Article Text


Angled shots onto body armour using 9 mm ammunition: the effect on potential blunt injury
  1. Alison Lyall1,
  2. D J Carr1,
  3. C Lankester1 and
  4. C Malbon2
  1. 1Impact and Armour Group, Centre for Defence Engineering, Cranfield University at the Defence Academy of the UK, Shrivenham, UK
  2. 2Mechanical Engineering, Material Science and Civil Engineering, Centre for Applied Science and Technology, Home Office Science, St Albans, Hertfordshire, UK
  1. Correspondence to Dr D J Carr, Impact and Armour Group, Centre for Defence Engineering, Cranfield University at the Defence Academy of the UK, Shrivenham SN6 8LA, UK; d.j.carr{at}


Introduction Some military specialists wear body armour that is more similar to police armour and provides protection from ammunition fired from pistols. During ballistic testing, these armours are mounted on a standardised type of modelling clay and the back face signature (BFS; depth of depression) formed as a result of the non-perforating impact event on to the armour is measured. This study investigated the effect of impact angle on the BFS and on the deformation of the bullet.

Methods Two commonly worn types of armour (HG1/A+KR1 and HG1+KR1) were considered that provide protection from pistol ammunition and sharp weapons. Armours were tested against two types of pistol ammunition (9 mm full metal jacket and 9 mm hollow point) at eight different impact angles (0°, 15°, 30°, 45°, 60°, 70°, 75° and 80°).

Results Increased impact angles resulted in smaller BFSs. Impact angle also affected whether bullets were retained in the armour; as the impact angle increased, the probability of a round exiting the side of the armour increased. Bullet deformation was affected by impact angle.

Conclusions Understanding the deformation of bullets may assist with recreating a shooting incident and interpreting forensic evidence.

  • body armour
  • wounding potential
  • bullet deformation

Statistics from

Key messages

  • Behind armour blunt trauma may be less severe for larger impact angles.

  • Increasing the impact angle increases the probability of a round exiting the side of the armour.

  • Full metal jacket bullets are likely to result in more severe blunt injuries compared with hollow point bullets.

  • Bullet deformation is affected by impact angle, which could be important during a crime scene reconstruction.


Selection of appropriate body armour is based on a risk analysis that considers the threat to be protected against, the area of coverage required and the logistical constraints that it would impose.1–5 Military body armour typically comprises of a multiple-layer fabric-based garment covering the torso that provides protection from fragmentation and ceramic plates that provide protection from rifle bullets.2 ,3 However, some military specialists wear body armour that is more akin to police armour which provides protection from pistol ammunition and sharp weapons. Such armour is tested with reference to the Home Office's suite of test standards for police body armour, rather than Allied Engineering Publication (AEP)-2920 which assesses protective capability against fragments.1 ,6 ,7

Two armour specifications are typically considered: HG1/A offers the lowest protection from pistol ammunition, but is the lightest, thinnest and most concealable armour described in the Home Office standards and HG1 which is general duty armour and provides slightly higher levels of protection compared with HG1/A;1 ,6 both are considered as being suitable for covert use. To provide protection from pistol ammunition, multiple layers of a high-performance fabrics are used, and if protection from sharp weapons is required (eg, KR1 armours), two common methods are used which are laminating the individual layers of body armour fabric or using a layer of (chain)mail in front of the fabric.3 ,8

When a non-perforating pistol bullet hits body armour, the armour deforms due to the kinetic energy deposited into the armour on impact and the body armour is accelerated into the human torso. As a result, the armour wearer may suffer a behind armour blunt trauma (BABT) injury, defined as ‘the non-penetrating injury resulting from the rapid deformation of armours covering the body’.9 Body armour deformation after impact is tested against the UK Home Office standard which measures the depth of deformation (the back face signature (BFS)) into a block of Roma Plastilina no. 1 (a type of modelling clay) located behind the armour.1 The maximum allowable BFS is 44 mm for HG1/A armours and 25 mm for HG1 armours. BFS is not correlated with BABT but is a pass/fail criterion used during testing.1 ,10

In the 2007 Home Office test standards, the armour is mounted on a block of Plastilina and tested perpendicular to the barrel of the weapon system (0°) and at 45°, as this is thought to be worst-case scenario.1 If the bullet is not retained in the armour, that is, it perforates the armour or exits via the side of armour, the shot is deemed not to have met the requirements of the standard;1 previous versions of the standard required the armour to be shot at 30° and 60°.11 ,12 A justification for angle selection does not appear to have been previously discussed in the literature and the effect of impact angle on BFS has also not been previously explored.

Bullets, particularly if they have a soft core such as lead, deform during the impact event and can fracture resulting in forensically important evidence which may be retained in the armour, or if the armour is perforated, penetrate into the body. The bullet deformation is affected by the construction, particularly the jacket,13 which suggests that examining deformed bullets after impact may be forensically important. The effect of impact angle on the shape of bullet deformation has not previously been considered, but might be of interest if recreating a shooting based on forensic evidence.

