testing. Raising density is often the wrong first move: it can increase transmitted force and reduce fit.

Does EPP still work in cold weather?

Yes, but it stiffens, and a stiffer protector transmits more force for the same strike. The standard family handles this with optional high (+40 °C) and low (-10 °C) impact tests, and a label that claims them is telling you something genuinely useful. For winter sports and cold-climate riding, ask specifically whether the low-temperature claim was tested and on which thickness, rather than assuming room-temperature data transfers.

Can EPP protectors be washed?

The foam itself tolerates water and mild detergent at 30–40 °C. What fails first is everything around it: adhesives, printed graphics, laminated comfort layers and some textile bindings. No solvents, no chlorine bleach, no tumble drying, and air dry away from direct heat. If the protector is bonded into a garment, the garment care label should have been written with the foam in mind — that is a design decision, not a laundry instruction.

How many impacts will a protector take?

There is no clean number, and any supplier who quotes one without a test is guessing. EPP recovers far better than EVA, but each strike removes a little thickness and stiffness margin. A practical retirement rule is a 10–20 % loss of original core thickness, visible cracking at flex lines, or any hard crash — whichever comes first. Specify the nominal thickness on the drawing so the check is objective.

Which parts of a protector programme can you actually supply?

The moulded EPP core and, where the design calls for it, the moulded protector assembly: the core with channels, segmentation, edge radii and pockets, in the density your testing selects. We can support the pattern work, the first-article dimensional report and the samples that go to your test house. The certification, the garment integration and the claim on the label stay with you — and we will say so in writing, because that division is what keeps a programme honest.

Working on a protector core?

Send us the coverage area, the level you are targeting, the thickness budget and the garment interface, and we will come back with a density and geometry proposal plus a first-article plan. If you are still at the material-selection stage, start with the moulded liner and impact-absorption parts below.

Molded EPP Helmet Liner EPP for helmet impact absorption Ski and hockey guard liners Talk to our engineering team