Scenarios: Unrest: Home invasion Pt7: Physical protection: Fences: Hostile vehicles

Anti ram barriers


  • The UK standard for hostile vehicle mitigation (‘HVM’), or ‘anti ram barriers’ in common parlance, is BSI’s PAS68, which can certify speed from 16-112kph for six vehicle types from 1.1t to 30t and includes a debris rating. Such rating does not automatically also achieve CPNI rating as it might be defeated by common tools in advance of a hostile vehicle arrival. Vehicles are categorised as 1.5t cars (M1), 2.5t 4×4 (N1G), 3.5t flat bed (N1), 7.5t lorry (N2), 18t lorry (N3) or 30t lorry (N3).
  • Unless you control the approach to stop cars accelerating head on, you can probably ignore PAS170, as that is a simulated test with a 2.5t trolley simulating an N1G vehicle type 4×4 for 10/20mph ram raid bollards. The USA has a similar standard, F3016-19
  • There is also an international standard, ISO’s IWA14, designed by picking bits out of PAS68 and ASTM 2656 (an American standard with six vehicle categories from 1.1t to 30t for 48-100kph given P ratings, with penetration bands and no debris rating or angles included in the rating, and requiring low cohesive compacted soil) by ISO, CPNI and US DoS, which can certify speeds from 16-112kph from 7.2 to 29.5t for 9 vehicle categories but with no debris rating.
  • Be aware that American standards use some different vehicle types as, for example, their lorries have cabs behind (type N2B) instead of over the engine (type N2A), in fact all but one ASTM vehicle class are different to European ones. This can push the datum point back from the bumper to make the rating read like a short penetration.
  • IWA and ASTM (before 2020) measure penetration from a datum line deemed to be the impact face of a barrier, whilst ASTM since 2020 and PAS and CWA measure from the back.
  • There used to be a standard called DoS SD-STD-02.01, with ‘K’ ratings, which dates back to 1985 from the USA, which was replaced by ASTM with ‘M’ ratings in 2007. You still see vendors flogging these tired previous century products, typically boasting of ‘K12’ when in new money it would be ‘M50’. This originally tested stopping a 6.82t pickup at 30/40/50mph within 3/20/50 feet, or within 1m at three all speeds from 2003.
  • There also used to be the European equivalent to PAS68, CWA 16221, since 2010, which rated from 16-112kph and rated debris, but that was abolished in 2018.
  • A new standard is brewing, called ISO 23243.
  • Tests certify products for a vehicle class, weight, speed, angle and penetration distance for vehicle (and debris over 25kg in PAS68). Penetration is only measured from the cabin of a car or the bed of a lorry so the bonnet or cab is allowed to come through without scoring badly in any way, in fact it allows manufacturers to boast of negative penetration if deceleration stops in the length of the bonnet or cab.
  • Look for products with a performance rating. Ignore marketing speak like ‘tested to’, ‘engineered to’ or ‘designed to’ – which is code for ‘failed’ or ‘never been anywhere near a damn test lab’.
  • In the UK tests are done at proving grounds run by TRL, Millbrook or Horiba Mira.
  • To give you an idea using London iconic bollards:
    • Westminister bollards are rated 40mph for 7.5t head on, sometimes combined with retractable bollads which allow 0.5m penetration,
  • City of London bollards are rated 80kph for 7.5 head on but with 22m penetration, sometimes these are shallow mounted allowing 19m penetration,



