The segment nobody else owns
Everything on this site up to now has been about water your utility treats, samples and reports on. This page is about the last stretch, which is yours.
Two mechanisms operate there, and they are unrelated to each other. Materials in contact with water can leach into it. Plastic pipe buried in contaminated soil can be permeated from outside.
The first is governed by a law with a number in it. The second is governed by what is in the ground.
“Lead free” is a threshold, not an absence
The Safe Drinking Water Act sets the line. EPA states it as “a weighted average of 0.25 percent calculated across the wetted surfaces of pipes, pipe fittings, plumbing fittings, and fixtures and 0.2 percent for solder and flux.”
A compliant fixture is therefore permitted to contain lead. That is the whole reason first-draw water is treated differently from water that has been running.
The line moved twice, so the date of the work matters
Use of plumbing that is not lead free has been prohibited since June 1986. But the definition then was not the definition now.
| Period | Solder and flux | Pipes and fittings |
|---|---|---|
| Before June 1986 | No federal limit | No federal limit |
| June 1986 to 2011 | 0.2% | 8% |
| 2011 onward | 0.2% | 0.25% weighted average |
The 2011 Reduction of Lead in Drinking Water Act is what lowered the figure from 8 percent to 0.25.
So a house plumbed lawfully in 1995 could carry fittings at up to 8 percent lead, and a house plumbed in 1975 was under no federal content limit at all.
What the rule does not cover, in its own words
40 CFR 143.16 exempts a named list. It is worth reading because it shows where the law draws its potable line.
143.16(b) — “Toilets, bidets, urinals, fill valves, flushometer valves, tub fillers, shower valves, fire hydrants, service saddles, and water distribution main gate valves (provided that such valves are 2 inches in diameter or larger).”
143.16(c) — “Clothes washing machines, emergency drench showers, emergency face wash equipment, eyewash devices, fire suppression sprinklers, steam capable clothes dryers, and sump pumps.”
Read together, these say the rule follows drinking, not plumbing. A shower valve is out of scope because the water leaving it is not for ingestion.
Which means the rule reaches further than people expect, too
The same logic works in the other direction. 40 CFR 143.11 defines potable uses as including “drinking, cooking, food preparation, dishwashing, teeth brushing, or maintaining oral hygiene.”
And it defines a fixture to include point-of-use treatment devices, coffee makers, refrigerator ice and water dispensers, water heaters, water meters, water pumps and water tanks, unless those are not used for potable uses.
Your filter housing and your fridge dispenser are inside the rule. Your washing machine is outside it.
Eight certifiers, not one
Most packaging carries a certification mark, and most people read it as meaning NSF. EPA is explicit that it is one of several.
There are “eight American National Standards Institute (ANSI) accredited third-party certification bodies” certifying to the lead free requirement: ALS Global, CSA Group, IAPMO R&T, ICC-ES, Intertek, NSF, UL Solutions and WQA.
Each publishes its own product listing directory, so a product absent from one is not uncertified.
One date decides what an NSF/ANSI 61 mark means
This is the detail that catches people out. EPA states that “In October 2017, NSF/ANSI/CAN 61 was updated to require compliance to NSF/ANSI/CAN 372 for all products other than those exempted in the Safe Drinking Water Act.”
Before that date, only 372 or the now-retired Annex G indicated lead free compliance. EPA notes older stock can still be on shelves or online.
So the standard printed beside the mark is the thing to read, not the mark.
What the standard behind the mark does and does not cover
NSF/ANSI 61 is about leaching, and only leaching. The issue paper EPA publishes describes it as establishing minimum health effects requirements for contaminants “indirectly imparted (via leaching) to drinking water” and states it “does not establish performance, taste and odor, or microbial growth support requirements.”
The same paper notes the programme began under a cooperative agreement with EPA in 1985, so a great deal of installed material predates any of it.
Permeation: the pipe wall is not a barrier to everything
The second mechanism is the one almost nobody knows about, and the evidence for it is in a paper EPA publishes.
More than 100 incidents of drinking water contamination from permeation of subsurface mains and fittings have been reported in the United States.
Plastics were involved in 98 percent of them. The paper states flatly that “No reported incidents of permeation through metal-based pipe were identified.”
What gets through, and what does not
89 percent of reported incidents involved diesel and petroleum products. Volatile chlorinated solvents accounted for 5 percent.
The paper also records what did not get through: “Strongly polar pesticides (e.g. paraquat, malathion, and atrazine) and long-chained (high molecular weight) hydrocarbons were not permeation threats.”
The calculated frequencies, by material
Incidents per million service-connection-years, as calculated by Holsen and others and reproduced in the paper:
| Material | Frequency |
|---|---|
| Metal | 0 |
| Polyvinyl chloride | 2.2 |
| Polyethylene | 3.6 |
| Polybutylene | 16.5 |
These are derived from reported incidents, so they understate unreported ones. They describe a population, not your service line.
Why it concentrates at your end of the pipe
Permeation is worst where the surface-to-volume ratio is highest and the water moves least. The paper puts it at “small-diameter mains and service lines”.
It adds that the effect is made worse by the greater likelihood of spills “on a customer’s property and consequently, closer to the point of withdrawal or consumption.”
Which is to say: the risk is concentrated in the segment you own, next to the place fuel and solvents actually get spilled.
Two findings that change what you would do about it
The first is about detection. Contamination “was generally identified by the customer, and indicated by an unusual taste and odor” — but for benzene, vinyl chloride and dichloromethane, “the taste and odor threshold is well above the drinking water Maximum Contaminant Level.”
You would smell a problem. You would not reliably smell it before it mattered.
