A paucity of accurate data on the physics and dynamics of VBIEDs results in calculations and assumptions being based on the well researched dynamics of military grade explosive devices.  Terror events around the world have shown that the efficacy of VBIEDs is generally significantly lower than that of military grade explosives.

When analyzing blast load from explosive devices we assume that we know the quantity (weight) and quality (type and gravitational weight) of the explosive. Without this data it is impossible to assess the explosion's output. With military type devices our assumptions and calculations are based on intelligence and historical data, which may not be relevant in the war against terrorism. Difference in gravitational weight and quantity may partly account for the 15-40 fold differences in the explosive size estimates described above. There are more factors to consider however when we analyze or calculate the reflected blast pressure generated from improvised explosive devices that were not built to the best military standards. The following factors may account for the huge gap between expected and actual output:

• The purity of the explosive (TNT) and percentage of "dirt" or inert material which can be expected to be 5%-10% higher in an improvised device.
• The shape and geometry of the explosive device which affects the blast result and may even result in total failure of the explosion to reach detonation velocity (as probably happened in the Istanbul event described above).
• Location and strength of the initiation mechanism (detonator and booster) since a poorly designed initiation sequence may result in poor detonation or even total failure.
• Explosive devise casing since confinement, the way that the explosive charge is packed, may dramatically change the explosive reaction and the blast load and fragmentation distribution, especially in improvised explosives.
  

As mentioned previously, a terrorist tries to ensure that the bomb will not be discovered and is less concerned with its purity, shape or initiation mechanism which is usually home-made and not the most effective. The resulting explosive device usually will give a less than optimal explosion effect (pressure and impulses).

 1. Explosive weight

Calculating the explosive weight based on volume is usually done by multiplying by 1.64 g/cm³, the specific gravitational weight for cast TNT however this ignores the fact that the explosive may not be TNT. Military type explosives tend to be denser then commercial explosives. For example, ANFO density is usually between 0.8 - 1.0 g/cm³, with an average of 0.9 g/cm³, compared to 1.64 g/cm³ for TNT so the actual weight of ANFO explosive in the same volume of space within the VBIED could be as little as 55% of the calculated weight. Even if the explosive is TNT, this can come in various shapes and density/gravitational weights – from powder or flakes to cast TNT. The specific gravitational weight for powder or flaked TNT is only 0.9 g/cm³, so here the actual weight of powder or flaked TNT explosive in the same volume of space within the VBIED could be only 55% of the calculated weight..

Therefore the difference in gravitational weight between military and improvised explosive devices can result in up to 45% reduction in explosive weight.

2. Velocity of detonation

Detonation velocity is one of the most important factors in assessing the peak blast pressure and impulses. The speed of detonation is translated into shock waves in the air that are measured by peak incident pressure and impulses. As can be seen in the Figure 7 below, the detonation velocity of a TNT explosive charge varies from under 4800 m/sec for low density (powdered) TNT to around 6750 m/sec for the maximum density of a good quality cast TNT charge with the optimum initiation and geometry shape. This 40% increase in detonation velocity will increase the reflected pressure on a building at any distance, depending on the pressure wave shape and size.

Test data from the early 1960’s that was confirmed in the 1970's indicated that commercially manufactured ANFO in relatively large quantities of above 120 kg had an average detonation velocity of 4200 m/s and an average equivalent weight, on a pressure basis, of 0.82 compared to standard TNT charge with explosion velocity of 6750 m/sec. The positive duration and impulse of the blast wave, although not stated in terms of equivalence, appeared to be somewhat less. It should be mentioned that this was long distance target testing and so was measured a long distance away from the blast. With VBIEDs we may expect a greater difference as the distance is much less and the reflected pressure from nearby buildings may influence the explosion's wave shape. This difference in velocity speed between ANFO and TNT of between 4200 m/sec to 6750 m/sec is responsible for a reduction of 18% in pressure and impulses in long range and even more in short distance targets. Assuming home made explosive devices use even lower velocity explosive materials, the loss of pressure and impulses will be more than 20%.

A reduction of 20% of the TNT equivalent pressure from the 55% left after the volume reduction (See 4.1 above) will leave only 44% equivalent weight, on a pressure basis. This is without taking into account other facts such as the level of professionalism of the bomb maker, the effect of camouflage and the challenges of initiating a VBIED.

