1 An experimental investigation of two different methods for swirl induction in a multivalve engine E Pipitone and U Mancuso Department of Mechanics (DIMA), University of Palermo, Palermo, Italy The manuscript was received on 3 September 2004 and was accepted after revision for publication on 23 September 2004. DOI: 10.1243/146808705X7365 Abstract: This paper describes an experimental investigation aimed at comparing the swirl effect induced by unbalancing the mass flow through the two intake ports of a multivalve engine head using two different methods: the first one reduced the curtain area of one of the intake valves [different lifts (DL) method]; the second one adopted a sluice-gate-shaped valve, installed upstream of the intake valves [swirl control valve (SCV) method] in order to cause a pressure drop. A steady-flow test rig (equipped with instrumentation for the discharge coefficient and swirl intensity measurement) was realized in order to compare and evaluate the results of both methods and determine their respective validity and limitations; the procedures used for both experimental methods are duscussed in detail. The flow characteristics were analysed through changes of lift difference or SCV position; it was found that both DL and SCV methods are effective in swirl induction but the DL mechanism, acting on the valve curtain area, is more effective in flow unbalancing between intake ports, since the flowrate depends linearly on the curtain area. The SCV method, instead, controls the port flowrate, inducing a localized pressure drop, whose intensity depends on the flow velocity in a non-linear manner. For this reason the SCV method can achieve strong swirl intensity only with high obstruction levels, in a narrow regulation window close to full-obstruction conditions. Keywords swirl induction, multivalve engine, DL method, SCV method 1 INTRODUCTIN formance. Several design approaches are used to create swirl during the induction process [1]. Some- times flow is discharged into the cylinder tangen- It is well known that turbulent flow patterns are a critical factor in the combustion process and in tially: a directed port brings the flow towards the valve opening in the tangential direction [Fig. 1(a)]. In determining the extent of mixing between fuel and air [1–3]. During the induction stroke the incoming other designs the swirl is generated within the inlet port, forcing the flow to rotate about the valve axis flow is shaped by the geometrical characteristics of the induction port/intake valve assembly. The before it enters the cylinder: a helical port [Fig. 1(b)] can be used to obtain this flow structure. In other large-scale turbulent motions within the cylinder generated by the inlet port arrangement start to designs a non-uniform distribution of flow around the circumference of the inlet valve is forced in order decay after closure of the intake valves, at a rate that strongly depends on their structure. A turbulent flow to obtain a net angular momentum about the cylin- der axis: deflector ports [Fig. 1(c)] and shrouded wall organized as rotation about the cylinder axis, known as ‘swirl’, is the most favourable in this respect, since ports are used to force the flow preferentially in a tangential direction. its decay rate is lower than those of other turbu- lent flows; as a result to this, many induction systems Another technique for swirl induction is based flow unbalancing between intake ports. Several swirl are designed to produce swirl [4]. Moreover, exper- imental investigations [5–7] have shown a strong induction systems based on total or partial obstruc- tion of one of the inlet ports, obtained through a influence of swirl intensity on overall engine per- butterfly-type swirl control valve [7–10], are reported. * Corresponding author: Department of Mechanics (DIMA), Using one of these methods, several experimental investigations were performed in order to find out University of Palermo, Viale delle Scienze, Palermo, 90128, Italy. email: pipitone@dima.unipa.it geometrical parameters for swirl induction [6, 11, 12]. JER04203 © IMechE 2005 Int. J. Engine Res. Vol. 6 No. 1 EN00004203 09-12-04 09:03:28 Rev 14.05 The Charlesworth Group, Huddersfield 01484 517077