WO2018025007A1

WO2018025007A1 - Double sided wire mesh security fence

Info

Publication number: WO2018025007A1
Application number: PCT/GB2017/051823
Authority: WO
Inventors: Shaun MOORES; Robert KELEMAN
Original assignee: Crh Fencing & Security Group (Uk) Limited
Priority date: 2016-08-01
Filing date: 2017-06-21
Publication date: 2018-02-08
Also published as: GB2552658A; GB201613210D0; EP3491208B1; EP3491208A1

Abstract

A security fence (500) comprises spaced, first (100) and second (300) panels of wire mesh arranged respectively on the front (503) and rear (504) sides of the posts (501, 502), the mesh of the second panels (300) having an aperture size (D1, D2) small enough to impede the insertion of bolt cutters (80). The first panels (100) may be less resistant to attack than the second panels (300), and effectively restrict the use of tools to defeat the second panels so that the fence as a whole is more resistant to attack than it would be if the more resistant panels (300) were arranged on the attack side. The first panels (100) may be made from a single sheet of 4mm prison mesh and the second panels (300) from two sheets of 4mm prison mesh welded together at intervals. The fence may be constructed to satisfy the requirements of Security Rating SR3 as defined in Loss Prevention Standard LPS 1175.

Description

Double sided wire mesh security fence

This invention relates to security fences made from wire mesh. Security fences are commonly used to secure the perimeter of a protected area against intruders, and are commonly made from panels of wire mesh fixed to posts set in the ground.

Figs. 1A - 1C show part of a panel 100 of welded steel mesh commonly known as "358 mesh" or "prison mesh" which is often used for security fences, comprising a plurality of first round steel wires 101 arranged in spaced parallel relation and a plurality of second round steel wires 102 welded to the first wires, the second wires 102 being arranged in spaced parallel relation and orthogonal to the first wires. The first and second wires are regularly spaced apart respectively by a first distance DDI between the axial centrelines of adjacent first wires 101 and a second distance DD2 between the axial centrelines of adjacent second wires 102.

When considered from either flat side of the panel 100, as shown in Fig. 1A, it can be seen that the wires 101 and 102 are connected together in mutually intersecting relation to define an array of apertures 90 between the wires, each aperture having a maximum dimension Dl or D2 when considered in a direction normal to a respective adjacent one of the wires. Like most other commercially available welded mesh panels, the first and second wires of the panel 100 of prison mesh are connected together by resistance welding at each point of intersection, with each of the first wires passing over each of the second wires on the same side of the panel, so that a flat plane can be drawn through all of the welded points of intersection to substantially separate the first wires on one side of the panel from the second wires on the other.

In the typical configuration as illustrated, each of the first and second wires 101, 102 has a diameter D3 of 4mm, the distance DDI is 12.7mm and the distance DD2 is 76.2mm, so that the maximum dimension of each aperture is the dimension D2 of 72.2mm between adjacent ones of the second wires 102, with the minor dimension Dl of 8.7mm being defined between adjacent ones of the first wires 101. The spacing between the first wires 101 is advantageously small enough to prevent an intruder from inserting their fingers into the mesh so as to scale the fence.

In this specification, wire diameters are expressed as the diameter of the structural metal wire excluding any protective coating, and should be construed as nominal or approximate; thus in practice a 4mm wire for example may have a diameter fractionally greater or smaller than 4mm, although always within the range from 3.5mm to 4.5mm. Similarly a round wire may depart from perfect circularity, with the average transverse dimension being taken as the diameter.

The solidity ratio of the panel 100 of prison mesh, defined as a proportion of the total area of the panel occupied by the wires when considered from either flat side of the panel, is approximately 36%. That is to say, when viewed from one flat side, or when projected from one flat side onto a parallel plane (as in a two dimensional drawing as shown), 36% of the total area of the panel appears to be occupied by the wires, and the remaining 64% by the apertures 90. A low solidity ratio is advantageous in a security fence because it improves visibility through the mesh, making it difficult for an intruder to hide behind the fence, as well as reducing wind resistance and, of course, reducing the quantity of steel and hence the cost of the panels.

The resistance of a security fence to different forms of attack may be indicated by a Security Rating determined by testing as defined in Loss Prevention Standard LPS 1175 : Issue 7.3 : Requirements and testing procedures for the LPCB approval and listing of intruder resistant building components, strongpoints, security enclosures and freestanding barriers : BRE Global Limited, September 2015, referred to hereinafter as LPS 1175, available online at www.redbooklive.com and incorporated herein by reference. Security Ratings 1, 2 and 3 as defined in LPS 1175 represent progressively higher levels of attack resistance and are referred to hereinafter respectively as LPS 1175 SRI, LPS 1175 SR2, and LPS 1175 SR3.

LPS 1175 SR3 provides a maximum working time of 5 minutes during a maximum test duration of 20 minutes to create an aperture in the fence of sufficient size to pass through the aperture a test block at least 300mm long and having an elliptical cross section with a major axis of 400 mm (-0 mm/+3 mm) and a minor axis of 225 mm (- 0 mm/+3 mm), using category A, B and C tools as defined in LPS 1175, including inter alia: 1 Axe - 350 mm long/1.5 kg

1 Bolt cutter - 400 mm long

Brick bolsters - 250 mm long x 75 mm wide blade

Cold chisels - 250 mm long x 25 mm wide blade

1 Crowbar - 700 mm long/2.5 kg

1 Hacksaw plus 2 HSS blades

1 Hammer - 400 mm long/1.5 kg

1 Pad saw plus 2 HSS blades

1 Scissor jack - 1500 kg capacity, 100 mm minimum retracted, 200 mm stroke. Tools of this type are likely to be used by a determined intruder to attack a secure perimeter, both because they are quieter than power tools and because they do not require a power supply. In tests, of the available tools, those most useful for cutting rapidly through a panel of prison mesh have been found to be the brick bolsters and cold chisels, which in use are held with the cutting edge in contact with one of the wires and then struck with a hammer with sufficient force to shear the wire, and the bolt cutter.

The bolt cutter 80 is a tool comprising a pair of hardened cutting jaws which are brought together in opposed relation by manual pressure on a pair of handles. The jaws are connected to the handles by an articulation providing a very limited jaw opening with a large mechanical advantage, and in order to apply the requisite pressure to cut the intended workpiece, which might be for example a steel rod or bar up to about 4mm in thickness for a 300mm bolt cutter or about 5mm in thickness for a 350 mm bolt cutter, are necessarily of a heavy construction and relatively wide, blunt profile when viewed from the side, as illustrated in Figs. 3A and 3B. Longer handles provide more mechanical advantage, so that a longer bolt cutter will generally cut more quickly and easily than a short one. However, the longer the handles, the wider the jaws must be (when viewed from the side) to deliver the pressure to the workpiece.

