How We Created FlexGuard — The Full Story

Nature spent 400 million years solving body armour. Here's what we learned from it.

By Harry Huang, Lead Material Engineer at ArmorLite, maker and retailer of body armour — Last reviewed: July 2026

1. The Problem Most Body Armour Asks You to Accept

For most of its history, stab-resistant body armour has asked the wearer to accept a compromise. If it protects well, it is usually stiff, heavy, or both. If it is comfortable enough to wear through an eight-hour shift, something has usually been given back in protective performance. The two requirements — genuine protection and genuine wearability — tend to pull in opposite directions. Most products choose one and manage the other as well as they can.

ArmorLite's founding question, was whether that compromise was actually necessary. Could you build a protective system someone would not only trust, but actually want to put on? Not just pass a certification test with. Not just photograph well on a product page. But wear for a full shift, sit in a vehicle with, move freely in, and forget it's there until the moment it matters.

The mission — to help people feel safe and confident in everyday activities, through reliable protection that doesn't weigh you down — became an engineering constraint, not a tagline. Because designing to both requirements simultaneously is harder than designing to either one alone. Every decision made over the following two decades had to answer to both.

We were not trying to build the hardest panel on a test bench. We were trying to build something people would actually wear. Everything else followed from that.

2. Why We Started With Biology

When we began working on FlexGuard, we were not looking for a metaphor. We were looking for a structural clue.

The problem was already clear. Protection and mobility were pulling in opposite directions, and most conventional approaches solved that by giving more of one and less of the other. It seemed worth asking whether that trade-off was non-negotiable we first started.

Biology was a sensible place to start because natural armour systems have been working on the same problem for far longer than we have. An animal does not wear protection that only works when standing still. It has to move, bend, breathe, twist, hunt, escape, and still sustain attacks from predators' teeth and claws, all in the same armour. That makes natural armour an unusually honest kind of engineering. It does not have the luxury of solving one requirement and ignoring the rest. If a design only protects well but restricts movement too much, the animal carrying it does not survive long enough to pass it on. The trade-off we were struggling with had already been tested, under real circumstances, for hundreds of millions of years.

What stood out was not one species, but a pattern. Across very different animals — fish, reptiles, mammals — we kept seeing variations on the same underlying structural logic (Yang et al., 2013; Meyers et al., 2013). Not a single rigid shell, not a soft layer alone either. Instead: hard protective elements, a flexible supporting layer beneath them, and overlap between the rigid but relatively small elements so that coverage stayed continuous even when the body moved.

Each part was doing a different job. The hard surface resisted the initial strike. The softer structure behind it kept it an integral system and absorbed what came next. The overlap preserved coverage as the body flexed, so that the vital organs remain covered at all times.

Figure 01. An 18th Century Indian Armour Coat Made From Pangolin Scales. Wikipedia.

That recurring pattern mattered. When the same solution appears independently across species that are not closely related, evolutionary biologists treat it as a sign that the solution is doing real work — that the underlying problem only has a small number of viable answers, and nature has found them more than once (Naleway et al., 2015).

For us, that was useful information. It meant the design space was narrower than it first looked. It also meant that the answer was probably not going to come from a single miracle material. It was going to come from architecture: from how materials are divided, layered, joined, and allowed to move relative to one another. In other words, the question was not only what the armour was made from. It was how the armour was put together.

That shift in framing — from material to structure — enabled the work that followed. Once we stopped looking for the perfect substance and started looking for the right architecture, the design problem opened up. We could ask sharper questions. How small should the hard elements be? How should they overlap? What should sit behind them? How should they be joined so that the system behaved like one continuous panel under load, but drapes like fabric in motion? Those are engineering questions, and engineering questions can be quantified and validated.

None of this meant copying animals directly. A pangolin scale is not a carbon-fibre tile. An Arapaima lives in water and weighs nothing on its own structure the way a person standing upright does. The biology gave us inspirations, not blueprints. The work of turning those inspirations into something a human could wear through a working shift — and trust while wearing it — was still ahead of us. But we had a starting point.

3. Nature's Best Lessons: Pangolin, Arapaima, Crocodile, and More

Once we accepted that the answer was about architecture rather than material, the next question was which natural systems to study most closely. Not every armoured animal had something useful to teach us. Some solutions were too specialised to their environment. Others relied on biological processes — continuous regeneration, for example — that no engineered product can replicate. We were looking for principles that could survive the translation from biology to a wearable panel. Luckily, Mother Nature is very generous, and there is a large spectrum we could draw inspiration from.

The Pangolin: Flexibility That Locks Under Load

The pangolin is the only mammal covered entirely in keratin scales. What makes it interesting is not the scales themselves, but how they behave depending on what is happening to the animal. In motion, the scales slide over one another with very little resistance. The animal walks, climbs, and curls without fighting its own armour. But under concentrated pressure — a predator's bite, a sharp impact — the geometry of the overlap changes. The scales push against each other, interlock more tightly, and stop behaving like independent plates. They start behaving like a single constrained surface.

Figure 02. A Pangolin

This is the principle we kept coming back to throughout development. Body armour does not need to be rigid all the time. It needs to be flexible when the wearer is moving and rigid when something is trying to get through it. The pangolin showed us that a single structure could do both, depending on how the elements were shaped and arranged.

The Arapaima Gigas: Layered Toughness and Crack Deflection

The Arapaima Gigas is a large freshwater fish from the Amazon basin. It shares its habitat with piranhas, which means its scales have to resist concentrated bite force from teeth sharp enough to crush anything in their way. What the Arapaima evolved is one of the most studied examples of natural composite armour.

