When the particles forming the filling material are fairly spherical and of equal size, the original water capacity of a bed amounts to about 38 per cent. of the space occupied by the filling material; but as in practice it is difficult to obtain spherical particles of uniform size, the original water capacity is found to range from 35 to 45 per cent. of this space.
Original water capacity with particles of different sizes.
When, however, the particles are of materially different sizes, and when the smaller ones fill up the spaces between the larger ones, the original water capacity may sink down to as low as from 5 to 10 per cent. of the space occupied by the filling material.
Size of particles of filling material does, under certain conditions, not affect original water capacity of bed.
It has been further demonstrated that the water capacity of a bed is not affected by the size of the particles, provided the latter are spherical and of uniform size. In other words, the water capacity of two beds filled with material of different sizes is the same, provided the particles are spherical and of uniform size throughout each bed.
Silting up of bed during regular work.
Rapid initial decrease of capacity.
Consolidation of bed.
This original water capacity is, however, not maintained in regular work, as has been pointed out already. Basing the observations on regular work only, the original capacity decreases at first, after a new bed has been started or after an old reconstructed bed has been taken in hand, rapidly for some time and afterwards more slowly. Graphically expressed, this decrease is not represented by a straight line but approaches more nearly a parabolic curve. This initial rapid decrease is chiefly due to the consolidation of the bed.
Disintegration of filling material.
In connection with the movements in the bed tending towards its consolidation, it is also clear that the continual filling and emptying operations cause the smaller particles to be washed out of their original position and to be placed in the larger passages between the filling material, and if this process is assisted by the gradual disintegration of the particles composing the filling material, it is clear that the pores must become smaller and smaller in time, i.e. choked.
From these observations it follows that the filling material should be a hard substance, which will only to a limited extent be subject to this crumbling away process.