Explosive impact welding is a process to produce bi-metallic plates and tubes. Whilst well established it has been essentially an empirical process. This thesis describes work carried out to numerically analyse the two plates welding process. Using finite element and finite difference engineering packages, most aspects of the explosive welding process are modelled. A notable advance is the inclusion of a 'Williamsburg' type equation of state for various mixtures of the low speed ANFO explosives used by most commercial explosive welding companies. The numerical simulations showed that the level of strain induced in the plates may be a factor in determining the bond strength. In most cases, the measured weld shear strength either remained relatively constant or increased slightly with the computationally predicted values of contact pressure, shear stress, plastic strain and impact angle. The shear stress in the flyer plates which welded was predicted to be of opposite sign to that in the base plate. The formation of interface waves and the phenomenon of jetting were computationally reproduced. The computational models were validated by explosive welding trials and by impact laboratory experiments using a pneumatic gun. A new semi-empirical theoretical analysis to predict the velocity and impact angle of the explosively driven flyer plate during its acceleration phase is developed and experimentally validated. This is an advancement on past theories which only considered a non accelerating flyer plate to impinge freely on the base plate. The equations were used to develop a code which includes the effect of explosive density, flyer plate thickness in the calculation of the flyer plate velocity during its acceleration phase. The results of the computer program were validated by experimental measurements. A method for calculating flyer plate velocity, based on the substitution of pressure equations into the 'Williamsburg' equation of state for reactive explosive and into the 'Muranghan' equation of state for the non-reactive explosive assuming non-ideal detonation behaviour is presented. The linear phase mixing rule is assumed for all thermodynamical properties. These equations provided a theoretical framework for the change of energy and pressure with respect to time in the reaction zone. The equilibrium and reactive adiabatic equations most commonly used are either empirical (e. g. JWL equation of state) or simplistic (e. g. polytropic EOS). In addition, they are usually inconsistent with the detonation EOS (e. g. they assume constant Gruneisen gamma or constant heat capacity), and are relatively inflexible. Early treatments used the polytropic equation for both unreacted explosive and detonation product gases, often with the same polytropic index. The analysis showed that the pressure predicted along the expansion adiabat by the 'Williamsburg' EoS based on an ideal calculation is too high, possibly by a considerable amount. Consideration of the length of the detonation/reaction zone (i. e. of order of 50mm) makes it necessary to take into account the variation of pressure along this zone
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:525492 |
| Date | January 2001 |
| Creators | Akbari, Mousavi A. A. |
| Publisher | University of Manchester |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0022 seconds