A Blast and Ballistic Resilient Air-beam Shelter System

  • Upload
    fogdart

  • View
    221

  • Download
    0

Embed Size (px)

Citation preview

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    1/14

    A BLAST AND BALLISTIC RESILIENT AIR-BEAM SHELTER

    SYSTEM

    D.V. Ritzel1, S.A. Parks

    2, J. Crocker

    3, H.A. Warner

    4

    1Dyn-FX Consulting Ltd., 19 Laird Ave. North, Amherstburg, Ontario N9V 2T5, Canada;2ORA Inc., 71 Commerce Pkwy, Suite 107, Fredericksburg, Virginia 22406, USA;

    3Martec

    Ltd., 1888 Brunswick St., Suite 400, Halifax, Nova Scotia B3J 3J8, Canada;4Dynamic Air

    Shelters Ltd., 220 4441 76 Ave. S.E., Calgary, Alberta T2C 2G8, Canada

    ABSTRACT

    Lightweight temporary shelters such as tents and work-trailers are sometimes required by armed forces, civil

    authorities, or emergency responders in areas at risk from enemy action, terrorist attack, or accidental explosions.

    An extensive program of full-scale trials, computational modelling, and component testing has been conducted todevelop and validate a novel deployable shelter system based on air-beam technology having high resilience to

    blast and ballistic threats. The shelter has no hard framing, paneling, or shear connections in its construction but

    is self-supporting by means of air-beam arches of large diameter polyester-fabric tubing which are lightly air-

    pressurized. Free spans to 40m can be enclosed. Such a structural system flexes and rebounds when subjected

    to blast, impact, or seismic action ultimately taking loads as membrane and tensile stresses for which the

    materials are inherently strong. The mode, rate, and extent of wall deflection can be largely controlled by flexing

    lateral supports. An optional self-supporting geotextile curtain-wall can be incorporated with the shelter which

    has been designed to provide high ballistic protection using minimal local soil fill. The integration of the air-

    beam structure, flexural support system, and ballistic curtain-wall is the basis for the Integrated Blast Resilient

    Shelter (IBRS) concept. The IBRS design is modular to meet a range of requirements and constraints.

    Due to the highly responsive nature of the fabric surfaces under blast, computational modelling of the blastencounter requires an approach using fluid-structure interaction (FSI). FSI modeling which links the FE code

    LS-DYNA with the blast CFD code CHINOOK has been applied to analyze the complex response dynamics and

    optimize the IBRS design. Full-scale blast field trials including the use of instrumented manikins have validated

    the blast resilience and occupant protection of the current IBRS prototype to blast levels of 40kPa x 36ms. Field

    trials have proved the standard geotextile curtain-wall will withstand the combined blast/fragmentation from155mm artillery at 5m standoff as well as military 50-cal rounds using only 300mm thickness of soil fill.

    INTRODUCTION

    Lightweight relocatable shelters (LRS) include a wide range of both soft- and hard-skinned

    structures such as trailers, tents, and various pre-fabricated field-assembled structures

    intended for expedient environmental protection of personnel or materiel during temporary

    deployments. Such shelters are used extensively at expeditionary military camps, industrial/

    commercial sites under construction or maintenance, and during emergency response

    operations. LRS are applied in wide-ranging roles including accommodations, offices,

    workshops, messing, vehicle-bays, and stores; specially adapted units are used for field

    hospitals or housing specialized equipment including aircraft. Large tents, such as those ofthe pavilion style, are also used for social functions or displays at public venues.

