PART I: WELL WATER



WELL WATER

Well sighting and drilling

There is a lot that goes on behind the scenes that owners never hear much about. Selection of well site is a good example. Presumably, when land is subdivided, the developer has to determine reasonable building envelopes and well sites. We never thought to investigate these factors before buying our land. Our builder and the local health department found an appropriate well site, and we were never consulted.

Our well drillers only had to drill 145 feet to get twenty gallons per minute (“gpm”) - an excellent result. Some of our neighbors had as little as two gallons per minute at depths in the hundreds of feet. A “Well Completion Report” was submitted by the drilling company to the Orange County Health Department, giving the yield at 100 feet (fifteen gpm) and 130 feet (an additional five gpm), and stating the casing depth (63 ft). Another company installed the well pump, a three-quarter horsepower, eight gallon Red Jacket®, and the water line was sized at one inch to the pressure tank. Our pressure tank is 40-gallon size, supposedly “oversized.” We later added a boiler drain in order to measure the gpm at the pressure tank, and found we had ten gpm at the pressure tank. This ten gpm was important as far as our options for water treatment, since some filtration media require more gpm (particularly for backwashing – see below) than others.

I never knew to specify in the building contract the size of the well pump, well line or pressure tank. We were therefore dependent on our builder to specify these, and we were fortunate that our builder had high standards. As the owner of the well pump company told me, “I always put oversized pressure tanks in Mr. Hartley’s houses.” Ideally, however, I should have specified in our contract that a reasonable gpm must be achieved at the pressure tank.

Water testing

After our well was drilled, chlorine was added to disinfect the well, an application was made for well testing, and some time later the Orange County Health Department came to take water samples. Private companies will test well water for a fee, and in North Carolina, there is water testing at low cost from the State Laboratory of Public Health, through the local county health department. I opted for the maximum analysis which the state offered, including testing for pesticides and coliform bacteria.

Our first water testing report showed typical results in that the water was modestly “hard” (73 milligrams per liter of calcium and magnesium, which translates [divide by 17.1] to 4.27 grains per gallon). The harder the water, the harder to bathe and wash clothes in, and the more tendency to form “scale” in water heaters, on coffee pots, ice makers and the like. Below is one chart I found which rates hardness[1]:

Classification Mg/L or PPM (parts-per-million) Grains/Gallon or GPG

Soft 0-17 0-1

Slightly Hard 17-60 1-3.5

Moderately Hard 61-120 3.5-7.0

Hard 121-180 7.0-10.5

Very Hard 180+ 10.5+

Our water showed high levels of iron and manganese (3.59 mg/l and 0.52 mg/l respectively), and we were given a notice that these levels of iron and manganese were “not within acceptable or recommended limits.” We had very low levels of arsenic and trihalomethanes. Our pH was 6.8, slightly acidic. No coliform bacteria or pesticide residues were found.

When our untreated well water comes out of a spigot, it looks clear. But let it stand for just a few minutes and it quickly becomes orange. This is the result of the oxidation of (clear) ferrous (dissolved) iron to (orange) ferric iron. The orange ferric iron will stain clothes – you cannot do laundry in it. It will also quickly stain toilets, tubs, and showers. In my case, it even stained the exterior of our house, when I watered bushes with a sprinkler using untreated water. Iron oxides later clogged the sprinkler.

Subsequent water tests showed even higher levels of iron: 4.93 and 4.21.

Researching water treatment options

I researched water treatment on the Internet, read Jim Dulley’s bulletins, compared notes with neighbors, and spoke with many salesmen of local water treatment companies. Most salesmen of water softeners could not tell me how salt-efficient their systems were (see discussion below). I also spoke to a few non-local salesmen who were kind enough to take some time with me, and I eventually bought a system which I had installed by my plumber, on the advice of a long-distance water filter salesman whom I found through a distributor of healthy paints and other household products.[2] Later, I found a local water treatment installer, Chuck Lewis, who was knowledgeable, honest and willing to help me substantially modify my system according to my own research.

I found a good number of Internet sites to be well-written and helpful, but no one site had all the pieces of the puzzle. This appendix is my attempt to integrate what I have learned from these sites and my other research.

I ran across some Internet sites which discussed large municipal systems. Reading about municipal systems taught me that individual residential water treatment is an inefficient and much more expensive way to treat water. After years of having municipal water and sewer, I can now appreciate what a bargain they were.

One experienced water treatment installer told me that many individuals pay around $6000 to $7000 for a residential water treatment system (sold on commission), and then finance that system at very high interest rates, resulting in a real initial cost of $13,000 to $14,000. Then they pay a monthly fee to maintain it.

I was looking for a system which would have a reasonable installation cost, but most importantly, have a low ongoing maintenance cost.

Water treatment options

We considered the following residential water treatment systems: 1) water softener (ion-exchange/salt-based)/regeneration systems (also called water conditioners), 2) water filtration systems, 3) reverse osmosis (“RO”) systems, 4) ultra-violet (“UV”) light systems, 5) magnetic systems.[3]

Magnetic systems

Magnetic systems, which use magnets to alter the shape of calcium and magnesium ions in much the same way as catalysts (see KDF below), are controversial, yet they are used successfully in many industrial applications. However, there was no magnetic option for us, since magnets cannot be used with water which contains significant amounts of iron.

Ultra-violet (UV) systems

Since we didn’t have any bacterial contamination, I saw no need for a UV system, which is only used to eliminate “microbiological contaminants” (such as bacteria and viruses).[4]

Reverse osmosis (RO) systems

I resisted RO systems, since these are “point of use” (under sinks or in showers) rather than whole-house systems. RO also uses a lot of water and produces a very small service flow (some only fifteen gallons of purified water PER DAY, and requiring three to ten gallons of untreated water to make a single gallon of purified water[5]). Using an RO system can affect the function of an ice-maker in refrigerators, if it lowers the water pressure too much.[6] The RO membrane usually has to be replaced every two years[7]. RO has the greatest range of contaminant removal[8], and can remove arsenic and salt, which are otherwise not well-removed by other kinds of filtration.

Water softeners: ion-exchange (salt-based) systems[9]

At first, I assumed that we would use a water softener, since these were really the only systems with which I was familiar. Most of my neighbors had some sort of water softener, sometimes combined with other systems, such as UV. As my research progressed, I learned that water softeners are ion-exchange systems which work by passing the water through specially treated polystyrene beads (“resin” or “zeolite”), charged with sodium or potassium ions. As the water flows over the resins, an ion exchange takes place so that the sodium or the potassium replaces the calcium and magnesium ions in the water. Essentially, one is trading calcium and magnesium in the untreated water for either sodium or potassium in the treated water.

In these systems, the salt used definitely gets into the drinking and bathing water. I did not want sodium in my water, since it has negative health effects for humans and gardens, and I also wanted, if possible, to retain the calcium and magnesium, which have beneficial health effects. Salt can also damage a refrigerator icemaker and lead to poor quality ice.[10] Salt also hastens the corrosion of anode tubes in water heaters.[11]

If I had chosen a water softener, I would have wanted to use potassium chloride instead of sodium chloride, since added potassium is generally much to be preferred in the diet and in the garden (assuming I watered with treated water) than sodium. Using sodium makes the treated water a bit “slimy” in feel, whereas using potassium does not. Using potassium chloride would have increased my operating costs, however, since potassium chloride costs two or three times more than regular sodium chloride. There are also sometimes problems of caking, mashing or bridging of potassium chloride in some systems.

Periodically, the resins in water softeners must be “regenerated” with (salt) brine which reverses the ion exchange, and the waste water with the manganese and calcium is washed into the sewer line along with excess salt regenerate. Most systems are “co-currently regenerated”, meaning that the service flow and the regeneration flow go in the same direction, and untreated water is used for the regeneration. Better is a “counter-current system in which the regeneration flow runs in the opposite direction from the service flow. Best is a system which is both a counter-current system and a duplex system, which uses two tanks controlled by one valve, only one of which tanks regenerates at any given time, using soft water from the non-regenerating tank, rather than untreated hard water, for the better regeneration of the resins. A duplex system also eliminates the necessity for a reserve water capacity to get the unit through the day so it can regenerate at night.

If I had chosen a salt-based system, I would also have looked for a system which:

1) operated at a capacity of at least 4000 grains of hardness per pound of salt;

2) used fine mesh (40 to 50 mesh) resins which require less gpm to backwash;

3) contained a resin bed depth of a minimum of 24 inches over sufficient under-bedding (the material at the bottom of the tank, which covers the basket/distributor);

4) had a “free board” of 50%. (The free board is the empty space in a tank above the resin bed);

5) used a demand-initiated regeneration (DIR) valve, since it ensures that regeneration is initiated as needed;

6) used counter-current regeneration in a twin-tank (duplex) configuration.

Based on these criteria, I would almost surely have chosen a Hague® or Kinetico® brand.[12]

Water softeners are not water purifiers, and would not remove bacteria or solids. If I had had bacteria in my well water, I might have combined, as many systems do, the water softener with a UV system. If I had had high levels of arsenic, I would have had to use RO systems under my kitchen sinks and in my shower area. Water softeners can remove iron[13]. In addition, many water softeners do not handle well hydrogen sulfide (rotten egg smell) which often accompanies high iron levels.

Water filtration systems

Water filtration systems do not use salt and ion-exchange; instead, they use a variety of filtration media in backwashing or non-backwashing tanks. With the exception of carbon block, I reason that all media should be backwashed at some interval to remove solids and prevent the compaction or channeling of the media.

Iron

Iron in water presents special filtration challenges. Our situation is a common scenario: ferrous iron (appears clear in water) in the well water, which quickly turns into bright-orange ferric iron (oxidized at point of use in toilets, tubs, showers, taps). Iron is more easily filtered in its oxidized, ferric state. Iron can be oxidized by aeration (exposure to air) and chlorination (exposure to chlorine). Aeration appealed to me as the least toxic alternative, until I learned that aeration was complicated by the necessity to have a certain level of dissolved oxygen in the water (which we did not have). To raise the level of dissolved oxygen, one can inject soda ash, which will raise the pH which in turn can improve the effectiveness of the filtration media. Chlorination is cheap, simple and fairly innocuous, since it is used in highly diluted concentrations. It does require subsequent filtration in a way which effectively removes the chlorine as well as the iron precipitate.

Service flow

Backwashing (regeneration) flow

Tank size and sequencing

Both salt-based (ion-exchange)/water softener systems and water filtration/backwashing systems usually involve one or two fiberglass tanks holding some kind of either filtration media or resins (in the case of water softeners). One or both of the tanks are topped by a valve which triggers regeneration and/or backwashing in the middle of the night (when presumably little water is otherwise being used). I could see no advantage to getting more than the basic fiberglass tank, along with a basic Fleck 5600 backwashing valve.

The size of the tanks is critical to “service flow” (the rate of water used in the house) and “backwashing flow” (water used to backwash the filter media to remove the filtered elements and keep the media clean and loosely packed). The diameter (not the total volume) of the tank is a key element contributing to service flow. It is also important that the “riser tube” inside the tank (which transports the filtered water from the tank) and the porting and bypass valves be of a one-inch diameter.[14] The difficulty is that “most filter media require a significantly higher backwash flow than the service flow they support.”[15] Most houses need at least 5 to 7 gpm of “service flow” (the rate of water used in the house).[16] But to achieve that rate of service flow for some filter media would require a tank for which the backwashing might require 10 gpm. The solution is often to install two smaller (8 to 10 inches in diameter) tanks instead of one large tank, and to install them in parallel (half of the water flowing through each tank) rather than serially. Thus it is necessary to specify in the construction contract that at least 10 gpm be achieved at the pressure tank, in order to accommodate most service flow and backwashing flow requirements.

