Reading Approach and Range: How to Tell If Your Cooling Tower Is Actually Underperforming

Most cooling towers that get replaced in Malaysia did not need replacing. They needed the fill cleaned, the fan pitch corrected, or the air inlet unblocked. The reason the diagnosis goes wrong so often is that the two numbers everyone quotes — approach and range — are routinely read the wrong way round.

Range gets the attention because it is the easy number. Both thermometers are already there. Approach is the one that actually tells you whether the tower is doing its job, and it is the one most plants either never calculate or calculate against a wet-bulb figure they pulled off a weather app.

This is a practical guide to reading both correctly, taking the measurements in a way that survives scrutiny, and converting the result into ringgit.

The two definitions, and why one of them is a trap

Range is the temperature drop across the tower: hot water entering minus cold water leaving.

Approach is how close the tower gets to the theoretical limit: cold water leaving minus the wet-bulb temperature of the air entering the tower.

The wet bulb is the floor. An open evaporative tower cannot produce water colder than the ambient wet-bulb temperature, no matter how large it is or how hard the fan runs. Everything a cooling tower does is a fight to get close to that line, and approach measures how much of that fight it is winning.

Here is where the trap sits. Range is not a measure of tower performance at all. Range is fixed by the heat load and the water flow rate:

Heat rejected (kW) = flow (L/s) × 4.18 × range (°C)

Rearrange it and range is simply load divided by flow. If your range drops from 5°C to 3.5°C, the overwhelmingly likely explanation is that production slowed down or somebody opened a bypass valve — not that the tower degraded. A tower with a fouled fill and a slipping fan belt will still show its full design range if the load and flow are unchanged. It will just do it at a higher cold water temperature.

Range tells you what the system is asking the tower to do. Approach tells you how well the tower is answering.

Neither number means anything without wet bulb and flow

This is the part most sites skip, and it is why so many cooling tower assessments are worthless.

You cannot assess approach without a trustworthy wet-bulb reading taken at the tower’s own air inlet. You cannot assess range without knowing the water flow rate. Most Malaysian plants I would expect to walk into have neither: no flow meter on the condenser water loop, and no psychrometer anywhere on site.

If you are going to invest in one thing before doing this properly, make it a clamp-on ultrasonic flow meter and a calibrated aspirated psychrometer, or hire someone who arrives with both.

Measuring wet bulb properly

Use a sling psychrometer or, better, an aspirated (motorised) psychrometer. A cheap handheld hygrometer with a slow-responding humidity chip will read high in the saturated air around a tower and quietly ruin your calculation.

Take the reading at the air inlet face, roughly 1.5 metres out from the louvres, and sample at several points across both the height and the length of each inlet. Average them. On a multi-cell tower, do every cell. Readings should be taken repeatedly across the test period rather than once.

Do not use airport or weather-station data. It can be twenty kilometres away and it cannot possibly capture what is happening at your louvres.

That last point matters more than it sounds, because the inlet wet bulb is itself a diagnostic. Measure the general ambient wet bulb well clear of the tower and upwind of it, then compare with the inlet reading. If the inlet is running more than about 1°C hotter, warm saturated discharge air is being pulled back into the intake. That is recirculation — a siting, spacing or discharge-velocity problem — and no amount of fill cleaning will fix it. In crowded industrial estates where towers sit shoulder to shoulder on a rooftop, one unit ingesting its neighbour’s plume is a common and badly under-diagnosed fault.

Measuring water temperature properly

Take the hot water temperature in the inlet riser or header, not in the hot water distribution basin, where evaporation has already begun cooling it.

Take the cold water temperature in the cold water basin near the outlet, or in the pump suction line. Avoid measuring downstream of the pump, which adds a small amount of heat, and avoid long uninsulated runs.

Then calibrate. Put both sensors in the same water bath before you start and check they agree. On a design approach of 4°C, a 0.3°C offset between two instruments is a 7.5% error in your headline number — enough to condemn a healthy tower or clear a sick one.

Making the comparison valid

Approach is not a constant. It varies with wet bulb, with load, and with flow. A tower running at 60% load will show a noticeably tighter approach than the same tower at 100% load, which means comparing today’s part-load reading against last year’s full-load reading will tell you the tower has improved when it has not.

CTI ATC-105, the field acceptance test code for water cooling towers, requires test conditions to fall within defined bands of the design point before results are considered valid: wet bulb within ±8.5°C, dry bulb within ±14°C, range within ±20%, circulating water flow within ±10%, and fan driver power within ±15%. You do not need to run a formal code test to borrow the discipline. If your operating point is outside those bands, do not draw conclusions from it.

Run at steady state for at least an hour with a stable load. Avoid testing in rain or gusty wind.

Turning the reading into a verdict

The professional answer is thermal capability, expressed as a percentage. Rather than a pass or fail, ATC-105 calculates what fraction of the predicted thermal capability the tower is actually delivering, using measured flow, hot and cold water temperatures, wet bulb and fan power against the manufacturer’s performance curves.