In general it is found that in comparison with a small bolt cutter with jaws of a conventional design, similar to that shown in Figs. 3A and 3B, and about 300mm in length, manual cutting tools of the type comprising a pair of hardened jaws operable by pressure on a pair of handles, but providing less mechanical advantage and having jaws of a more slender or tapering profile, including for example various types of snips and so-called side cutting or diagonal cutting pliers, tend to be much slower and much more difficult to use in cutting mild steel wires of 3mm or more in diameter.

As shown in Fig. 3B, the spacing between the first wires 101 of the conventional panel 100 of prison mesh is also too small to accommodate the jaws of a small bolt cutter 80 in an orientation so as to position one of the first wires 101 between the jaws to cut the wire, although the bolt cutter can be inserted and operated with difficulty by angling it relative to the face of the mesh or impacting it to deform the individual wires. However, as shown in Fig. 3A, the wider spacing between the second wires 102 allows the jaws of the small bolt cutter 80 to be inserted easily in the correct orientation between adjacent ones of the first wires 101, or at least up to the surface of the first wires 101, and opened to accommodate one of the second wires 102 between the jaws. In this manner the second wires 102 of the panel 100 of prison mesh can be cut with bolt cutters along a line parallel with the first wires 101, and then the cut edges of the panel bent away from one another to form a gap so that the bolt cutters can be turned through 90 degrees and fairly easily inserted between the opposed edges of the cut line to begin cutting the first wires 101. An intruder can thus use a small bolt cutter to form an aperture in the panel 100, even when it is fully supported between adjacent upright posts of a fence. Figs. 2A - 2C show another panel 300 of wire mesh, formed as known in the art from a first sheet 100 and a second sheet 200 of wire prison mesh. Each sheet 100, 200 is of identical construction to the panel 100 of Figs. 1A - 1C, comprising a plurality of first wires arranged in spaced parallel relation and a plurality of second wires welded to the first wires, the second wires being arranged in spaced parallel relation and orthogonal to the first wires, with the first wires being more closely spaced than the second wires. In the figures, the first and second wires of the first sheet 100 are indicated respectively by reference numerals 101 and 102, and those of the second sheet respectively by reference numerals 201 and 202. The two sheets 100, 200 are laid face to face and welded together in superposed relation with the first wires 101 of the first sheet orthogonal to the first wires 201 of the second sheet, and with each of the second wires 102 of the first sheet arranged in parallel relation between two adjacent first wires 201 of the second sheet, so that the two sheets together define regularly spaced groups of three closely adjacent wires (201, 102, 201) separated by the more widely spaced wires 201 of the second sheet which are spaced apart between the groups. Each sheet may be formed by resistance welding the respective first and second wires together, and then the two sheets welded together at spaced positions, for example, at intervals along selected ones of the groups of three wires 201, 102, 201, to form a single panel 300.

In the illustrated example, where the second panel 300 is approximately 2500mm square, the two sheets 100, 200 are welded together by two horizontal lines, each comprising 31 equally spaced spot welds, about 50mm respectively above and below its upper and lower edges; by three principal vertical lines, each comprising 15 equally spaced spot welds, arranged respectively centrally and about 50mm inwardly of its left and right hand vertical edges; and by two groups, each of three vertical lines, each line comprising 7 equally spaced spot welds, the three lines of each group being equally spaced apart between the respective left hand or right hand principal vertical line and the central vertical line.

In this configuration, the second wires 202 of the second sheet 200 form the opposite face of the panel from the first wires 101 of the first sheet 100. In the illustrated examples, these second wires 202 are arranged opposite and parallel with respective ones of the first wires 101 of the first sheet when viewed from one flat side of the panel 300, so that the solidity ratio of the panel 300 is minimised, and in the illustrated configuration is approximately 58%. However, the first and second sheets 100 and 200 could alternatively be positioned so that the second wires 202 are arranged between adjacent pairs of the first wires 101 when viewed from one flat side of the panel 300, in which case a higher solidity ratio is obtained.

Fig. 2D shows how the intersecting first and second wires 101, 102, 201, 202 of the panel 300 define apertures 90 of various sizes between adjacent ones of the wires when considered from either flat side of the panel or as projected onto a parallel plane, with the largest apertures 90 having equal dimensions Dl, D2 in a direction normal to each of the adjacent first wires 101 and 201. In the illustrated examples, the dimensions of each of the first and second panels 100, 200 forming the composite panel 300 are the same as that of the panel 100 as shown in Figs. 1A - 1C, so that the first and second wires of each of the panels 100, 200 have a diameter of 4mm, and the first wires 101, 201 of each of the panels 100, 200 are regularly spaced apart by a distance DDI of 12.7mm. The maximum dimension Dl or D2 of any of the apertures 90 of the panel 300 is thus 8.7mm.

In tests it is found that the small size of the apertures 90 of the panel 300 makes it difficult to use bolt cutters to cut the first wires 101 or 201 from either face of the panel, but the panel remains vulnerable to the use of brick bolsters and cold chisels which can rapidly shear through the first wires 101 and 201 to create an aperture. As detailed below, the tests showed that when the second panel 300 was exposed on the outer face of the fence, a cold chisel was capable of cutting the wires of the second panel almost twice as quickly as a bolt cutter. For this reason, a panel of mesh 300 is not found capable of resisting attack up to the level required by LPS 1175 SR3.

By way of example, a security fence constructed with panels 300 is commercially available under the trade name HiSec DualSkin (TM) from Zaun Fencing Limited (TM) of Wolverhampton, United Kingdom (www.zaun.co.uk).

It is known to make a single panel of prison mesh more resistant to attack by providing a wire spacing similar to that of the panel 100, but with larger diameter wire. For example, the first wires can be 4mm in diameter with the second wires 8mm or 10mm in diameter, providing a solidity ratio of approximately 39% or 40% respectively.

It is also known to configure a wire mesh panel with a pairs of second and third wires arranged opposite one another and respectively on opposite sides of a set of more closely spaced first wires, so that the first wires separate the second wire of each pair on one side of the panel from the corresponding third wire on the other. For example, so-called "868 mesh" is another popular pattern, providing first, 6mm diameter wires at a 50mm spacing between pairs of second and third, 8mm diameter wires arranged opposite each other at a 200mm spacing. The larger apertures of this pattern of mesh however provide finger holds which make the mesh easier to climb when compared with prison mesh, so-called because its closely spaced first wires make it very difficult to climb and so ideal for use as a security perimeter for a prison or the like.

