Blast Modelling – Getting More Bang For Your Buck

    Bomb blastSecurity managers, and others, ask the quite legitimate question, “What would a bomb do to my building?” The answer can be provided by blast modelling. The problem is that explosive effects are particularly complex, being a combination of gas and hydro dynamics that create thermal increases, pressures, impulses and physical responses well outside the scope of normal structural engineering. This brief article provides guidance for those seeking blast modelling on what to ask for, how to discuss the requirement with the modeller and how to ensure the response is useful.

    When an explosion occurs, the material, usually a solid, is converted into a gas at speeds measured in thousands of metres a second. As gas takes up more volume than a solid, the gas expands, compressing the air around into a hardened wall of air travelling away from the seat of the explosion at roughly the speed of sound. The initial pressure that hits a surface is referred to as the ‘peak incident pressure’ and, if the surface does not collapse immediately, the pressure will continue to build until either the surface fails or the pressure is reflected at the ‘peak reflected pressure’. As the reflected pressure is applied over time, the effect against the building is actually an impulse which is described in terms of megapascals of pressure applied in milliseconds. Depending on the amount and type of explosive and the distance involved, very few materials can withstand this type of assault, which is why explosives are common tools in the mining, agriculture, construction, demolition and military sectors.

    In simplified terms, the three primary products of an explosion are: blast, which is the expanding wall of air; fragmentation, which is dependent on the casing of the improvised explosive device (IED) and the objects nearby; and heat, which near the centre of the explosion can be thousands of degrees. Blast is well researched and the existing models can provide good approximations of expected pressure and impulse effects. Fragmentation is difficult to model as there are many variables, particularly in relation to an IED, and the thermal effects are often omitted as being less relevant than the pressure effects.

    There are a number of reputable providers of blast modelling and there are others that may be able to operate the software but do not fully comprehend the intricacies of blast. Blast modelling services range from simple assessments of explosive effects to complex computational fluid dynamic (CFD) analysis of specific structural elements. The issue is determining what level of information is needed and framing the request for modelling to match.

    In some cases, modelling is not worthwhile. For example, “What will happen if a 500kg charge explodes 5m from my building?” (an actual question posed) does not need modelling. The building will suffer catastrophic damage and people will die. It may be of interest to know at what point progressive collapse of the building may cease. The building will be so badly damaged that it will be uninhabitable, as well as it being a complex crime scene. The response now rests with business continuity, staff support, media management and legal aspects of the business.

    Blast modelling relies on three main inputs: the size of the donor charge (the IED), the type of explosive used and the distance to the target. The structure can then be compared to the calculated blast effects and the responses predicted. Using CFD modelling, how steel beams will bend like plastic under the stresses from an explosion and how a shard of glass can travel hundreds of metres and penetrate a brick wall can be clearly demonstrated. In most cases, the client only wishes to know if the wall or fixtures will fail. The exact manner in which the beam will deform may be of academic rather than practical interest.

    It must be noted that all assumptions related to the use of IEDs are inaccurate as they are, by definition, ‘improvised’. It is not possible to predict what explosive or what quantity will be used, how it will be primed or detonated, how it will be encased or when and where it will be detonated. Intelligence information and consideration of potential motives and attack vectors can assist in determining what is probable. The design of the site, aligned with appropriate policies and procedures, can limit, within certain parameters, where an IED can be placed. A change of even a few metres will have a significant impact on the pressures felt by the receiving surfaces. As a result, any modelling related to IEDs can, at best, only be indicative.

    The first input, the size of the donor charge (IED), can be realistically scoped by visualising that a 5kg weight can be held with an outstretched arm, 10kg can be carried by the side of the body, 20kg is a heavy two-arm carry and anything above that will be transported on wheels. A review of open source media reports on bombing incidents around the world over 15 years suggests that most IEDs are less than 5kg, most vehicle-borne IEDs are less than 30kg, a few in the hundreds of kilograms and very few in the tonnes. Consideration can be given to what size device can be brought into which areas of the site. How close to the building can a vehicle approach? Can anyone enter the public area carrying or wheeling any size item? Are there controls over what can be brought into access controlled areas? The size of the donor charge(s) should be based on what is probable given the existing operating and security environments. The often quoted charge weights of 23, 225 and 500kg are at the higher end of what experience would indicate is probable and their use may provide unwarranted results.

    Related to the size of the charge is the type of explosive. Most modellers use trinitrotoluene (TNT) as it is the standard against which other explosives are measured. Unfortunately, TNT does not reflect the reality of IED construction. TNT is difficult to obtain as it is rarely used in the commercial and military sectors other than as a component of other explosives. Consideration of the size of the expected IED would assist in selecting an appropriate explosive to model; for example, pentaerythrite tetranitrate (PETN) based explosives for small devices and ammonium nitrate, fuel oil (ANFO) or other nitrate-based explosives for larger IEDs. Models using PETN or ANFO will provide different, and probably more realistic, results than those using TNT.

    When discussing the modelling with the provider it is worth defining what assumptions, in addition to those about the IED, will be used. Will fragmentation and thermal effects be excluded? Other aspects which may be ‘assumed out’ include detailed façade fixtures, door recesses, windows, overhangs and vents to the basements, surrounding terrain features, voids under the roads and so on; all of which may be quite legitimate exclusions, as long as the client is aware. It is also important to know if the modeller is calculating only the peak incident pressure, which the structure may withstand, or the larger peak reflected pressure and related impulse.

    In many instances it is not possible to prevent an IED from being introduced to the site, although it may be possible to limit the size. The value of blast modelling may be to identify where the critical services are vulnerable and to make the building and the tenants’ businesses survivable.

    If recommendations are sought as part of the modelling, be aware that most modellers are engineering companies or engineering departments within universities and therefore the answers may be structural – more concrete, larger barriers, stronger fittings – all of which may be relevant to the problem. The client may wish to consider procedural solutions such as access controls, or environmental factors such as creating distance through landscaping. Recommendations should consider the operating environment and image of the site. A friendly, open, family-orientated environment with a 5 star green rating may not be best served by the addition of large expanses of concrete.

    Suggested discussion points in relation to blast modelling include:

    • the size of the donor charge (IED) and why this is appropriate to the particular location
    • where the IED will be located and how this matches the operating environment and controls at the site
    • which explosive will be used in the models and why
    • the assumptions that will be made in the model
    • what level of information is needed as a result of the modelling; enough to know that the wall or column will collapse or detailed analysis of how specific elements will fail
    • whether peak reflected pressures will be provided as well as peak incident pressures and if the impulse loadings are needed
    • if recommendations are sought, will they consider the operating environment and image of the site as well as physical treatments

    Modelling of blast effects is a very useful tool in helping to understand a complex problem, but it is of real value only if the client and the provider both understand the questions and the limitations.