How to Take a Dew Point and Surface Reading Before Coating

Why this matters

The coating rule everyone quotes, keep the surface at least 5 F above the dew point, is only as good as the three numbers behind it. Most bad readings are not instrument failures. They are a humidity reading taken in the truck cab, a surface reading taken on the sunny side of the work, or one reading at 8 a.m. standing in for a whole day. Each of those passes a surface that is about to collect a film of water you cannot see, and the coat fails weeks later with nothing on the ticket to show why.

This card is the method: what to measure, where, in what order, and how often. Why the margin is 5 F and when the data sheet replaces it is owned by The Surface Five Degrees Above Dew Point Rule for Coatings. What dew point is in general is owned by What Dew Point Predicts.

What you need on the truck

• A humidity source. Either a digital thermo-hygrometer that reads air temperature and relative humidity (RH) and usually computes dew point, or a sling psychrometer: two glass thermometers on a handle, one bare (the dry bulb) and one in a wetted cotton wick (the wet bulb). Spinning it evaporates water off the wick, which cools the wet bulb in proportion to how dry the air is. The gap between the two readings, with a psychrometric chart or slide calculator, gives RH and dew point. The sling needs no battery and is the reference a digital meter can be checked against.
• A surface thermometer. A contact probe (a magnetic-base gauge or a thermocouple with a surface tip) or an infrared (IR) thermometer. Contact is more trustworthy on bare or shiny metal; IR is faster across a large area.
• A log. Paper or the job record, with time, location, each reading and the spread written next to it.

The steps

1. Let the instruments acclimate where the work is, before you read anything. Carry the meter and the psychrometer to the work area and give them about ten minutes in the shade at the surface you will coat. A hygrometer brought out of a heated cab reads the cab air for a while and then drifts toward the truth, and the readings in the middle of that drift look perfectly stable. What breaks if you skip it: the first reading of the day, the one that decides whether the crew starts, is the least trustworthy one.

2. Read the air at the work, in shade, at the height of the surface. Stand where the coating will go, not at the tailgate. Air over a wet lawn, under a deck or against a north wall carries different humidity from air in the driveway. Keep the sensor out of direct sun and away from your breath and body. What breaks: a driveway reading on a dry breeze can pass a shaded wall that sits in its own damp air.

3. If you use a sling, wet the wick with clean water and spin until the wet bulb stops falling. Use distilled or clean tap water, not water from a pail with paint in it. Whirl at arm's length for about a minute, read the wet bulb first (it starts warming the moment you stop), whirl again and read again. When two readings agree within half a degree, take them, then read the dry bulb. A dry wick reads the same as the dry bulb, and a zero gap reads as saturated air: 100 percent RH, with the dew point equal to the air temperature. Below freezing a sling becomes unreliable because the wick may be ice rather than water; use a digital meter rated for the temperature.

4. Get the dew point. The digital meter displays it. For a sling, enter the dry bulb and the gap (dry minus wet) on the chart or calculator that came with it; charts are printed for a stated barometric pressure, and at high altitude use the chart or setting for your elevation. Write down air temperature, RH and dew point, all three. What breaks if you log only the dew point: nobody can check the reading later or catch a meter that was off.

5. Find the coldest surface on the job, and read that one. The surface temperature that governs is the coldest spot that gets coated, not the average. Look for:

• Shaded faces, especially north and anything under an overhang.
• Sky-facing horizontal surfaces after a clear night. A deck board, a handrail top or a landing radiates heat to a clear sky and can run several degrees below the air before sunrise. See related: Radiational Frost: Why a Surface Freezes Above 32.
• Metal in contact with something cold: a base plate on a slab, a post where it enters the ground, the underside of a stringer.
• Thin metal versus thick: thick steel lags behind the air in both directions, so on a warming morning it is colder than thin sheet beside it.

Read several of these and log each by location. With an IR thermometer, bare or galvanized metal reads low because shiny surfaces reflect rather than emit; read through a small piece of masking tape or a painted spot on the same part, or use the contact probe. What breaks: the reading taken on the warm sunny tread passes a job whose base plates are wet.

6. Compute the spread at the coldest point, and show the subtraction. Spread equals surface temperature minus dew point. Write it as a subtraction in the log ("54.0 minus 49.6 = 4.4"), not just the result. Compare it with the sheet's margin, or 5 F where the sheet gives none. Check the sheet's RH maximum and minimum surface temperature at the same time; a reading can pass the spread and fail either of those.

7. Read again on a schedule, and before the last coat. Conditions turn. A morning reading says nothing about a surface that loses the sun at 2 p.m. A common default is a full set at start, then every two hours, plus one before any coat that will still be curing when the light goes. The rule protects application and early cure, so the reading that matters most is often the one taken while there is still time to stop.

