Why Warm, Stagnant Water Is a Biological Problem

Why this matters

Every system you service holds water somewhere it is not moving. A branch past the last recirculation tee, a riser in a warm shaft, a tower basin on a mild weekend, a spa that was filled for a showing and then left. That water is not sitting still chemically. It is growing a film on the inside of the pipe, and the film is where the organisms that cause a reportable illness actually live. The reason this belongs in a fundamentals card rather than a specialty one is that the conditions are set at each POINT in a system, not by the label on the pipe, so the water heater setpoint you check on every call can be perfect while two branches in the same building sit in the growth window for days.

Before you flush, sample or reset anything

Three actions in this article create hazards that land on someone other than you.

• Flushing a line that has been standing throws an aerosol. Remove the aerator, open the valve slowly and run at low flow into a container or a drain rather than a wide-open spray, and where an aerosol cannot be avoided use respiratory protection selected and fit-tested under a written program per 29 CFR 1910.134. A glove does nothing for an inhalation route.
• Raising stored water temperature to control growth creates a scald hazard at every fixture downstream. Water near 140 F can produce a serious burn on adult skin in a few seconds, so raise storage only where an approved thermostatic mixing valve controls the delivered temperature (the common listings are ASSE 1017 for a master valve and ASSE 1070 at the point of use), and verify delivered temperature at the furthest fixture before you leave.
• Taking a temperature at a fixture on a hot system means putting a probe into water that may be at storage temperature. Run into a vessel and read the vessel, not the stream over your hand.

What is actually growing, and where it lives

The organisms that matter in building water are not floating freely in the middle of the pipe. Within days of a surface being wetted, dissolved organics adsorb onto it and bacteria attach, then secrete a polysaccharide film that holds them there. That biofilm is a habitat: it traps sediment, it shelters protozoa, and several waterborne pathogens survive and multiply INSIDE those protozoa, which is why they tolerate conditions that would kill them in open water.

Two consequences follow, and they explain most of the confusing results in this work.

First, the biofilm is the reservoir and the bulk water is the sample. A negative bulk-water result tells you what was shedding into the stream at the moment you drew it, not what is living on the wall. Second, a disinfectant residual that reliably kills organisms suspended in water does not necessarily penetrate a mature film at the same concentration, because the film consumes the oxidant at its outer layer. That is why chlorinating to a residual and then finding the problem back in six weeks is a normal outcome rather than a sign the work was done wrong.

The three conditions, stated at a point

Growth needs a temperature window, residence time, and a surface with something to eat. State each one at the point you are standing, not for the system as a whole.

Temperature. The commonly cited amplification range for the waterborne organisms of concern in building systems is roughly 77 F to 113 F, with survival extending well outside it in both directions. That range is a property of the water at that spot at that moment. A shaft that holds a cold riser at 80 F has put "cold" water in the window without anything being heated.

Residence time. Turnover, not volume, is the variable. A gallon that sits for two days is a worse problem than fifty gallons that turn over hourly. Residence time is computed per branch: the water the branch holds, divided by the volume drawn through it per day.

Surface and nutrient. Sediment, scale, rubber and some elastomers, and corrosion products all give the film something to hold and something to consume. A branch with a debris bed at its low point has a head start on an identical clean branch.

One gate, run against two branches in the same building

The gate is this: at this point, is the water inside the amplification range, and is the turnover longer than a day? Both true means you have an amplification site regardless of what the equipment is called. Run it against the two cases below, which sit in one four-story building and resolve in opposite directions from what the labels suggest.

Case one: the hot system that is cold enough to be a problem

Storage is 140 F. The recirculation return reads 128 F at the mechanical room, so the loop itself is above the window on both ends and is not an amplification site. Past the last recirculation tee, a branch runs 42 feet of 3/4 inch copper to one lavatory in a rarely used office. Three-quarter inch type L copper holds about 0.025 gallons per foot, so that branch holds about 1.05 gallons. The building's own fixture log shows that lavatory used twice a day at roughly 0.3 gallons a use, so about 0.6 gallons a day passes through it. Turnover is 1.05 divided by 0.6, about 1.75 days.