The aims of this work were to investigate the effect of angled ballistic testing of body armour on BFS measurements and the resulting deformation of bullets.


Eighteen 400×400 mm armour panels were used, nine panels were HG1/A+KR1 (chainmail construction) and nine were HG1+KR1 (laminate construction). The exact construction of the two armours and the manufacturers cannot be reported due to commercial confidentiality, but both types of panels were procured from armour manufacturers listed on the Home Office approved suppliers list. The panels comprised of the protective materials encased in an outer cover as would be used in the actual armour. Two types of pistol ammunition were used: 9 mm full metal jacket (FMJ; 9 mm Luger; 9×19; DM11A1B2)1 and 9 mm hollow point (HP; 9×19; FED P9HST1), both of which are readily available within the UK and overseas, and thus represent a real threat. Ammunition was fired using a proof housing fitted with the appropriate barrel. The impact velocity was measured using a Doppler radar.

Armour panels were mounted on calibrated Roma Plastilina No. 11 and eight impact angles were tested: 0° (perpendicular to the armour), 15°, 30°, 45°, 60°, 70°, 75° and 80°. The barrel was kept in a fixed position throughout testing and the panels and backing were rotated to the required angle (Figure 1). A maximum of 10 shots were conducted on each panel; each shot was located a minimum of 50 mm from the edge of the panel, 50 mm from each other and not located on the same horizontal and vertical axis as described in the Home Office standard.1 A random shot pattern was generated so that not all shots of the same angle and bullet type were located on a single panel, nor same location on a panel. Five replicates of each bullet type/armour type/impact angle were completed. The BFS was measured after each shot using calibrated callipers at the deepest part of indentation.1 The effect of impact angle and bullet type on BFS for each armour type was determined using analysis of variance (ANOVA; IBM SPSS V.22); equality of variance and normality of residuals were checked. The Tukey honest significant difference (HSD) test was used to identify which levels within variables were significantly different (IBM SPSS V.22). Bullets were recovered for postfailure examination.

Figure 1

Impact angles (vertical line=direction of travel of projectile, horizontal line=position of armour when located at 0°, solid block=position of armour at various impact angles).


The mean impact velocity for the 9 mm FMJ projectile was 369 m/s (SD=4 m/s) and for the 9 mm HP ammunition was 360 m/s (SD=6 m/s). Summary BFS data are presented in Table 1. When BFS data were collected, they met the requirements of the Home Office standard.1 Mean BFS was smaller for impacts with HP bullets compared with FMJ bullets irrespective of armour type, or impact angle. Increasing the impact angle resulted in a smaller BFS irrespective of armour or bullet type.

Table 1

Descriptive statistics for measured back face signature (BFS) (mm) due to the two bullet types shot at different angles into two types of armour

From a body armour test perspective, it is useful to statistically analyse the BFS to ascertain whether any variables significantly affected measured BFS. However, due to unbalanced groups, not all combinations of variables could be analysed. The effect of impact angle and bullet type on BFS for each armour type for impact angles ≤45° was considered. According to the ANOVA conducted, the impact angle and bullet type affected BFS for both armours (HG1/A+KR1 F3, 32=19.82, p≤0.001; F1, 32=8.23, p≤0.01 and HG1+KR1 F2, 22=16.07, p≤0.001; F1, 22=6.45, p≤0.05). For HG1/A+KR1 armour, Tukey HSD analysis identified three groups; impact angles of 0° and 15° resulted in similar mean BFS (31.8 mm; 29.8 mm), an impact angle of 30° resulted in a mean BFS of 25.9 mm and an impact angle of 45° resulted in the smallest mean BFS (21.8 mm). For HG1+KR1 armours, 0° and 15° impact angles resulted in similar mean BFS (13.6 mm; 13.2 mm); for 45° impact angle, the mean BFS was 8.6 mm. Thus, BFS was significantly smaller for larger impact angles. For both armours, FMJ bullets resulted in significantly larger mean BFS than HP bullets (HG1/A+KR1: FMJ=28.8 mm, HP=25.9 mm; HG1+KR1: FMJ=13.1 mm, HP=11.0 mm).

Table 1 shows there were a number of shots for which a BFS was not recorded (NR) because these shots resulted in the bullet moving sideways within the armour between fabric layers and exiting the side of the armour pack. This phenomenon is commonly observed during angled testing of body armour, but has not been previously described in the literature. Overall more FMJ rounds exited the side of armour compared with HP rounds. A greater number of bullets exited the side of the armour at higher impact angles and for the laminated armour.