  • As energy is mass x speed squared, your priority is to slow down vehicles more than worrying about what size they might be. You also want them to hit a barrier at a shallow angle.
  • Energy ratings vary from 200KJ for a car doing 40mph to 8MJ for an HGV doing 50mph. The maximum ratings go up to about 1.2M foot pounds. An example of kinetic energy is a 7.5t lorry at 30mph hits with 667KJ, a 2.5t 4×4 at 90mph hits with 1.2MJ, whereas a 15t lorry at 60mph hits with 10MJ. Yet if you could slow them down to 10mph the 4×4 only hits with 15kJ and the lorry with 296kJ. A 1.5t car hits as hard at 90mph as a 30t lorry at 20mph.
Impact table by Marshalls
  • Beware that ratings are for the tested soil and width, not yours.
  • Barriers just need to be strong enough to deal with the design basis threat energy.
  • Beware buying a lorry-proof barrier that lets a car slip underneath it.
  • You need to control the runup so you can bend it to prevent high speeds – which means you might need to own more land. Ways to slow down vehicles include reducing radius and reducing friction, so the ideal approach is a downhill tight bend to force attackers off the road or to skid or brake.
  • Vehicle acceleration is assumed to be 6-11 feet per second per second.
  • Friction coefficient is normally assumed to be 0.6, but you can design to 1.0 if you fear someone with a Mythbusters workshop fitting tank tracks.


  • The faster the vehicle decelerates, the higher the impact force, eg a 1t indestructible car travelling at 10m/s decelerating within 0.01s would hit a barrier with 1MN force, equivalent to 102t static force.
  • A car hitting a genuinely immovable rigid barrier at 120mph would peak at about 400G. That is worth seeing.
Concrete crash test
  • In reality, barriers and vehicles crumple and/or move, not least because an indestructible barrier is only attached to compressible earth, so deceleration is slower than even crumple zones allow, and kinetic energy is delivered more slowly with a lower impact force.
  • Even rigid barriers only decelerate vehicles within 0.1-0.2s, meaning the 1t car travelling at 10m/s only hits with 50-100kN force, equivalent to 5-10t static force, which only needs three bog standard parking bollards.
  • Vehicles typically decelerate on hitting a barrier at an average of 28G, peaking at up to 100G; they typically need a design force of about 60t.
  • Barriers absorb energy through movement and/or or elastic and/or inelastic deformation.
  • Undeformable immovable barriers bounce back half the energy into the vehicle, destroying it, and absorb the other half.


  • Barriers with foundations bounce back some energy, absorb some and send the rest to soil, although impacts are so short that they rarely have time to rotate foundations, although it can happen slightly with enough energy.
Compressible soil reduces impact force on concrete wall
  • Foundations should be able to hold 75kN/m2.
  • Surface dropped barriers without foundations, like some blocks and planters, rely on friction and possibly damage to the barrier and vehicle to absorb the energy. So as long as standoff is maintained after displacement, a surface dropped barrier relying on friction might be acceptable, but you have no way of testing it and it might be totally negated by a bigger vehicle and may simply become a missile, so barriers with foundations are advised.
Think those concrete lego blocks will stop a jihadi truck? Think again!


  • Ratings typically allow penetration by 1-15m or even 30m, although normally only 1m is acceptable, so normally, rather than bounce off, a vehicle is crumpled while it stretches the barrier, meaning some displacement or penetration is often tolerated where standoff allows.
Lorry penetrates PAS68 planter, but destroyed, although with terrible debris