    In historical military applications, LRS were intended for use behind lines beyond risk of

    combat threats; in general, such shelters are deployed with the primary consideration being to

    provide expedient environmental protection during short-term operations. However, in

    modern applications LRS are often required in areas at risk from explosions, ballistic impacts

    including military small-arms fire, accidental impact from industrial equipment or vehicles, or

    seismic actions. For the military, even homeland bases can no longer be considered behind-

    lines from terrorist attack. Military expeditionary camps are at risk from attack by enemy

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    2/14

    explosive munitions or insurgent truck-bombing; such camps also have risks from explosive

    accidents at their own ammunition or fuel-storage compounds. Emergency response

    operations require LRS for field hospitals, accommodations, and stores; however, in the

    aftermath of a terrorist attack, these shelters may be vulnerable to second hit bombing or

    small-arms attack specifically targeting responders. In deployments to earthquake-stricken

    areas, temporary shelters are at risk from after-shocks.

    Although LRS are usually designed for some nominal wind or snow-load capacity, theirprimary design criteria focuses on transportability, breakdown size/weight, and the ease and

    speed of setup. Therefore, such structures are typically highly vulnerable to blast since they

    present large surface areas relative to their lightweight, low-rigidity framework in

    combination with weak ground fixity. A weak incident blast of 10kPa overpressure, which

    would not inflict serious ear-drum injury to personnel in the open [1], will impart peak

    reflective loading exceeding 10-fold that from a 200km/hr wind and cause serious damage to

    LRS. Even with the benefit of continuous full-sized framing elements and rigid fixture to

    concrete foundations, standard steel-framed light industrial buildings will typically fail

    catastrophically when subjected to long-duration blasts of 25kPa [2], hence blast failures for

    LRS can be expected at a fraction of this level. Extensive defence-research studies conducted

    in the US [3] and Canada [4] investigated the blast response dynamics and injury risk to

    occupants for a range of military LRS designs and confirmed this vulnerability. Figure 1

    shows a typical deployment of tents at a military encampment and frames from internal high-

    speed imaging of the response dynamics to low-level blast [4].

    Figure 1. (Upper) Typical deployment of tent accommodations at a military expeditionary

    camp. (Lower Left and Right) Blast effects within a military tent installed with manikins and

    fittings typical of habited shelters (1: clip-on fan; 2: clip-on lamp; 3: helmet; 4: circuit panel;

    5: lamp fixture; 6: 5KW heater). The deflecting framework, projected items and whipping

    wires present a significant injury risk to occupants (from [4]).

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    3/14

    Typical LRS framing is not designed for severe side-loading, especially applied as a shock or

    step function, and frame members often fail at their fixtures or joints if not buckled or

    broken outright. Blast-induced deflections, break-up and projection of structural elements or

    wall attachments, and ultimately structural collapse will usually present far greater injury risk

    to occupants than had they been exposed to the blast in the open. Whereas humans can

    survive a free-field blast exposure of about 100kPa overpressure [1], standard hard-framed

    LRS structures such as tents or work-trailers are typically destroyed at blast conditions 1/5th

    that level. This injury risk became tragically highlighted by the casualties of the Texas City

    petrol-chemical blast accident in 2005 blast in which all 16 fatalities and seriously injuries

    were inflicted on occupants of work trailers within the blast-hazard zone [5].

    Similarly, by virtue of their primary basis for design, standard LRS are intrinsically

    vulnerable to ballistic threats. Although appliqu ballistic panels have been proposed for

    LRS [6], these are typically expensive, limited in protection to non-military threats, and in

    fact can increase injury risk and damage form blast. When deployed to areas at risk of

    significant ballistic threats, LRS will usually be enclosed by a separate ballistic barrier wall of

    concrete segments or earthworks such as shown in Fig. 2 which demand considerable

    resources and time for their installation and ultimately for dismantling or relocation.

    INITIAL STUDIES

    The recognition that injuries to LRS occupants from blast events were inflicted primarily by

    the impact or projection of the hard framing or hard sheathing of the shelter itself led to the

    investigation of a novel alternative air-beam shelter for deployments having blast risks,

    effectively a soft-framed soft-sheathed structure. Figure 2 shows two models as

    commercially available at the onset of the study in 2007 designated as the DSI-10 and DSI-19.