All ammunition deformed on impact irrespective of impact angle or armour type. Figure 2 shows typical examples of bullets that had been shot at armour containing chainmail. At low impact angles, both types of bullets mushroomed. FMJ bullets shot at 70° and 75° were slightly deformed into a wedge; bullets shot at 80° had small broken pieces of chainmail embedded in the tip of the projectile (Figure 2A). HP bullets shot at angles of ≥70° typically had embedded fibres and yarns and broken chainmail links (Figure 2B).

Figure 2

(A) Typical examples of full metal jacket bullets shot into chainmail armour at different impact angles (left to right: 70° (bullets 19, 10), 75° (bullets 30, 27, 21), 80° (bullets 16, 9)). (B) Hollow point bullets shot into chainmail armour at different impact angles (left to right: 75° (bullet 25), 70° (bullet 14), 80° (bullet 11)).

Figure 3 shows typical bullets that had been shot at laminate armour panels. FMJ bullets shot at ≤30° flattened on impact to a familiar mushroom shape (Figure 3A); in comparison those bullets shot at angles ≥30° formed a wedge shape (Figure 3A, B). Figure 3B shows FMJ bullets after impacting laminate armour at 60° compared with those after impacting chainmail armour at 60°. All of these rounds wedged slightly on impact. HP bullets deformed on impact, and fibres/yarns from the body armour were typically found embedded in the bullet tip (Figure 3C). The tips of the HP bullets deformed more than the FMJ projectiles as expected. Figure 3C compares HP bullets shot at 45° and 60°, increased levels of deformation were observed for larger impact angles.

Figure 3

(A) Full metal jacket (FMJ) bullet before testing (bottom left) compared with FMJ bullets shot at various angles into laminate armour (left to right: 45° (bullet 75), 30° (bullets 87, 77, 69, 58)). (B) FMJ bullet before testing (bottom left, not numbered) compared with FMJ shot 60° into chainmail and laminate armour (left=chainmail armour (bullets 37, 23, 4), right=laminate (bullets 89, 79, 65, 54)). (C) Hollow point (HP) bullets shot into laminate armour at various angles compared with an HP unfired bullet (bottom left) (60° (bullets 82, 60), 45° (bullet 78)).


The velocities and associated variability recorded in this study are typical for the ammunition used1 but the resulting BFS and resulting bullet deformation require discussion.

BFS was smaller for HP bullets because this type of bullet mushrooms more readily as it has an exposed core rather than an FMJ; the HP ammunition is less penetrative and thus would result in injury sooner in the unprotected body than the FMJ round. In comparison, the FMJ restricts mushrooming (the bullet does still mushroom when it impacts armour) and would result in injury later in the unprotected body.13 Thus, BFS is expected to be larger for FMJ ammunition compared with HP ammunition. That increasing impact angles resulted in smaller BFS might result in armour wearers not showing significant visible wounding such as a blunt trauma injury if the impact angle is relatively large, which does not appear to have been previously reported.

Some unique characteristics were observed for bullet deformation which may be forensically important, and thus recovery of ammunition when possible is important. Deformation changed with impact angle. Such information could assist in the determination of the location of the firearm in relation to armour. High impact angles onto armours that use chainmail to provide sharp-weapon protection may result in chainmail fracture. These chainmail fragments can become secondary projectiles that may perforate the armour and penetrate the human torso, which does not appear to have been previously reported and is of interest to armour designers and medical personnel. The presence of such chainmail segments may also be of interest to crime scene investigators as they may assist in the recreation of a shooting with respect to impact angle. The increased levels of deformation observed with impact angle onto laminate armour may also be of interest during the recreation of a shooting.


Impact angle and bullet type affect the BFS for two commonly worn body armours that provide protection from pistol bullets. Larger impact angles typically resulted in smaller mean BFS measurements and thus less severe BABT injuries would be expected. It might be possible for the severity of a BABT injury to assist with ascertaining the angle with which a bullet struck an armour. The type of ammunition affected BFS, with 9 mm FMJ ammunition resulting in larger BFS measurements than 9 mm HP ammunition; thus, it is likely that a more severe BABT injury will be observed with 9 mm FMJ ammunition. Caution is required in this interpretation, as many factors can affect BABT, and BFS is not correlated with BABT. Furthermore, this work has only considered two types of pistol ammunition and two types of armour and so the results from this study cannot necessarily be carried over to different ammunition/armour combinations.

This study has identified that the deformation of bullets could provide evidence as to at what angle a body armour was hit, or the direction from which the shot was taken (eg, elevated). This may be important when collecting forensic evidence and recreating a shooting of an individual who was wearing body armour.

Increasing impact angle resulting in an increased probability that ammunition would not be retained by body armour, but might exit out of the side of the armour and be located elsewhere within a crime scene. This is important information for crime scene investigators to consider.


View Abstract


  • Contributors CM suggested this study. AL planned and completed this study with assistance from DJC and CL. AL drafted this manuscript from her MSc thesis; DJC, CM and CL commented on the manuscript draft.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The MSc thesis from which this paper is drafted can be accessed via DJC (

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.