  • Perimeters can be protected from vehicles using gates, bollards, fences, cables or blocks. The aim is to stop vehicles blasting through your facade, and force an adversary to lug tools by hand and hopefully be left without an escape vehicle. In increasing order of strength, the main options are bollards, cables, planters, rising barriers and sliding gates, and ditches and berms.
  • To prevent ramming whilst preserving aesthetics you could use disguised concrete planters, fountains or walls or just trees or giant rocks, either way with a maximum 4 feet gap to walk through, or you could use a tiger trap, perhaps backed up by a mound or ditch – although they take up about 20 feet.
  • Portable barriers have the disadvantages of rarely working on anything but flat surfaces, needing to be stored and potentially not being deployed in time.
  • Barriers can be a single line, or interlocked using two lines that trap visitors, and/or a final denial barrier nearer the building.
  • Dedicated HVM barriers costs thousands and the electronic accessories can add tens of thousands to that.
  • Because intruders may want a getaway vehicle and to not get injured they are likely to ram raid at lower speeds than the maximum HVM standard ratings which have suicidal terrorists in mind. This means you might get away with cheaper, smaller barriers and perhaps even just a few parking bollards, especially if you can keep approach speeds right down.
  • Consider whether a barrier could be bridged, removed or disabled by a sacrificial vehicle before another glides past.
  • Ideally, keep vehicles far from the home so thieves have a long walk carrying any heavy loot.
  • You will need a survey to check the ground for obstructions and ideally also vehicle performance. To be sure a rating will work at your site needs expertise in traversible routes, foundations, vehicle dynamics and topography. There are HVM courses run by NPSA for engineers and levels of certification from peer reviewed Principal down to Technician Member. You can find advisers from the Register of Security Engineers and Specialists; top HVM consultants are at Arup, DJ Goode, Explora, MFD and SJN (Steve Newbold). Arup also have the top qualified forced entry consultant.

Tyre spikes

  • Especially if you can disguise them, caltrops may be an option, scattered on your driveway.


  • Frontier Pitts do a SR2 IWA14 (USA standard) 2.4m tall 7.2t 30mph anti ram Terra Bifold gate that could work for a driveway (subject to possible planning permission). The gate stretches inwards a few feet to catch a hostile vehicle like a net, finally letting the posts bend to take some of the strain. They do a non-LPS rated version which would be better value as the mid rail lets someone climb over, rendering the penetration rating potentially undermined. They also do 7.2t 50mph gates such as the Terra V, Terra Ultimate Hinged Gate, Terra Hinged Gate and Terra G8 Sliding Cantilevered.


  • For domestic use you can get 18mm 20 ton aircraft arrestor cables U-bolted to 30” high posts, hidden with foliage or a dummy wooden fence outside it, with the ends bolted in concrete deadman anchors about 1m square and 0.5m deep buried 1m within ten feet of fence ends.
  • Jacksons do a PAS68 & SBD terror rated Linebacker covert cable anti ram picket fence with 50cm foundation allowing 7.3m penetration from 7.5t head on at 80kph.
  • Remember that cables can be cut.
Meet Cable Guy


  • Zaun do anti ram PAS68 fences:
    • Hisec Super (2.7-3.6m high 4mm & 6mm wire 12.7mm x 76mm pattern reinforced 358 mesh with metal crash straps at bumper level),
  • RDS PAS68 temporary fence (90x80x45cm 750kg blocks with 1.8-3m high 4mm wire 12.7 x 76.2mm pattern 358 mesh bolted to the front, rated for 7.5t head on at 30mph with 11.6m penetration),
  • Multifence temporary fence (TVCB blocks with cables rated for 7.5t head on at 40mph, with a choice of fence up to 5.2m high bolted on top), and
  • SecureGuard temporary fence of 1.8-3m high 4mm wire (or 6mm vertical wire or Armaweave upgrade) mesh on the front of 80cm x 40cm x 12m 850kg metal blocks.
  • Barkers do:
    • Stronguard RCS 1.8-3.5m high 3mm palisade on 120x80mm posts to stop 2.5 or 7.5t (3.5m penetration) at 30mph, and
  • Stronguard Beam (adding a 4’1″ high crash barrier with 25m anchorage behind the fence) to stop a 7.2t lorry at 80kph with 1.5m penetration.
  • Eagle do EAG08010PL PAS68 planters with versions to stop a 3.5t vehicle at 30mpg or a 7.5t truck at 40mph – their size and stainless steel finish will certainly create a dramatic garden feature.
Eagle EAG8010PL PAS 68 planter


  • Reinforced concrete walls need to be about 3’ high and buried 4’ deep and about 2’ thick.
US DoD concrete ram wall design