    As shown in the figure, the structures are formed from large columns of reinforced vinyl

    tubing formed into arches and lightly pressurized by low-power air blowers. The system of

    arches is tied together by cross-cabling and the entire assembly is enclosed by a tough fly

    covering. Although the shelter is self-supporting, guy-lines run from hug-straps around the

    girth of each column at various points along each arch to ground stakes in order to providelateral restraint from wind action. In fact, the shelter is stabile in winds exceeding 100km/hr

    using ground stakes at the column base alone. The DSI shelters have been well-proven in

    industrial applications including extended performance in arctic conditions.

    Figure 2. The original DSI-10 and DSI-19 air-beam shelters.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    4/14

    The initial investigation of the blast response of air-beam shelters and the development of

    concepts leading to the current Integrated Blast Resilient Shelter system have been described

    in previous reports [7,8,9,10], but will be summarized here for completeness.

    Although the original DSI shelters had not been designed for blast, an exploratory full-scale

    blast trial was conducted in September 2007 to test the DSI-10 for blast response dynamics.

    The shelter was subjected to the blast from a 2000kg TNT-equivalent charge at 100m

    standoff, yielding an incident blast wave of 30.6kPa amplitude, 45ms positive-phase duration,and 570kPa-ms impulse. From previous studies it was known a blast of this severity would

    have catastrophically damaged a conventional LRS design. As reported in detail in [7], the

    shelter performed remarkably well and rebounded fully from the blast after deflecting in a

    mode of elastic buckling action; the air-beam columns intruded 14% into the primary

    habitable internal space at their maximum deflection. The wall deflection would not have

    inflicted serious impact injuries, although occupants in the path of the air-columns would

    certainly have been knocked to the ground. A diminished and distorted pressure wave was

    transmitted into the interior space of lesser severity than had been measured for tents [4].

    The main conclusions from the exploratory trial were as follows:

    The DSI air-beam shelters demonstrated strong potential for further developmenttowards a military-grade blast-resilient LRS

    Certain components required upgrading to survive the high-acceleration conditionsimparted by the blast loading in most cases involving elimination of stress

    concentrations, distributing loads, or introducing shock-absorbing connections

    Significant revision of the tethering system was required to control the mode, rate, andextent of wall deflection and reduce the required ground footprint

    New measures would be required to ensure ballistic protection for the shelter as wellas ensure a safe-fail backup capacity to support the arches in some diminished mode

    even if all pressurization was lost

    Modular component design was required such that levels of blast and ballisticprotection can be adjusted depending on particular user requirements and constraints.

    Due to the highly responsive nature of the fabric surfaces, an FSI (fluid-structureinteraction) approach was required in the computational modeling by which thesolution for the structural response dynamics is coupled in time with that for the

    compressible gas-dynamics of the blast-wave flow

    Since the structure rebounds from blast through actions of elastic buckling andirregular flexure, a new performance criteria was required to assess designs based on

    injury potential rather than one based on traditional structural or material failure

    INTEGRATED BLAST RESILIENT AIR-BEAM SHELTER SYSTEM

    Following the initial exploratory blast trial, an intensive 3-yr R&D program was initiated to

    develop an advanced Integrated Blast Resilient Shelter (IBRS) system to meet the objectives

    identified above. Aided by the development and application of specialized FSI modeling [7],working prototypes of the necessary upgrades were developed and validated in full-scale blast

    and ballistic trials and use of large-scale blast simulator facilities.