Each scale has a hard mineralised outer layer and a tougher, more flexible inner layer beneath it. The two layers are not simply stacked. The fibres in the inner layer are arranged in a twisted-plywood pattern — known in the materials science literature as a Bouligand structure — in which each successive layer of fibres is rotated by a small angle relative to the one below (Meyers et al., 2013; Browning et al., 2013). At the moment of a strike, the super-hard outermost scale dulls the sharp teeth or claws, largely reducing penetration energy, and the twisting fibres underneath make it difficult for a crack to travel cleanly through the scale. As the crack tries to propagate, it has to twist to follow the changing fibre direction, and twisting absorbs energy.

Figure 03. Arapaima Gigas

There is also the stacking of the scales on top of each other: when one scale is being struck, it instantly spreads the impact energy to neighbouring scales. These three mechanisms work together and effectively reduce the impact into almost nothing. This is well-documented in published research from materials science groups at UC San Diego and UC Berkeley, among others.

Figure 04. Anatomy of Arapaima's Scale and the Bouligand Structure, Lawrence Berkeley National Laboratory

For us, the practical lesson was that toughness is not just about how hard a material is. It is about how the material is structured internally to handle force that has already entered it. A hard surface stops the first part of the strike. The layered architecture behind it manages what comes next. That three-stage logic — resist on the surface, spread the energy, and dissipate what remains beneath — became one of the design principles built into every generation of FlexGuard.

The Crocodile: Geometry as Protection

Crocodiles and alligators carry bony plates called osteoderms. The plates are not flat. They are slightly domed, with ridges and arched profiles that look almost architectural when viewed in cross-section. These plates are so strong they could even deflect bullets from small firearms.

Figure 05. Crocodile's Bony Plates

That geometry is not decorative. A domed surface distributes a point load outward across a wider area than a flat surface does. The same principle is used in bridges, arches, and the curved sections of vehicle armour (Yang et al., 2013).

The osteoderm taught us something that is easy to overlook when thinking about armour: shape does work. Two panels made from the same material, with the same thickness and the same weight, can perform very differently depending on their surface geometry. A protective tile is not just a piece of strong material. Its shape would make a significant difference. This influenced how FlexGuard tiles were profiled and embossed. The anti-slip surface geometry on the panel is not only there to keep the tiles from sliding against clothing — it also contributes to how the tile handles a sharp point arriving at an angle.

Supporting Influences: Armadillo, Tortoise, Horseshoe Crab

Three other animals shaped the thinking without becoming direct templates.

The armadillo demonstrates banded articulation — rigid sections joined by flexible bands that allow the body to curl. The principle is similar to the accordion articulation between FlexGuard's tile rows, where rigid protection is joined by sections that allow controlled movement.

The tortoise demonstrates how a curved shell handles concentrated load, and how a fused composite structure can outperform what a single material of the same weight could achieve on its own. The shell is not a single thing — it is many bones bonded into one functional surface.

The horseshoe crab is one of the oldest armoured animals still alive — the lineage goes back more than 400 million years, which is where the subtitle of this post comes from. Its persistence is itself a kind of evidence. Whatever structural principles it embodies have been tested by a longer span of time than any laboratory can replicate.

Figure 06. Horseshoe Crab

What These Lessons Had in Common

None of these animals solved exactly the same problem we were solving. A pangolin does not need to sit in a patrol vehicle for eight hours. An Arapaima does not have to be wearable under a uniform shirt. The point was never to copy any one of them.

Across very different species, evolving under very different pressures, the same architectural ideas kept appearing: hard elements on the surface, tougher structure behind, overlap to maintain coverage, geometry that distributes impact energies, and behaviour that changes between motion and impact. The recurring principles were found independently in animals separated by hundreds of millions of years of evolution. That convergence gave us confidence that we were looking in the right direction. Our job was then about turning those principles into something measurable, manufacturable, and wearable.

4. Twenty Years, Six Generations: The Engineering Journey

The architecture described in the previous section did not arrive in one piece. It arrived through six generations of prototypes, years of numerous iterations and testing, and a long sequence of problems that had to be solved before the panel could behave the way it does now. Some of them are obvious in hindsight. Most of them were not so obvious at first sight.

4.1 The Substrate

It sounds naive now when we look back, but we started with foam. The reasoning was straightforward: foam absorbs energy, is lightweight, and doesn't press against the body abruptly. On paper it looked like a reasonable starting point.

In testing it was not. Foam compressed under repeated loading and did not recover its original geometry. After a series of strikes, the tile positions had drifted relative to one another. The offset pattern that was supposed to guarantee coverage no longer held. It also cracked too easily.

We then moved onto textile substrates, but the trade-off changed shape rather than disappearing. Some textiles were strong enough to anchor the tiles but were too heavy — it would feel like wearing a castle and the wearer's energy is drained quickly. Others were light and comfortable but would tear or were difficult for the outer shells to attach onto. Being too soft is also problematic: the material would deflect around a sharp point rather than resist it.

The substrate had to do four things at once: anchor the tiles in their intended geometry, survive repeated loading without permanent deformation, contribute its own resistance to penetration, and stay wearable across long durations in real life. The eventual answer was a continuous woven aramid layer, but identifying aramid as the right material family was only the first step. The weave pattern, the thread count, and the surface treatment all had to be tuned for the specific job the substrate was being asked to do. A substrate optimised for ballistic protection does not necessarily perform well under stab loading, because the failure modes are different — ballistic threats arrive fast and spread their energy, while stab threats arrive slower and concentrate it.