Chlorination/filtration systems

We have now installed two water treatment systems, the second system replacing and being a modification of the first.

Both systems use chlorination as the first step in oxidizing and precipitating out the iron. Water from the well is pumped to the pressure tank in the garage. From the pressure tank, untreated water passes through the first 25-micron polypropylene string sedimentation filter and enters a 120-gallon mixing tank. At the point of entry into the mixing tank, a chlorine solution is injected into the untreated water supply by a Chem-Tech chemical pump from a 25-gallon chemical tank filled with a mixture of simple liquid bleach and water.

The injection of chlorine is triggered by a sensor at the pressure tank which is itself activated by a drop in the water pressure in the pressure tank. I monitor the water in the mixing tank with a chlorine pool-testing kit so that the chlorine concentration will be kept at between 0.5 and 1.5 parts per million (“ppm”). I take the water sample from a boiler drain at the top of the mixing tank. In the mixing tank, the ferrous iron is oxidized to ferric iron, causing the ferric iron to precipitate out. Some of the ferric iron precipitate accumulates at the bottom of the mixing tank. A boiler drain at the bottom of the mixing tank is used periodically (every three to six months) to flush the tank by attaching a garden hose to the boiler drain and allowing the precipitated iron to flow out to the ground outside. A good deal of the ferric iron precipitate remains to be filtered out in the backwashing filtration tanks.

The chlorine also serves as a sterilizing agent. If my untreated well water had contained high levels of organic compounds, however, the chlorine would also have produced carcinogenic trihalomethanes which would then also required removal by filtration (probably more coconut shell carbon); however, the more layers of filtration, the more potential for drops in water pressure.

Original filtration: KDF55 plus coconut carbon shell

plus three polypropylene string filters

In our original system, the water from the chlorine mixing tank passed first to a second 25-micron polypropylene string filter and then in sequence to two fiberglass media tanks, each 36 inches high and eight inches in diameter.

Originally, the first tank had a Fleck 5600 backwashing valve and was filled with “KDF55”. KDF55 is a 50-50 copper-zinc alloy material which is one of two different copper-zinc media used in “Kinetic Degradation Fluxion” (“KDF”). KDF is also called “Redox” (Oxidation/Reduction) for its electrochemical oxidation effects.[17] KDF media comes in two varieties: KDF 55 and KDF85. The copper-zinc KDF media act in a catalytic manner to 1) remove chlorine, 2) kill bacteria, 3) transform (but not remove) calcium and magnesium ions to a chemical structure which does not form scale, and 4) remove hydrogen sulfide. It also acts mechanically to filter out solids, such as iron precipitate.

The second tank originally had no backwashing valve, and was filled with coconut shell carbon. In retrospect, it seems to me that the coconut shell tank should have had a backwashing valve, since the coconut carbon shell was subject to compaction and “channeling” as water passed through it

In the original system, chlorinated water with iron precipitate then went from the KDF backwashing tank (which backwashed every night), to the non-backwashing coconut shell carbon tank, through a third 25-micron polypropylene string filter and then into the house to be used.

After a few days of use, this system began to show “breakthroughs” of rusty water, particularly for high volume uses such as baths. Every filtration system has a service flow at which it operates efficiently. If the service flow is exceeded, the efficiency declines drastically, and some of the water which is forced through the system is essentially not treated. This is what was happening to us. I was also changing the string filters about every two weeks, since they quickly filled up with iron precipitate – a lot more maintenance than I had counted on.

Rejected filtration media

As I was casting about for alternatives, the best Internet site I found for understanding water filtration media options was , offered by Pure Water Products in Denton, TX. The site offers excellent charts showing the different filtration media and their characteristics, particularly their service and backwash flow rates at various tank diameters.

I considered many filtration media, particularly Birm, a manufactured medium (consisting of plastic coated with magnesium oxide) which is designed for iron and manganese reduction. But Birm produces only a 1.5 gpm service flow in an eight-inch tank. To use Birm, I would have had to increase my tanks to at least thirteen inches in diameter to get a 3.9 gpm flow rate (and a backwash rate of 10.1 gpm). And Birm does not have all of the other beneficial effects of KDF (“water softening” effects, chlorine removal, bacteriastatic properties).

