Backwash flow is an area calculation, not a tank size
Fluidising a bed of media means pushing water up through it fast enough to lift and separate the grains so the trapped material washes out of the top. What matters is the upward velocity, so the requirement is stated as a flow per unit of bed area:
Backwash flow (gpm) = rate (gpm per sq ft) × bed area (sq ft)
And bed area goes as the square of the tank diameter. A 10 inch tank is 0.545 square feet; a 13 inch tank is 0.922, which is 69 percent more. So moving up one tank size to fix a contact time problem also raises the backwash demand by two thirds, and that second consequence is the one people meet later, at night, when the pressure disappears.
At 12 gpm per square foot, a 13 inch tank wants about 11 gallons a minute for the whole backwash. That is more than most residential fixtures combined and it runs continuously for ten minutes or more. On a municipal supply it is a bill. On a well it is a question about the well.
| Tank | Bed area | Flow at 8 gpm/sq ft | at 12 | at 15 |
|---|---|---|---|---|
| 8 in | 0.349 sq ft | 2.8 gpm | 4.2 gpm | 5.2 gpm |
| 9 in | 0.442 sq ft | 3.5 gpm | 5.3 gpm | 6.6 gpm |
| 10 in | 0.545 sq ft | 4.4 gpm | 6.5 gpm | 8.2 gpm |
| 12 in | 0.785 sq ft | 6.3 gpm | 9.4 gpm | 11.8 gpm |
| 13 in | 0.922 sq ft | 7.4 gpm | 11.1 gpm | 13.8 gpm |
| 16 in | 1.396 sq ft | 11.2 gpm | 16.8 gpm | 20.9 gpm |
Read that table next to what your well actually sustains and the constraint becomes obvious. A well yielding 5 gallons a minute cannot properly backwash a 12 inch tank of media wanting 12 gpm per square foot, regardless of what the pump nameplate says, because the pressure tank contributes only its drawdown — typically twenty or thirty gallons — and then the whole demand falls on the formation. The well yield and drawdown calculator separates those two, and the pressure tank calculator tells you how many gallons the tank has to give.
What a failed backwash looks like
Nothing. That is the difficulty. A cycle that ran without enough flow still ran: the valve moved, the drain got wet, the control returned to service. The bed simply was not lifted, so the accumulated material stayed in it, and the filter continues to pass water at slowly increasing pressure loss and slowly decreasing performance.
Over months this compacts, and in some media it cements, at which point the fix stops being a setting and becomes replacing the bed. So the flow check is worth doing at installation with a bucket and a watch on the drain line rather than assumed from the control settings. If the measured drain flow is well under the calculated figure, the source is the constraint and no amount of adjusting the timer changes it.
The teaching point: the control type moves more water than any other decision
Take the default case — a 13 inch tank, 10 minutes of backwash at 12 gpm per square foot and 6 minutes of rinse at 5 — and each cycle sends about 138 gallons to drain. Run it every three days and that is 122 cycles and roughly 16,900 gallons a year. A household using 300 gallons a day uses 109,500. The filter is spending about 13 percent of everything the pump lifts on cleaning itself.
Now change one thing. A timer control runs on a fixed interval and knows nothing about how much water went through: on a week when the house was empty, it backwashes anyway. A metered control counts gallons and runs the cycle when the bed has actually processed its capacity. On a household whose real usage would justify a five day interval rather than three, that is 73 cycles instead of 122 — about 6,800 gallons a year saved, doing exactly the same job.
Nothing about the filter changed. The water saved is entirely the difference between running on a calendar and running on a measurement, and it is the same argument that makes demand-initiated softener controls worth their extra cost. It is also the lever with no downside: a metered cycle happens when the bed needs it, so the filter performs the same or better while using less.
The drain itself
The drain line has to carry the peak flow, which is the backwash flow rather than the rinse flow, and it has to do it without backing up into the control valve — a restricted drain line changes the flow through the bed and quietly undoes the whole cycle. Long runs, small diameter and lift all reduce what a drain will carry.