Figure 07. Study of Knife Penetration Behaviour.

By the end of this phase, we had a substrate that held the tiles where they were supposed to be, contributed meaningfully to the system's resistance, and did not punish the wearer for wearing the body armour for hours.

4.2 The Attachment

Bonding a piece of rigid carbon-fibre tile to a textile foundation isn't as trivial as we initially anticipated. It needs to stick onto something with very different physical properties, and stay there.

The instinctive approach is to use a strong adhesive — something that holds the tile firmly in place and does not move. The problem with that approach is that it creates a brittle interface between two materials with very different mechanical behaviours. Under flex, the rigid bond resists movement that the substrate wants to make. Under impact, the bond becomes a stress concentrator. Cracks initiate at the bond line and propagate outward. The tile may survive the strike. The bond may not.

The opposite approach is to use a more compliant bond — something that allows the tile and the substrate to move slightly relative to one another. This solves the brittleness problem but introduces a new one. Over time, with thousands of flex cycles in normal wear, a compliant bond can creep. The tile shifts. The offset geometry drifts. The coverage relationship between the two layers changes in ways that are not visible from the outside but matter under load.

And there is the problem of holding things together at high temperatures, and after multiple washing cycles. It took a long time and countless different chemical combinations. The eventual solution was not a single adhesive choice but a joining strategy that combined a primary bond with a secondary mechanical engagement between the tile and the substrate. The primary bond holds the tile in place under normal wear. The secondary engagement comes into effect under load, when the forces on the tile would otherwise concentrate at the bond line.

This is one of the parts of the FlexGuard architecture that looks simple in cross-section and is not. The challenge is to ensure the carbon-fibre tiles are attached sturdily on the substrate, across the full range of conditions a panel encounters in real use — from a desk worker leaning back in a chair to a frontline officer absorbing a direct strike.

4.3 The Geometry

The next question was the shape, size, and layout of the tiles themselves. This is a deceptively large design space. Tile size affects flexibility, coverage, weight, and the number of joints in the panel. Tile shape affects how the tiles tessellate, how they overlap, and how they handle impacts.

We tested square-only layouts, triangle-only layouts, and several hybrid arrangements. Each had trade-offs. Small tiles produced excellent flexibility but introduced more joints, and joints are where coverage gets complicated. Every joint is a place where the offset geometry has to do its work, and every additional joint is another place where the design has to prove that the layer beneath is doing its job. Small tiles also meant more glue would be used, and more glue means increased overall weight — and more weight is working against our design goal.

Large tiles reduced the joint count but conformed to the body less well. A panel made from large tiles tended to bridge across body curvature rather than follow it, which made the panel feel more like a piece of stiff board attached to the wearer than a part of the wearer's clothing.

The shape of individual tiles also mattered, as the crocodile example showed. Irregular shapes like those found on crocodiles' bony plates are effective in spreading out the impact force, but are difficult to manufacture in mass production. Sharp tile corners created stress concentrations under impact. Overly rounded tiles tessellated poorly and left awkward gaps that the substrate had to compensate for.

Figure 08. Development Stages of The Tiles

The final tile geometry is the result of resolving these competing pressures — not by choosing one priority over another, but by finding a shape and size that performed adequately on all of them. This was where simulation and physical testing had to work in parallel. Simulation let us screen geometries quickly and rule out arrangements that obviously did not work. Physical testing was the only way to find out which of the promising arrangements actually performed well under real loading, and where the simulation models were missing something. Several layouts that looked excellent on screen failed on the bench. A smaller number failed on the bench in ways that taught us how to improve the simulation models. That feedback loop — between digital prediction and physical reality — was one of the most extensively utilised development tools we had.

Figure 09. Study of Penetration Energy From Different Knife Tip Geometries.

4.4 The Movement

Protection that works on a test bench but restricts the wearer in the field is not a solution. It is a trade displaced from one place to another.

This sounds obvious. It was harder to act on than it sounds, because movement is difficult to measure in ways that are both precise and meaningful. For the first several years of development, we relied primarily on qualitative assessment — we put panels on people and asked them how the panels felt. That feedback was useful for identifying obvious failures, but it was not sensitive enough to distinguish between panels that were genuinely different in meaningful ways and panels that felt different for reasons that had nothing to do with their architecture.

The movement measurement work came later, and it changed how we thought about flexibility as a design target. Measuring bending resistance along a single axis — usually a front-to-back fold — captures something real, but it leaves out a great deal. A wearer does not move along a single axis. They sit, turn, reach upward, reach across, twist at the waist, lean into a vehicle, push against resistance. Each of those movements loads the panel differently, and a panel that performs well in a single-axis bend test can still feel rigid and restrictive in compound movements that combine bend and torsion simultaneously.

Figure 10. Different Ways of Accessing "Flexibility".

Over time, we developed our own internal protocol for multi-axis flexibility assessment, based on a set of reference movements drawn from analysis with operational users. The movements were chosen because they represented a broad range of loading conditions that a torso-worn panel actually encounters during a working shift — not just the movements that were easiest to measure.

A panel that scores well across all these movements feels different to wear than a panel that scores well on a bending test and poorly on compound movements. The difference is noticeable within the first hour of wear, and it becomes more noticeable the longer the shift goes on. This work produced the metric we now describe as our flexibility index — not an industry-standard figure, but our own measurement, using our own protocol, developed because the standard measurements were not capturing what mattered.

