This section is from the book "A Treatise On Beverages or The Complete Practical Bottler", by Charles Herman Sulz. Also available from Amazon: A Treatise On Beverages.
Too much cannot be said about this subject. New filters, which render service very well at first in removing micro-organisms from water, may, after they have been in use a short time, become breeding-places for the organisms, and if pathogenic germs are present, far from purifying the water, are indeed a source of pollution, and render it much more dangerous to attempt to filter it.
It is erroneously assumed that so long as it continues to filter the water clear it also purifies the water; but there never was a graver and more dangerous mistake. In a few days the purifying action of a newly set charcoal-filter is entirely exhausted, where thousands of gallons of polluted water are daily passed through it. Then the filter becomes choked and charged with impurities, contaminating the water instead of purifying it. Indeed, a putrid decomposition of the retained organic impurities sets in, the presence of phosphates in the animal charcoal will act as fertilizer and materially assist the production within the porous mass of the filter itself of objectionable vegetable and animal growth, thus adding to the accumulations of impurities, which can be only reliably dislodged by the rebuming of the animal charcoal - not, as frequently stated, by reversing the current, introduction of hot water or steam. Reburn-ing only restores the full activity of the charcoal. For these reasons, unless the charcoal is changed very frequently for restored reburned charcoal, these filters are more dangerous than ordinary filters in proportion to the unjustified reliance placed upon their virtue. The purifying action may have ceased long ago - nobody can tell when, except by analysis - while the continued mere percolating property of the filter is misleading. There is no absolute connection between the two actions.
If we take Gauttier de Claubry's statement, that animal charcoal purifies 136 times its weight of very impure water, as correct, we can easily calculate how long or how much water we can purify with a certain quantity of charcoal; say 136 pounds animal charcoal purify 136 pounds or 17 gallons of water.
What an immense quantity of charcoal would it take to purify 10,000, 50,000 or even 100,000 gallons daily, as required for large establishments.
Sand, once prepared for filtering purposes, and after having been in use for some time, must be washed out and thus its filtering capacity restored. The small amount of salts which sand has absorbed from water is removed by washing, and the suspended impurities that it has retained are thus removed likewise.
There is no insoluble matter retained in pores which need destroying or burning, like in animal charcoal, sand being of a different structure. The purifying action of sand, however, as we know already, is a mere mechanical one, while animal charcoal acts both mechanically and chemically. The same holds good for coke.
Whatever filter material be employed, charcoal, sand, asbestos, etc., a thorough cleansing or renewal must take place whenever it ceases to do its proper work, of which every carbonator should convince himself by making repeated tests.
The filter should be so arranged that it can easily be cleansed.
Where a large amount of filtering is carried on, several filters should be employed to permit an uninterrupted filtering process while one or two filters are being cleansed, or the filter medium is being renewed.
A charcoal filter for limited use should be cleansed, and the charcoal reburned or renewed, at least every month.
A sand or coke filter, and especially those filters that have the practical arrangement to reverse the current and agitate the filtering material, should be cleansed at least every other day.