Welded wire mesh panels are also available in many other mesh configurations.

Generally for fencing construction the wires may be round, mild steel wires, although it is possible to make the wires from high tensile steel and/or with a patterned profile, as known for example in concrete reinforcement mesh. In such cases the diameter of wire is taken to be the average diameter of the continuous circular section, ignoring the pattern. Unlike some reinforcement mesh however, welded mesh for fencing construction should preferably have a high weld strength, the yield strength of the welds being preferably more than 60%, more preferably at least about 70% of the tensile strength of the wires.

In principle, any wire mesh panel can be made more resistant to attack simply by placing the wires closer together. However, increasing the wire diameter while also reducing the wire spacing increases the cost and the solidity ratio of the mesh, so that a practical limit is reached above which a wire mesh may be regarded as relatively uneconomic when compared with a perforate or imperforate, solid sheet material. For this reason the 40% solidity ratio of the adapted single panel of welded prison mesh mentioned above may be regarded as the highest for any generally commercially available single panel of welded mesh, i.e. for any panel having only two layers of wires, and the 58% solidity ratio of the composite or double panel of prison mesh 300 as the highest for any generally commercially available welded wire mesh panel of any construction.

It is also known to make wire mesh in a woven configuration instead of a welded configuration, but generally similar practical limits apply to the spacing and diameter of the wires. Moreover, since in a woven mesh each wire must pass regularly over and under other ones of the wires, a high solidity ratio representing relatively heavy wires at relatively close spacing is particularly difficult and so particularly expensive to manufacture. For this reason, even if configured as a double or composite panel 300, a welded wire mesh is generally more cost effective and more readily available than an equivalent woven mesh.

In addition to increasing wire diameter or reducing wire spacing, the attack resistance of a wire mesh fence can also be enhanced by arranging wire mesh panels on both sides of the posts. For example, a wire mesh security fence comprising spaced panels of a heavy mesh made from woven high tensile steel wires is available under the trade name Armaweave Plus (TM) from Zaun Fencing Limited (TM) of Wolverhampton, United Kingdom (www.zaun.co.uk). It is also known to arrange a fill material between double wire mesh panels, for example, as taught by EP1925745 Al, which however (and disadvantageously in a security fence application) makes it impossible to see through the fence.

However, even when double panels of mesh are provided, it has generally been found impossible or prohibitively expensive to increase the wire diameter or the quality of the steel sufficiently to achieve the level of attack resistance required by LPS 1175 SR3 in a wire mesh fence. For this reason, wire mesh security fences are generally constructed to satisfy the substantially less stringent standards of security rating SR 1 or, at the most, SR 2 as defined in LPS 1175. In particular, it has not been found possible to construct a practical wire mesh security fence made from welded prison mesh (the material of choice for less secure wire mesh fences), even using an enhanced wire diameter or double or composite panels such as the panels 300, to meet the standard of LPS 1175 SR3. More secure fences are made using other materials, which are usually heavier and more expensive than wire mesh. By way of example, one of very few commercially available security fences certified to LPS 1175 SR3 comprises panels of expanded steel sheet, available under the trade name ExMesh SR3 (TM) from The Expanded Metal Company Ltd (TM) (www.exmeshsecurity.co.uk) of Hartlepool, United Kingdom.

Accordingly it is a genera! object of the present invention to provide a wire mesh fence which provides a relatively higher resistance to attack relative to its cost of

construction. Embodiments of the invention are more particularly directed to provide a wire mesh fence which is capable of meeting LPS 1175 SR3, and which preferably is capable of meeting LPS 1175 SR3 when made from a welded wire mesh, most preferably a prison mesh of relatively economical construction as typically used in less secure fences.

Accordingly the present invention provides a fence as defined in the claims.

In tests it is found that a fence having a double or composite panel 300 of ordinary prison mesh on the attack face does not satisfy LPS 1175 SR3, even when a single panel 100 of ordinary prison mesh is added on the rear face. Surprisingly however it is found that when the panels are reversed so that the double or composite panel 300 is arranged on the rear face and the much less resistant single panel 100 on the attack face, the fence does exhibit sufficient resistance to attack to satisfy LPS 1175 SR3.

Proceeding from this surprising result, the invention is based on the realisation that a fence can be constructed to satisfy LPS 1175 SR3 using two spaced panels of wire mesh, even ordinary prison mesh, when the rear panel has a sufficiently small aperture size to substantially impede the use of bolt cutters providing sufficient mechanical advantage to penetrate it rapidly from the front. An aperture size of 15mm or less is considered small enough to substantially impede the operation of a typical bolt cutter of this type, forcing the attacker to use what would otherwise be slower and less effective tools to penetrate the rear panel. The novel fence provides substantially improved attack resistance even in a relatively economical construction.

The principles underlying the invention together with more specific objects, features and advantages will be better understood from the following description of illustrative embodiments of the invention, which is provided purely by way of example and without limitation to the scope of the claims, and with reference to the accompanying drawings, in which:

Figs. 1A - 1C show part of a panel 100 of prison mesh, from one flat side (Fig. 1A) and edge on, respectively from the top (Fig. 1C) and from the right (Fig. IB) with respect to the view of Fig. 1A; Figs. 2A - 2C show part of a panel 300 of wire mesh formed from two panels of the type shown in Figs. 1A - 1C, in respectively corresponding views;

Fig. 2D is an enlarged view of a portion of Fig. 2A;

Figs. 3A and 3B illustrate a small bolt cutter applied to cut the wires of the panel 100 of prison mesh shown in Figs. 1A - 1C;

Fig. 4 shows a first fence in accordance with an embodiment of the invention, comprising panels 100 as shown in Figs. 1A - 1C and panels 300 as shown in Figs. 2A - 2C;

Fig. 5 is an end view of the first fence;

Fig. 6 is an enlarged view of part of Fig. 5;

Fig. 7 is a top view of an intermediate post of the first fence;

Fig. 8 is a front view of an upper part of the first fence;

Fig. 9 is a top view of an internal corner post of the first fence; and

Fig. 10 is a top view of an external corner post of the first fence.

Reference numerals appearing in more than one of the figures indicate the same or corresponding parts in each of them.

