Selamat siang,

There is no need for the 4" elbow; you should raise the nozzle position to feed straight into the cone. If the cone is full of nitrogen then you can drill sparge holes anywhere over the conical face and if you don't like the result you can plug them and drill elsewhere. Only the outer cone face needs to be sort of smooth afterwards.

I agree that your present nitrogen inlet is suspect: unless there is a delay in getting the nitrogen into the tower at process start up: which there certainly shouldn't be.

Your intended modification is self cleaning by the way. There should be no need to consult ASME VIII because the cone is not a pressure part. (And you are venting anyway, as I've just remembered.)

Baik! ■

In order to secure efficient gas diffusion through a moving bed of bulk solids it is necessary to have a uniform velocity across the flow stream, an even state of bulk density and a constant pressure differential over the gas flow path. The objective in this instance appears to be the stripping of volatiles from the product voids during the flow through a hopper. Unfortunately, the hopper is very unlikely to be a mass flow design, as a 60 degree wall angle from the horizontal is below the inclination of slope that most materials will slip on a conical wall surface.

It is also necessary to ensure that the draw-down is uniform around the periphery of the flow annulus. Symmetry of constrution does not mean uniformity of flow and it is suspected that the material is travelling faster down one side that the other, with uneven radial velocity and the reverse gas flow is similarly uneven. The type of extracting device is therefore important. In this case it is probably a rotary valve, which tends to have a biased draw so, in combination with the off-set gas injection, the uniformity of the solids flow in the cone may be suspect.

The first step is therefore to check the wall friction of the product on a surface similar to the material of construction of the resent cone. The material of construction of the inner cone is also important, but this situation is different because flow is diverging, rather than converging. There is hope. Presuming that stainless steel is a suitable contact material, it is practical to construct a boundary cone that will extend up into the wall of the hopper to generate a mass flow pattern of virtually uniform velocity.

If the drawing is to scale, the vertical section where the nitrogen is currently shown to be injected can serve a a standpipe to create a coherent plug flow, which can be split with segmented dividers to extract a balanced flow from around the annulus. A dispersion cone for the nitrogen will ensure even gas diffusion. The situation calls for a thorough technical appraisal, but the general principles are quite clear and can be expounded by direct contact with lyn@ajax.co.uk ■