Acta Polytechnica

The paper presents a performance characteristics of a pneumatic launcher, which is an important element of the split Hopkinson bar set-up (SHPB) at the Department of Military Engineering and Infrastructure (the Military University of Technology in Warsaw) for the purpose of dynamic strength tests of construction materials. The process of experimental calibration of the launcher for selected loading bar-projectiles is shown. Two types of compression during direct impact tests were also used simultaneously to investigate the behaviour of metallic samples with the use of this launcher as well as the Hopkinson measuring bar: the first — a short cylindrical sample, including a miniature (small diameter) sample, and the second — a long cylindrical sample (Taylor test). The relationships describing the stress and strain state as a function of strain rate for the first type of the experiment and engineering empirical formulas for the second type of the research were given.


Introduction
The beginnings of the development of gas launchers are related to a military technology. The design of the pneumatic guns used to launch large-diameter missiles was first presented by D. M. Medford in 1883 in Fort Hamilton (USA). However, the first land-based air cannons designed by the American inventor born in Kórnik (Poland), Major Edmund Żaliński, were installed in 1894 at Sandy Hook Fort in New Jersey. It was a three-gun battery of 15-inch (381 mm) coastal artillery guns operating in a similar way to an air gun: compressed air was used to fire a projectile (explosive charge) [1]. In 1900, Żaliński's triple air gun was installed on the USS Vesuvius, and a twin 8.425 in (214 mm) on the Holland IV submarine, known as Zalinski Boat, designed by Edmund Żaliński and John Holland. The rapid development of fuel and missiles in the late 1890s and early 1900s led to the creation of gunpowder guns and caused pneumatic guns to be substantially phased out from the US Army starting in 1905.
Currently, pneumatic guns are designed to carry out various research impact tests, with various energy possibilities, limited by the diameter and working pressure [2]. Small-diameter systems (up to 70 mm) allow for higher velocities for smaller-mass projectiles, while in medium-diameter solutions (70-150 mm) for objects with larger mass, the muzzle velocity is lower. The greatest drop in speed is recorded for large-diameter devices (over 200 mm).
The small-diameter pneumatic launcher is an important element of the stationary test stand called the split Hopkinson pressure bar SHPB [3]. This stand is intended for testing the behaviour of samples of construction materials, including construction materials subjected to impact loads [4] and in the field metals [5][6][7][8][9] as well as for concretes [10], polymers, wood, soils [11][12][13][14][15] and other materials [16][17][18]. Understanding the dynamic strength characteristics of these materials is important for the design of protective structures for buildings, especially in the conditions prone to industrial accidents [19], to ensure safety [20]. With the use of compressed air, the launcher on this test stand allows for throwing projectiles, such as a bar or Hopkinson measuring bars, directly loading the tested metallic sample, U. S. Lindholm used a spring and then a pneumatic 0.5 inch (12.7 mm) launcher in the SHPB test stand for the first time at the beginning of the 1960s [21]. Until then, blasting shots have been used to generate a stress pulse in Hopkinson measuring bars.
The stationary pneumatic launcher, which is an element of the SHPB, intended to test the behaviour of material samples subjected to dynamic (shock) loads, does not meet the statutory definitions of "firearms", "gun" and "pneumatic weapons" [22]. This means that pursuant to the Act of June 13, 2019 on the performance of economic activity in the field of production and trade in explosives, weapons, ammunition, and products and technology for military or police purposes [23] it is not considered a weapon. However, if the pneumatic launcher in question was designed and intended solely for the production or certification, qualification or testing of products included in Part IV -WT of the Annex to the Regulation [24], then it would be subject to regulation resulting from the provisions of the Act [25]. The contractor of such a pneumatic launcher would have to have a license granted by the Minister of the Interior and Administration, at least in the scope of manufacturing and trading in products for military or police purposes specified in WT XIII section 1 or 2 depending on: • a type of equipment specially designed or modified for the production of products covered by Part IV -WT, and specially designed components thereof; • the type of a specially designed facility for conducting environmental tests and the type of specially designed equipment for the purpose of certification, qualification or testing of products included in the list contained in Part IV -WT.
At present, the disadvantages of split Hopkinson bars, such as the high air operating pressures to obtain high strain rates, the noise due to the instantaneous air expansion, and a large overall station length, have been eliminated in the electromagnetic Hopkinson bar. It uses the intense pressure created in the magnetic field created by the passage of an electric current pulse through a series of coils. The magnetic field behaves like the release of air from a high-pressure vessel and can impart a high initial velocity to the bar-projectile to obtain very high compressive and strain rates of metallic materials, more than 10 4 1/s. However, for low and medium impact velocities of this projectile, pneumatic launchers are still useful for conducting physical experiments in the range of deformation rates 10 2 -10 3 1/s. They ensure a good reproducibility of obtaining the value of the impact velocity for individual set values of the deformation rate. However, from the point of view of objectivity of dynamic tests, it is necessary to conduct preliminary tests to validate the pneumatic launcher, a so-called calibration procedure, before a series of physical experiments to obtain empirical relationships between the working pressure and the impact velocity for the geometric parameters of the bar-projectile used in further tests, which characterize the performance of this launcher. This is especially important during various schemes for material impact tests with the use of the pneumatic launcher. The subject of this work is devoted to these issues.