    In the development and appraisal of various design upgrades both from computational and

    experimental studies, a new performance criteria or measure-of-effectiveness was required

    as noted in the last item of the previous section. That is, since the structure largely responds

    elastically and ultimately rebounds despite sometimes large irregular deformations, traditional

    engineering design criteria based on limiting rotation of shear connections, beam deflection,

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    5/14

    or material failure are inappropriate. Since injury risk to personnel or damage to housed

    materiel is affected most by the extent and rate of wall deflection into the occupied space, a

    performance criteria based on the maximum cross-sectional area intrusion of the deflecting

    wall into the primary habitable space was adopted as depicted in Fig. 3. This criteria was

    also used to define the iso-damage curves for the P-I diagrams generated from the

    computational studies; that is, rather than having a pass/fail iso-damage curve, a family of

    curves was generated by which any user can define an unacceptable level of elastic intrusionof the air-beam columns. The three primary upgrade developments are described in the

    following subsections with regard to their incorporation in baseline DSI-19 model. However,

    all concepts can be applied equally well to both larger and smaller DSI shelter models.

    Tethermast

    The key upgrade for the IBRS involves an entirely revamped tethering system including the

    use of a line of tethering masts or tethermasts along the wall of the shelter as shown in Fig.

    3. The tethermast has several critical roles in the enhanced capabilities of the IBRS. The

    exact design and materials for the tethermast to optimize its performance under blast load

    when coupled to the air-beam column are IPsensitive (Intellectual Property) and will not be

    detailed here. In the computational simulations shown, the tethermasts are external to the

    shelter wall and in-line with the air-beams; however, this is only one option for the tethermastarray and convenient for illustration purposes. For most installations the tethermasts will be

    set between the columns and beneath the fly such that they are within the profile and normal

    architectural lines of the shelter as shown in Fig. 4. The introduction of tethermasts to replace

    the original ground guy-lines greatly improves the effectiveness and control of wall deflection

    as well as reducing the ground footprint of the shelter installation.

    The line of tethermasts allows incorporation of an auxiliary curtain-wall for the shelter which

    serves as a barrier for blast and fragmentation protection. Although the design details and

    materials of the revised tethering system and curtain-wall are proprietary, Fig. 3 shows that

    blast deflections can be reduced by factors of 5-fold by the upgrades; the pressure wave

    transmitted to the interior is also greatly diminished in severity compared to the case without

    the curtain-wall. The tethermasts are modular and can be broken-down to five componentsfor ease of shipping and handling. Each of the primary components, such as the base-flexure

    unit, has variants or adjustments which can be substituted to meet particular performance,

    cost, or weight constraints.

    A final optional role for the tethermasts is to allow a safe-fail mode for suspension of the

    arch roof in the unlikely event of total loss of pressure to the entire system. Although each

    air-beam column is independently pressurized and incorporates a check valve, it is possible

    (for reasons not evident at this time) that all air-beam arches might abruptly and

    simultaneously lose pressure. In such an event, the deflating air-beam arches will be

    restrained from total collapse by the use of cross-cables spanning the width of the shelter as

    shown in Fig. 4. For the configuration shown, the cross-cable acts as a catenary between the

    tethermasts across the shelter and will maintain head-room of over 2m at the centre of the

    shelter to nearly 3m at the inner wall. The internal cross-cables also allow the option ofsuspending lightweight partitions to divide work-spaces within the shelter interior.

    Air-Beam Restraint

    The original method of tether attachment and lateral restraint of the air-beam columns had

    been by means of the hugstrap connection described from Fig. 2. This attachment is in fact

    extremely ineffective for load transfer under impulsive forces in particular. The hugstrap

    restraints were revised to a hugsheet arrangement as shown in Fig. 5.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    6/14

    Figure 3. Computational modelling of maximum air-beam shelter deflections subjected to a

    20kPa blast of 100ms duration, comparing results for standard ground tethering, tethermast

    support, and tethermast with fabric curtainwall.

    Figure 4. Sketches of optional configurations for tethermasts set between the air-beam columns

    for the cases without and with external geotextile curtain-wall. The curtain-wall option

    provides ballistic shielding or ballast as required.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    7/14

    Curtain-Wall

    As described previously, the line of tethermasts allows incorporation of an optional geotextile

    curtain-wall to provide ballistic protection, enhanced blast mitigation, as well as stabilizing

    ballast for deployments not having other means of secure ground fixture. The same curtain-

    wall design can be configured in various modes to meet a range of performance specifications.