4.5 The Impact Behaviour Problem

By this point, the problem was no longer only whether the panel could stop a point load from breaking through. It was whether the structure could take that localised load and spread it before too much of it reached the body underneath.

The useful lesson from biology was not any one animal taken as a template. It was a structural relationship: overlapping hard elements take the first contact, while a softer backing holds them in place, spreads the load, and allows the whole system to move (Martini et al., 2017; Rudykh et al., 2015). That became the working model for FlexGuard. The carbon-fibre tiles were doing the first part of the job. The backing had to do the second.

The practical requirement was straightforward. A point strike should not be allowed to remain a point event for very long. The struck tile has to resist the initial penetration attempt, neighbouring tiles have to begin sharing load quickly, and the backing has to absorb and spread what remains while keeping the tile field geometrically stable. That combination of overlapping hard elements and a compliant supporting layer is what gave the panel both its protective behaviour and its usable flexibility. Remove either part and the system stops behaving its best.

4.6 The Joining and the Measurement of What We Had Built

By Generation 5, the architecture was stable. The substrate material and weave had been optimised. The tile shape, size, and layout were fixed. The attachment method was performing reliably across accelerated wear cycles. The multi-axis flexibility index was being measured consistently.

What remained was the joining geometry between the two layers — the question of how the upper and lower layers were connected to each other, and how that connection affected both the protective performance and the wearable behaviour of the completed panel.

Connecting two flexible layers so that they move together without becoming a single rigid unit is another version of the problem that runs through the entire development history: the requirement for the panel to be one stiff armour against impact and a flexible one around the torso that adjusts itself with body movements.

We experimented with several approaches. A full-perimeter bond between the two layers produced a panel that was stiffer than the individual layers by more than the increase in thickness would predict — the bond was adding stiffness across the full panel area, not just at the edges. A spot-join approach — connecting the layers only at defined points — produced better flexibility numbers but allowed the layers to separate under certain loading conditions, which changed the geometry of the offset pattern in ways that mattered.

Finite-element analysis helped in filtering out the best candidates from different joining designs, for actual validations on test rigs. We found the Z-type and cylindrical connections to be the best compromise between stab resistance and mobility.

Figure 11. FEA assessment of stab-resistance at the joints.

The triangular extruding shapes on each one of the carbon-fibre scales were initially intended to take on and spread impact energies, in the same way as a ridge on a crocodile's bony plate. But by placing these triangles slightly off and over the edges of the square scales underneath, they lock and stop the next scale from rotating inwards towards the body, and as a whole the panel effectively becomes a stiff plate resisting strikes from causing the panel to bend inwards.

By this stage we had the measurement tools to quantify what we had built with a level of precision that was not available in the earlier generations. The flexibility index, the instrumented impact testing, and the accelerated wear protocol together gave a complete picture of panel behaviour.

4.7 What the Numbers Finally Said

By the Gen 5 development cycle, the architecture had reached the point where it could be judged as a finished protective system rather than a concept. ArmorLite's carbon-fibre stab-resistant material achieved a unit-areal-density puncture work of 5.08 × 10⁴ J/(g/m²), with average energy dissipation reaching 0.82 J/mm, representing an improvement of at least 30 percent over other stab-resistant materials currently on the market.

Figure 12. The Amount of Energy It takes To Penetrate Differernt Stab Resistant Materials.

It's also worth noting that, in instrumented testing, the panel keeps backface deformation to ≤10 mm — half the 20 mm threshold specified in the GA68-2024 stab-resistance standard. In everyday terms: the vest doesn't just stop the blade from going through; it stops the blunt force behind the blade from doing damage on its own.

Those figures mattered because they were not being achieved by a rigid plate or by a single continuous layer that simply accepted stiffness as the cost of protection. They were being achieved by a dual-layer articulated structure designed to remain wearable across long shifts, variable posture, and ordinary movement. That was the point at which the project stopped being only an interesting concept. It's far from the perfect answer, but it is ready as a product practical enough to take on real world challenges.

5. What FlexGuard Is, and How It Behaves Under Load

The previous section described how we got there. This section describes what we ended up with — and how it behaves once it is on a body.

The Architecture in Brief

FlexGuard is built from two identical articulated layers, stacked and offset. Each layer is made of small carbon-fibre composite tiles bonded to a continuous woven aramid substrate. The tiles are arranged in rows, with equal spacing between them so the layer can bend and flex along defined lines. The aramid substrate holds the system together, anchors the tiles in their intended geometry, and contributes its own resistance to penetration.

Figure 13. Single Layer FlexGuard Stab Resistant Composite Material.

The two layers are then stacked so that the tile rows in the upper layer are offset from the tile rows in the lower layer. Wherever there is a gap between tiles in one layer, there is solid coverage behind it in the other. No straight-line path passes through both layers without meeting carbon-fibre material somewhere along its travel. That offset is the single most important geometric feature of the panel. It is what makes continuous protection possible without making the panel continuous itself.

Why Two Layers Rather Than One

A single layer of tiles, no matter how carefully arranged, has to make a compromise between flexibility and coverage. Push the tiles closer together and the layer becomes stiffer. Space them further apart and the layer becomes more flexible but starts to open up vulnerable gaps.

Two offset layers resolve that compromise. Each individual layer can be spaced for flexibility — the gaps between the carbon-fibre tile rows on the top layer are backed by rows of the same kind of carbon-fibre tiles beneath on the bottom layer. The system as a whole behaves like a continuous protective surface, while each component layer behaves like a flexible articulated one. The same logic appears in the natural systems described earlier: a pangolin is not protected by a single scale. It is protected by many scales whose overlap pattern ensures that no straight line into the body misses all of them.

