Why French Drains Fail: Five Failure Modes

A homeowner calls in March. Water is standing over the drain line a contractor installed three years ago. He's not returning calls. She wants to know if the whole thing has to come out.

Most of the time it doesn't. French drains fail in a small number of predictable ways, and four of the five leave evidence you can find in under thirty minutes with a shovel, a hose and a tape. The fifth requires digging, and you should say so before you start.

What follows is how to work through those five on a service call, what each one costs to fix relative to a full replacement, and how to quote the repair without either overselling a rebuild or promising a $400 fix on a system that's dead.

Start by proving water still enters the pipe

Before you theorize, run water. Put a hose at the upstream end of the trench, on the surface directly over the stone, and open it to somewhere around five gallons a minute. Then go stand at the outfall.

You're timing three things: how long until water shows at the outfall, how much comes out relative to what you're putting in, and whether it runs clear or carries fines. That gives you the first fork in the diagnosis.

• Water at the outfall within a few minutes, roughly matching the hose. The pipe and the stone are still conveying. Your problem is upstream — the water in the yard isn't getting into the system — or the system is undersized for real rain.
• Water at the outfall, but a trickle, and late. Partial blockage or inadequate fall. Something is throttling it.
• Nothing at the outfall at all. Either the outfall is buried, the pipe is fully silted, or the line is broken or crushed somewhere between.
• Water surfacing mid-run, upstream of the outfall. The line is blocked at that point and backing up. Mark it. That's where you dig.

Do this first, every time. It costs you ten minutes and it keeps you from writing a rebuild proposal for a system whose only problem is that the daylight end got covered by mulch.

Failure one: fabric wrapped the wrong way around silt

This is the most common one, and the hardest to explain to a homeowner without sounding like you're bad-mouthing the last guy.

Filter fabric belongs between the stone and the native soil — a full trench wrap, stone inside, dirt outside. What a lot of installs use instead is sock pipe, or a fabric wrap sized just to the pipe, sitting in stone that's sitting directly against clay. Every wet season, fines migrate into the voids in the stone. Two or three years in, the stone is a solid mass of silt-filled gravel and the sock is glazed over.

How you confirm it: dig a test hole down to the stone somewhere in the middle of the run. Pull a handful. Clean washed stone rattles and you can see daylight between pieces. Silted stone comes out as a clod. If you can smear it between your fingers, the aggregate is done.

A second tell — if the water at the outfall ran milky during the hose test, the system is actively moving fines. That means the soil-to-stone interface was never separated properly, and it will silt again no matter how well you jet the pipe.

This is the failure that genuinely requires replacement of the affected run. You can't rinse a trench. Be direct with the homeowner: the pipe may be fine, the stone is not, and the stone is most of the labor.

Failure two: the stone was never washed, or it was the wrong size

Related but distinct. The first failure is soil getting in. This one is the stone arriving pre-loaded with fines, or being too small to hold useful void space.

Clean drainage stone — a washed 3/4 inch to 1 1/2 inch, no fines — gives you roughly 30 to 40 percent void space depending on gradation and how it's placed. Crusher run, road base, or anything with a dust component in the mix gives you almost none once it's compacted and wet. It sets up like weak concrete.

The tell on a service call is simple. Dig your test hole, wet the excavated stone, and squeeze. Washed stone stays loose. Anything with fines will bind. You can also just look at the gradation: if the pieces range from an inch down to sand in the same shovelful, it wasn't washed drainage stone.

Void space matters more than contractors usually credit. A 100-foot trench, 12 inches wide, 18 inches of stone depth, is 100 × 1 × 1.5 = 150 cubic feet of stone. At 35 percent voids, that's about 52 cubic feet of storage, or roughly 390 gallons sitting in the trench during a storm. At 5 percent voids, it's 56 gallons. That's the difference between a system that buffers a downpour and one that overflows in ten minutes.

Substitute your own trench dimensions. Width in feet, times stone depth in feet, times length, times your void fraction, times 7.48 gallons per cubic foot. It's a number worth having in your pocket when a homeowner asks why the cheaper bid used a narrower trench.

Failure three: the fall isn't there, or it reverses

Perforated pipe wants fall. The common target is 1 percent — an inch of drop per eight feet, or a foot per hundred. Some markets and some soils will run at half a percent successfully. Below that you're relying on a full pipe and hydraulic head, and silt settles instead of moving.