    Although specialized curtain-walls are being devised for requirements such as protection from

    rocket-propelled grenade attacks, the standard design has a tapered side profile with a nominal600mm base, 300mm width at the top, and height of 2400mm. Although generally intended

    to be a geotextile wall, that is, filled with local soil, gravel, or crushed stone as available, the

    simple fabric curtain-wall provides a high degree of ballistic protection against severe non-

    military threats such as tornado-borne debris. Figure 6 shows an image sequence from testing

    of the empty curtain-wall conducted to validate protection level for the highest kinetic-energy

    threat specified in ASTM E1886 for tornado-borne debris.

    Figure 6. Impact of a 6.8kg steel bar at 35m/s on an unfilled prototype curtain-wall as part of

    testing to validate protection from explosion-borne debris due to accidental explosions at

    petrol-chemical sites. The projectile is highlighted in the first frame as it was launched by

    means of a specially designed gun-barrel. No damage was inflicted to the curtain-wall in this

    test due to the combined membrane action of the fabric and flexure of the supports.

    Figure 5. Revision of the air-beam restraint from hugstrap to a hugsheetconfiguration such

    that the columns are restrained as if in a lateral sling. The line-of-action of the forces and

    the load distribution at the connection reduce stresses over 1000-fold as depicted at right.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    8/14

    Once filled as a geotextile barrier, the curtain-wall has been designed to provide maximum

    ballistic protection for its thickness of fill using a simple but effective new technique to

    restrict the cavity growth from high-speed projectiles. Due to the incorporation of the curtain-

    wall with the tethermast supports, the combined system is highly resilient to the synergistic

    effects of combined blast and ballistic loading from close-proximity detonations of military

    munitions. As shown in Fig. 7, arena tests proved the performance of the standard curtain-

    wall against near-field explosions of military munitions to the severity of a 155mm artilleryshell at 5m standoff. NATO STANAG 4569 specifies qualification testing for fragmentation

    from 155mm shells at 25m standoff, since it is generally not expected that ballistic panels also

    perform well in blast resilience. In separate tests, 50-cal BMG rounds (Federal American

    Eagle XM33C, 42.8gm, steel-core) fired at 30m standoff with nominal muzzle velocity

    884m/s were successfully stopped by the curtain-wall; this performance significantly

    surpasses STANAG 4569 Level 3 ballistic protection intended to cover Russian AK rounds.

    As previously noted, this level of ballistic protection is afforded by the standard geotextile

    curtain-wall tested at its minimal 300mm thickness. A double-wall barrier with burster-screen

    is being developed to defeat the shaped-charge warhead of a rocket-propelled grenade

    including follow-through effects of its spent motor casing.

    BLAST TESTING

    12RQ Full-Scale Field Trials

    By arrangement with the Suffield laboratory of Defence R&D Canada (DRDC), a full-scale

    configuration of a prototype blast resilient shelter system was deployed in blast field trials

    being staged under the DRDC 12RQ Defence Research program. The blast trials were

    conducted on the Experimental Proving Ground of DRDC Suffield in Alberta September-

    October 2009 under direction of the 12RQ program manager Dr. J. Anderson.

    A series of three tests of escalating blast intensity were conducted subjecting the prototype

    shelter to the blast conditions summarized in Table 1. The instrumentation layout for the

    trials is depicted in Fig. 8 and included both H-III and H-II ATDs (Anthropomorphic Test

    Devices). The H-III was fitted sensors for assessment of head/neck injury due to impact of the

    air-beam columns, while the un-instrumented H-II was used to assess gross motions from air-

    beam impact. At the time of the trials, a complete prototype curtain-wall was not ready for

    deployment around the shelter, hence a crude simulation of its effect was made by suspending

    Figure 7. Arena tests of the prototype standard curtain-wall have verified protection against

    the combined blast/fragmentation effects of the munitions as shown. The specifications

    shown are nominal values for design as provided in US DoD manuals [11]; the asteriskdenotes that the designated wall thickness relates to protection against ballistic penetration

    without consideration of blast effects.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    9/14

    the equivalent layers of fabric material from the tethermasts as shown in Fig. 8. Therefore, it

    was not possible in these trials to assess the full efficacy of the curtain-wall for blast

    attenuation, including its optimal soil-filled (geotextile) mode.

