(40e) The Influence of a Cone-in-Cone Insert on Flow Pattern and Wall Pressure in a Full Scale Silo

An insert is sometimes used to improve the flow pattern in a funnel flow silo with a shallow hopper. This may occur in a new installation when the headroom available is limited, or it may occur as a retrofit to address a flow problem in an existing silo. The common types of insert include an inverted cone, a double cone or a cone-in-cone. An insert is normally placed inside the hopper section and if correctly designed and positioned, it can significantly reduce stagnant zones in the whole silo, leading to a flow pattern approaching mass flow. However the design of an insert is still much more art rather than science and there is a genuine need for research to develop a more rigorous basis for the design and placement of an insert. This paper describes a recent experimental investigation of the wall pressure and the flow pattern observed during filling and emptying of a full-scale silo with and without a cone-in-cone insert. The experiments were conducted at the Tel-Tek Research Institute, dept. POSTEC in Norway. The tests were performed in a cylindrical steel silo with a capacity of 35 m3. The cylindrical section was approximately 8 m in height and 2.51 m in diameter. The axisymmetric conical hopper was 1.26 m high and had a hopper half angle of 44º. The outlet diameter was 0.1 m. The wall pressures were measured at carefully chosen locations using a total of 10 pressure cells mounted along the wall, with 7 in the hopper section and 3 in the cylindrical section. The cells measured both normal and tangential loads. Each pressure cell has a diameter of 120 mm and was made by cutting an appropriate piece from the silo walls. Thus, the surface of each cell had the same roughness and the curvature as the silo walls. The filling material used for the tests was quartz sand with a loose bulk density of 1400 kg/m3. The sand had an average angle of repose of 34.6º and a moisture content varying between 0.11% - 0.26% for the silo tests reported here. More then 90% of the particle sizes are between 0.25 mm and 1 mm. The wall friction angle between the stainless steel wall and the sand was measured to be 17º. The silo was filled and emptied symmetrically. The surface profile of the sand was measured every 10 min from the top of the silo during filing and emptying. These measurements were used to determine the angle of repose and the rates of filling and discharge. To investigate the pattern of flow, markers with a diameter of 30 mm were placed at fixed positions in the silo during filling using a carefully installed system of locating tubes. During discharge, markers passing through the outlet were detected, and the residence time for each marker was noted. The flow profile during discharge can then be deduced from the residence time measurements for the markers and the surface measurements during discharge. Flow tests were carried out in the silo with and without an insert. The whole test programme includes studying three cone-in-cone inserts, positioning of an insert and using one or a combination of inserts. Due to space constraint, this paper will focus on the conein-cone insert with a height of 1 m and an inclination angle of 21º. A total of eight variations were tested and several tests were performed for each variation to check repeatability. The results show that without an insert, the silo exhibited funnel flow with a narrow flow channel. With the insert installed, the flow channel widened considerably but mass flow was never achieved in any of the eight positions. The insert had a strong influence on the pressures in the hopper. A significant increase in the wall pressures at certain locations were observed at the commencement of flow. The results of this study have some significant implications on the use of inserts in silos.