What Wet Bulb Tells You That Dry Bulb Cannot

Why this matters

The temperature split across a coil is the most-used number in the trade and one of the least reliable, because it answers a question nobody asked. A dry bulb thermometer sees only the heat that changes temperature. Every bit of work a coil does pulling water out of the air is invisible to it. That means a healthy system on a humid day reads like a sick one, a tech adds refrigerant it did not need, and the second visit is longer than the first. One extra thermometer, wet, converts a guess into a capacity measurement.

Before you open a rooftop unit

Roof access first: general industry fall protection is triggered at 4 feet under 29 CFR 1910.28, and construction work is triggered at 6 feet under 29 CFR 1926.501. A service shop can fall under either depending on the job, so know which one you are working under before you walk near an unprotected edge or a skylight.

The blower is the hazard inside the cabinet, and it is a mechanical and stored-energy hazard, so it goes to 29 CFR 1910.147: open the unit's disconnect, apply your own lock and tag, and confirm the blower has coasted to a full stop before any hand goes past the panel. A belt-drive blower will restart on a timer or a thermostat call if the disconnect is not locked. Take air readings through test ports drilled in the duct from outside the cabinet, never with an arm in the airstream.

Where a reading has to be taken with the unit energized, exposed live work is permitted under the troubleshooting exception at 29 CFR 1910.333(a)(1), which covers testing where de-energizing introduces additional hazards or is infeasible. Use a meter and leads rated for the circuit category, CAT III for branch-circuit and distribution work. An under-rated meter across a line-voltage fault is the failure that kills.

The call, and the number the first tech wrote down

A small commercial suite, single-zone rooftop unit, nominal 3 tons. Complaint: not holding setpoint on hot afternoons. Space at 78 F at four in the afternoon against a 72 F setpoint.

The first tech measured return and supply dry bulb, got 78 F and 58 F, wrote "20 degree split, low on charge," added refrigerant, and left. The space still did not hold. The second visit found the charge above the manufacturer's target and had to recover the surplus, which is two visits and a refrigerant recovery to correct a diagnosis that the first reading never supported.

Nothing about the measurement was sloppy. Both thermometers were accurate. The reasoning on top of them was wrong.

Why the split rule failed him

The familiar 18 to 22 degree split is not a property of a coil. It is what a coil produces at a particular entering condition, and the entering condition it assumes is moderate humidity. Change the humidity of the air entering the coil and the split changes without anything changing inside the equipment.

The physics is simple once it is said out loud. A coil removes total heat. On humid air, part of that total goes into condensing water vapor out of the airstream, and condensing water changes no temperature at all. So on a humid day a fully healthy coil delivers a smaller dry bulb split, because a share of its capacity is doing invisible work. On dry air the same coil delivers a bigger split while doing less total work.

The split rule, used alone, therefore fails in the most dangerous direction: it condemns healthy equipment in humid weather and passes weak equipment in dry weather.

The second visit: two thermometers instead of one

The second tech took four readings instead of two: dry bulb and wet bulb, entering and leaving, plus airflow.

• Entering: 78 F dry bulb, 66 F wet bulb.
• Leaving: 58 F dry bulb, 56 F wet bulb.
• Airflow, measured rather than assumed: 1,200 cubic feet per minute.

Wet bulb is the temperature a thermometer settles at when its bulb is wet and air is moving over it, and evaporation cools it until the heat arriving by conduction balances the heat leaving by evaporation. That balance point is set by how much water the air can still take, which is why wet bulb tracks the air's total heat content and dry bulb does not.

What the enthalpy said

Read enthalpy off a psychrometric chart or an app for each condition. At 78 F dry bulb and 66 F wet bulb the air holds about 30.7 Btu per pound. At 58 F and 56 F it holds about 23.8. The difference is 6.9 Btu per pound.

Total capacity is 4.5 x airflow x the enthalpy difference:

4.5 x 1,200 x 6.9 = about 37,300 Btu/hr.

Sensible capacity is 1.08 x airflow x the dry bulb difference:

1.08 x 1,200 x 20 = about 25,900 Btu/hr.

So the sensible heat ratio is 25,900 divided by 37,300, about 0.70. Latent capacity is the remainder, about 11,400 Btu/hr.