Tiles, Substrate, Carrier

The tile geometry, the substrate weave, the attachment method, and the joining geometry between the two layers are the parts of the panel that took the longest to get right.

The tiles are sized and shaped to conform to the curvature of the human torso without buckling, to maintain offset coverage across the full panel area including at the edges, and to allow the panel to flex along different axes when the wearer is sitting, reaching, turning, or running. The tile surface carries an embossed triangular geometry that stops knives from sliding and slashing, and contributes to how the tile handles a sharp point arriving at an angle. It also locks the tiles in one direction but not the other, as described in the joining section.

The substrate anchors the tiles in the intended arrangement, absorbs energy that has passed through or around the tile layer, and routes force across the panel under impact rather than concentrating it at the strike point. The substrate is part of the protective system, not just a carrier for the tiles.

The completed panel sits inside a zip-access pocket in the carrier garment and can be removed for inspection, replacement, and reinstallation without specialised equipment.

A Note on the Tile Material

Carbon-fibre composite is the right material for the tile layer because of what the tile layer is being asked to do, not because carbon fibre is one of the strongest or hardest materials available. The tiles need to be strong, lightweight, easily machined into desired shapes, attachable to textile substrate, and not noisy.

A steel plate of equivalent protective performance would be simply too heavy for the wearer to tolerate over long hours, and a steel tile small enough to articulate would lose much of steel's advantage at the joints. A ceramic tile resists penetration well but fails catastrophically once cracked (Chen et al., 2003) — which is acceptable in a monolithic plate replaced after a single event, but unacceptable in a multi-tile system that needs to keep working across thousands of flex cycles and repeated lower-energy contacts. A fabric-only approach — aramid or UHMWPE without rigid elements — handles flexibility well but does not provide the hard surface resistance the architecture depends on.

Carbon-fibre composite is stiff enough to act as a discrete protective element, light enough to be tolerable in the numbers the panel uses, and tough enough to absorb repeated impact without the brittle failure mode that rules ceramics out. It is fit-for-architecture rather than best-in-class, and that distinction matters: the question we were answering was not which material is strongest, but which material does the job the architecture needs it to do (Crouch, 2019).

Figure 14. Study of Carbon-fibre Material Behaviour Under Stabbing.

Behaviour Under Motion

The panel is intended to be flexible while the wearer is moving and to resist penetration when something is trying to get through it. Those two states come from the same geometry, depending on how the panel is loaded.

Under normal wear, the spacing between tile rows allows each layer to bend along defined lines. The articulation between rows behaves like an accordion: the panel as a whole stretches and contracts, and it also wraps around the body. The multi-axis flexibility work described in Section 4 was about making sure this articulation worked along the lines a body actually uses. The two stacked layers track each other closely enough during motion that the offset geometry stays in place, but independently enough that the combined panel does not become stiffer than either layer alone. That was the joining problem described earlier, and it is also why the panel does not feel like two layers when it is worn — it feels like one.

Behaviour Under Impact

When a concentrated force arrives at the panel — a sharp point, a heavy strike — the geometry shifts. The tiles in the strike zone push against their neighbours. The tiles brace against each other, through the triangles that overlap and bridge across the square tiles underneath. What was a series of articulated rows under motion becomes a coupled surface under load. We call this dynamic locking, because the resistance is not a fixed property of the material — it is a behaviour that emerges from the interaction between elements of the geometry once force is applied.

Beneath the tile in the strike zone, the offset layer is already in position. Force that arrives at a tile edge in the upper layer — the failure mode that causes the most trouble for single-layer systems — is met by the second set of tiles underneath. Force that passes around or through the tile layer is picked up by the aramid substrate, which dissipates impact energy along the tile joint lines, spreading load across a wider area of the panel than a flat backing would.

Figure 15. Testing Stab Resistance of Carbon-fibre Composite Material On Test Rig.

The combined effect is a panel that distributes incoming force across the system rather than concentrating it at the strike point, reduces backface deformation behind the strike, and resists penetration through the layered architecture rather than through any single component of it. The 30 percent improvement in unit-areal-density puncture work is the measurable consequence of that combined behaviour.

What the Wearer Experiences

Described in this much detail, the architecture can start to sound complicated. In practice, what the wearer sees is a slim, flexible insert that naturally wraps around their torso, and weighs less than conventional plates. The complexity is in the design and in the years of iteration that produced it. The experience of wearing it is meant to be simple.

6. Who Wears FlexGuard

FlexGuard has been in operational use for years across policing, military, private security, medical, corrections, and other public-facing roles in several countries. The specific jobs differ considerably — a patrol shift, an emergency department shift, and a close-protection assignment do not resemble each other in tempo, environment, or threat profile — and we do not design as if they do.

What these roles share is the combination the panel was built for: long wear duration, variable posture and movement, and a threat profile in which puncture and stab resistance matter at least as much as ballistic protection. That combination is what brings otherwise dissimilar users to the same category of equipment.

The more important point, from our side of the work, is what the accumulated field use has done for the product. Years of wear across these roles have produced a volume of operational feedback that no laboratory programme could have generated on its own. Wearers have reported how the panel feels at the end of a long shift, how it sits after a year of daily use, how it responds to being pulled on and off in a hurry, how it interacts with duty belts and load-bearing vests and clinical scrubs, and how it feels during the kinds of sudden physical engagements that testing rigs cannot replicate.