  • One technique which mainly needs labour and perhaps relatively cheap landscaping materials is to dig ditches or pile up berms, although these need over 20 feet of approach.
US military ditch schematics

Blocks & planters

  • Surface dropped concrete blocks are available without foundations which can stop a lorry at 30-50mph, depending which side it hits, although they need a big standoff as they allow the wreckage to glide over them until it grinds to a halt.
  • An example of a concrete block with shallow foundation is Townscape‘s Counter Terror Block VED0002, which is sunk 10cm, 50cm high, 149cm x 99cm and weighs 2.2t. They are available in smooth, exposed aggregate, coloured and/or marble finish, so it is by no means impossible to incorporate them into domestic landscaping. They also sell wood slat seats to help disguise them.
  • Townscape also do 1.2m tall HVM planters that are easier to visualise in a residential setting than exposed concrete or steel or the largest planters. Their Toledo planter wraps their Counter Terror Block in hardwood and sits a steel tray on the block to take a flower bed. Their 1.2m tall Dekton VED0021 bollard sleeve is a 18cm deep flower pot that sits on a 1m high Counter Terror Bollard wrapped in a 1.2m tall skin.
Westminster International WG4400


  • Bollards need to be at least 2.5 feet high and 2.5 feet buried, and even planters need foundations (about 3 feet deep by 3 feet thick), and best of all a row of them with a shared foundation. The latter is how shallow versions work, like you see on bridges, although they use a much shallower metal plate or rebar instead of deeper concrete.
  • Ideally bollards use at least half inch steel 8” diameter concrete-filled tubes rising 30” and buried 4’ into reinforced concrete. Bollards are stronger than cable. NPSA maintain a guide to bollard specification classified as official sensitive.
US DoD 2t 30mph bollard foundation design


  • Planters cost £1,000-2,000 each. To build one you use reinforced concrete 3’ thick 18” deep foundations, with a 1’ thick attack wall and 6” retainer wall, or you just use a genuine planter to hide a bollard.
US DoD concrete HVM flowerbed design
US DoD concrete HVM flowerbed design, old version showing rebar spec

Rising kerbs


Vehicle arrestor nets embed in the ground and are launched in front of vehicles, such as the AVB system.


  • If you want to stop a vehicle ramming through your garage door, Charter Global’s Obexion MDXR CPNI roller shutter is rated VADS and IWA14, although apparently only to 20mph for a 4×4, and only in a special frame. It happens to be the strongest roller shutter against burglary too, rated against terror and based on a model rated up to SR5 against manual attack.
Charter Global’s humungous Obexion MDXR IWA14 roller shutter
  • Other options that could cope with low speed ram raid attacks include installing bollards behind a shutter or building steel reinforcements with a framework of 6mm 75x75mm box sections.

Temporary portable barriers

  • NPSA counter terror options which you could DIY-replicate include their cheaper removable temporary event versions of vehicle security barriers for repeated ramming under the ‘vehicle attack delay standard’ (‘VADS’). These include surface mounted spikes, bollards, concrete blocks, sandbags in metal cages, two or three skips, a dustcart, securipods and barges. Bear in mind that you would need at least five to eight tonnes of concrete blocks to achieve even this delay standard for just the width of the block.

Coming technology

  • In 2019 government started looking at new technologies such as novel materials (perhaps glue) and non kinetic smoke, pulse or hacking originally intended for stopping Russian tanks in Europe.
  • The military already use Qinetiq X-net vehicle arrest system which slows vehicles at less than 1G but stops them rolling as far as spikes would, and which you can buy in 3t and 10t ratings. It is ludicrously non user friendly, but maybe worth knowing the concept.

Fun Fact: NPSA PIDS testing is done by the MoD’s Security Services Group’s Technical Services Area’s Protective Services Advice Team at RAF Valley (where security infamously broke down with sentries asleep on the job). The same area designed and built the Almond alarm GUI and AC12M systems.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: