The Difference Between an Open Loop and a Closed Loop Chemically

Why this matters

The same treatment product, at the same dose, does two completely different jobs depending on which kind of system it goes into. In one it is a consumable that has to be fed continuously against a stream that is carrying it out of the building. In the other it is a charge that stays put for years and is lost only by dilution. Shops that treat both the same way either overspend chemical on a closed loop or quietly starve an open one, and in both cases the water numbers look fine on the visit and drift between visits.

The distinction that decides it is not whether the system has a lid. It is whether dissolved solids have an exit.

Before you touch a blowdown valve or a feeder

Opening a blowdown line sends hot, treated water somewhere it was not going a minute ago. Confirm the receiving drain can take the flow and the temperature before you crack the valve; a discharge above the temperature rating of plastic drainage piping deforms it, and a drain sized for a trickle backs up onto a mechanical room floor that also holds energized equipment. Route it, do not aim it, and stand clear of the discharge point while you open the valve.

Anything that atomizes water in an open evaporative system creates an inhalable aerosol, and that route needs respiratory protection issued under a written program with fit testing (29 CFR 1910.134) or the equipment shut down and the water management program consulted before it is opened (ASHRAE Standard 188). Gloves are irrelevant to that route.

Chemical feeders hold concentrated product. Get the glove class, eye and face protection and any respiratory control from Section 8 of that product's safety data sheet, which your employer has to make available (29 CFR 1910.1200), and never charge an acid product into a feeder, sump or drain that has held a hypochlorite product, because that combination releases chlorine gas into the room.

Isolating a pump or a pressure vessel to work on a feeder means isolating stored energy and locking it out under 29 CFR 1910.147; the electrical disconnect for that pump is a separate matter under 29 CFR 1910.333(b)(2), since 1910.147 excludes electric utilization installations at 1910.147(a)(1)(ii)(C).

Any feeder or makeup connection that can push treated water back toward a potable line is a cross-connection, and that is a public-health event rather than a plumbing detail. The backflow protection the local plumbing code requires for that chemical class goes in first.

The definition that matters is the exit, not the lid

An open evaporative system loses water as vapor and keeps everything that was dissolved in it. Cooling towers, evaporative condensers, humidifiers, boilers making steam. Dissolved solids can only leave in liquid water, so unless liquid is deliberately discharged, the concentration rises without limit.

A closed system loses water as liquid or not at all. Hydronic heating loops, chilled water loops, glycol loops, condenser loops on a closed circuit. When water leaves through a leak, a weep or a purge, it takes its dissolved load with it, and the replacement arrives with a fresh load at makeup concentration.

That single difference generates every other difference on the list.

The mass balance, one line each

Open: makeup = evaporation + blowdown + drift. The concentration ratio, the number of times the water has been concentrated relative to the makeup, is makeup divided by the sum of blowdown and drift. Since evaporation is the term you cannot control, blowdown is the only lever, and cutting it raises concentration on everything.

Closed: makeup = losses. There is no evaporation term, so there is no concentration mechanism. What a closed loop accumulates is not concentration but delivered mass - the total hardness and the total oxygen that entered over the life of the system, which is the makeup concentration multiplied by the total makeup volume. A companion article works that arithmetic out; the point here is that it is a different quantity from the number on a report.

Say the direction plainly, because this is the one people invert. In an open system, chloride and conductivity climb over time. In a closed system, they stay near the makeup value forever, even while the loop is being damaged, because a leak exchanges water rather than concentrating it. A tech looking for rising chloride in a closed loop as evidence of a problem will never find it, and will conclude the loop is healthy.

Cycles of concentration, and its one derivation condition

Concentration ratio is set by choosing how much liquid to discharge. At a ratio of 5, blowdown is one fifth of the makeup, which means 20 percent of the makeup leaves as liquid and 80 percent leaves as vapor.

Evaporation is estimated with a rule worth knowing and worth conditioning: roughly 1 percent of the circulating flow evaporates per 10 F of range, where range is the temperature drop across the tower. The derivation is simple - cooling one pound of water by 10 F removes about 10 Btu, the latent heat of vaporization is on the order of 1,000 Btu per pound, and 10 divided by 1,000 is 1 percent.

The condition is in that derivation: it assumes essentially all of the heat rejection is latent. In hot, humid weather that is close to true and the rule is good. In cold, dry weather a meaningful share of the rejection is sensible - the air simply gets warmer - so actual evaporation is lower and the rule overstates it. Which end you are on decides whether a makeup figure calculated this way is a reasonable estimate or an alibi.

Raising cycles saves water and chemical and moves every dissolved species closer to its solubility limit. The limit is whichever species saturates first, commonly calcium and alkalinity together, and on some supplies silica well before either. That ceiling is water-specific and program-specific; get it from the water analysis and the treatment supplier rather than from a general number.

