Andy Smith
Mine action specialist
 Developing body armour for demining

I first designed PPE for demining in the mid 1990s and have been credited with inventing the 'demining apron' concept. The HALO Trust demining group did use a simple form of frontal armour before I did but their design did not cover the shoulders and throat and was rather short, so it was very different.

Halo armour and AVS Mk1 armour

The picture shows the minimalist HALO armour and a deminer wearing the AVS Mk1 armour apron.

In fact the Mk1 apron was not my first design of body armour. My goal was to produce something that provided full frontal coverage, was cheap, easy to make without specialist skills, and as ventilated as possible (for use in hot climates). To ensure its availability, it should also be possible to make in a simple workshop.

I did some research into historical armours and was fascinated by the glass armour made during WW1 (an early insight into ceramics) before finally deciding to use materials and skills that I would find anywhere in the world in a vehicle repair workshop.

In 1996, I made and tested demining shields with a polycarbonate window on top for the user to look through. Nothing could be better ventilated that a shield resting on the ground and held in one hand.

demining shield

The shield could be positioned so that it was always between the user and the threat, whatever the working position.They performed well in blast tests in Cambodia (even against fragmentation mines) but the 'experts' did not like the idea and the Cambodian deminers wore no PPE at the time and did not want to do so. Lessons learned, I put the shields aside and made my first demining body armour using 3mm mild steel panels backed with 5mm polycarbonate. The panels overlapped upwards (because AP mine blasts come from below) and allowed unrestricted movement. I did not understand explosives at the time so tested it on Salisbury Plain using 230g PE at 25 cms. This was a harsher test that necessary but it performed well. It was too heavy for me to take forward after I had worked as a deminer and realised that I myself would not have wanted to wear this armour.

steelbody armour

I now know that I could have achieved the required blast protection with thinner and lighter steel, but there was never any real interest in the steel armour so I have not revisited this approach. However, I did use a shaped shoulder plate to carry the weight in later apron designs.

I finally made my first body armour aprons using ballistic aramid, which is much cheaper than Kevlar and widely available, but has the disadvantage of weighing significantly more. There was no standard protection requirement for demining body armour at the time, so I settled on 360 m/s after testing a dozen of my 360 m/s aprons against large anti-personnel blast mines and having no penetrations from mine casing or bits of the ground. In fact it was rare for a single layer of the material to be penetrated, and each apron had 16 layers. The material came in rolls so it was simply folded, then stitched, and the end product was weighed to ensure that no one had miscounted layers. Simple, and capable of being made anywhere in the world, my Mk 1 and Mk2 body armour designs were ridiculed as being a 'ballistic blanket with a hole in it' by one of the big commercial body armour manufacturers of the time. This was true, if unkind. My apron was also effective, cheap and popular at a time when there were no rules and many deminers were given no armour at all.

deminers in Viewtnam wearing the Mk1

The picture above shows deminers in Vietnam wearing my Mk1 armour. While it is a bit more than a blanket with a hole in it, I knew that it was not exactly good looking. That said, some of it is still being worn today.

Andy sewing armour on the street

This picture shows me sewing a Mk2 armour apron with a treadle machine on the streets in Rawalpindi, Pakistan in 1998. At that time, the UN provided a single flak-jacket to be shared by the deminers in each team working in Afghanistan and I wanted to prove that armour could be made in the region at low cost. I am not a skilled tailor but I made one, then got the street tailor to sew several more. I proved my point, but could not convince the UN that armour was necessary and without the promise of a customer, I could not find a reliable partner organisation to teach how to do it.

deminer wearing Mk2 armour

The main change that I made for the Mk2 was to sew a collar standing from the chest which overlapped with the visor and prevented any blast debris getting beneath. This involved sewing through double the thickness of aramid. Given that ballistic aramid is designed to stop penetrations, sewing through 32 layers was a predictable challenge but I was able to hand sew it using a cobbler's tool. Even today, I know of no manufacturer that attaches panels together like this. They make separate panels for the collar, shoulder and groin areas and hold them in place by sewing each panel's cover together, meaning that the join is only as strong as the cover material. Evidence from accidents shows that although the cover and stitching often tears, that does not necessarily affect the outcome in terms of injury, so I was perhaps being too cautious - but a collar that folds over the bottom of the visor in a blast is now the norm for most demining organisations.