You can shoot this on a service call without much gear. A laser level and a rod is fastest. A string line and a line level across a long run works, and a four-foot level with a measured shim works for spot checks.

What you want is the invert elevation at the upstream end, the invert at the outfall, and at least two points in between. Two points is not enough. Trenches that were dug by a machine operator eyeballing it often have a belly in the middle — fall at both ends, a low spot at 40 feet, and that low spot is where every fine in the system has been collecting since day one.

The arithmetic: total drop divided by total run. A 120-foot line dropping 9 inches is 0.75 feet over 120, or 0.625 percent. Workable. The same line dropping 2 inches is 0.14 percent, and it's a horizontal storage tank with a pipe in it.

Reversed fall is worse and more common than you'd think, usually because the outfall was located for convenience rather than elevation. If the daylight end is higher than a point mid-run, water will never leave. This is the same elevation question that decides whether a yard needs a gravity outlet at all, and it's covered in more depth in catch basin vs dry well — the outfall elevation check is the same check.

Failure four: no cleanout, so nothing was ever maintainable

A French drain with no cleanout is a system nobody can service. You can't camera it, you can't jet it, and you can't tell a homeowner what's wrong without excavating.

A cleanout is a riser tee at the upstream end, and ideally a second at any major bend or every 75 to 100 feet on a long run. Cost to add at install is small — a tee, a length of solid pipe, a cap, maybe fifteen minutes. Cost to retrofit is a dig at each location.

Here's the honest part for a service call. If the system has silted stone, adding cleanouts doesn't save it. If the pipe silted but the stone is still clean and the fall is good, jetting through a new cleanout can genuinely restore it, and that's a few hundred dollars of work instead of a few thousand.

So the sequence matters: confirm stone condition first, then decide whether cleanouts and jetting are worth proposing. Selling a cleanout retrofit on a system whose aggregate is dead is the kind of thing that gets you called back in eight months.

Failure five: the outfall is buried, crushed, or freezes shut

Always walk to the daylight end before you write anything. This is the cheapest failure to find and the one contractors skip because it feels too obvious.

What you'll find, roughly in order of frequency: mulch or sod grown over the grate, a pop-up emitter with a spring that seized, the discharge buried by a landscape crew regrading a bed, a crushed end from a mower or a vehicle, a rodent screen packed solid with leaves, or the outfall sitting below the grade of the swale it discharges into so it backs up every storm.

In cold markets there's a sixth: the last few feet freeze. Shallow discharge, slow trickle, cold snap, and the line ices from the outlet backward. Then the February thaw arrives, the system is plugged, and the basement takes it. The mechanics are the same ones that cause winter callbacks on sump pump discharge — shallow cover, low flow, and an exit that sits where snow piles.

Fixing an outfall is usually the smallest ticket you'll write. Clear it, re-set the emitter, extend it to better grade, add a rodent guard that can actually be opened. Charge for the trip and the diagnostic, do the fix, and be honest that you haven't verified the rest of the system. Then offer the full diagnostic separately.

Pricing the fix without guessing

Once you know which failure you're looking at, the repair scope roughly sorts into three tiers. Your own numbers go in the right column — these are relative magnitudes, not prices.

Two things to add to any rebuild quote that contractors routinely eat. First, disposal — silted stone is not reusable and it's heavy. Second, the extra excavation difficulty of digging out compacted, wet, fines-bound aggregate versus virgin soil. It's slower than a fresh trench, not faster, and if you price a rebuild at your new-install per-foot number you'll lose on it.

If you price by the foot as a sanity check, a rebuild should sit above your new-install band, not below. The reasoning for why per-foot numbers break down on nonstandard conditions is in pricing a drainage job by the foot, and a failed-system rebuild is exactly the nonstandard condition that breaks it.

How to say it to the homeowner

The awkward part of a failed-install service call is that someone already took this person's money. You don't need to trash them, and you shouldn't. You need to describe what you found and what it means.

"I dug two test holes. The pipe is fine. The stone around it is packed with clay — that's what's stopping the water. There's no way to clean stone in the ground, so the fix is to pull that section and rebuild it with washed stone and fabric against the soil. I can show you the sample I pulled."