The second is about the remedy. In one Montana case from March 2000, benzene reached 527 micrograms per litre in a black polyethylene service line, which the paper describes as over 100 times the limit.
And flushing does not fix it. The pipe “will retain its status as a swollen, highly permeable medium.” Calculated penetration times fell from weeks for new pipe to minutes for pipe that had been contaminated and flushed.
Pre-1977 PVC is a separate, dated case
One material has a leaching history of its own. PVC mains made before 1977 contain elevated vinyl chloride monomer.
In a rural Kansas system with over 100 miles of it, 55 percent of 53 samples taken from four sites over six years exceeded the 2 microgram per litre limit, reaching as high as 14.
The paper states no violations from post-1977 PVC were cited in the literature it reviewed. That is absence of citation, not a demonstration of safety.
The decision framework
1. Date the work before guessing at the material
The federal content limit has been three different numbers. What was lawful to install depends entirely on when the installing happened, so the age of the house and the age of any repipe are the first two facts to establish.
2. Read the standard, not the mark
A certifier’s logo tells you who tested it. The identifier text beside it tells you what against. For lead free, that means 372, or 61 from October 2017 onward.
3. Treat the exemption list as scope, not as a warning
Nothing on the exempt list is exempt because it is dangerous. Each is exempt because it is not a potable-use fixture. The useful reading is that your kitchen tap and your outdoor spigot are not governed the same way.
4. Ask where the service line runs and what it runs through
This is the segment where lead and permeation both concentrate. According to EPA, where a home has a lead service line that pipe is typically the most significant source of lead in the water at that home, and the permeation paper puts its highest calculated frequencies on service connections rather than on mains.
5. Do not repipe on the strength of a category
Material class is a prior, not a measurement. Plastic is not a finding and copper is not a clearance. If the question is what is in the water, the answer comes from a test of your tap.
The decision path
| What you know | Next step | What it settles |
|---|---|---|
| House predates 1986 | Check the service line material in your utility’s inventory | Whether lead is a live question at all |
| Plumbed or repiped 1986–2011 | Treat fittings as permitted up to 8 percent lead | Why first-draw and flushed samples differ |
| Buying a fitting now | Read the identifier text beside the certifier’s mark | Whether it is certified lead free or only tested for leachate |
| Fuel station, spill or tank near the line | Ask the utility about the service line material | Whether a permeable material sits in suspect ground |
| Persistent solvent or fuel taste | Contact the utility rather than buying a filter | Whether the cause is inside the pipe wall |
| Nothing unusual, no old plumbing | Do nothing | A material class on its own is not a finding |
Choose to identify materials if
Your home predates 1986, your utility’s inventory lists your service line as lead, galvanized or unknown, or you are about to open a wall and can look.
Choose point-of-use filtration if
You have a measured figure to act on, and a claim exists for it. A certified lead claim is the well-populated case; there is no permeation claim to buy.
Choose to contact your utility if
The taste is of fuel or solvent, it does not clear on flushing, or you know the ground near your line has been contaminated. None of that is a purchase decision.
Mistakes that cost the most money
Reading “lead free” as lead-free. It is a content ceiling of 0.25 percent by weighted average. A compliant fitting can still contribute.
Assuming the mark is the answer. The certifier’s logo says who tested. The standard printed beside it says against what — and before October 2017, a bare NSF/ANSI 61 mark did not carry the lead free requirement.
Repiping on category alone. Plastic is not a verdict. Metal pipe has never been reported to permeate, but it leaches lead and copper, which plastic does not. Swapping one material for another trades one mechanism for the other.
Flushing a permeated line and calling it fixed. The paper is explicit that the pipe stays swollen and permeable after flushing.
Buying a whole-house filter for a service line problem. It sits downstream of the segment in question, and there is no certified claim for what permeates.
Three situations, and what changes
A 1960s house, original plumbing. The content limits did not exist when it was built, so the age is the finding. Start with your utility’s service line inventory, then a first-draw test.
A 2005 build, never touched. Lawful at the time at up to 8 percent lead in fittings. Not a defect, and not nothing. Running the tap before drinking costs nothing and changes the sample.
A plastic service line near a former fuel station. This is the one case on this page where material class genuinely matters. Ask the utility what the line is made of and what is known about the ground.
The objections worth answering
“My water is fine, I would know.” Not for this failure mode. The paper’s own point is that for benzene, vinyl chloride and dichloromethane the taste and odour threshold sits above the drinking water limit.
“This was a 2002 paper, it is out of date.” It is, and that is stated on every figure drawn from it. What it establishes is mechanism and direction, not today’s counts. The regulatory half of this page is current law.
“Plastic pipe is obviously worse, then.” Only against permeation. Against leaching the comparison reverses, and the pre-1977 PVC case shows plastic has its own history. Neither material is clean on both axes.
“Surely someone checks my plumbing.” Your utility’s obligations effectively end at the property. The materials law binds manufacturers and installers, not your existing pipe. Nobody samples your particular tap unless it is one of the sites chosen for lead and copper monitoring.
Who this is for
Households on city water who want to understand the last segment before the glass, owners of pre-1986 homes, anyone reading a certification mark on a tap or fitting before buying it, and anyone whose service line is plastic and whose ground has a history.
It is not for anyone looking for a repiping recommendation. Nothing here supports one.
What to settle first
→ Lead in drinking water: the action level and your service line → Copper and blue-green staining → How to read your water quality report → Testing city water: what your utility already measured → Why your water tastes, smells or looks wrong → Main breaks and new pipe: what a repair does to your water → Water age: why the same supply differs from street to street → How to check a certification claim yourself → What sitting water does to the pipe it is sitting in