How a treatment device is connected to a drain, and what has to sit between the two, is set by the code your jurisdiction has adopted and by the inspector who enforces it. This page states none of that. What it gives you is the flow and the volume the drain has to carry, so the person making the connection is working from a number rather than a guess. Where the discharge is permitted to go is a separate jurisdictional question, and on a property with a septic system it is also a hydraulic load question: 16,900 gallons a year is a material addition to what the household wastewater flow calculator counts, and it arrives in concentrated 138 gallon slugs rather than spread out. Take that to your local health department rather than to a calculator.
What this page is not
Every water quality number on this page — hardness, iron, solids, transmittance, a concentration, a target dose — is something you enter from a laboratory test of your own water. Use a state-certified laboratory; your state or county health department maintains the list and will usually tell you what a well in your area is commonly tested for. This calculator has no idea what is in your water, will not tell you whether a result is acceptable, and does not select equipment. Treatment follows the report, and a device that removes one thing does not remove another.
It also does not cover brine regeneration. A softener uses a chemically different cycle with a brine draw and a slow rinse, and the salt and water arithmetic for it lives in the water softener sizing calculator. A media filter that backwashes without brine — carbon, sand, an oxidising media, a neutralizer bed — is what this page describes. If the bed is a consumable that dissolves rather than one that is cleaned, the media refill calculator is the other half of the picture, since backwash frequency and media consumption are separate clocks running on the same tank.
Questions people ask
How much water does a backwashing filter use?
It is the bed area multiplied by the backwash rate multiplied by the duration, plus the rinse that follows. A 13 inch tank has 0.92 square feet of bed; at 12 gallons per minute per square foot for 10 minutes that is about 111 gallons, and a 6 minute rinse at 5 gpm per square foot adds another 28, so roughly 138 gallons a cycle. Every three days that is around 16,900 gallons a year, which for a household using 300 gallons a day is about 13 percent of everything the pump lifts. The rate and duration figures come from your media supplier, since different media need very different treatment.
My well is 5 gpm. Can it backwash a media filter?
It depends entirely on the bed area and the rate the media needs, and for many combinations the answer is no. At 12 gallons per minute per square foot, 5 gpm supports about 0.42 square feet of bed, which is roughly a 9 inch tank. Anything larger is not being properly fluidised no matter what the control says, because the pressure tank supplies only its drawdown — commonly twenty to thirty gallons — before the whole demand lands on the well. The usual answers on a low yield well are a smaller diameter tank, a media that backwashes at a lower rate, or an atmospheric storage tank and a booster pump so that the backwash draws from storage rather than from the formation.
Does a failed backwash show up anywhere?
Not immediately, and that is what makes it expensive. A cycle that ran with insufficient flow looks identical to one that worked: the valve cycled, water went to the drain, the unit returned to service. What did not happen is the bed lifting, so trapped material stayed in it. Over months the bed compacts, pressure loss climbs slowly, and performance falls slowly, and by the time either is obvious the fix is often replacing media rather than adjusting a setting. The check is a bucket and a stopwatch on the drain line during a manually started backwash, done once at installation.
Is a metered control worth it over a timer?
On water use, almost always. A timer runs a full cycle on a fixed interval regardless of what the household actually did, so it backwashes a filter that treated nothing while you were away and it backwashes at the same rate whether use was heavy or light. A metered control counts gallons and cycles when the bed has genuinely processed its capacity. On a household whose usage would justify five days between cycles rather than three, that is roughly 6,800 gallons a year, and the filter performs the same or better because the cycles land where they are needed. On a well the saving is pump energy and wear rather than a bill, but it is the same water.
Why does backwash work worse in winter?
Because cold water is more viscous, and viscosity is part of what lifts the bed. The same media in the same tank needs a different flow to fluidise properly at 45 degrees than at 70, and a control set once in summer can be pushing too little or too much when the groundwater is at its coldest. Media suppliers publish backwash rates against temperature for exactly this reason. Where a system was commissioned in warm weather and the filter seems to have gradually stopped performing over a winter, the backwash rate against the current water temperature is the first thing to check rather than the last.