Pull a sample and leave it in a bucket on the driveway. A homeowner who can squeeze silted gravel in their own hand doesn't need to take your word for anything. Photograph the test hole, the sample, the outfall and the level reading, and put those photos in the proposal.

Then give them a real choice — a minimum scope that fixes the immediate symptom and a full scope that fixes the cause, with what each one does and doesn't cover. Documenting the failure this thoroughly is also what protects you when they compare your number to a cheaper rebuild bid that skips fabric and washed stone. That conversation is its own skill, and it's laid out in holding your price against a cheaper bid.

Photos, measurements and scope language belong on the proposal itself, not in a follow-up text that gets buried. If you're building these in ScapesHQ, the test-hole photos and the fall readings go into the proposal options alongside the price, so the homeowner is looking at the evidence and the number on the same page.

Build the diagnostic into your service call

Make it a fixed sequence so you don't skip the cheap checks. Same order every time:

1. Walk to the outfall. Clear it if it's buried. Note the elevation relative to where it discharges.
2. Run the hose test from the upstream end. Time it, watch clarity, mark where water surfaces if it does.
3. Shoot fall at the two ends plus two or more intermediate points. Write down the inverts.
4. Dig test holes — one mid-run, one at any surfacing point. Pull stone. Squeeze it.
5. Look for cleanouts. If there are none, note that as a scope item regardless of what else you find.
6. Photograph everything and leave a stone sample where the homeowner can handle it.

That's an hour to ninety minutes of real work on a property where the existing system is under the ground and you have no as-built. It's worth charging for, and the case for charging is the same one that applies to any drainage diagnostic — the value is in the answer, not the shovel. If you don't already charge for this, what to charge for a drainage site visit walks through pricing it from your own loaded cost and crediting it against the job.

The contractor who shows up, finds the buried outfall in four minutes, fixes it for a small ticket and tells the truth about the rest of the system is the one who gets the rebuild call in two years when the stone finally gives out. That's a better business than selling a rebuild today on a system that only needed its emitter cleared.

FAQ

How long should a French drain last before it fails?

A properly built system — washed stone, full trench fabric separation between stone and soil, adequate fall and accessible cleanouts — should run 20 years or more in most soils. Systems that fail at three to five years almost always failed on materials or fall, not age. Heavy clay and high silt loads shorten the life of any install, so the honest answer varies by soil and by how much surface sediment reaches the trench.

Can a silted French drain be jetted clean?

You can jet the pipe if there's a cleanout to get a nozzle into. You cannot clean the stone surrounding the pipe, and in most failed installs the stone is the problem. Jetting is worth proposing when a test hole shows the aggregate is still loose and washed, and it's a waste of the homeowner's money when the stone comes out as a clod.

What minimum slope does a French drain need?

One percent — about an inch of drop per eight feet — is the common target. Half a percent will work in some markets and soils if the pipe stays clean, but below that fines settle rather than move. Shoot at least four points along the run, not just the two ends, because a belly in the middle is invisible from either end.

How do I tell whether the stone or the pipe is the problem?

Run a hose at the upstream end and watch the outfall. If flow comes out reasonably fast and clear, the conveyance path works and the issue is upstream or capacity. If flow is slow, milky, or absent, dig a test hole mid-run and pull stone. Loose, rattling, washed stone means the pipe is the suspect. Stone that comes out bound together with clay means the aggregate is done.

Should I charge for diagnosing another contractor's failed drain?

Yes. You're spending an hour or more with a level, a hose and a shovel on a buried system with no as-built, and the value you deliver is the answer. Price it from your loaded hourly cost plus travel, and credit it against the repair if they book. Free diagnostics on failed installs attract homeowners shopping for a second opinion they have no intention of paying anyone for.

The short version

French drains fail five ways: silt migrating into stone that had no fabric separation, unwashed or wrong-gradation stone with no void space, insufficient or reversed fall, no cleanout to service the line, and a buried, crushed or frozen outfall. Check the outfall first, then run a hose test, then shoot fall at four or more points, then dig a test hole and squeeze the stone. Silted stone means rebuild; clean stone with a silted pipe means cleanouts and jetting at a fraction of the cost.

Book a walkthrough and we'll look at how your diagnostic photos, fall readings and repair options come together on one proposal page.