Filtration ~ facts, fallacies and futures
The following article first appeared in Splash, the official and very popular magazine of the World Waterpark Association. It begins with fundamentals well known by most PPOA members, but proceeds to bust a number of popular myths that have been misleading operators since the Romans invented filtration…
Filters, everywhere in our environment, allow little stuff to pass through and keep the big stuff out. They’re in cigarettes and coffee pots, automobiles and air conditioners. They appear as window screens, fish nets, and accordion-pleated rolls of paper. Filters are so critical to the function of swimming pools and water features that a room has been named in their honor. No matter how important we think these appliances are, however, filters are often assigned virtues well beyond their capabilities. It’s important to know what they will do and what they won’t; with reasonable expectations we can get the very most out of them ? or into them ? and leave sanitation, oxidation and other miracles of pool care to different devices.
First, let’s look at some FACTS:
Obviously, removing suspended solid material from recreational water to make it clear and “clean” is our goal. To accomplish this particulate removal through filtration, there are many alternatives open to designers and builders. Sand, diatomaceous earth, cartridge and more ? there exists a variety of types, sub-types and sizes in the world of pool filtration even among the three major categories. Which one should we choose?
There’s been plenty of sparring for “best” between pool owners and operators based on individual preferences and performance comparisons. However, all three types, assuming they are of good design, deliver effluent of equal quality. It is little more than an intellectual exercise to debate which of them will remove the smallest particles in terms of microns or of water clarity. Any seasoned operator, when presented with a crystal-clear pool, would be challenged to determine which category, much less type, of filter was on line. She or he would have a one-in-three chance; polished water looks like polished water, no matter how you got it that way.
Of the two most commonly used filter categories the sand system is the easiest to understand so will be our focus here. The specific technology called “high-rate” sand filtration ? now gaining dominance world-wide ? will be included in most of this conversation. Let’s first look at the concept called particulate “entrapment”.
Just what is big and what is little, in the context of water turbidity and particle size? Good filters stop solid particles from around 5 microns to 20 microns in dimension, and all sizes larger. (One micron, incidentally, is .00004 inches…) Here are some interesting size comparisons in terms of known materias. Note that a typical human hair is 70 microns in diameter and the lower limit of human visibility is about 40 microns, so 5 or even 20 microns is pretty tiny indeed.
ONE AVERAGE GRAIN OF TABLE SALT 100 MICRONS
HUMAN HAIR 70 MICRONS
THE LOWER LIMIT OF HUMAN VISION 40 MICRONS
WHITE BLOOD CELLS 25 MICRONS
TALCUM POWDER 10 MICRONS
AVERAGE BACTERIA 2 MICRONS
Turbidity, however, is based on the density (numbers in a specified volume) of particles even more than on particulate size. Filters must remove virtually all of the larger material and most of the smaller in order to render the water “sparkling clear.” (Even then chlorine or another oxidizer must be in the equation for that polish you’re looking for…)
Soil is captured in the sand filter bed by a combination of two processes: First, solid particles lodge in the extremely small spaces and voids between the sand particles and, second, gelatinous and mucous-like substances and oils tend to cling to the grains of filter sand. The latter, then, enhances the mechanical effectiveness of the former, as “soil entraps soil” and the filter efficacy actually improves over time and usage. (In the high-rate version, this phenomenon works its way deep into the sand bed rather than existing only on the surface ? hence the term “three-dimensional” filtration.)
As these two dirt-collecting mechanisms work together, the filter stores up more and more of a gelatin-and-soil-filled bed. The filter is now extremely effective in removing suspended soil and the resultant water quality is optimum for many days.
Finally, however, the cycle comes to an end. The loaded medium becomes increasingly dense and, ultimately, resistant to water flow. At some point adequate flow can no longer be sustained; the filter must be cleaned. We know the time has come because the pressure in ? influent ? goes up, pressure out ? effluent ? goes down (up to 15 psi differential), and the reading on the flow meter drops ten percent or more. The cleaning process, called backwashing, is accomplished by re-directing the water flow backwards through the filter, generally from the bottom up, in order to expand and scrub the sand. The entrapped soil is thus carried out to waste in no more than about three minutes per tank.
Valves are switched back, and a new filter cycles begins all over again.
Now let’s expose some FABLES:
Most any sand will work in a high-rate-sand filter.
No. Sand grade and grain type are very important. Crushed, angular silica sand works well while rounded beach sand (typical of sandblasting material) is very poor for entrapment. The sharper the sand the more likely particles will be held while allowing the water to pass. Filter companies are particular about the sand sources for their filters.
The size of the sand particles serving as the filter bed is an important consideration too. If the sand is too course, the voids between the particles are too large to efficiently stop the passage of fine solids. If it’s too fine, the bed will be too dense and very little space is left for dirt to accumulate between the sand grains. The sand particle size best for pool filters has been established in a range of 0.4 to 0.6 millimeters, or, in grade, number 20. Number 30, a finer sand grade, is sometimes recommended for indoor pools when the soil volume is inadequate to effectively load the more common grade.
It is best to change the filter’s sand every year.
No. Well cared for sand never needs changing. Backwashing does not “wear out” sand during the brief 100 to 300 minutes per year that sand scrubbing occurs. After all, it took nature hundreds, even thousands of years to create and round off that sand on the beach. Calcified or oil-laden sand may have to be replaced, but that’s your fault, not that of the filter or the media.
If the water gets cloudy, we need to backwash.