When I was amongst those co-opted into developing International Standards for demining most of those involved were ex-military and they wanted the standard for body armour to be the same as that for NATO combat armour, which was protection against fragments travelling at 450 m/s. This level was not enough to protect against bounding fragmentation mines and was irrelevant for measuring protection against blast, but we knew that it was wearable because some demining organisations used armour made to meet that standard. I use the word 'irrelevant' because blast can find a way through loose weaves of Kevlar in a way that it does not through a tightly woven ballistic aramid, but even loosely woven Kevlar stops fragments much better than the the same weight of cheaper aramids. To make my aprons using enough layers of aramid to give 450 m/s fragment protection would have made them unacceptably heavy.

Andy setting up an armour test

I had done many blast tests as shown in this picture, which shows me measuring a 30cm distance from the centre of a PMD6 mine to the base of a Mk3 armour apron and one of my visors. I had also gathered and studied many accident reports from around the world. I knew that anti-personnel blast mines often sprayed fragments of the ground, stones and the mine casing, but not at a velocity high enough to penetrate the 360 m/s aprons already in use. But compromise was needed if we were to to agree what the International Standards requirement should be and there was no other formal test to prefer, so I agreed that the NATO STANAG 2920 V50 standard test for fragmentation should be adopted. Some users of my Mk1 and Mk2 armour then had chest panels of polycarbonate inserted to raise the V50 of their aprons but I did not like this because the inflexible panels were uncomfortable and added significantly to the weight.

fragment firing rig

I later made a rig to fire fragments at my designs to determine the V50 (a V50 is the Velocity at which 50% of the strikes compromise the armour) but I did not test any Mk1 and Mk2 aprons with polycarbonate inserts. I knew from tests and real accidents that the MK1 and Mk2 apron did protect against the effects of blast mines reliably, and I seriously doubted that the polycarbonate insert made much difference to the V50. The insert was not my idea, so I let that rest.

I needed money to make a new MK3 apron using lighter material (Kevlar) and to work on its looks so that it was more attractive to the demining purchasers who often judged everything by appearances. By that time I was becoming known and was actually invited to submit an application for funding support. This meant that I was able to buy a heavy sewing machine and stocks of Kevlar to design and test the Mk3. See some general comments on Kevlar body armour.

ABS Mk3 armour

My Mk 3 above included a shaped piece of 5mm polycarbonate that fitted over the shoulders and allowed the entire weight of the apron to be supported without applying pressure to the back of the neck. All other aprons have straps which I wanted to avoid so that the armour could hang loose and allow free air circulation. I also added an additional flap behind the collar so that the front of the throat was protected despite the armour having a large head hole. This is probably my best ever design of armour but the shoulder support needed more work (because it was not comfortable for everyone) and the main body panel was over-sewn too much, which made it stiff. Ironically, although this was the first time I had any funding support for armour design, my contract with the donor obliged me to be in other places so I had no time to make refinements. I left my designs being made by a small company in Zimbabwe.



Having funding support meant that I was able to take twenty of the Mk3 armours shown above (with my design of blast visor) to Afghanistan for field trials, which went very well. I was known by many of the demining groups in Afghanistan by that time, having traveled around with them on several occasions while conducting tests for others, studying demining methods and investigating accidents.

Afghan trial of PPE

After a user trial, which got entirely positive feedback, the armour was blast tested using PMN mines. The apron and visor sets were presented on simple frames of bent reinforcing rod so that the blast could push them backwards as it would a deminer.

testing the armour in Afghanistan

The mines were put on the ground surface and given the worst kind of environmental fragmentation by covering them with stone chippings.

Ideally I would have put weighed sandbags on the support frames, buried the mines flush with the surface, only presented one apron and visor to each mine, weighed the extra initiation charges and photographed each test before and after... but I was not there. Demining in Afghanistan at the time was controlled by the UN Mine Action Coordination Centre for Afghanistan (UNMACCA) which was based in Islamabad in Pakistan for security reasons. The UN's Chief Technical Advisor (CTA) had approved the tests but then decided that he would not allow me to be present because I had embarrassed him by letting it be known how poor the PPE provided to Afghan deminers was at that time. I tried reasoning that, after all, I would not have been there at all if there was no need for PPE, but he was not in a listening mood. When I refused to write a message to the world saying that Afghan deminers had good PPE, he refused to allow me to conduct the tests myself. At that time I could only get into Afghanistan if I had his approval so the tests were conducted by a UN Technical Advisor who promised to follow my written instructions. He brought the aprons and visors back for me to examine afterwards.