Figure 16. Different Shapes of The FlexGuard Composite Material.

Much of what is now built into the current generation of the panel exists because a user in the field told us something we had not seen on the bench. Some of that feedback has driven refinements, including revisions to the geometry and to the way the complete vest is designed — the Stealth Guard Covert Vest's pullover design and the Trekker Coat's civilian appearance both trace directly to operational user requests.

What comes back across all of this feedback, in roughly the same language from users whose jobs otherwise look nothing alike, is a fairly narrow set of requirements: the panel should do what it is supposed to do, it should not draw the wearer's attention away from their work, and it should be ready for unexpected impacts at any time.

7. Where FlexGuard Fits, and Where It Does Not

A protective system is only as useful as the match between what it does and what its wearer needs. We feel it is important to tell who FlexGuard is built for, what it is built to do, and, equally important, what it is not.

Who the Panel Is For

FlexGuard was developed for users who wear protective equipment as part of long working shifts in unpredictable environments, and who need that equipment to remain wearable across the full duration of the shift without compromising the protection it delivers. In practical terms, that includes several distinct groups:

Operational policing roles, where the wearer is moving through different environments across a shift — vehicle, foot patrol, indoor scenes, extended periods seated, sudden physical engagement. The mix of postures and activity levels in a typical shift demands protection that does not fight the body across any of them.

Close protection and security roles, where the wearer needs to remain alert, mobile, and unobtrusive for extended periods, often in contexts where visibly bulky armour would be counterproductive. The slim profile and conformity of the panel matter here for reasons that go beyond comfort.

Custodial and corrections environments, where the threat profile is dominated by edged and improvised weapons rather than firearms, and where staff often wear protection for shifts that span eight, ten, or twelve hours. This is a use case where puncture and stab performance per unit weight is the most relevant single metric, and where long-shift wearability determines whether the equipment is actually worn.

Certain emergency response and frontline civilian roles, including some medical and outreach contexts in environments with elevated risk of edged-weapon assault. These are users for whom traditional rigid armour has historically been impractical, and for whom an articulated covert panel may make wearing protection feasible at all.

What these roles have in common is the combination of long wear duration, variable posture and movement, and a threat profile in which puncture and stab resistance is important.

What the Panel Is Built to Do

FlexGuard is a covert protective insert built to resist stab, spike, and edged-weapon threats, and optimised for movement and comfort. It is designed to deliver that performance at a weight and flexibility that makes long-shift wear realistic — not in the sense that wearing armour for ten hours is comfortable, but in the sense that the panel does not progressively make the shift harder as the hours accumulate. It is designed to be worn under ordinary clothing, in a carrier that allows the insert to be removed for inspection and maintenance, and to be reinstalled without specialist equipment. It is designed to perform consistently across repeated use, with substrate and attachment systems chosen for long-term geometric stability rather than only first-impact performance.

These are specific claims about specific capabilities, supported by the development work described throughout this post.

What the Panel Is Not Built to Do

There are four limits worth being explicit about — two that concern the product itself, and two that concern the context the product sits inside.

FlexGuard is not a ballistic plate. It is not designed, tested, or certified for protection against firearm threats. A user whose threat profile is dominated by ballistic risk needs a different category of equipment, and in many cases needs that equipment in addition to, rather than instead of, a stab-rated panel. We are clear about this distinction with every operational user we work with, because conflating the two threat categories has real consequences.

FlexGuard is not a permanent piece of equipment. Protective panels have a service life. They should be inspected after incidents, replaced according to the manufacturer's guidance, and not relied upon indefinitely on the assumption that performance does not degrade. The removable-insert design is partly there to make that lifecycle visible to the wearer and to the organisation responsible for the equipment.

FlexGuard is not a substitute for situational judgement, tactical training, or appropriate use-of-force protocols. No protective panel is. Armour reduces the consequences of certain failure modes. It does not prevent the failures themselves, and it does not change the wearer's responsibility to operate within the procedures and constraints of their role.

FlexGuard is not optimised for environments dominated by high-velocity fragmentation, sustained heavy load-bearing carriage, or specialist threat profiles that lie outside the design envelope described above. Users in those environments should be specifying equipment built for those specific conditions.

8. What This Means for the People Who Wear It

The engineering described in the previous sections may not be of interest to everyone, and that's why we try to explain here in contexts that people could relate to more easily.

A protective panel that is uncomfortable enough to be loosened, removed during breaks, or left in a locker on a hot afternoon is not protecting anyone during those hours. The non-wear problem is not a failure of discipline on the part of the wearer. It is a design failure on the part of the equipment, and it is the failure FlexGuard was designed to address.

A panel that conforms to the torso, distributes its weight across a larger area, and does not fight against the wearer's movement is a panel that is more likely to be worn for the full shift. For covert users, the practical challenge is that protection becomes compatible with operating in plain clothes without the panel announcing itself through the line of a shirt or the way the wearer moves. For overt professional users in uniform, the challenge is more about fatigue and posture across a long shift — the panel does not progressively make the working day harder as the hours accumulate, and the wearer arrives at the end of the shift feeling fresher than they would in heavier, stiffer equipment.

The Trekker Coat is a useful illustration of what that means in practice. A person wearing stab-resistant protection inside what reads as an ordinary weatherproof coat is not thinking about the fact that they are armoured. They are commuting, or walking their route, or moving through a crowd, in clothing that does not mark them out. The protection is present and doing its job, but it is not occupying any part of the wearer's attention, and it is not announcing itself to anyone else either.