Manganese Greensand is another filtration media often used to filter out iron precipitate. There are both natural and synthetic forms of greensand. Both have to be regenerated, usually with potassium permanganate, a highly toxic chemical, although there are apparently ways to regenerate it continuously with chlorine. Also, the service flow for greensand is only 1.2 gpm for an eight-inch diameter tank, with a 4.7gpm backwash rate. At twelve inches, the gpm would increase to only 2.7 gpm, and the backwash required would jump to 10.6 gpm (the maximum on my pressure tank).[18]

My goal was to have a relatively simple, low maintenance system with as few elements as possible, at as little additional cost to convert. My conversion cost was minimized by the use of my existing tanks so that I only had to add an additional Fleck 5600 backwashing valve and change the filtration media.

I wish I could have kept a tank with coconut shell carbon filtration media, because I like the sweeter taste of coconut shell carbon filtration. I also like the extra protection which carbon filtration provides against trihalomethanes produced when water containing organic matter is treated with chlorine. However, I felt I had to make my system as simple as possible. I also read my water analysis as meaning that my well water had low organic matter content, so not many trihalomethanes were likely to be produced.

Second filtration system: KDF85 in backwashing tanks

plus polypropylene string filters

Based on what I learned about filtration media from the Pure Water site, I designed a new system and implemented it with the help of Chuck Lewis. He ordered an extra Fleck 5600 backwashing head and KDF85 (85-15 mix of copper-zinc) media (along with the necessary underbedding garnet). He filled both of my tanks with KDF85 media, on an underbedding of eight by twelve mesh garnet with a 22-inch freeboard. He then installed the tanks in parallel, rather than serially. In this way, we increased our service flow capacity, since each eight-inch tank could produce a flow rate of 5.2 gpm in service flow (for a total of 10.4 gpm in service flow), and they could backwash on alternate nights – important, since the heavy KDF85 media requires 10.5 gpm to backwash each of my 8-inch tanks.

In November 2007 (a little over two years after we moved in to the house on April 1, 2005), I got Chuck to install in parallel two more eight-inch fiberglass tanks with Fleck backwashing heads, and to replace/refresh the KDF85 in all four tanks. I did this because it was clear that the media were in need of replenishment from the state of the water coming directly from the tanks which I was using to fill up the 25-gallon chemical tank. By having a total of four tanks, I am hoping to have to replenish the KDF85 media less often (perhaps only every four years, instead of every two years).

The KDF85 filtration system is a big improvement over the original KDF55 plus carbon shell system. KDF85 is most often used for iron, manganese and hydrogen sulfide reduction. At first we eliminated one of the original three polypropylene string filters – the one just after the holding tank, which had to be changed so often. I retained the string filter at the beginning of the system (for sediment) and the string filter at the very end of the system (which catches the iron precipitate not trapped by the KDF85 media). Later I added back the third string filter cartridge into the system, located just after the KDF85 tanks, but before the boiler drain which allows me to fill the 25-gallon chlorine tank, in order to keep the water cleaner in the chlorine tank.

In addition to changing the string filters every six months (inexpensive and quick), I have to add liquid chlorine and water to the 25-gallon chemical tank when it runs low, about once a month, depending on usage. Occasionally, I have to check the chlorine levels in the 120-gallon mixing tank, using the pool kit (a simple procedure of taking a water sample from the boiler drain at the top of the mixing tank, pouring it into the test tube, adding five drops of detection fluid, and using the color guide on the test tube to identify the chlorine level.) The only other maintenance is occasionally to flush out the iron from the bottom of the 120-gallon mixing tank by attaching a garden hose from the outdoors to the yard outside. That means some chlorinated water gets on the mulch, but I assume the chlorine evaporates fairly quickly. When I drain the mixing tank, I also drain a bit of water from my two water heaters (one solar preheater and one Marathon electric heater).