Compare the total against the manufacturer's expanded performance table at the actual entering wet bulb and the actual outdoor temperature, not against the nominal tonnage on the nameplate, which is a rating-point figure at a standard entering condition. Read that way, the unit was performing where the table said it should. It was not weak. Thirty percent of everything it was doing was invisible to the first tech's thermometers.

Note the constant: 4.5 assumes standard air at 0.075 pounds per cubic foot. Correct it for elevation and for high temperatures before you trust the third digit, because at 6,000 feet that constant is roughly a fifth optimistic.

The same unit on a dry day, and the inversion that proves the point

Run the identical unit at the identical airflow in dry weather to see how badly split alone misleads.

• Entering: 78 F dry bulb, 60 F wet bulb, about 26.4 Btu per pound.
• Leaving: 55 F dry bulb. The entering air's dew point here is around 48 F, which is below the coil's leaving temperature, so no water condenses at all and the air leaves with the moisture it arrived with, about 20.8 Btu per pound.

Total: 4.5 x 1,200 x 5.6 = about 30,200 Btu/hr. Sensible: 1.08 x 1,200 x 23 = about 29,800 Btu/hr. Sensible heat ratio: essentially 1.0, the small residual being the rounded constants rather than any real latent work.

Now put the two side by side, same equipment, same airflow.

The bigger split came with about 81 percent of the total capacity. That inversion is the entire argument for the second thermometer: the number that looked better described the day the equipment did less work.

Where the capacity was going

Full capacity and a space that will not hold setpoint means the load grew, so the second tech went looking for load rather than for a defect.

The outdoor air damper was stuck at roughly 40 percent open against a design position near 15 percent. At 1,200 cubic feet per minute total, that is about 480 cubic feet per minute of outdoor air where 180 was intended, an extra 300.

Outdoor conditions that afternoon: 92 F dry bulb, 76 F wet bulb, about 39.4 Btu per pound. Space air, as measured, about 30.7. The extra outdoor air therefore imposes:

• Total: 4.5 x 300 x 8.7 = about 11,700 Btu/hr.
• Of which sensible: 1.08 x 300 x 14 = about 4,500 Btu/hr.
• And latent, the remainder, roughly 6,800 to 7,200 Btu/hr depending on which constants you use.

About 11,700 Btu/hr of added load against a unit measuring 37,300 total. Something close to 30 percent of the machine's output was being spent conditioning air the building did not ask for, and the majority of that added load was latent, which is why the space felt worse than the thermostat reading suggested.

The repair was a damper actuator and a linkage, and it took a fraction of the hours the two refrigerant visits had already consumed. What made it findable was measuring capacity rather than inferring it.

Getting a wet bulb you can trust

A wet bulb reading is only as good as the wick, and a bad wick fails in the direction that flatters the equipment.

Wet the wick with distilled water, not tap. Mineral deposits from tap water build a crust that slows evaporation, which makes the reading run high, which makes the entering enthalpy read high, which makes you calculate more capacity than the machine is making.

Get real air movement over it. Either sling it or use a powered psychrometer. A still wick reads high for the same reason a dirty one does.

Replace a wick that has stiffened or discolored. This is a consumable, and a shop that treats it as permanent has an instrument that reads optimistically for years without anyone noticing.

Let it stabilize. Wet bulb takes longer to settle than dry bulb. Watch for the reading to stop falling rather than taking the first number that appears.

Measure airflow rather than assuming it. Every capacity number above is proportional to airflow, so a 400-cubic-feet-per-minute-per-ton assumption on a system running 300 puts a 25 percent error straight into your total, and it will not be obvious because the enthalpy numbers will still look sensible.

Cross-check yourself before you write a conclusion: sensible capacity must be less than or equal to total capacity. If your arithmetic gives a sensible heat ratio above 1.0 by more than a rounding margin, one of your four temperatures or your airflow is wrong, and the wick is where to look first.

References

• ASHRAE Handbook, Fundamentals, psychrometrics chapter, for enthalpy, wet bulb, and the sensible and total capacity relationships
• Manufacturer expanded performance tables, indexed on entering wet bulb, entering dry bulb, and outdoor temperature, as the governing comparator for a specific unit
• 29 CFR 1910.147, OSHA general industry, for lockout of the blower's mechanical and stored energy before opening a cabinet
• 29 CFR 1910.28 (general industry) and 29 CFR 1926.501 (construction), OSHA fall protection triggers for roof work
• See related: What a Latent Load Does to Equipment Selection; How Altitude Changes Combustion; Psychrometrics Fundamentals