What a Foot Valve and a Suction Strainer Each Cost You

Why this matters

Both of these parts are fitted to protect the pump, and both of them charge for that protection in the one currency the suction side cannot spare: feet of available suction head. That would be a fair trade if the charge were the number on the manufacturer's curve. It is not. The number that actually lands in the budget is the drop at the dirtiest condition you are willing to let the part reach before somebody cleans it, and nobody writes that condition down. In the lift worked below, the clean drop of the two parts together is a bit more than half the suction friction, and the fouling allowance nobody ever agreed on eats 85 percent of the installation's entire margin. That makes both parts a maintenance decision wearing a design decision's clothes.

Before you open a strainer or pull a foot valve

Isolate both sides of the strainer, then prove the isolation. Close the upstream and downstream valves, open the strainer's vent, and watch the body gauge for a full minute before a cover bolt moves. A valve that is passing shows up in that minute and nowhere else.

If the line is under vacuum, break the vacuum deliberately at a high-point vent, not by cracking the cover. A joint opened under vacuum pulls air and loose debris inward, straight toward the impeller. On a hot line it is worse: the moment atmospheric pressure gets in, liquid that was stable under vacuum flashes to steam.

A strainer cover on a hot line is a scald source. Break it from the side rather than standing over it, with a face shield rather than glasses, and drain to a routed hose rather than to the floor.

Assume the basket contents are biologically active if the source is a wet well, a cooling tower basin or any open sump. Wet, that is a glove and eye problem; once the debris dries and is knocked about, the route becomes inhalation, and the control is respiratory under the employer's program at 29 CFR 1910.134 rather than a glove. If anyone has to enter a wet well or a tank to reach a foot valve, that is a permit-required confined space under 29 CFR 1910.146 in general industry, or 29 CFR 1926 Subpart AA if the work is construction, and the entry is isolated, tested and attended before a head goes below the opening.

What each one buys

A foot valve is a check valve at the bottom of a suction lift, almost always with an integral screen. It buys two things: the suction line and the pump stay full of liquid when the pump stops, so the pump can start without being re-primed, and coarse solids stay out of the line entirely. On a lift with no priming system, it is not an accessory, it is what makes the installation work at all.

A suction strainer is a removable basket or screen in the suction line, upstream of the pump. It buys one thing: particles above its mesh opening do not reach the impeller, the wear rings or the seal faces. That matters more than it looks. A seal's faces run on a liquid film thinner than the particle sizes a suction line routinely carries, and the fastest way to score a seal face is to feed it grit.

Neither part is buying reliability in general. Each one is buying protection against a specific failure, and each one adds a component that itself fails.

What each one charges, and the currency

Both charge a pressure drop, and on the suction side a pressure drop is subtracted directly from available suction head.

There is no generic loss coefficient worth quoting for either part. A foot valve's drop is dominated by its screen's open area and its poppet design, and a basket strainer's by its open area ratio and the mesh; two parts of the same nominal size from two makers can differ by a large multiple. The manufacturer's own loss curve at your flow is the only source for these numbers. Every figure used below is labelled illustrative for that reason.

Both charges scale with roughly the square of flow, like every other velocity-driven loss in the line, so a ledger written at one flow is a statement about that flow only.

The foot valve carries one charge a strainer does not: it is a check valve, so when its seat stops sealing the line back-drains between runs and the pump loses prime. That failure presents as a pump that will not start reliably, and it is routinely misread as a motor or a control problem.

The number that belongs in the budget is the dirty one

A clean strainer drop is a commissioning number. The strainer spends most of its life somewhere between clean and the point where somebody notices, and the budget has to survive that point.

So the design decision is not the strainer, it is the fouling allowance: how much of the clean drop you will let it multiply by before it gets cleaned. Common practice is to allow a strainer to reach two to three times its clean drop and to alarm there, and that allowance is what gets carried in the suction budget. The same applies to a foot valve's screen, which nobody puts a gauge on and which fouls from the outside, where you cannot see it.

An allowance you cannot measure is not an allowance. A differential gauge across the strainer is what converts the allowance from a hope into a rule, because it is the only thing that tells anyone where in the allowance the basket currently sits. Without it the cleaning trigger becomes a calendar, and a calendar is wrong in both directions: it opens clean baskets and it leaves loaded ones.