The armour and visors were damaged but there were no holes so they were not compromised. They had performed exactly as I wanted them to do but I could not really prove that because he had failed to record the tests on camera or on paper. The photograph above is almost the only one I have. Throughout the history of Humanitarian Demining this is the only time that multiple demining PPE sets have been blast tested using real mines and although the results are well recorded, that really is not enough. In retrospect, I can see that the friendly Technical Advisor did exactly what he was told to do by his boss, the CTA. The CTA had obvious problems. An ex-officer faced with someone like me felt obliged to show me who was in charge. He wasted my effort and a significant amount of my donor's money without even realising what he had done. I know this because he later claimed that I owed him 'friendship'. I blame the alcohol.

Whatever, I had other demands on my time and a lot more to learn, but I will always regret having missed an opportunity to help the Afghan deminers back then.

Over the four years to 2000, the most common request I had from Afghan deminers was for genital protection. From the accident record I knew that severe genital injury did occur more often in Afghanistan than in other countries, probably because Afghan deminers often squat rather than kneel to excavate, as shown in the trial pictures above and in the picture below.

Afghan deminer squatting

WIth no body armour, this man has turned sideways when squatting to avoid presenting his genitals to the worst of any blast. My Mk3 apron design with no straps was intended to allow the deminer to either kneel or squat face on to the working area with the apron hanging to the ground in front of him, so protecting thighs and genitals. But there was no sign of the Afghans being given frontal armour at that time. At close to $400 each, giving each man body protection was considered too expensive.

However, genital protection was the deminers' priority and could be low cost. Wanting to protect the femoral arteries as well as the genitals, I could not suggest wearing the 'cup' or 'box' used in some sports, so I improvised two simple designs of genital protectors made using 5mm polycarbonate sheet.

polycarbonate genital protectors

One was designed to be worn inside the trousers (held in place by the trousers) and one to be worn outside (held on the belt). The inside version was actually more comfortable but 'one-size' definitely did not fit all. Attaching to the belt gave more leeway in making one size fit all, but even that design was not comfortable for everyone. These genital protectors provide the same level of protection as that provided by blast visors, so are a lot better than nothing. Twenty were made for an Afghan group but I don't know whether anyone is wearing them now.

Deminers in several countries had told me that blast visors were too hot and uncomfortable to wear for long periods. They were said to distort vision, give the wearer a headache and steam up on cold mornings. I had worn them a lot so I knew that, when properly adjusted, they could be comfortable and I had no problem seeing through my own visors, but I had to agree that the lack of airflow behind them did feel claustrophobic. I have tried several ways to overcome this over the years, and one of these was to attach visor material to the armour rather than the wearer's head. This allowed the wearer to have a flat piece of polycarbonate to look through (the chest-mounted visor was cold-pressed so had a flat front) and left the head unencumbered.

a visor attached to armour

I tried several variants like the one shown above. It was necessary to use more polycarbonate material so that the face was protected whichever way the wearer moved his head. Trying them myself (a good rule) I found that the greater distance from the eye to the visor material made it more difficult to see through and the flat face frequently reflected the sun and blinded me. I tried sun-tints and reducing the size but I would not have wanted to wear it myself, so I rejected the idea.

It was almost a decade before the people who made my first armour designs in Africa asked me to come up with a new design because the armour they were making was not selling well. I may not have been the best person to ask, but I was the only one who would be cost-free. As a result, I made two more armour designs, one in 2009 and one in 2011.

designing armour in a garage in Wales

My untidy domestic garage is my workshop and it has never been well equipped. I drew sketches, made paper templates and dusted off my massive old sewing machine (since donated to the charity 'Tools for Self Reliance' because I really will never make armour again) and set to work.