    Key results from the trials are shown in Figs. 9-11 for which the incident blast, transmitted

    overpressure, and wall deflection into the internal space are presented for each trial. Being the

    case of most severe blast, further detailed results from Trial 3 are presented in Figs. 12 &13.The main conclusions from the trials were as follows:

    The response of the prototype shelter met or exceeded expectations for controlled walldeflection and minimized injury risk to occupants. For Trial 2 (20kPa x 40ms), the

    maximum extent of air-beam intrusion was less than 1% of the habitable space; for

    Trial 3 (40kPa x 36ms), the maximum intrusion was 4% of the habitable space. The

    shelter fully rebounded from all blasts, although several non-structural seams and

    connections of the covering fly were torn in the final test. It should be considered that

    the same structure was exposed to all blasts in succession without intermediate repairs,

    hence incipient damage was likely accumulated from the prior tests.

    In both tests yielding measurable wall deflection, only the zone immediately adjacentto the wall was affected such that personnel, or their furnishings they occupy, would

    have to be abutted directly against the air-beam column for any significant effects.

    Partly due to the yielding nature of the air-beam impact, a seated manikin in a chair

    with its back abutted to the wall was imparted an average velocity less 0.7m/s.

    The pressure transmitted to the interior was significantly diminished in effect asquantified by its amplitude, impulse, and rise-time. The degree of amplitude reduction

    increased with blast strength to 50% for the strongest blast; impulse reduction

    decreased with incident blast strength from 38 to 28%. Very importantly, in all cases

    Table 1. Summary of 12RQ blast field-trial test conditions.

    Trial Charge Incident Blast Conditions

    P (kPa) Duration (ms) Impulse (kPa-ms)

    1 165kg TNT Eq 12.4 24 150

    2 500kg TNT Eq 20.7 40 330

    3 1000kg TNT Eq 40.6 36 690

    Figure 8. Instrumentation layout for the 12RQ blast trials.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    10/14

    the shock front of the incident blast became diminished to a less-injurious compression

    or set of staggered shocks spanning about 4ms of rise-time.

    Although not affecting occupants, unexpected damage was sustained by someequipment due to the significant upward rebound of the shelter well after the passage

    of the blast itself (~ 200ms). Ultimately, much of the elastic energy stored from the

    blast encounter with the shelter is recovered as a powerful upward rebound of the air-

    beam columns. Straightforward measures were identified to secure the column bases

    against this action, as well as introduce shock absorption, flexure, and reduced mass of

    attachments at the base of the air-columns.

    Component Testing

    Certain components, fittings, and seams were identified from the most severe 12RQ blast test

    of 40kPa as warranting further assessment and possible modification. Although not damaged

    by the blast itself, the heavy and rigid blower fitting attachments to the base of the air-beam

    columns were damaged by the powerful and abrupt late-time vertical rebound action of the

    columns. The blower bulkhead assembly was redesigned to minimize its mass and stiffness

    and allow for flexing action with its various connections. In addition, a check-valve wasintroduced such that in the event of a failure of the blower attachment or any rupture of

    external feed lines, pressure is not lost from the columns. The revised blower assembly was

    subsequently re-qualified in tests using a large-scale blast simulator as shown in Fig. 14.