Referring to the figures, a fence 500 comprises a plurality of upright posts 501, 502 arranged in spaced relation. A plurality of first panels 100 of wire mesh as described above are supported, each between respective adjacent ones of the posts to define a front, attack side 503 of the fence. A plurality of second panels 300 of wire mesh as described above are supported, each between respective adjacent ones of the posts in opposed, spaced relation to a respective one of the first panels 100 to define a rear side 504 of the fence. The lower edges of the first, second, or (most preferably) both first and second panels are preferably supported, e.g. by fixing them to rails or bars (not shown), preferably connected to the ground, or by burying them in the ground.

Additional rails or bars (not shown) may be provided above ground level to further stiffen and support the panels. The rear side 504 of the fence will typically face towards a protected area 505, which may be surrounded by a perimeter defined by the fence, in which case the front or attack side 503 of the fence will face outwardly from the perimeter. Alternatively, where the fence surrounds a prison or the like, the protected area may be the area outside the perimeter defined by the fence, in which case the front or attack side 503 of the fence will face inwardly from the perimeter.

As exemplified by the illustrated embodiment, the first and second panels may be connected to the posts by fixings, each fixing comprising a shank, a head integral with the shank, and a fastener formed separately from the shank and engaged with the shank to secure the fixing to the post. In such arrangements, the heads are

conventionally arranged on the front or attack side 503 and the fasteners on the rear side 504 of the fence. Typically the fixings are designed or adapted (e.g. deliberately damaged) so as to resist disassembly once installed.

In the illustrated example, the fixings 506 are bolts comprising a (partially) threaded shank 507 with a head 508 which is rounded to resist attack when exposed on the front side of the fence, and a shear nut fastener 509 which after engagement with the shank separates into a disposable portion (not shown) which in use is engaged by a spanner to tighten the fastener, and a conical nut which threadedly engages the fastener and which, lacking any surfaces which can be engaged by a spanner, is difficult to remove after installation. The conical nut is positioned on the rear side of the fence so that an attacker armed with self locking pliers will have great difficulty in gripping and manipulating it through a hole formed in the mesh panels.

In the illustrated example, the corner posts 502 comprise square hollow steel sections with outwardly extending steel angle sections 510 defining respective front and rear surfaces of the posts against which the edge regions of the panels are positioned and then clamped to the posts using flat bars 511 secured with the fasteners as known in the art. Adjacent panels are overlapped at their edges and secured in a similar way to the intermediate post 501 (Fig. 7), which like the end posts 501 comprises a rectangular hollow steel section.

It will be noted that by fixing the second panel 300 to the rear surface of the post, it is not possible to cut the last wire which is arranged on the rear facing surface of the post and which unites and runs orthogonally to those more closely spaced wires of the same sheet between which the fastener extends. This makes it impossible to lever the edge of the second panel 300 away from the post, merely by cutting the last wire proximate the fasteners.

In the fence 500 as tested, each post was mounted on a baseplate 512 fixed to the ground as shown in Fig. 4, although it could alternatively be sunk into the ground, e.g. in a concrete block 513 below the ground surface 514, as shown in Fig. 5. In the illustrated example, the first panel 100 on the front side 503 of the fence is arranged to extend to the top of the posts 501, 502, and, particularly when the posts are sunk into the ground, and as illustrated in Fig. 5, may also extend some distance below the ground surface 514, while the second panel 300 on the rear side 504 of the fence is arranged to terminate a short distance below the top of the posts. In the illustrated example, the fence 500 is constructed to satisfy at least LPS 1175 SR3, although it will be appreciated of course that the plan layout, height, fixings, post dimensions and other constructional details of the fence can be selected as required. A sample fence 500 was constructed and independently tested as defined in LPS 1175 SR3, and further constructional details of the tested fence 500 are set out in detail below.

Preferably as shown no fill material is provided between the opposed pairs of panels 100, 300, so that wind resistance is minimised and any intruder will be visible though the wire mesh of the panels. In the illustrated example, the panels 100 are fixed to the front faces of the posts and the panels 300 to the rear faces of the posts, so that each pair of panels 100, 300 are spaced apart by a distance D4 which is approximately equal to the horizontal depth dimension of the posts normal to the plane of the panels. In the illustrated example this horizontal depth dimension of each post normal to the panels is 120mm, which is typical for a security fence of this type.

When the sharpened cutting edge of an impact tool such as a bolster, cold chisel or hand axe is held against the wires of a mesh panel and struck with a hammer, or alternatively when a hand axe is swung at the wires, the tool must impact against the wires at angle that reacts the force of the blow to minimise elastic rebound, so that the energy of the blow is absorbed by plastic yielding of the wires. Since elastic rebound is greatest when the blow is directed in the least rigid direction of the fence, i.e. at 90 degrees to the plane of the panels, this requires the tool to be held or swung at an acute angle to the surface of the panel in order to obtain an effective cut.

The approximately 120mm spacing D4 between the panels is not enough to allow a bolster or cold chisel to be effectively used between the facing surfaces of the panels, but is too great to allow the partially cut front panel, in a worst case scenario within the test parameters of LPS 1175 SR3, to be bent inwards until it lies flat against the undefeated rear panel while a bolster or cold chisel is applied to the rear panel and struck with a hammer which is swung at an optimal angle of attack, just in front of the flattened face of the front panel. Accordingly the partially defeated front panel effectively obstructs the operation of the tools used to defeat the rear panel, including in particular cutting tools such as a bolster, cold chisel or axe, but also bolt cutters which, because the apertures are too small to allow the jaws to be inserted at an optimal angle with the tool held at 90 degrees to the panel, must be held at an acute angle to the face of the panel in order to engage the closely spaced wires. In order to give enough room to use the tools effectively to defeat the second panel 300, it is therefore necessary to make a much larger aperture in the first panel 100 than would otherwise be required, which increases the time required to penetrate the fence.

The panels may be connected to the posts in any desired way and the dimension D4 adapted accordingly within reasonable limits determined by trial and error

experimentation, and depending inter alia on the stiffness of the front panel and any supporting rails and the spacing between the posts, to ensure that the partially cut front panel still interferes in this way with the tools applied to the rear panel during testing in accordance with LPS 1175 SR3. By way of non-limiting example, the dimension D4 = 120mm might be reduced or increased by about 25% and then tested to confirm that a sufficient interfering effect is still obtained.

As previously described in detail, each panel 100 or 300 comprises a plurality of wires 101, 102, 201, 202 connected together in mutually intersecting relation to define when considered from the front or rear side of the fence an array of apertures 90 between the wires. The respective wires and apertures form at least most of each panel, and as shown in the illustrated embodiment, may form the entirety of each panel, although parts of the panel may alternatively be formed e.g. from supporting or stiffening solid plates or bars forming fixtures or attachments or the like. When considered from the front or rear side of the fence, the said at least most of each of the first and second panels has a total area, and each panel has a solidity ratio defined as a proportion of said total area occupied by the wires.