Measured at the fixture before any run, that branch reads 104 F. Inside the window, turnover approaching two days, and it is fed by a system whose setpoint is exactly right. The tech who checked the tank passed this building.

Name the other end of the range before you accept the mechanism: if someone drops storage to 120 F because of a scald complaint, the return leg lands near 110 F and the TANK moves into the upper part of the window, so the fix for the complaint creates a second site. That is the direction check that tells you the temperature rule is real and not asserted.

Case two: the cold system that is warm enough to be a problem

Same building. Incoming water at the meter reads 62 F. The cold riser shares a shaft with uninsulated heating mains, and at the top floor the riser reads 79 F flowing and 84 F standing. It gained 17 F in the shaft, from 62 F to 79 F, without anyone heating it on purpose.

Standing at 84 F, this is inside the same window as the 104 F branch, and its turnover is set by whatever the top-floor tenants draw. The gate resolves the same way for both, and the two fixes have nothing in common: the hot branch needs its dead leg shortened, recirculation extended, or a logged flush; the cold riser needs insulation and separation from the heat source in the shaft. A program aimed only at the water heater fixes neither.

Why a clean bulk-water sample proves so little

A single negative sample from a fixture answers one narrow question: was this organism shedding into the stream, above the method's detection limit, at that outlet, at that moment. It does not tell you the wall is clean, and it does not survive a change in flow, because a disturbance releases film into the water and the next sample from the same tap can read differently. Treat sampling as a trend instrument tied to fixed locations and a fixed procedure, not as a pass/fail on the building.

The corollary is the useful part: temperature and turnover are cheap, repeatable and physical, and they are the conditions you actually control. Measure those on every visit, and use sampling to check whether your control of them is working.

Worked example: which branch gets the money

Same building, four low-use branches found on the walk. Their held volumes and logged daily draws are 1.05 gallons against 0.6 (turnover 1.75 days), 0.4 against 2.0 (0.2 days), 2.2 against 0.5 (4.4 days), and 0.9 against 1.8 (0.5 days). Two of the four clear the one-day gate and two do not.

The 4.4-day branch is the worst by a wide margin, and it is 2.2 gallons of standing water, so a flush has to move at least that volume plus the run it feeds before the water at the outlet is fresh. That is the argument for cutting the branch back rather than flushing it forever: at a weekly flush the branch stands for the other six days no matter how thorough the flush is, so flushing manages the outlet and does nothing for the film in the middle of the run.

The 1.75-day branch is a candidate for extending recirculation, because it is on the hot side and its temperature deficit is what makes its residence time matter. The two branches under a day get logged, not worked, and re-checked if occupancy changes. Nothing about that ranking came from a lab result; it came from two measurements and a fixture log.

The failure mode is treating all four the same. A shop that quotes a flush program covering every low-use outlet in the building spends recurring labor forever, reports full compliance, and still has 2.2 gallons standing six days a week in the branch that most deserved a pipe change.

How to verify you read a point correctly

• Take the temperature after a defined stand, not after a run. Note the stand time. A branch flushed by the person ahead of you reads like the main.
• Re-read the same point twice on different days at the same clock time. A shaft riser swings with the heating load, so one reading can miss the warm end of its daily range entirely.
• Get the turnover from the building's use, not your guess. A fixture counter, a badge log, or the tenant telling you which rooms are empty beats an assumption about a floor plan.
• Confirm the branch length and diameter, then compute the held volume. People consistently underestimate dead legs because the run is above a ceiling and only the last three feet are visible.
• Write down the point. A temperature with no location is not data. The same tap, described the same way, is what makes the second reading mean anything.

References

• CDC guidance on controlling Legionella growth in building water systems
• ANSI/ASHRAE Standard 188, Legionellosis: Risk Management for Building Water Systems (pin the edition your client or jurisdiction references; 188-2018 is the widely cited one), and ASHRAE Guideline 12 for the technical background
• 29 CFR 1910.134, respiratory protection program requirements including fit testing
• ASSE listing standards for thermostatic mixing valves (1017 master, 1070 point of use)
• See related: What a Water Management Plan Is For; How to Reduce Biological Risk in the Systems You Service