Characteristics of the pneumatic launcher of the SHPB stand
The subject of the work is a pneumatic launcher included in the split Hopkinson bar testing stand (shown in Figure 1), which is located at the Department of Military Engineering and Infrastructure (DMEI) of the Military University of Technology (MUT) in Warsaw. The pneumatic launcher consists of a pressure chamber with a capacity of 10 dm 3 with a smooth barrel with a diameter of 20 mm and a length of 2700 mm. Figure 2 shows a general view of the launcher and shows the valve with a digital pressure gauge that feeds the launcher chamber and the valve supplying the space behind the bar-projectile in the barrel.
The launcher in question is fed with compressed air from a compressor to a maximum working pressure p max =8 bar. The important elements of this pneumatic system are:       The standard deviation for each cycle of five experiments was calculated according to the formula:

Pneumatic launcher calibration procedure
The results of the calibration of the launchers are shown in Table 2.
The results contained in Table 2 are presented graphically as diagrams of the muzzle velocity v 0 of the bar-projectile depending on: a) the working pressure p 0 of the pneumatic launcher and b) the number of the experimental attempt in three variants of the initial pressure p 0 for the length L bp of the bar-projectile: • L bp1 =100 mm - Figure 6a

Impact tests with the use of an SHPB pneumatic launcher
Using the SHPB pneumatic launcher, it is possible to use the schemes of two types of compression impact tests for the purpose of testing the behaviour of  Table 2. Summary of the obtained muzzle velocities v0i of the bar-projectiles for different variants of the length L bp of the bar-projectile and the working pressure p0 of the pneumatic launcher.        [26] are presented.
(1.) Two variants of direct compression of the first type: (a) variant I of a miniature sample -with the use of a loading bar-projectile, which, at the moment of impact, has accumulated kinetic energy many times greater than the work of elasto-plastic deformation of this sample; in this case, the speed of the bar-projectile is constant or changes slowly during the entire process of elasto-plastic deforma-tion of the sample on the front of the Hopkinson measuring bar ( Figure 9); where: • L 1 is initial length of the short cylindrical specimen; Figure 9. Scheme of direct compression in variant I -miniature short cylindrical sample.
• ε t (t) is elastic positive incident strain pulse in the measuring Hopkinson bar, registered by a strain gauge after passing the compressive loading wave through the specimen; • D H and D 1 are initial diameters of the measuring Hopkinson bar and the specimen, respectively; • E H is Young's modulus of the measuring Hopkinson bar; • c H is sound velocity in the measuring Hopkinson bar; • σ s is engineering stress in the specimen obtained as a function of time in the assumption of equality of forces at the ends of the specimen during the entire deformation process; • ε s is engineering strain in the specimen tested; • dotε s is engineering strain rate of the specimen tested.
(b) variant II -with the use of a bar-projectile with a much lower mass than in variant I (about 50 % mass of a bar-projectile in variant I); the process of dynamic loading of the sample is wavelike in the case of the bar-projectile -sample -Hopkinson measuring bar system ( Figure 10); where: signs and symbols as in point ((1.).a.). (2.) Taylor impact test -a long cylindrical sampleprojectile fired by a pneumatic launcher hits directly on the front of a Hopkinson measuring bar and undergoes an inhomogeneous elastic-plastic deformation ( Figure 11).
L pl [12] (10) where: • L 0 is initial length of specimen; • L 1 is compressed (final) length of specimen; • L pl is length of a section of the specimen along its axis where only plastic deformation occurred; • L el is length of a section of the specimen along its axis where only elastic deformation occurred; • σ y is engineering yield stress (upper index Taccording to the Taylor formula, upper index W G -according to the Wilkins and Guinan formulation); • v 0 is impact velocity; • ρ is mass density of specimen.
The above formulas describing the real (true) measures of stress σ t , strain ε t and strain rateε t of the tested samples in variants I and II have the following forms, respectively:   The direct compression test of a short cylindrical sample (variant II) also allows to determine the dynamic Bauschinger effect δ of the tested metal. The paper [20] presents a method of determining the measure of this effect using this shock test, using the registration of the compressive elastic deformation in time in the Hopkinson measuring bar for this purpose. Figure 12 shows a typical course of an elastic strain pulse in this measuring rod. From this diagram, the initial dynamic yield limits can be determined: σ L y during loading and σ U L y during unloading -the differences between the points: between 0 and B, and between B and C. The calculation of the dynamic Bauschinger effect δ is determined from the relationship:

Conclusions
The characteristics of the operation of a pneumatic launcher for the purpose of conducting direct impact tests with the use of a bar-projectile and a Hopkinson measuring bar are presented. The obtained results of the experimental calibration of the pneumatic launcher -loading bar-projectiles characterize the performance of a given essential element of the SHPB set-up, which is significant for conducting various impact direct tests, including the Taylor one.
Schemes of two impact tests with the use of the pneumatic launcher and the Hopkinson measuring bar are presented as noteworthy for dynamic testing of metals in various typical compression modes, which allow to determine their real impact resistance -two variants of direct compression of a short cylindrical sample, including miniature, and a long cylindrical samples, which can also be used as projectiles in the Taylor test. For these tests, relationships were given that allow to determine the nominal (engineering) and real (true) values of stress σ, strain ε and strain rateε as well as the initial dynamic yield strength σ y in the version proposed by Taylor σ T y and Wilkins and Guinan σ W G y . Direct compression of a short cylindrical sample also allows to measure the dynamic Bauschinger effect δ.