As already described in detail, each aperture 90 has a maximum dimension Dl or D2 when considered in a direction normal to a respective adjacent one of the wires. In the illustrated example, the maximum dimension of the apertures 90 of the first panels 100 on the front side of the fence is D2 = 72.2mm, and that of the apertures 90 of the second panels 300 on the rear side of the fence is Dl = D2 = 8.7mm, which is small enough to substantially impede the use of a bolt cutter to rapidly penetrate the panel from the front. In alternative embodiments, the maximum dimension of at least most, preferably all of the apertures of the second panels may be 15mm or less, which is considered small enough to substantially impede the operation of a typical bolt cutter otherwise capable of rapidly penetrating the panel from the front. In particular, it is small enough to require the bolt cutter to be offered to the panel at an acute angle to the surface of the panel, rather than at 90 degrees to the surface of the panel, so that the presence of the partially defeated first panel (constraining the movement of the tool relative to the second panel) prevents the effective operation of the tool.

It will be understood of course that the maximum aperture size of the rear panel can be reduced to any desired value to impede the use of a bolt cutter of unusually slim design, although as mentioned above, the mechanical advantage provided by such a tool is likely to be commensurately smaller than a bolt cutter of more conventional shape and so the tool is likely to be slower and less effective in penetrating the panel.

For example, a maximum aperture dimension of not more than 10mm may be small enough to impede the use of slower cutting tools offering more than an insignificant mechanical advantage if used to cut wires of 4mm or less in diameter, while a maximum aperture dimension of 12mm may be small enough to impede the use of slower cutting tools offering more than an insignificant mechanical advantage if used to cut wires of 4mm or more in diameter.

Advantageously, the novel fence may be constructed more economically when the first wire mesh panel forming the front (attack) side of the fence is less resistant to attack than the second wire mesh panel forming the rear side of the fence.

As shown in the illustrated example, the maximum dimension of at least most, preferably substantially all of the apertures of the second panels, e.g. second panels 300 may therefore be less than the maximum dimension of at least most, preferably substantially all of the apertures 90 of the first panels, e.g. first panels 100.

Advantageously, the solidity ratio of each of the first and second panels may be not more than 60%, in which case the solidity ratio of each of the first panels may be not more than 40% and less than the solidity ratio of each of the second panels. Further advantageously, at least most of the wires of each panel may be round steel wires of not more than 4mm in diameter. In the illustrated example, all of the wires of the first and second panels are 4mm in diameter, and the solidity ratio of each first panel 100 is 36% and that of each second panel 300 is 58%.

The first panel on the front side of the fence is configured so that even when partially defeated it obstructs the use of a bolster or cold chisel to shear the wires of the second panel on the rear side of the fence, sufficiently to extend the time required to penetrate the fence beyond the standard defined by LPS 1175 SR3.

As known in the art, and as verified by the test summarised in Table 5 below, a panel 100 of standard prison mesh when supported between suitable fence posts is found to resist attack (independently of any other panel) to the level defined by LPS 1175 SRI. Accordingly it may be expected that other wire mesh panels capable of satisfying LPS 1175 SRI when used on the front (attack) side of a fence without any further panels may be suitable for use as the first panels of the novel fence, subject to verification by testing the whole fence in accordance with LPS 1175 SR3. Furthermore, since the panel 100 comprises wires of 4mm diameter and a solidity ratio of 36%, it may be expected that any panel of wire mesh, preferably welded wire mesh but alternatively woven wire mesh, having wires of at least 4mm diameter and a solidity ratio of more than about 30% may be a suitable candidate for testing to determine its suitability for use as the first panel of the novel fence. It will be understood that the front and rear panels can be made from other types of mesh than those illustrated, for example, a woven steel mesh. Advantageously however the front and rear panels are made from a welded mesh, which may comprise steel, e.g. ordinary mild steel wires. In this construction, as illustrated, and when considered from the front or rear side of the fence, each panel 100, 300 includes a plurality of first wires 101 of steel arranged in spaced parallel relation and a plurality of second wires 102 of steel arranged in spaced parallel relation and orthogonal to the first wires, and the first wires are welded to the second wires, typically all crossing the second wires on the same side of the panel so that a flat plane can be drawn to pass through all the welded intersections and substantially separate the first wires on one side of the panel from the second wires on the other.

First or second panels having more closely spaced or harder or stronger steel wires of less than 4mm in diameter, e.g. 3.5mm or even 3mm, may also be effective when combined to form the novel fence and tested in accordance with LPS 1175 SR3.

Preferably however at least most of the wires of the first and second panels are steel wires, e.g. mild steel wires, having a circular section, or a transverse sectional area equivalent to that of a circular section, with a diameter of 4mm or more. Of course, first or second panels having wires greater than 4mm in diameter, for example, 4.5mm or 5mm, are likely also to be effective when the wire spacing is not reduced or even when the wire spacing is somewhat increased compared with that shown in the illustrated examples. If a somewhat heavier fence than the illustrated example is desired, providing a still economical construction when compared with prior art fences meeting LPS 1175 SR3 but with additional attack resistance further in excess of that required by LPS 1175 SR3 at the cost of somewhat higher wind resistance and reduced visibility through the fence, then the mesh may be adapted so that the solidity ratio of each of the first and second panels is somewhat increased but still not more than 70%. In this case the solidity ratio of each of the first panels may be not more than 50% and less than the solidity ratio of each of the second panels. Further advantageously, at least most of the wires of each of the first and second panels may be round steel wires of not more than 5mm in diameter.

Of course, if desired, although providing a less economical construction, the first panels could be selected to provide an attack resistance similar to or greater than that of the second panels. Advantageously, as exemplified by the illustrated embodiment and shown most clearly in Figs. 2A - 2D, each of the second panels may be made from two sheets of mesh, for example, from two panels 100, 200, each sheet or panel having a set of widely spaced wires welded to a set of closely spaced wires, with the two sheets being welded together to form a composite construction having closely spaced groups of three parallel wires in which the closely spaced wires of one sheet, preferably on the attack side of the composite panel, are directly welded only to the central wire of the group.

