Thermoelectric cooler condensation: when a cold plate drips

A thermoelectric cooler drips when the surface it chills falls below the dew point of the air it touches. The water lands on the cold plate and the fins the internal air flows over, never on the warm electronics that the chilled air then reaches. If you size the cooler for a modest temperature drop, control the setpoint, and keep humid outside air out of the box, a Peltier-cooled enclosure stays dry. Get one of those three things wrong and you find water where it has no business being.

This article covers the condensation side of thermoelectric cooling. The condensation guide walks through dew points, heaters and drains for sealed enclosures in general, and the same physics applies when the cooler is a Peltier module.

What condensation actually requires

Condensation needs three things at once: water vapor in the air, a surface colder than the air's dew point, and time. Remove any one of them and no water forms.

Air holds a limited amount of vapor at a given temperature. Cool the air and that capacity drops; cool it past the dew point and the excess vapor condenses on whatever surface did the chilling. The dew point depends on how much moisture the air carries, not on the temperature alone.

As a working rule, indoor electronics air at 77 degrees F (25 degrees C) and 50 percent relative humidity has a dew point near 56 degrees F (13 degrees C). At 80 percent humidity the dew point climbs to about 73 degrees F (23 degrees C). A cold plate in humid, unconditioned air does not have to be very cold to start dripping.

The condensation forms on the cold surface itself, not downstream. Air that passes over the cold plate leaves it drier and slightly cooler, and every surface that air then touches is warmer than the air. That is why a Peltier cooler rarely wets the board it is protecting; it wets its own cold side, or the housing around it, if anything.

Where the water shows up in a real enclosure

In a sealed enclosure with a thermoelectric cooler mounted through the wall, the riskiest spots follow the cold path:

• The cold-side plate and its internal fins, where vapor condenses first.
• The wall of the enclosure around the cooler mount, chilled by conduction.
• Anything mounted close enough to touch the cold-side airflow.
• Door gaskets and low corners, if the enclosure breathes humid air through leaks and the cooler cycles off overnight, letting the trapped air cool below its dew point on the same surfaces.

The last case is the one that surprises people. The cooler itself never added water; a leaky gland plate pulled humid air in, and the overnight temperature swing squeezed it out as droplets. The condensation in a sealed enclosure article covers why a watertight rating alone cannot stop this.

Why Peltier coolers make the problem easy to create

A thermoelectric module is a heat pump with no thermostat of its own. Feed it constant current and its cold side keeps dropping until the heat leaking in balances the heat pumped out. The temperature difference across the module can reach 120 degrees F (about 65 degrees C) in ideal conditions; a typical small module under real load settles around 40 to 60 degrees F (20 to 30 degrees C) below its hot side.

That is far more delta-T than most electronics need. If your only goal is to hold a panel a few degrees below ambient, an uncontrolled Peltier can easily drag its cold plate well under the dew point on a humid day. The module does not know where the dew point is; it only knows current in, heat moved.

The modules are also inefficient. They pump roughly as much heat as they consume in electrical power, and dump both onto the hot side. A 60 W module cooling a 20 W load puts about 80 W into the hot-side heat sink, which needs real airflow to stay alive.

So the failure mode is predictable: an oversized or unregulated module in a humid box drives the cold plate below dew point and condenses the trapped moisture onto the coldest metal in the enclosure.

What a rating test does and does not cover

Manufacturer ratings for sealed enclosures, such as our stainless boxes rated NEMA 4X-type / IP66 by the manufacturer; not UL listed, describe resistance to water and dust ingress from outside. They say nothing about what happens to moisture already inside, and nothing about any cooling hardware you add.

Drilling and cutting the enclosure for a cooler mount is your work, and the rating of the finished assembly depends on how you seal it. A through-wall cooler with a proper gasket keeps the ingress rating; one bolted to an unsealed cutout does not. For entry sealing generally, see when to specify a stainless steel enclosure with a gland plate.

Sizing and controlling so the plate stays dry

The design goal is simple: the coldest internal surface stays above the interior dew point. Three habits get you there.

Size for a small delta-T. Electronics in a sealed box rarely need more than 10 to 15 degrees F (5 to 8 degrees C) below ambient to stay reliable. Pick a module whose capacity at that small difference comfortably exceeds the enclosure's heat load, rather than one that promises a dramatic temperature drop. The sizing method for the enclosure side is the same as any sealed cooler: find the real heat gain first, as in finding the real heat gain of an outdoor enclosure.

Control the setpoint, do not run open loop. A thermostat or PWM controller holds the interior in a band, say 68 to 77 degrees F (20 to 25 degrees C), and never lets the cold side overscroll. If your controller reads only interior air, remember the plate runs colder than the air; leaving a few degrees of margin between the setpoint's lower limit and the expected dew point is the margin that keeps the plate dry.

Keep the cold side sealed from outside air. In a properly designed thermoelectric enclosure cooler, the cold side and its fan face the interior; the hot side and its own fan face outside. Nothing humid from outdoors ever crosses the cold plate. Condensation in a laptop-cooler style clip-on, where the cold plate sits exposed in open room air, is a different and much wetter situation — that plate sits directly in the path of humid air and drips onto whatever is under it.

For context on the other sealed-cooling option at humid sites, closed-loop enclosure cooling where fumes are in the air explains why isolating the refrigerant circuit from the interior matters.

A worked example

Take a 16 x 16 x 6 in 304 stainless wall-mount enclosure on an outdoor wall, holding a small controller and a power supply that together dissipate 40 W. Ambient peaks at 95 degrees F (35 degrees C). A sealed Peltier unit rated around 100 W at a 25 degrees F (13 degrees C) difference holds the interior near 85 degrees F (29 degrees C) — comfortable for the electronics.

Now the dew point. On a humid summer day the outdoor air can sit at 80 degrees F (27 degrees C) and 70 percent relative humidity, a dew point near 70 degrees F (21 degrees C). The interior air is 85 degrees F, above dew point, so nothing condenses on the electronics. The cold-side plate, though, can run 20 degrees F (11 degrees C) below the interior air — about 65 degrees F (18 degrees C) — below dew point. Any humid air reaching that plate condenses on it.

The fix in practice: the cold side is ducted and gasketed inside the cooler so interior air crosses its fins, the controller holds 85 degrees F as the floor, and the cable entries are sealed so overnight cooling cannot pull in a fresh charge of humid air. Dry the interior at assembly with a desiccant pack and the loop stays closed.

What to specify

When you order an enclosure you intend to cool with a thermoelectric unit:

• Give the real heat load in watts and the maximum ambient, so the cooler is sized for a small delta-T, not a dramatic one.
• Order the cutouts for the cooler mount on the drawing, with gasket surfaces, rather than cutting them in the field.
• Plan a controller with a setpoint floor above the site's expected dew point, and say what that dew point is.
• Ask for the mounting panel position that keeps electronics out of the direct cold-side airflow.
• Include desiccant access or a drain provision if the site is humid year-round.

Our boxes are made to order: set the material, the dimensions and every cutout, and it is delivered in 10 business days, counted from your order or, for custom work, from your approval of the drawing.

Condensation from a thermoelectric cooler is not mysterious. It is the dew point, reached on the coldest surface in the box. Size and control for a small temperature drop, seal the cold side and the cable entries, and the drips never start.

Enclosures for this job

Catalog sizes that suit the job this article answers, picked from its subject; open one for its full specifications.

Questions