    Figure 9. Summary of key results from 12RQ Trial 1, 12.4kPa x 24ms blast. (Upper left

    and right) Overview of the trial layout showing the fireball shortly after detonation and

    comparison of incident blast and transmitted overpressure waveforms. (Lower left and

    right) Interior view of the shelter immediately prior to blast arrival and at time of maximum

    air-beam deflection at 36ms. The seated and propped manikins shown abutted to air-beam

    columns were unaffected by the blast.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    11/14

    Figure 10. (Upper left and right) Overview of the Trial 2 layout showing the fireball shortly

    after detonation and comparison of incident blast and transmitted overpressure waveforms.

    (Lower left and right) Interior view of the shelter immediately prior to blast arrival and at

    time of maximum air-beam deflection at 37ms. The seated manikin shown abutted to the

    air-beam column was unaffected by the blast.

    Figure 11. (Upper left and right) Overview of the Trial 3 layout showing the fireball shortly

    after detonation and comparison of incident blast and transmitted overpressure waveforms.

    (Lower left and right) Interior view of the shelter immediately prior to blast arrival and at

    time of maximum air-beam deflection about 80ms after blast arrival. The manikin shown

    seated in the chair abutted to air-beam columns slid from the chair at about 0.7m/s.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    12/14

    Figure 12. Deflection of the inner wall of the

    central air-beam column during the first 100msof the blast encounter. The initial shape is

    shown in blue with the deformation tracked at

    10ms intervals. In the final stages of the

    response, the base of the columns rebounded

    backwards and upwards. The general mode of

    elastic buckling closely followed computational

    predictions as previously shown in Fig. 3.

    Figure 13. Prostrate manikin on cot

    abutted to air-beam column in Trial 3comparing positions pre- and post-blast

    (upper and lower respectively). The

    manikin itself was not displaced,

    although the cot was shifted about

    15cms beneath it by the action of the

    air-beam column on its frame.

    Figure 14. 1.8m Blast Tube facility used to qualify IBRS components such as the

    revamped air-blower configuration and air-beam check valves. From lower-left to right:

    installation of the test column and blower fittings; setting of the concrete-wall closure; and

    view from inside the Tube after the test column has been strapped to the reflecting wall.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    13/14

    DISCUSSION AND CONCLUSION

    Lightweight relocatable shelters such as tents, trailers, and pre-fabricated huts are intended for

    rapid deployments with the primary role of providing weather protection for personnel or

    equipment during temporary operations spanning weeks to a few years. However, in many

    applications these shelters are at risk from severe loadings such as due to blast or ballistic

    threats including terrorist bombings or small-arms attacks, accidental explosions, extreme

    winds, or seismic action such as earthquake aftershocks. Ironically and tragically, it is mostoften the shelter in these events which inflicts the gravest of injuries to occupants in

    comparison to personnel being exposed to such threats in the open. Whereas the traditional

    design approach for blast or ballistic protection of structures usually involves the hardening,

    thickening, or stiffening of components, this is not feasible for lightweight deployable

    shelters. In this regard a novel approach has been taken to allow significant but controlled

    flexure in the event of severe loads where this deformation is readily absorbed by the

    materials and can be exploited to maximize the overall resilience of the system and ultimately

    minimize injury risk to occupants or damage to materiel being housed.

    The development of the Integrated Blast Resilient Shelter system has been described

    involving a comprehensive 3yr R&D program involving advanced computational modelling

    full-scale field trials, and component testing. The basis of the design is an underlying low-pressure air-beam structure with flexing lateral supports such that shock and impact loads are

    taken by membrane and tensile stresses for which the materials are inherently strong and

    energy-absorbing. The air-beam arches allow open spans of 40m to be enclosed offering

    large working areas for equipment or as required for assembly areas such as lunchrooms. In

    the case of blast loading, although some degree of intrusion of the air-beam wall into the

    internal space is incurred, this deflection can be tailored to present very low injury risk to

    occupants. Overpressure is transmitted to the interior, although the amplitude, impulse, and

    rise-time of the transmitted wave are greatly mitigated; all blast levels of consideration here

    are non-lethal to personnel in the open.