Worked example: one water, two systems, same building

Take a supply with hardness 224 mg/L as CaCO3, alkalinity 180, chloride 45, conductivity 620 uS/cm.

Into the tower. Circulating flow 500 gallons per minute at a 10 F range, held at 5 concentrations.

• Evaporation: 1 percent of 500 = 5 gallons per minute (with the latent-heat condition above attached).
• Blowdown, to hold 5 concentrations: evaporation divided by (5 minus 1) = 5 / 4 = 1.25 gallons per minute.
• Makeup: 5 + 1.25 = 6.25 gallons per minute, and the check closes, because 6.25 / 1.25 = 5.

Now the chemistry. Chloride is conservative, so it lands at 5 x 45 = 225 mg/L. Conductivity, near enough conservative at this level, lands around 5 x 620 = 3,100 uS/cm. Hardness and alkalinity have a nominal 5 x figure of 1,120 and 900, and they will not reach it: carbon dioxide is stripped in the tower, calcium carbonate saturates, and the excess leaves the water as deposit rather than staying in solution. The measured shortfall against the nominal is the deposit, which is exactly the divergence a companion article uses as a diagnostic. Inhibitor here is being discharged at 1.25 gallons per minute continuously, so it is a feed, permanently, and if the feed stops the reserve is gone within a few turnovers.

Into the closed loop. Same water, a 500-gallon hydronic loop, taking 100 gallons of makeup a year.

• That is 20 percent of the system volume per year.
• Chloride in the loop: still about 45 mg/L. Conductivity: still about 620 uS/cm, give or take what corrosion products and inhibitor contribute. Nothing concentrated, because nothing evaporated.
• Inhibitor, if it is not replenished, is diluted by that makeup. Arriving as a slow continuous drip into a well-mixed loop, the fraction remaining after one year is the exponential form: e raised to minus 0.20, which is about 82 percent. Arriving as one drain-and-refill of the same volume it would be a straight 80 percent. The two models differ by about 2 percentage points here, so the method distinguishes a loop losing a fifth of itself a year from one losing half, not 82 from 80.

What that comparison teaches. The tower's chemistry is a flow problem and its treatment is a rate. The loop's chemistry is a batch problem and its treatment is an inventory. A tower left unfed for a month is untreated. A loop left unfed for a month at 20 percent annual makeup has lost under 2 percent of its reserve and is fine, which is why annual service is defensible on one and negligent on the other.

What flips it: a closed loop that is not actually closed. At 100 gallons a week rather than a year, the same loop passes about ten system volumes annually, the inhibitor reserve is effectively gone between visits, and it now needs the treatment cadence of an open system while still reading like a closed one on chloride and conductivity. That is the case that gets missed, and the only measurement that catches it is makeup volume.

The failure mode: a shop puts a closed loop on a tower's service schedule, tests conductivity every month, finds it steady, and signs off. Steady conductivity in a closed loop is not evidence of health. It is the expected reading whether the loop is tight or hemorrhaging.

The three things that change over a closed loop's life

• Oxygen is consumed and not replaced, so a genuinely tight loop goes anoxic within weeks and stays there. That is the healthy end state and it is why dissolved oxygen near zero is the target rather than a concern.
• Inhibitor falls by dilution and by consumption, and only one of those is proportional to makeup. Test the reserve rather than assuming the dose from the last charge is still present.
• Microbiology establishes in the low-velocity legs. A closed loop is not sterile and it is not immune. Stagnant runs, seasonally idle branches and dead legs are where sulfate-reducing bacteria set up, and no amount of concentration arithmetic will show it.

How to verify which one you are actually dealing with

Find the makeup connection and follow it. If there is an automatic fill valve, the system can take water without anyone knowing, and its label does not tell you whether it is closed in practice. Then take a conductivity reading on the makeup and on the system water on the same visit with the same meter. A ratio near 1.0 says nothing is concentrating and you are looking at a closed system. A ratio meaningfully above 1.0 says water is leaving as vapor somewhere, and on a system that is supposed to be closed, that is a finding rather than a reading.

References

• 29 CFR 1910.134 for respiratory protection where evaporative equipment generates aerosol, and ASHRAE Standard 188 for whether that equipment is opened at all
• 29 CFR 1910.1200 (Hazard Communication) and Section 8 of each product safety data sheet for glove class and eye and face protection at a chemical feeder
• 29 CFR 1910.147 for stored energy isolation on pumps and pressure vessels, and 29 CFR 1910.333(b)(2) for the electrical disconnect, which 1910.147 excludes at 1910.147(a)(1)(ii)(C)
• ASHRAE handbook guidance and treatment supplier documentation for the concentration limit applicable to a specific supply, which is water-specific and not a general figure
• See related: What Conductivity and Total Dissolved Solids Actually Measure; Why Makeup Water Is the Variable That Decides Everything; How to Find Out How Much Makeup Water a System Is Really Taking