Although many alternative wire configurations may be adopted, it is surprisingly found that when the rear panel 300 is formed with this composite construction it may further increase the time required to penetrate the panel under the test conditions defined in LPS 1175 SR3. When attacked by a bolster or cold chisel, it is difficult to insert the cutting edge between the three closely spaced wires 201, 102, 201 of each group of three, and so the location of the cut is spaced apart by at least one wire diameter from the position of the weld which unites the respective target wire 101 only with the central wire 102 of the group. As a result, it is believed the target wire tends to bend away from the blow rather than cuttng cleanly. Optionally, and as further explained below, this advantageous effect may be further enhanced by arranging the first wires 101 of the first sheet 100 to face towards the rear side 504 of the fence rather than towards the front side 503 as illustrated. In the illustrated example, each of the second panels 300 comprises first and second sheets 100, 200 of wire mesh, each sheet comprising a plurality of first wires 101, 201 of steel arranged in spaced parallel relation and a plurality of second wires 102, 202 of steel welded to the first wires, the second wires of each sheet being arranged in spaced parallel relation and orthogonal to the respective first wires, the first wires being more closely spaced than the respective second wires. The two sheets 100, 200 are welded together in superposed relation with the first wires 101 of the first sheet 100 orthogonal to the first wires 201 of the second sheet 200, and with each of the second wires 102 of the first sheet 100 arranged in parallel relation between two adjacent first wires 201 of the second sheet 200.

Optionally and as shown, this mesh configuration can be implemented using a wire configuration in which the first wires 101, 201 of each sheet are regularly spaced apart by a first distance DDI and the second wires 102, 202 of each sheet are regularly spaced apart by a second distance DD2, the second distance being between four and eight times the first distance and a multiple of the first distance. In conventional prison mesh as illustrated, the second distance is six times the first distance so that the groups of three closely spaced wires repeat after every six regularly spaced wires across the sheet. Advantageously, as illustrated, the first panels may be made from a similar mesh configuration, e.g. conventional prison mesh, which is readily available and relatively economical as well as being advantageously configured to prevent climbing of the attack face of the fence. In this case the first wires 101 of each of the first panels 100 are more closely spaced than the second wires 102 of each of the first panels, so that each of the apertures 90 of the first panels has said maximum dimension D2 normal to the second wires 102 and a second dimension Dl smaller than the maximum dimension and normal to the first wires 101.

In the illustrated example, each second panel 300 comprises two sheets of prison mesh 100, 200 of approximately equal area, with one of the sheets being slightly wider than the other so as to provide a neater overlap and to accommodate the shank of the fastener where the panel extends over the face of the post 501 or 502 behind the flat bar 511. In alternative embodiments both sheets 100, 200 could be exactly the same size, or one of the sheets couid be higher than the other so as to form an additional, third panel integral with the second panel 300 but extending above its upper edge. This configuration provides a heavier construction in the lower region of the fence and a lighter construction in the less vulnerable upper region, and was adopted in the sample fence 500 which was tested as detailed below. In the illustrated embodiment the second panel 300 on the rear side of the fence comprises a set of close spaced horizontal wires 101 facing the front (attack) side 503, a set of close spaced vertical wires 201, 102 behind them, and a set of wide spaced horizontal wires 202 facing the rear side 504. In alternative embodiments however the second panel 300 on the rear side of the fence could be turned through 90 degrees to provide a set of close spaced vertical wires 101 facing the front (attack) side 503, a set of close spaced horizontal wires 201, 102 behind them, and a set of wide spaced vertical wires 202 on the rear side.

As mentioned earlier, the second panel 300 on the rear side of the fence may also be reversed and optionally also turned through 90 degrees, to provide a set of wide spaced horizontal or vertical wires 202 facing the front (attack) side of the fence, a set of close spaced vertical or horizontal wires 201, 102 behind them, and a set of close spaced horizontal or vertical wires 101 on the rear side. Thus, in this configuration, each second panel 300 is arranged with the first wires 101 of the respective first sheet facing towards the rear side 504 of the fence and the second wires 202 of the respective second sheet facing towards the front side 503 of the fence.

In this last mentioned configuration, the rearmost layer of close spaced wires 101 are unsupported in the direction towards the rear side 504 of the fence except by their welded connections to the wide spaced wires 102 which are arranged in front of them and in-between respective pairs of close spaced wires 201 to form groups of three as shown in Figs. 2A - 2D. Without wishing to be bound by theory, it is believed that after cutting through the closer of the two sheets of mesh forming the panel 300, the absence of any support for the first wires 101 on the rear side of the fence other than the welds which unite them at intervals to the second wires 102, in combination with the spacing of at least one wire diameter between the locus of each cut and the position of the nearest supporting weld, may further enhance the tendency of the close spaced wires 101 of the final sheet to bend away from the force of the blow when attacked with a cold chisel from the attack side of the fence via the partially defeated first panel. For this reason they may tend to deform to a greater extent than when arranged facing towards the attack side 503. Despite the greater vulnerability of the second wires 202 to attack by bolt cutters offered at 90 degrees to the face of the panel, this may further increase the time required to penetrate the fence. The sample fence 500 was constructed with the second panels mounted on the posts in this last mentioned configuration, and with the first wires 101 of the first sheet of each second panel oriented horizontally and the first wires 201 of the second sheet 200 oriented vertically, and tested in accordance with LPS 1175 SR3, as will now be described.

Summary of independent test results

A sample fence 500 substantially as illustrated, but with some small differences as earlier mentioned and further explained below, was constructed and independently tested in accordance with LPS 1175 SR3. As shown, the sample fence 500 comprised four pairs of panels 100, 300 supported between five posts 501, 502 to define when considered from either of its first and second sides one 90 degree internal corner and one 270 degree external corner. The longest straight portion of the fence comprising two pairs of panels was 5010mm in length between the outer edges of the respective end and corner posts. The next portion was 2530mm long and the last portion was 2660mm long. The top of the fence was 3060mm above ground level.

The two corner posts 502 were made from 5mm thick steel hollow section 120mm square. The steel angle sections 510 were 60mm x 30mm x 5mm thick and were secured to the square hollow sections by 40mm long welds at 600mm maximum centres. The two end posts and one intermediate post 501 were made from 5mm thick steel 120mm x 80mm steel rectangular hollow section. The 120mm dimension of each post extended in the depth or thickness direction of the fence between the first and second panels to define the dimension D4 between the panels.