    A geotextile curtain-wall is incorporated into the shelter system which has been specially

    designed to maximize performance of local soil as fill for ballistic protection. The standardcurtain-wall of nominal 300mm thickness has been field-tested against a range of munitions

    including 155mm artillery detonation at 5m standoff and 50-cal rounds; the ballistic

    protection can be readily increased as required. The curtain-wall is also functional as a

    separate rapidly deployable stand-alone barrier to protect other structures or areas. The

    combined shelter system meets all the normal criteria for a lightweight deployable shelter yet

    has many attributes of traditional hardened bunkers including blast protection exceeding

    35kPa overpressure and ballistic protection greatly exceeding NATO STANAG 4569 Level 3.

    R&D continues at this stage to provide optional overhead protection and defeat of special

    threats such anti-armour weapons including rocket-propelled grenades with shaped-charge

    warheads.

  • 7/31/2019 A Blast and Ballistic Resilient Air-beam Shelter System

    14/14

    REFERENCES

    [1] Bowen, I.G., Fletcher, E.R., Richmond, D.R., 1968, Estimate of Mans Tolerance to the Direct

    Effects of Air Blast, Progress Report DASA-2113, Defense Atomic Support Agency, US Dept. of

    Defense, Washington, DC.

    [2] Glasstone S., and Dolan, P.J., (Eds), 1977, The Effects of Nuclear Weapons, Chapter V: Structural

    Damage from Air Blast, 3rd Ed., US DoD and the Energy R&D Administration, Washington, DC.

    [3] Stevens, D., Marchand, K., Young, L.A., Moriarty, R., Cropsey, L., 2003, Evaluation of Structural

    Response and Human Injury in Shock-Loaded Expeditionary and Temporary Shelters, Proc. 11th

    Intl Symp. on the Interaction of the Effects of Munitions with Structures, Mannheim, Germany.

    [4] Ritzel, D.V., Crocker, J., 2006, Blast Response of Hemi-Cylindrical Tents, 19th Intl Symposium

    on Military Aspects of Blast and Shock (MABS19), Banff, Alberta, Canada.

    [5] US Chemical Safety and Hazard Investigation Board, 2007, Final Investigation Report: BP Texas

    City Refinery Explosion and Fire, Report No. 2005-04-1-TX.

    [6] Horak,K.,Quigley, C., Devine, R., 2009, Ballistic Protection for Tentage,Logistics, US Army

    Natick Soldier RD&E Center, pp.56-60, Natick, MA.

    [7] Ritzel, D.V., 2008, Blast Response of an Air-Beam Shelter, Dyn-FX Contract Report CR040801

    for Dynamic Air Shelters Ltd, Dyn-FX Consulting Ltd, Amherstburg, Ontario, Canada.COMMERCIAL-IN-CONFIDENCE.

    [8] Ritzel, D.V., 2008, Blast Response of the DSI-19 Air-Beam Shelter, Dyn-FX Contract Report

    CR100801 for Dynamic Air Shelters Ltd, Dyn-FX Consulting Ltd, Amherstburg, Ontario, Canada.

    COMMERCIAL-IN-CONFIDENCE.

    [9] Ritzel, D.V., Parks, S.A., Crocker, J., 2009, Blast, Ballistic, and Earthquake Response of a

    Deployable Air-Beam Shelter System, 8th

    Intl Symp. Shock & Impact Loads on Structures,

    Adelaide, Australia.

    [10] Warner, H.A., Ritzel, D.V., Parks, S.A., Crocker, J. 2010, Development of a Lightweight

    Relocatable Shelter System with High Resilience to Blast, Ballistic, and Seismic Threats, 1st

    Intl

    Conf of Protective Structures, Manchester, UK.

    [11] US Dept of Defense, 2008, Structures to Resist the Effects of Accidental Explosions, Unified

    Facilities Criteria, UFC 3-340-02, Washington, DC.