A worked example: the north stair

A crew is recoating a galvanized-and-painted steel exterior stair on the north side of a small office building. The stair has an upper landing, two stringers, and base plates bolted to a concrete pad. The sheet for the coating says, for this example (illustrative values; use your sheet): surface at least 5 F above dew point through application and until dry to touch, RH 85 percent maximum, dry to touch about 90 minutes at the expected temperatures. The night was clear and calm with a low of 44 F.

NORTH STAIR - SURFACE AND DEW POINT LOG
time   dry  wet  RH  dewpt  landing stringer base
       F    F    %   F      F       F        F
08:30  58   52   67  47.0   50      49       48
11:00  62   55   64  49.6   60      56       54
12:30  66   57   57  50.4   64      59       57
15:00  64   56   60  50.0   60      56       54
spread at base (coldest): 1.0  4.4  6.6  4.0

8:30. Sling after the meter acclimated: dry 58 F, wet 52 F, which the chart gives as 67 percent RH and a 47.0 F dew point. The landing reads 50 F, the stringer underside 49 F, the base plates 48 F. Coldest point 48.0 minus 47.0 = 1.0 F. Hold. Nothing looks wet, and the landing alone, at 50.0 minus 47.0 = 3.0 F, would also fail.

11:00. Air 62 F, wet 55 F, 64 percent, dew point 49.6 F. Note what happened: the air warmed four degrees and the dew point rose 2.6 F, because the sun is evaporating the night's dew off the lawn and pavement into the air. A crew watching only the air temperature would expect the spread to open fast. At the base plates, 54.0 minus 49.6 = 4.4 F. The landing passes at 60.0 minus 49.6 = 10.4 F, but the base still fails. Hold for the base; the crew could start the landing.

12:30. Air 66 F, wet 57 F, 57 percent, dew point 50.4 F. Base 57.0 minus 50.4 = 6.6 F. All points pass, RH is under 85 percent. Go.

The sequencing call. The lead knows the north side never gets direct sun and the building's shadow lengthens after 2 p.m. So the coldest parts go first, while their spread is widest: base plates and lower stringers 12:30 to 1:15, upper stringers next, landing last because it runs warmest. Every coat started by 1:30 is dry to touch by about 3:00.

15:00. The before-the-light-goes reading: air 64 F, wet 56 F, 60 percent, dew point 50.0 F. Base 54.0 minus 50.0 = 4.0 F, under the margin. The crew stops. The base plates were coated first and dried by 2:45, so no coat is curing below the margin. Had the crew worked top-down, landing first, the base plates would have been coated around 2:30 and would now be in early cure at 4.0 F of spread, falling.

What the log shows a year from now. Four timed reading sets, the coldest point named each time, and the stop at 15:00. If a base plate rusts through early, the record shows it was coated at 6.6 F of spread and dried before the margin closed.

Where readings go wrong

• One meter, never checked. A cheap hygrometer can drift several percent RH, and in the 60 to 80 percent range a few percent moves the dew point roughly a degree or more. Check the digital meter against the sling at the start of each season and when a reading surprises you; if they disagree by more than a couple of degrees of dew point, trust the sling until the meter is checked or replaced.
• The surface reading on the wrong side of the job. Logging "surface 60 F" without a location is how a passing landing covers for a failing base. Every surface figure gets its location.
• Reading the forecast instead of the surface. The hourly forecast tells you which way the day is going and when to take the next reading. It does not replace the reading. See related: Reading a Forecast for the Work, Not the Weekend.
• Standing water that the spread cannot see. A passing spread says new condensation will not form; dew that collected overnight in an inside corner, under a bracket lip or in a cupped plate can still be sitting there at noon. Wipe and look in those pockets before coating them.

When the method changes

• Interior work with heaters. A propane or kerosene heater adds water vapor as it burns and can raise the dew point inside an enclosure faster than it warms a cold slab. Read inside the enclosure, not outside it, and vent unvented combustion heaters as their instructions require, with a CO monitor running.
• Cold, dry days. The spread may pass by twenty degrees while the surface fails the sheet's minimum temperature. The spread is one gate. See related: Air, Surface and Material Temperature: Three Different Limits.
• Specified work. Industrial and commercial specs may require a calibrated instrument, readings at set intervals, and a signed log. The spec's method then governs this one.

References

• See related: The Surface Five Degrees Above Dew Point Rule for Coatings
• See related: What Dew Point Predicts
• See related: Air, Surface and Material Temperature: Three Different Limits
• See related: The Coating and Sealant Conditions Log SOP
• Coating manufacturer's product data sheet for the margin, RH limit and minimum surface temperature that govern the job