Worked ledger: one lift, clean and allowed-dirty

An end-suction pump lifting from a below-grade sump. Water at 70 F, specific gravity 0.998, so 2.315 ft per psi. Station barometric 14.3 psia. Design flow 250 gpm. The liquid surface sits 12.0 ft below the pump centerline. Published required suction head at 250 gpm is 9.0 ft.

Loss figures at 250 gpm, each taken from its own manufacturer's curve and used here as illustrative values: foot valve with integral screen, 2.8 ft clean; basket strainer, 0.6 ft clean; all other suction pipe and fittings, 3.2 ft.

Clean: 13.66 available against 9.0 required, a ratio of 1.52, which sits at the top of the 1.2 to 1.5 band specifications commonly use, with the pump manufacturer's guidance governing where it differs.

Allowed dirty: 9.66 against 9.0, a ratio of 1.07. The pump is not past the required value, and required value is defined at a 3 percent head drop, so 1.07 is a pump sitting just outside its own documented onset with no margin left for the tank level, the flow or the temperature to move.

Where the money went. The two protective parts contribute 3.4 ft of the 6.6 ft of clean suction friction, which is 52 percent of it. At the allowance, they contribute 7.4 ft of 10.6 ft, or 70 percent. And the allowance alone, 4.0 ft, is 85 percent of the 4.66 ft of clean margin the installation had. The parts did not take the margin with their clean drop. They took it with a permission nobody wrote down.

Now move the flow. At 312 gpm, 25 percent above design, every velocity-driven loss scales by roughly the square: 1.25 squared is 1.5625, so 6.6 ft of clean friction becomes 10.31 ft, and available falls to 9.95 ft with a spotless strainer. Required rises with flow at the same time, so a curve that gives 9.0 ft at 250 gpm will give appreciably more at 312. A clean installation at 25 percent over design is already tighter than a fouled one at design, which is why the ledger has to state its flow on the same line as its answer.

The failure mode, and it is the ordinary one: the strainer gets cleaned on the callback, the pump quietens, and the ticket says the strainer was blocked. Cleaning recovered 1.2 ft out of the 4.0 ft the allowance permits, because the foot valve screen sitting in the sump was never touched and nobody has ever seen it.

When each one comes out

Delete the foot valve on a flooded suction. Its whole purpose is holding prime on a lift. If the source is above the pump centerline, it is 2.8 ft of pure charge with nothing bought, and it is a check valve waiting to leak by. On a lift, deleting it means fitting a priming system instead, which is a different trade, not a saving.

Take the temporary startup strainer out. A conical strainer fitted at commissioning to catch weld slag and construction debris is meant to come out after a stated running period. Left in, it becomes a permanent restriction with no basket to inspect, no differential gauge and no entry on any drawing, and it is one of the most common explanations for a pump that was never right from the day it was started.

Do not go finer than the pump needs. The largest particle the impeller passages, the wear-ring clearances and the seal faces can tolerate is a number the pump manufacturer states. That number sets the mesh. A finer mesh charges more head at every flow and loads faster, so it buys a shorter cleaning interval in exchange for protection the pump did not ask for.

How to verify you got this right

Fit a differential gauge across the strainer and record its reading on the day the basket is known clean. That single number is the datum for the whole allowance, and without it the allowance has no zero.

Then, at the next clean, record the differential immediately before opening and immediately after closing. The before-and-after difference is a difference from one instrument, so its fixed zero offset cancels and what survives is the proportional term on the difference itself. If the two numbers are the same, either the basket was clean and did not need doing, or the gauge tappings are plugged, which is common on a strainer and is checked by tapping the lines rather than by trusting the gauge.

Finally, get the foot valve on the same schedule as the strainer, or accept in writing that its screen is unmonitored. It is the part in this pair with the larger clean drop, the larger allowance and no instrument on it at all.

References

• 29 CFR 1910.146, permit-required confined spaces, in general industry, and 29 CFR 1926 Subpart AA for construction, where reaching a foot valve means entering a sump, wet well or tank
• 29 CFR 1910.134, respiratory protection, where dried debris from a basket is disturbed and the SDS or the material calls for a respiratory control
• Manufacturer loss curves for the specific foot valve and strainer, which own every pressure-drop figure used above, and the pump manufacturer's maximum passable particle size, which owns the mesh selection
• See related: How to Work Out NPSH Available at the Pump You Are Standing At; How Suction Conditions Fail in Practice; Why the Suction Side Decides Whether a Pump Works