Two different chlorine solution-injection pumps

For the first two and one-half years of use, our water treatment system relied on a Chem-Tech eccentric cam displacement pump. The pump was reliable for the period we used it, but very noisy.

In December 2007, we replaced the Chem-Tech with a Stenner peristaltic pump, which was much quieter, and, we thought, longer-lasting than the Chem-Tech eccentric cam displacement pump.

In the end, we re-installed the Chem-Tech, when it developed that the Stenner pump was injecting the chlorine solution at a lower rate, due to a smaller diameter injection tube. Rather than modify the tube, we re-installed the Chem-Tech .

When the Chem-Tech does wear out, we may modify and install the Stenner pump, which also requires yearly replacement of a flexible feed tube. I have conflicting information about the suitability of the Stenner pump to injecting soda ash solutions, however (see below).

Addition of Soda Ash

When we began to design our rainharvesting system in 2008, I became more aware of issues surrounding the pH of well and rainwater, since rainwater is often very acidic. Chuck remarked that he thought our chlorine precipitation of the ferric oxide would be much more efficient if we raised the pH from 6.5 to 7.0 or a bit higher by adding soda ash to the chlorine solution. Chuck brought a 50-pound bag of light soda ash (sodium carbonate anhydrous) and we began adding about 20 cups of the soda ash to the 25-gallon chlorine chemical tank. The soda ash, being difficult to dissolve, is first mixed in a bucket of hot water (taken directly from the boiler drain of the Marathon water heater located conveniently nearby), before being added to the chemical tank.

Adding soda ash has noticeably reduced the amount of ferric iron precipitate making it through the filtration system. Adding soda ash does put some sodium into the water, however. I have not been able to find a source of potassium carbonate to substitute for the sodium carbonate. Sodium carbonate is readily available at pool supply stores in 50-pound bags selling for around $30. I find a 50-pound bag lasts about 6 months, adding 20 cups of soda ash about once a month to the chemical tank.

-----------------------

[1] (March 25, 2008).

[2] Shelter Ecology: (March 25, 2008).

[3] I also ran into “mysterious” water treatment systems. These are advertised on the Internet, come in impressive tanks and their salespeople explain to you that they are very qualified professionals who are not permitted to give away trade secrets and divulge the exact filtration mechanism.

[4] (March 25, 2008).

[5] With newer “auto-shutoff” features this is more like 3 than 10 gallons of untreated water to every gallon of purified water.

[6] “Kenmore Top-Mount Refrigerator Use and Care Guide,” for Model 106.74252400, ©2004, Sears Roebuck and Co., p. 5.

[7] Some RO membranes can last longer with proper pretreatment and filter replacement schedules.

[8] (March 25, 2008).

[9] See Mike Keller, “Maximizing Water Softener Efficiencies,” Water Conditioning & Purification (March 2004).

[10] “Kenmore Top-Mount Refrigerator Use and Care Guide,” for Model 106.74252400, ©2004, Sears Roebuck and Co., p. 15.

[11] “Water heaters: Hot tips for a better buy,” (March 25, 2008).

[12] One Hague system is touted to remove up to 5800 grains of hardness per pound of regenerate, to meet the strict requirements of the state of California, and to use potassium chloride pellets. Hague also prides itself on not using toxic glues on (RO) systems using plastic parts.

[13] The Hague “Water Max” is said to handle very high iron levels – up to 15 ppm.

[14] Because of retooling costs, many manufacturers may not have increased the diameter of the riser tube and the porting and bypass valves to the better one-inch diameter, from the older, smaller diameters.

[15] “Flow and Backwash Chart for Various Filter Media,” (March 25, 2008).

[16] (March 25, 2008).

[17] (March 25, 2008).

[18] Chuck Lewis says that his company makes a greensand tank 10 inches in diameter and 54 inches tall which produces a service flow of 8 gpm and requires a backwash flow of 5 gpm.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download