By that time, the range of demining aprons being made and sold around the world was extensive. Many of them were complicated, including genital and thigh protection but there was no evidence that they prevented injury any better in accidents. In fact, the accident record still showed that heavy or complicated protection was often laid aside when supervision was lax. So I designed new aprons that fell in between the full frontal coverage offered by my first simple designs and the more popular 'good looking' designs of others.

Mk apron armour design

Remembering my Afghan friends, my Mk4 above included an optional soft genital protection flap that folded inside the cover when not wanted. Whether or not the genital protection was used, the flap over the groin could be folded up (held with Velcro) when walking around.

Mk 5 armour design

My Mk5 above was designed to be easy to manufacture in different sizes. The thigh flaps have a polycarbonate former inside them to keep them in place and the small collar is augmented with a second collar inside the visor. The small collar makes it easier to look down and see your feet and still interfaces with the visor when doing so, but if you were looking up, blast could get to the throat if there were not a second collar in the way. This is the lightest armour I ever designed, so it is the armour I take with me when I know I might need it.

Kevlar weaves have got tighter and cheaper so it is easier than ever to achieve the NATO STANAG 2920 V50 of 450 m/s required in the standards. While that is still an inappropriate measure of an armour's ability to protect against blast, it does provide a measure for protecting against fragmentation. However, 450 m/s is not enough to stop the fastest fragments from the threats deminers face. Ceramic or steel plates could provide that, but only to small areas and these plates are both heavy and uncomfortable to wear so no one uses them routinely in demining. Some attempts at making flexible ceramic armour for combat troops have been publicised, not always politely, but the designs are very complex and expensive and not available for manufacture near to the area of need. Because I felt that someone should, I decided to have a go at making flexible ceramic armour in my garage.

Hard ceramic materials are expensive so I started with the cheapest which is ceramic alumina. Before long I had bitten the bullet and gone for the far harder but also much more expensive 'ytria stabilised zirconia'. I wanted to avoid using ceramic plates as fish-scales or 'dragon skin' because the overlap is either more than required, or less, which must make a good end product heavier than necessary. I decided to use ceramic balls that are widely manufactured for use in grinding mills, so widely available.

No longer having access to a working STANAG 2920 fragment firing rig, I had to test using bullets. I did not have easy access to guns either, so most of the testing was done in Africa. Over three years (this was done in my spare time) I made close to thirty different test panels before finally making a flexible panel that could stop multiple strikes of 5,56 NATO US rifle ammunition moving at close to 1000 m/s at a distance of five metres.

Flexible ceramic armour

Having started my tests with 25mm ceramic balls (which the bullets tended to break and push aside) I used smaller and smaller balls before finally choosing 3mm diameter balls as shown. The orange panel is made by stacking 6 flexible layers that slide over each other as the panel flexes.

the test setup

The final testing was conducted in UK. A marksman fired through a double chronograph set up to measure the velocity, hitting each panel with two shots. The test panels had a plasticine backing to record blunt trauma, and a sandbag behind (in case of failure). The orange flexible ceramic panel shown above right was held in place with black tape.

The fragments that deminers face are much easier to stop than bullets because they are an irregular shape and 'tumble' through the air. My flexible ceramic material could be made in large sheets and it is likely that a few layers added to existing armour would be able to stop almost all of the fragment risks that deminers face, but it is heavy. My full thickness test panels are almost as heavy as the hard ceramic panels they were designed to replace.

I had successfully made a flexible product but I had failed to make it cheap, light or to make it in a process that could be readily transferred to the areas of need. My process of hand manufacture is far too slow and really should be automated. Some 10k down, I lacked the means to continue so I published what I has done for others to use. See the Flexible Ceramic Composite page and the link to how to make it.

See also my work on developing safer tools for demining, developing face and eye protection for demining, on area preparation machines, and on disrupters. For a more complete story of my work in HD and HMA, click here.


The following people have provided practical help in my body armour work: Matthew Chambers, Matthew Smith, Trevor Thomsen, Christopher Marizani, Cosmos Mutemba, Ed Pennington-Ridge, retired Brigadier John Hooper and Cris Chellingsworth. The Territorial Army in UK, Roscoe in Zimbabwe, and many deminers and demining managers have also helped to get things tested and sometimes offered design advice. Thanks are also due to retired Colonel George Zahaczewsky from US ARMY CECOM NVESD who gave deminer protection the same priority as I did.