Figure 17. The ArmorLite FlexGuard Trekker Stab Proof Coat

That is close to the condition the design work is aiming for across the whole product line: protection that is available when it is needed and effectively invisible the rest of the time — invisible to the wearer, in the sense of not intruding on the day, and, where appropriate, invisible to the people around the wearer as well. The panel underneath, and the carrier around it, are engineered so that the wearer is free to concentrate on the work in front of them. When the equipment succeeds, the wearer notices it least on the days it matters most.

9. Closing

This post opened with a question about whether the compromise between protection and wearability was actually necessary — whether a panel could be built that a wearer would not only trust, but actually want to put on.

The protective equipment that fails, most often not at laboratory tests, but by being left in the wardrobe and not worn by the wearer, for whatever reason. The protective performance of a panel is the reason the equipment exists. What the design work is for is the conditions under which that performance is actually present — on the body of the wearer, correctly fitted, for the full duration of the shift.

The effort required to achieve that condition is considerable, and it should be. It belongs with the people engineering the panel, the people testing it, and the people manufacturing it. It should not belong with the wearer. The measure of a successful protective panel is how little of itself it asks the person carrying it to think about.

This is an ongoing project rather than a finished one. ArmorLite exists to continue that work — not to declare it complete, and not to sell a finished answer to a problem that does not have one, but to keep narrowing the gap between what the engineering can deliver and what the wearer actually experiences across a working day.

In Summary

The compromise is not mandatory

For most of its history, stab-resistant body armour has forced a trade-off: protect well and be stiff and heavy, or be comfortable and sacrifice performance. FlexGuard was built on the premise that this trade-off is a design choice, not a law of physics. By looking to biology — where the same problem has been solved independently by pangolins, arapaima, crocodiles, armadillos, tortoises, and horseshoe crabs over 400 million years — we identified a recurring architectural pattern: hard elements on the surface, a flexible layer behind, and overlap to maintain continuous coverage during movement.

Architecture beats material

The six-generation development journey revealed that the solution was not a miracle fibre. It was how ordinary materials were divided, layered, joined, and allowed to move relative to one another. The dual-layer offset architecture — carbon-fibre composite tiles on a woven aramid substrate, with the gaps in one layer backed by solid tiles in the other — is what delivers both the protective performance and the wearable flexibility. Carbon fibre was chosen not because it is the strongest material, but because it does the specific job the architecture needs: stiff enough to act as a discrete protective element, light enough to be worn all day, and tough enough to survive repeated loading without the catastrophic failure of ceramics.

Dynamic locking: flexible in motion, rigid under impact

The panel's defining behaviour — articulated and conforming during normal movement, coupled and rigid when a concentrated force arrives — emerges from the geometry of the tiles and their triangular interlocking features. It is not a fixed material property. It is a behaviour the structure exhibits depending on how it is loaded, and it is the same principle the pangolin's scales demonstrate.

Measurable results

FlexGuard achieves a unit-areal-density puncture work of 5.08 × 10⁴ J/(g/m²) with average energy dissipation of 0.82 J/mm — a 30% improvement over comparable stab-resistant materials on the market. These figures were achieved by an articulated, wearable panel, not a rigid plate that accepts stiffness as the cost of protection.

Field feedback completes the loop

Two decades of development, six generations of prototypes, and years of operational feedback from police, corrections, security, and medical users have shaped the current product. The Stealth Guard Covert Vest's pullover design and the Trekker Coat's civilian appearance both came directly from users who told us what the bench could not. What they ask for, in roughly the same language regardless of their role, is simple: the panel should do its job, it should not draw attention away from their work, and it should be ready at any moment.

What FlexGuard is not matters as much as what it is

It is not a ballistic plate. It is not a permanent piece of equipment. It is not a substitute for training or situational judgement. These limits are not weaknesses — they are part of an honest relationship with the people who trust their safety to the product. Knowing what a protective system cannot do is as important as knowing what it can.

Somewhere, right now, someone is putting on a vest before a shift. They are not thinking about Bouligand structures or finite-element models or unit-areal-density puncture work. They are thinking about the day ahead — and whether the panel they are wearing will let them get through it without fighting their own equipment. Twenty years of engineering, six generations of prototypes, and 400 million years of evolutionary biology have gone into making sure the answer is yes.

For the underlying science of how stab-resistant armour stops blades — the five mechanisms and three penetration stages that FlexGuard's architecture is built to defeat — read: How Does a Stab-Proof Vest Actually Work?. For a detailed comparison of the materials used in stab armour, see: Kevlar vs Dyneema vs Chainmail: Stab Armour Materials Compared. For practical guidance on choosing the right vest, read: Which Stab Vest Should I Choose? A Buyer's Guide.

Frequently Asked Questions

How is FlexGuard different from a regular stab vest?

The most common stab-rated vests on the market use a single continuous protective layer — woven aramid, chainmail, or a rigid composite plate — to cover the torso. That construction works, but it forces a trade-off: the stiffer the layer, the better it resists puncture, and the worse it conforms to the wearer's body across a long shift. FlexGuard takes a different approach. The protective layer is built from articulated carbon-fibre tiles, arranged in two offset layers so that the seams of one layer are covered by the body of the next. The result is a panel that behaves like a continuous protective surface under a stab or spike threat, but moves like a flexible material when the wearer is breathing, twisting, sitting, or running. The short answer: a regular stab vest asks the wearer to put up with the equipment for the sake of the protection. FlexGuard is built so that the wearer does not have to make that trade as often, or as severely, across the working day.