3. The explosion sequence

An explosion sequence starts with the firing mechanism which can be a chemical, mechanical or electronic device that will usually cause an explosion of several grams of highly sensitive explosives within the detonator. The small explosion of the detonator will start the detonation sequence by initiating the primer – accelerator, and with or without a booster will start the explosion of the main charge in a steady state velocity of detonation. Any problem or poor performance of one of the elements in the explosion sequence will result in a poor explosion and failure to reach the theoretically calculated explosion results in terms of energy, heat and pressure load. Theoretically, a gram of TNT releases 980–1100 calories upon explosion. A ton of TNT was defined by arbitrarily standardizing it whereby 1000 thermo-chemical calories = 1 gram TNT = 4184 J. Explosive energy is normally calculated using the thermodynamic work energy of detonation, which for TNT has been accurately measured at 1120 calth/g from large numbers of air blast experiments and theoretically calculated to be 1160 calth/g. These values assume all other factors of the explosive charge are to the best possible design and manufacturing condition, and of course not improvised. Choosing a detonator that is too weak or placed in a bad location in the explosive charge or using the wrong primer explosive material or distance between the detonator and the primer, or failure to reach a steady state velocity of detonation will all result in poor detonation or even in failure to explode.

Testing to establish the optimum initiation method for ANFO showed that a poor initiating sequence due to the use of the wrong primer shape resulted in massive losses in the released energy. The size of the primer needed to reach a steady state velocity of detonation in ANFO should be no less then 8 times its diameter. Figure 9 shows that the first 20-30 cm does not really reach detonation speed and therefore the expected pressure and impulse is low. The expected pressure will be reached only after 60 cm in this test. Assuming the length of the charge is only 120 cm, we can expect to lose 25% or more of the total energy realized in this explosion. The combination of detonator and primer give the explosive charge the ability to achieve its optimum, or stable velocity of explosion. Starting with high quality primer that will give a minimum pressure of 5000 mpa in a contact blast will assure a good detonation result in an ANFO charge. Below that pressure, the result will be much less then the optimum, and in an improvised charge will probably result in total failure.

Velocity of explosion of ANFO in various primer detonation pressures	Velocity of explosion of ANFO in a poor initiation sequence

Not enough information was found on improvised charges to provide reliable numbers on the level of losses that we may experience in the reflected pressure and impulses in a VBIED event, but we do know that a number of terror related VBIEDs failed to explode and were found inert as a result of a poor explosion sequence. We believe that this area should be studied more thoroughly to provide reliable data.

4. Purity of the explosive

Explosive purity is a significant element when calculating performance. Inert materials and "dirt" are a direct result of the manufacturing process and can reduce the efficiency of the explosive by about 5%-10% in a commercially available explosive compared to the same type of explosive at its top performance.

Improvised bombs will usually use home made explosive material such as TATP, ANFO, ANS or improvised military type explosives like Nitroglycerin and C4. In many events the bomb manufacturer at his home will use a mixture of military ammunition and military and improvised explosives to build a bomb, particularly in the case of vehicle borne or other large bombs. The fact that the explosive charge is not homogenous affects its performance to the extent that parts of the bomb may not even be initiated while other parts of it will detonate with a relatively good performance. That phenomenon may happen when the less sensitive explosive is inside a casing, for example old ammunition, that will break the detonation wave to the point below the sensitivity point. Introducing inert metal (or other inert material) particles to explosives results in weak detonations. The non-ideal behavior of the explosive is caused by failure of some of the individual detonation wavelets between the metal particles. Subsequent decomposition of the partially decomposed explosive occurs behind the detonation front. This purity/manufacturing quality issue in improvised explosives has not been investigated enough, but is most certainly one of the major reasons for failure in terror related explosions.

Even a very conservative figure of a 10% loss in explosive force in improvised ANFO as a result of inert materials and bad production compared to industrially manufactured ANFO, results in 40% efficiency of the VBIED charge compared to original calculated value (after the volume, velocity and purity issues have been taken into account).

5. Geometry of the explosive device

Geometry is crucial in a commercial, less sensitive explosive material like ANFO, and less so but still important in military type high explosives. ANFO is usually about 6% fuel oil and 94% ammonium nitrate with a density of 0.8 to 1.0 g/cc. Algot Persson has shown that the detonation velocity of ANFO at 0.88 g/cc increases from about 0.35 cm/μsec in 3.5 cm diameter charges in rock or steel to the “ideal” velocity of 0.55 cm/μsec in 26.8 cm diameter charges in rock. He reported detonation velocities for 0.8-g/cc ANFO of 0.325 cm/μsec in 5.1-cm-diameter copper cylinders, 0.389 cm/μsec in 10.2-cm-diameter, and 0.455 cm/μsec in 29.2-cm-diameter cylinders. He also observed that the cylinder wall velocities scale for 10.2- and 29.2 cm diameter cylinders but failed to scale for the 5.1 cm diameter cylinder, which is below the unconfined failure diameter of about 8 cm. This result suggests that the energy release is about the same for 10 and 29 cm diameters even though the C-J velocity changes by more than 15%. The performance of ANFO is strongly dependent upon the diameter of the charge and its confinement.