No. Cloudy water can occur for a variety of reasons, but it’s rare that simple backwashing will improve things. Backwashing places the sand filter, temporarily, in its least effective mode ? hardly a help for your cloudy water. You need to do it, but only when you need to do it. The dirtier the filter the better, until the filter begins to stop water. Then and only then is when you need to backwash. Early in the new cycle the filter’s efficacy is not optimized; it takes from a few hours to a day or more to establish that natural “floc” of gelatinous material and fines in order to propagate the process. Those operators who backwash by the calendar or clock rather than by the gauges are guaranteeing that their filters run less efficiently than they could. Frequent, unnecessary backwashings can make sand filters run downright poorly! Don’t backwash your filters along with the Saturday bath; they probably don’t need it as badly as you might.
( With indoor pools and water features, “breakthrough” of fines is possible, requiring what appears to be premature backwashing ? where the differential was never achieved. This is unusual and often is the result of an improper sand choice.)
If we can’t backwash at design flows, lower values for longer periods work just fine.
No. Backwashing must be done at the design flow, usually 15 gallons-per-minute for each square foot of filter surface area. A slower rate, for inadequate sewer lines or whatever reason, is certain to allow conglomerated sand to develop. Also, with hair or lint often serving as a structure, balls of organic material ? “mudballs” in pool terminology ? will form and become imbedded in the top layer of sand. If the backwash velocities are insufficient to break up and wash the clumps to waste, they will work their way deeper into the filter bed, helping to create “channels”. Ultimately such ant-nest-like passages will permit unfiltered water to take the path of least resistance through the medium ? and our systems fail to function.
In poorly flushed sand, living organisms can propagate, too. This colonization of pathogens can be dangerous. Fix the waste line or install a holding tank for backwash water so the full, backward flow can be used to keep your sand healthy.
Iron, manganese, copper, and calcium can be removed by filters.
No. Nothing but suspended solids is removed by filtration. Suspensions and solutions are very different. Dissolved solids like salt and intentional calcium hardness (remember TDS?) cannot be seen by the filter media. Iron pipes, copper heater elements, calcium plaster and grout, impeller metal or any other affected solids, when dissolved or “in solution”, are solids no more; essentially, they are part of the liquid and will fly right through our filters. Ultra-fine suspension won’t stop on its way through either; “colored” water, otherwise transparent, is an example.
If you raise the pH accidentally to well over 8.3, you may in fact precipitate your formerly dissolved solids. Your whole pool will be a mess, so this is to be avoided. For the most part, dissolved solids are invisible and harmless to your system, your patrons and your chlorine’s work potential. Don’t expect your filter even to notice them.
Filters stop bacteria and sanitize water.
No. There are many who confuse the filtration function with the disinfection function. Disinfection is the chemical process of killing disease-causing bacteria and other micro-organisms, accomplished in the body of pool water itself. Bacteria are generally far too small to be noticed by your filter media. Most pathogens sail right through the sand and back to the pool. Remember it is chlorine (or equivalent) that has the job of sanitizing; the filter’s assignment is to remove the “big pieces”, mostly those inorganic suspended particles that the chlorine can’t oxidize.
Filters do remove soils that, if not removed, might impede the sanitation effort. Those who think of filtration as part of the disinfection process, however, because a filter may be capable of removing some pathogens, have been misled. This removal is not particularly beneficial, and can be a problem if relied upon.
Cryptosporidium, a large oocyst (six to eight microns in size), can in fact be trapped in a well-loaded filter. Recently, a focus on filtration for crypto removal has shown up in fecal-accident recommendations. It is not reasonable to expect the filtration system to remove all such infestation, even in three or four cycles. Too much of the water never sees the filter in a day, or even two. Chlorine’s still got the primary job. (Crypto is another conversation for another time…)
And, finally, let’s imagine Filter FUTURES:
Automated and sequential backwashing, remote management, sand alternatives, new packaging and efficient configurations have been emerging for a while. The next step may be the intelligent breaking of a longstanding rule: Never floc a high-rate sand filter. While old, two-dimensional sand filters benefited from this addition of a coagulant or other filter aid, adding alum, polymer, PAC or another flocking agent to the three-dimensional process in an attempt to improve output has always “blinded” the bed. The water often looked better, at least at the very beginning hours of a filter cycle, but too much build-up created the need for premature backwashing. No amount of care in the trickling in of the micro-flock avoids this problem; eventually, either too little is added to do any good or too much collects and filtration grinds to a stop. In every case, we lose filter cycle time, labor, energy and water; we simply come out short. Flocking high-rate sand is a flop.
But what if we could somehow control a filter’s output qualitatively? That’s how our chemical controllers and our thermostats do it. Choose a desired outcome, seek it by measuring results, then stop the feed (in this case the polymer) before problems arise. Let’s use a turbidimeter on the filter’s effluent and trickle in a flocking agent while we watch results! If our output is, say, 0.3 NTU (Nephelometric Turbidity Units) and we set the instrument ? the controller ? to 0.1 NTU, the relay remains closed and a tiny pump feeds well-chosen polymer until the improvement in clarity is achieved (or no more improvement is observed by the meter). Then the feeder is shut off. The “clarity controller” patiently waits until the clarity degrades slightly ? to a point pre-selected to trigger feed again.
All successful automation uses qualitative management rather than quantitative, blind feed. A controller of this type will avoid high-rate sand’s only real drawback, the marginal quality during the beginning of each filter cycle. Such a subtle, monitored application of filter aids may in fact make high-rate-sand filtration better than it’s ever been before. Look for this leap in filter technology in our filters’ collective futures… Maybe there will be a reason to brag about one category of filter after all.