Is FlexGuard actually flexible, or is that marketing?

It is actually flexible. The articulated tile geometry is what makes the panel conform — it is not a single rigid plate, and it is not a continuous fabric layer that has been described as flexible because it bends a little. The panel flexes along the lines between tiles, in directions that match the way the torso moves. The honest qualifier is that "flexible" is a relative term in this category. FlexGuard is substantially more conforming than a rigid composite plate, and meaningfully more conforming than most multi-layer stab-rated panels. It is not as flexible as wearing a shirt. A wearer will know the panel is there. What they should not feel, and what the design is built to avoid, is the panel fighting their movement across a long shift.

What animals inspired the design?

Several biological examples informed the early design thinking, but no single animal was treated as a direct template. The useful lesson was broader: overlapping hard elements can take direct impact, while a softer backing can hold those elements in place, spread the energy of the strike, and allow the system as a whole to remain flexible. That is the principle FlexGuard took forward — overlap, load sharing, and compliant support — and then applied in carbon-fibre composite tiles and a woven aramid backing at the scale required for torso-worn protective equipment. The specific animals whose architectures were studied include the pangolin (scale interlocking), Arapaima gigas (Bouligand-structured layered scales), crocodile (domed osteoderm geometry), armadillo (banded articulation), tortoise (fused composite shell), and horseshoe crab (durability across 400 million years).

Does the panel help reduce blunt trauma as well as penetration?

Yes. The articulated tile structure distributes the energy of an impact across a larger area of the panel than a single point of contact, which reduces the peak force transmitted to the wearer's body underneath. The substrate layer behind the tiles contributes additional energy absorption. The panel is optimised for stab, spike, and edged-weapon protection rather than blunt-force protection specifically, but the geometry that delivers the puncture performance also delivers a meaningful reduction in blunt-trauma transmission as a consequence of how the energy is spread.

How does ArmorLite test stress distribution and impact behaviour in FlexGuard?

Testing happens in two layers. The first is instrumented laboratory testing, where standardised stab and spike threats are delivered at controlled energies into the panel mounted on a backing material that approximates human tissue response. Force, displacement, and penetration depth are measured directly, and the puncture work and energy dissipation figures reported in Section 5 come from this testing. The second layer is structural and durability testing across the panel's expected service life — repeated flex cycles, environmental conditioning, and post-conditioning re-testing to confirm that the performance figures hold up after the panel has been worn for the kind of duration a real working shift demands. A protective panel that performs well in a first-impact test on a new sample, and degrades silently over six months of wear, is not a panel we are willing to ship.

How does it hold up after multiple strikes?

Our guidance is the same as the guidance for any ballistic or stab-rated protective panel: after a strike event, the panel should be inspected and, in most cases, replaced. This is the standard position for serious protective equipment, and we do not soften it. That said, the practical behaviour of the panel under multiple strikes is worth understanding. FlexGuard is built from a large number of individual carbon-fibre tiles, and a strike to one region of the panel does not compromise the tiles elsewhere. In real-world incidents, repeated strikes very rarely land on the exact same spot. The articulated construction means that a panel that has taken a strike retains protective function across the rest of its surface area. That is a useful property in an incident, but it is not a reason to keep wearing a struck panel afterwards. The replacement guidance stands.

Can the panels be removed for washing?

Yes. The protective insert is housed in a carrier that can be opened, and the insert removed, without specialist equipment. The carrier itself can be washed according to the care guidance supplied with it. The insert should not be machine-washed or submerged; it should be wiped down with a damp cloth if needed and allowed to dry at room temperature. The removable-insert design also makes routine inspection straightforward, which is part of why we chose it.

How is the panel sized, and does it accommodate different body types?

The panel is supplied in a range of sizes that cover the majority of adult torso dimensions, and the articulated design is meaningfully more forgiving of body-type variation than a rigid plate of the same coverage area. The tile geometry conforms to chest and torso shapes that a single continuous plate would sit awkwardly against. For users at the edges of the standard size range, or for organisations with specific fitting requirements across a workforce, we work directly on sizing during the procurement process.

Is the panel noticeable under a uniform shirt?

For most wearers, under a standard uniform shirt, the panel is not visibly obvious. The slim profile and conformity to the torso are part of what the design was optimised for, and the covert use case has been a constraint on the engineering from the beginning. A close observer who knows what they are looking for may notice the outline; a casual observer in normal lighting generally will not. For close-protection and plain-clothes use, this matters operationally, not just cosmetically.

What standards is the panel tested against?

FlexGuard is independently tested and certified to NIJ 0115.0 Level 1 — the U.S. National Institute of Justice standard for stab and spike resistance. That means the panel has been subjected to a controlled 24-joule stab impact and a 36-joule spike impact using the prescribed NIJ test protocol, and has passed the penetration limit for both threat types.

In plain terms: 24 joules is roughly the energy of a firm one-handed thrust with a combat knife. NIJ Level 1 is the standard accepted by most police forces and security organisations worldwide for edged-weapon protection.

Is independent testing available?

Yes. Beyond our internal testing programme, FlexGuard has been submitted to independent laboratory testing for certification purposes, and we are willing to discuss the results of that testing directly with serious procurement contacts. We do not publish full test reports openly, partly because the reports contain information that is meaningful only with the testing context attached, and partly because the standards bodies whose protocols we test against generally prefer that certification claims be verified through their own channels rather than through summarised marketing extracts. The figures in Section 5 are real, they are reproducible, and we are prepared to demonstrate that to anyone with a legitimate reason to ask.

References