Assuming the ANFO is ideally mixed and the minimum diameter/thickness is more than 12 cm in a confined location, the energy released will reach the optimum; otherwise there will be massive losses of energy up to the point of 8 cm or less in non confined locations when detonation will stop spontaneously. Traditionally calculated pressure and impulses results for any given explosive charge are made assuming the charge shape is spherical to give the optimum performance. Failure to build an explosive charge in this shape and geometry will result in massive losses of pressure load and even the possibility of the pressure load focusing on the wrong direction.

Results of the test program conducted in May 1968 in the Rattlesnake Flats area about 18 miles southwest of Hawthorne, Nevada indicated that unconfined ANFO charges of about 260 1bs (about 125 kg) are required before stable detonation and blast conditions could be achieved. This was evidenced by the scalability of the pressure distance data for charges weighing from 260 lbs to 4,000 lbs and the leveling off of the TNT equivalence of ANFO for charges weighing more than 260 lbs (about 125 kg). These results confirmed the suspicion that the smaller charges used in the earlier bootleg tests were not sufficiently large in diameter or weight to permit steady state conditions to be realized.

To develop a more effective and efficient combined high explosive blast and shock source, the PRE-DICE THROW II Charge Development Program was conducted by AFWL in 1976. In this program, guided by previous large field trials, small shaped charge tests and hydrodynamic calculations were made consisting of 28 explosive detonations of C-4, ANFO, and TNT at 1-pound, 1/2-ton, and 5-ton yields. The objective was to develop a new charge configuration using ANFO that would provide blast and shock data compatible with the existing TNT surface tangent data base, while significantly reducing the associated fireball/air blast anomalies. The final charge design was a hemispherical capped 0.75:1 cylinder of bagged multiply detonated along the center line. Fireball/air blast anomalies were significantly reduced and air blast and ground shock records followed the TNT standard. While apparent crater profiles were in reasonable agreement, volumes were 23% higher and crater morphology and subsurface deformations were visibly different.

The average detonation velocity of the charge, as measured by a suitable measuring device (LLL) with three rate sticks was 4,790 m/sec (compared to the traditional large scale ANFO charge velocity of 4200 m/sec). This is 14% higher than that measured in earlier tests. There is apparently a direct correlation between detonation velocity and charge size. It is probable that the bulk density of the ANFO in situ increases with the size of the charger. The larger (i.e. taller) the charge, the more the lower layers of prills (spherical shape ANFO particles) are compacted by the weight of the upper layers. It is only reasonable to assume that the reverse will also be true if the charge is smaller and none standard.

Combining the possible failure of the charge's shape assuming it is larger then the minimum 125 kg, with the height /width/length ratio, minimum thickness and confinement problems, we can reasonably assume 10% losses in the explosive's charge performance as a minimum, but much more study is needed to verify this data with improvised explosive in small charges.

Conclusions

Starting from the optimum condition of 100% performance for a military grade explosive device, we have shown that after the volume, velocity and purity issues and shape/dimensional issues are considered, we are left with a very conservative 36% of the original TNT equivalent charge. This goes a long way to accounting for the phenomena that were seen in the two VBIED studies described earlier.

Charles L. Mader assesses our capabilities to calculate the precise blast reaction assuming we know all other factors:

“Some of the practical consequences of non-steady state behavior of explosives have been elucidated. Our previous attempts to describe an explosive with a single effective C-J pressure, regardless of run distance or the initiating system were as certain of failure as if we had used a single C-J pressure and velocity to describe all explosives”.

These words are particularly valid and relevant to our study of improvised explosive devises. There are too many variables which have not been adequately researched in the dynamics of a VBIED reaching steady state velocity of explosion and the resulting pressure and impulse, for us to be able to assess with relative accuracy the event's output.

Nevertheless, if we calculate the known energy losses in a VBIED threat based on the ATF or FEMA threat analysis method using very conservative assumptions, we can demonstrate a 64% loss. In other words, the explosion’s output will be only a third of the original assumed output. The difference between protecting against a 300 kg VBIED compared to a 100 kg VBIED threat in a structural and envelope wall enhancement project may represent millions of US $ of savings.

PURCHASE THE ONLINE SECURITY MANAGEMENT COURSE