CTI’s own certification reporting treats 95% capability as the meaningful threshold. As a working rule for an in-service tower:

• 95% or better — healthy, keep trending
• 85–95% — degrading, investigate at the next shutdown
• Below 85% — something specific is wrong, and it is costing you money now
• Below 70% — expect a substantial physical fault such as collapsed fill, a reversed fan, or major air bypass

If you cannot justify a formal test, the practical substitute is trending. Log the approach, the inlet wet bulb, the load and the flow once a shift. Plot approach against load, grouping readings by similar wet bulb. What you are looking for is drift: the same load and the same wet bulb producing a wider approach this quarter than last. A tower that has quietly given up 2°C of approach over eighteen months is invisible on any single day’s reading and glaringly obvious on a twelve-month plot.

What a widening approach is telling you

Once you are confident the approach has genuinely widened at comparable conditions, the fault is almost always in one of three places: airflow, water distribution, or the fill.

Work down that list before you accept any recommendation to replace the tower. A tower is a heat exchanger with an air pump attached. The failure modes are few and mostly visible.

What it costs, in ringgit

This is the number that gets budget approved.

Chiller power responds directly to condenser water temperature. The widely used industry rule of thumb is that every 1°F rise in entering condenser water temperature above design costs 1% to 2% in chiller efficiency, which works out to roughly 2% to 3.5% per °C. Treat that as an estimate for scoping only — for a real business case, pull the actual performance curves for your specific chiller, because the sensitivity varies with machine type and load.

Take a 500 RT water-cooled plant designed for a 4°C approach at a 28°C design wet bulb, so 32°C cold water. Suppose fouling has widened the approach to 6.5°C, putting 34.5°C into the condenser instead of 32°C.

• Extra condenser water temperature: 2.5°C
• Chiller power penalty at roughly 2.5% per °C: about 6%
• Chiller drawing 0.65 kW/RT at 500 RT: 325 kW, so about 20 kW extra
• At 4,000 operating hours a year: about 80,000 kWh

Under the RP4 tariff structure that took effect on 1 July 2025, the Peninsular base tariff rose from 39.95 to 45.4 sen/kWh. Using roughly 45 sen/kWh as an all-in figure, that is in the region of RM36,000 a year — from one tower, from a fault that a fill clean might fully reverse.

Then add the part most people forget. That 20 kW does not just cost energy. Under the restructured tariff, the old Maximum Demand charge was replaced by separate Capacity and Network charges, so persistently elevated demand carries its own cost on top of the kWh. Check your own bill rather than relying on the average figure above, since energy, capacity and network components are itemised separately and vary by voltage class.

There is a related opportunity in the same tariff change: for non-domestic medium and high voltage customers, Maximum Demand occurring during off-peak periods is not charged. If your tower and chiller scheduling has never been revisited against the current peak window, it is worth a look — but fix the tower first. Shifting load does nothing for a plant that is rejecting heat inefficiently around the clock.

Why this bites harder in Malaysia

Cooling tower degradation is a universal problem, but our climate changes both how fast it develops and how quickly it hurts.

There is no seasonal relief. In a temperate country, a fouled tower hides through the cooler months because the wet bulb drops and the tower still makes its target cold water temperature. Malaysian wet bulb sits high and stable year-round, with tropical design values commonly in the 25 to 28°C region. There is no cool season to mask the fault, and no cool season to give the chiller a break.

Design approaches here are already tight. When the design wet bulb is 28°C and you still need 32°C condenser water, you are asking for a 4°C approach, which is an aggressive selection requiring a physically large tower or additional cells. Starting from 4°C, losing 2°C means the approach has widened by half. The same 2°C loss on a temperate tower designed for a 7°C approach is a far smaller proportional hit.

Biofouling never slows down. Warm circulating water all year, with no cold season to suppress microbial growth, is close to ideal for biofilm. Film fill with narrow flute spacing is particularly vulnerable where the make-up water carries suspended solids.

The air is loaded. Haze episodes, plus airborne particulates and organics from nearby palm oil, rubber, cement and food processing operations, foul both the fill and the air inlets faster than a clean urban site would.

Coastal sites corrode. Towers at Port Klang, Pasir Gudang, Prai and similar locations face salt-laden air that attacks galvanised steel structure, fasteners and drift eliminators. Deteriorated drift eliminators cost you water and treatment chemicals, and in a tight cluster they worsen the plume that your neighbour’s tower then ingests.

Rooftop clustering is normal here. Industrial estate plots are tight and towers get squeezed together. Recirculation and cross-interference between adjacent towers is common enough that it should be checked before anything else, precisely because it is the one fault that no amount of maintenance will resolve.

A reasonable starting point

If you do nothing else after reading this, do these four things:

1. Calculate your current approach using a wet-bulb reading taken at your own tower’s air inlet, with a real psychrometer.
2. Measure the ambient wet bulb well clear of the tower and compare. A gap of more than about 1°C means you have a recirculation problem to solve first.
3. Find out what your actual condenser water flow rate is. If nobody knows, that is your first finding.
4. Open the top hatch with the pump running and look at the water distribution across the fill.

Steps 2 and 4 cost nothing and between them account for a large share of the underperformance we see in the field.