Each post was 3070mm high and was welded all round to a baseplate 512 as illustrated, comprising a steel plate 300mm square and 15mm thick. The fence was mounted on a concrete substrate and each baseplate secured to the substrate using four 20 ground anchors, which in practice would be deliberately damaged after installation to prevent disassembly. In the sample fence 500 as tested, the second panels 300 were arranged on the side identified in the figures as front or attack side 503, and the first panels 100 on the side identified in the figures as rear side 504, with the fixings 506, flat bars 511 and other features of the fence remaining exactly as illustrated, i.e. with the heads 508 of the fasteners on the front side 503 and the flat bars and fasteners engaged with the edge regions of the panels to secure them to the posts as illustrated.

Each second panel 300 was made from two sheets of prison mesh 100, 200 welded together at intervals as described and illustrated, but one of the sheets was larger than the other, so that the double or composite second panel 300 extended from ground level up to a height of 2450mm, and the larger sheet extended above the upper edge of the second panel 300 up to a height of 3060mm to form a third panel (not shown) defining the upper edge of the fence on its front or attack side 503.

Each second panel 300 was arranged with the close spaced wires 101 oriented horizontally on the front or attack side 503, the close spaced wires 201, 102 oriented vertically, and the wide spaced wires 202 oriented horizontally and facing towards the first panels 100 on the rear side 504 of the fence.

Each first panel 100 was a single sheet of prison mesh as described and illustrated and was arranged with the close spaced wires 101 oriented horizontally on the rear side 504 of the fence and the wide spaced wires 102 oriented vertically and facing towards the opposite, second panel 300 on the front or attack side 503 of the fence.

The fence was tested in accordance with LPS 1175 S 3 to determine the time required to penetrate the fence, both from the front or attack side 503, and from the rear side 504, with each test being carried out on a previously undamaged area of the fence. It will be appreciated that in the tests carried out from the rear side 504 of the fence, the fence was somewhat more vulnerable to attack than it would have been in the configuration as illustrated, because the flat bars 511, fasteners 509 and baseplate fixings and other constructional details of the fence 500 were arranged as illustrated to resist attack from the attack side 503 and not from the rear side 504. In other respects however, the tests carried out from the rear side 504 of the fence, i.e with the first panels 100 in front and the second pa nels 300 behind, were broadly representative of the behaviour of the attack side 503 of the fence 500, having the first panels 100 in front and the second panels 300 behind, as illustrated.

In the sample fence 500 as tested, the flat bars 511 were made from steel and were 60mm wide and 6mm thick on the attack side 503 and 40mm wide and 5mm thick on the rear side 504. Each fixing 506 was a galvanized steel M8 coach bolt secured using a steel M8 shear nut as illustrated, with the fixings 506 spaced apart at 300mm centres and extending through the posts, bars and panels as shown. The adjacent mesh panels 100 and 300 overlapped by 80mm at the intermediate post 501.

An unequal angle steel section (not shown) 6mm thick was arranged between the base plates of each pair of posts immediately behind the lower edge of each second panel 300, with its 70mm wide leg extending vertically to support the lower edge of the second panel and its 50mm wide leg flat on the horizontal substrate, and secured to the substrate using M12 ground anchors located at 450mm centres. A flat steel bar 60mm wide and 6mm thick extended between the base plates of each pair of posts just above ground level on the attack side 503 of the fence, with its width dimension oriented vertically, and was secured to the vertical leg of the unequal angle using M8 galvanized steel coach bolts located at 300mm centres, having heads on the front side 503 of the fence and shear nuts on the reverse side of the unequal angle, to clamp the lower edge region of the second panel 300 between the bar and the unequal angle. The lower edge of each first panel 100 was unsupported.

The tests were carried out in accordance with LPS 1175 SR3 by an experienced operative who freely selected from the full range of tools, including Category C tools as defined in LPS 1175 SR3 to defeat the fence so as to pass the test block through the fence as quickly as possible within the maximum working time defined by each test, which was recorded on a stopwatch. Preliminary attack tests were carried out to determine the vulnerability of the fence to various methods of attack before commencing the principal tests. The results of the principal tests carried out on the mesh panels are set out in Tables 1 - 4 below, indicating that the panels of the fence satisified LPS 1175 SR3 when attached from the rear side 504 (i.e. with the first panels 100 in front and the second panels 300 behind) but not when attacked from the front side 503 (i.e. with the second panels 300 in front and the first panels 100 behind). Table 5 shows the results of a further test on the first panels 100 in accordance with LPS 1175 SRI. Further tests, not shown, indicated that the fence as a whole satisified LPS 1175 SR3 when attached from the rear side 504.

Table 1

Preliminary manual attack test reference: P104691-A

Attacking the fence from the front side 503 (i.e. with the second panels 300 in front and the first panels 100 behind).

Objective: Cut the second panel 300 using various methods of attack.

Tool category : C

I Comments:

Attempts were made to cut the mesh using plate shears, NWS (TM) pliers, JCB (TM) pliers, i and Draper expert (TM) pliers, all of which were found to be ineffective.

Table 2

Preliminary manual attack test reference: P104691-E

Objective: Cut the first panel 100 using various methods of attack, working through the aperture previously created in the second panel.

Tool category : C

Attack Attack action Attack tools Working time Notes test # (minutes : seconds :

1/100 seconds)

Increment , Running

P10469 Cut the mesh Category B bolt 00:30:00 00:30:00 21 wires were cut 1-El using a bolt cutter

cutter P10469 Cut the mesh Category C bolt 00:30:00 00:30:00 18 wires were cut 1-E2 using a bolt cutter

cutter

P10469 Cut the mesh Club hammer and 00:30:00 00:30:00 10 wires were cut 1-E3 using a cold cold chisel

chisel

P10469 Cut the mesh Club hammer and 00:30:00 00:30:00 Method

1-E4 using a hand axe hand axe ineffective.

Table 3

Manual attack test reference: P104691-1

Objective: Create an aperture through the fence through which to pass the elliptical test block.

Tool category : C

Attack action Attack tools Working tirr ie 1 Notes

(minutes : s sconds :

1/100 secor ds)

Increment Running

Impact a cold chisel into the Club 01:47:64 01:47:64 Vertical cuts made; mesh of the second panel hammer, 40 wires cut per side 300 to create an aperture cold chisel 01:21:63 03:09:27 Horizontal cut made; through which to pass the and Category 31 wires cut elliptical test block C bolt cutter

Bend down the mesh of the Hands 00:07:06 03:16:33 Mesh bent down. second panel 300 to gain 500mm x 360mm access to the first panel 100 aperture created.

Cut the mesh of the first Category C 01:20:00 04:36:33 38 horizontal wires panel 100 using Category B bolt cutters and 8 vertical wires bolt cutters were cut

Bend back the mesh of the Hands 00:06:00 04:42:33 ; 470mm x 260mm first panel 100 aperture created

Comments:

It was possible to achieve the objective of this test within 4 minutes 42 seconds using this method of attack. In this configuration the fence therefore did not meet the requirements of LPS 1175 S 3.

It was considered that in this configuration the fence met the requirements of LPS 1175 SR2. This was because the overall working time exceeded 3 minutes, and Category B tools as used in LPS 1175 SR2 are less aggressive than Category C tools. Table 4

Manual attack test reference: P104691-12

Attacking the fence from the rear side 504 (i.e. with the first panels 100 in front and the second panels 300 behind).

Tool category : C

Comments:

It was not possible to achieve the objective of this test within 5 minutes using this method of attack. I n this configuration the fence therefore did meet the requirements of LPS 1175 SR3.

It was noted that the close spaced horizontal wires 101 on the rear side of the fence deformed rather than cutting cleanly, which significantly slowed down the cutting rate. Table 5

! Manuafattack test reference: P104691-13

Attacking the first panel 100 from the rear side 504 (i.e. with the first panel 100 in front and the second panel 300 behind) to determine whether the first panel on its own satisfies LPS 1175 SRI.

Objective: Create an aperture through the first panel 100 through which to pass the elliptical test block, local to a corner post.

Tool category : A

Comments:

It was not possible to achieve the objective of this test within 1 minute using this method of attack. The first panel 100 therefore satisfied the requirements of LPS 1175 SRI.

In summary, a preferred embodiment provides a security fence comprising spaced, first and second panels of wire mesh arranged respectively on the front and rear sides of the posts, the mesh of the second panels having an aperture size small enough to impede 5 the insertion of bolt cutters. The first panels may be less resistant to attack than the second panels, and effectively restrict the use of tools to defeat the second panels so that the fence as a whole is more resistant to attack than it would be if the more resistant panels were arranged on the attack side. The first panels may be made from a single sheet of 4mm prison mesh and the second panels from two sheets of 4mm prison 10 mesh welded together at intervals. The fence may be constructed to satisfy the

requirements of LPS 1175 SR3.

Further possible adaptations within the scope of the claims will be evident to those skilled in the art.

1 5

Claims

1. A fence (500) comprising:

a plurality of upright posts (501, 502) arranged in spaced relation;

a plurality of first panels (100) of wire mesh, each supported between respective adjacent ones of the posts to define a front, attack side (503) of the fence; and

a plurality of second panels (300) of wire mesh, each supported between respective adjacent ones of the posts (501, 502) in opposed, spaced relation to a respective one of the first panels (100) to define a rear side (504) of the fence;

each panel (100, 300) comprising a plurality of wires (101, 102, 201, 202) connected together in mutually intersecting relation to define when considered from the front or rear side of the fence an array of apertures (90) between the wires, the wires and apertures forming at least most of the panel;

each aperture (90) having a maximum dimension (Dl, D2) when considered in a direction normal to a respective adjacent one of the wires;

characterised in that the maximum dimension (Dl, D2) of at least most of the apertures (90) of the second panels (300) is 15mm or less.

2. A fence according to claim 1, wherein the maximum dimension (Dl, D2) of at least most of the apertures (90) of the second panels (300) is less than the maximum dimension (D2) of at least most of the apertures of the first panels (100).

3. A fence according to claim 1, wherein when considered from the front or rear side of the fence, the said at least most of each panel has a total area, and each panel has a solidity ratio defined as a proportion of said total area occupied by the wires; and the solidity ratio of each panel is not more than 70%.

4. A fence according to claim 3, wherein at least most of the wires of each panel are round steel wires of not more than 5mm in diameter.

5. A fence according to claim 4, wherein the solidity ratio of each first panel (100) is not more than 50% and is less than the solidity ratio of each second panel (300).

6. A fence according to claim 3, wherein the solidity ratio of each panel (100, 300) is not more than 60%.

7. A fence according to claim 6, wherein at least most of the wires of each panel (100, 300) are round steel wires of not more than 4mm in diameter.

8. A fence according to claim 7, wherein the solidity ratio of each first panel (100) is not more than 40% and is less than the solidity ratio of each second panel (300).

9. A fence according to any preceding claim, wherein when considered from the front or rear side of the fence, each panel (100, 300) includes a plurality of first wires (101) of steel arranged in spaced parallel relation and a plurality of second wires (102) of steel arranged in spaced parallel relation and orthogonal to the first wires, and the first wires are welded to the second wires.

10. A fence according to claim 9, wherein each of the second panels (300) comprises first and second sheets (100, 200) of wire mesh, each sheet comprising a plurality of said first wires (101, 201) of steel arranged in spaced parallel relation and a plurality of said second wires (102, 202) of steel welded to the first wires, the second wires of each sheet being arranged in spaced parallel relation and orthogonal to the respective first wires, the first wires being more closely spaced than the respective second wires;

and the two sheets (100, 200) are welded together in superposed relation with the first wires (101) of the first sheet (100) orthogonal to the first wires (201) of the second sheet (200), and with each of the second wires (102) of the first sheet (100) arranged in parallel relation between two adjacent first wires (201) of the second sheet (200).

11. A fence according to claim 10, wherein each second panel (300) is arranged with the first wires (101) of the respective first sheet facing towards the rear side (504) of the fence and the second wires (202) of the respective second sheet facing towards the front side (503) of the fence.

12. A fence according to claim 10 or claim 11, wherein the first wires (101, 201) of each sheet are regularly spaced apart by a first distance (DDI) and the second wires (102, 202) of each sheet are regularly spaced apart by a second distance (DD2), the second distance being between four and eight times the first distance and a multiple of the first distance.

13. A fence according to claim 12, wherein the first wires (101) of each of the first panels (100) are more closely spaced than the second wires (102) of each of the first panels, so that each of the apertures (90) of the first panels has said maximum dimension (D2) normal to the second wires (102) and a second dimension (Dl) smaller than the maximum dimension and normal to the first wires (101).

14. A fence according to any preceding claim, wherein the panels (100, 300) are connected to the posts by fixings (506), each fixing comprising a shank (507), a head (508) integral with the shank, and a fastener (509) formed separately from the shank and engaged with the shank to secure the fixing to the post (501, 502); the heads (508) being arranged on the front side (503) and the fasteners (509) on the rear side (504) of the fence.

15. A fence according to any preceding claim, wherein the fence is constructed to satisfy at least Security Rating 3 as defined in Loss Prevention Standard LPS 1175: Issue 7.3.