IEC 60068 temperature humidity test chambers standards and configurations
IEC 60068 spells out the climate a product has to survive, and it does so long before anyone quotes a box to deliver it. Temperature, moisture, timing, the punishment a part lives through, all of it sits under a part number and a letter.
The part number carries the choices. A 2-30 and a 2-78 are different animals, and a Db behaves nothing like a Cab, yet both get waved through as damp heat across a desk. Reading the number and the letter early settles the bulk of the purchase, and getting them loose buys the wrong machine far more often than a real spec slip ever does.
Severities written down to the number
Each climatic part lists severities. A severity is a fixed set of temperature, humidity and time, numbered so two labs run the identical test from one line on a drawing.
| Part | Variant | Severity ladder | Duration |
|---|---|---|---|
| 2-1 | Cold A | -65, -40, -10 C | 2, 16, 72, 96 h |
| 2-2 | Dry heat B | +175 to +30 C | 2, 16, 72, 96 h |
| 2-30 | Damp heat Db | 25 to 40 or 55 C, 93 to 95 percent RH | 1, 2, 6, 12, 21, 56 cycles |
| 2-78 | Damp heat Cab | 40 C, 93 percent RH | 4, 10, 21, 56 days |
Cold under part 2-1 sits at minus 65, minus 40 or minus 10. The cold chamber range and tolerance for 2-1 covers how tight the low limit has to hold.
Dry heat under part 2-2 mirrors it at the top of the scale. The dry heat chamber configuration for 2-2 covers the heater sizing the high severities call for.
Damp heat cyclic, variant Db, is the one buyers underestimate.
Part 2-30 runs a 24 hour cycle. The air climbs to 40 or 55 degrees, holds, then falls while the humidity stays at 93 percent or above, and the standard repeats it up to 56 times. The cycle is built to force condensation on the down ramp, and the damp heat cyclic conditions for 2-30 Db set the ramp rate and dwell the profile needs.
Steady state is variant Cab.
Part 2-78 drops the cycling and holds 40 degrees at 93 percent for up to 56 days. Simpler to run, harder to sustain, and the steady state parameters for 2-78 Cab punish a box that drifts across a long soak.
Three more parts stack stresses. Part 2-38 runs a composite temperature humidity cycle with a cold subcycle to minus 10. Part 2-39 folds in altitude as a combined cold, dry heat and low pressure run. Part 2-67, variant Cy, drives a steady accelerated humidity test for capacitor aging, and part 2-68 covers fluid contamination of components.
The variant letter changes everything
The letter is the part people skip on the order.
Da, Db, Ca, Cab each name a different severity family. Db cycles. Cab holds. They share the words damp heat and behave nothing alike, so an order that names damp heat without the letter buys the wrong box half the time. The letter, not the brand, is the first thing to confirm.
Severity is a choice
The standard offers the ladder. The product class picks the rung.
A board for a heated indoor cabinet does not need the 56 day Cab soak a part for a tropical outdoor enclosure does. Over-specify and the budget bleeds on a test the product never faces. Under-specify and the field finds the failure the lab missed.
A run is three phases
Pre-conditioning, exposure, recovery.
Pre-conditioning brings the sample to a known start state. Exposure applies the severity. Recovery returns the sample to standard atmosphere before the final measurement, and the standard fixes how long that window lasts. Measure inside the transient instead of after recovery and the number reads as noise.
Classify the field with IEC 60721
A test counts only when it mirrors the field.
IEC 60721 sits beside the 60068 series for exactly this. It sorts real environments into classes, a heated office as one, a sheltered outdoor cabinet as another, a coastal site as a third. The environmental classification for chamber selection under 60721 turns a vague field description into a defined class, and the class names the 60068 parts and severities the product has to clear.
Start from the class. Map it to the parts. Then spec the box.
The part number writes the configuration
Once the part and the severity are fixed, the configuration follows on its own.
A low humidity severity needs a thin film evaporator the steady soaks never call for. A long Cab run wants the dew point control method chosen for stability over speed. The humidifier rests on a deionised water specification and a routine that keeps that water from scaling the humidifier, whether the supply is softened or run through reverse osmosis, and the lowest set point on the test decides the humidifier type, pan, steam or ultrasonic, with an electrode steam unit kept to its own cylinder replacement schedule. Each detail has its own page. The part number only points at which one matters. Those same parts are where the search begins when a chamber stops holding ninety-five percent at 85.
Holding the band the standard names
Every severity carries a tolerance the chamber has to honour.
The damp heat parts allow roughly two kelvin on temperature and a few percent on humidity, held across the workspace and not the probe alone. Long soaks add a second risk: a sensor that reads true on day one can walk three percent low by day fifty. The acceptance check has to keep an eye on sensor accuracy under sustained damp heat as the soak stretches long. Cyclic profiles add a third, since condensation control through the down ramp separates a true Db box from a steady one wearing the badge.
Holding the band sounds simple and hides a knot of competing demands. The standard fixes a tolerance on temperature and another on humidity, and both have to hold across the whole working space rather than at one sensor, for the full length of a run that can stretch to weeks. The trouble is that the two are linked. Relative humidity rides on temperature, so a small thermal drift moves the humidity even when the water content has not changed, and a chamber that lets its temperature sag overnight watches its humidity climb off target without a drop of extra moisture entering the box. A degree of gradient between the top and bottom of a loaded space becomes a humidity gradient too, so the corner that runs a touch cool reads several points wetter than the sensor by the door. Meeting the tolerance, then, is less about raw capacity and more about how evenly the box moves its air and how tightly its two loops stay coordinated through the quiet hours of a long unattended run.
The band is in truth a statement about control as much as capacity. A box can reach ninety-three percent and still fail to hold it if its loops fight each other, if its humidifier adds moisture in slugs too coarse for fine trimming, or if a cool corner of the working space sits a few percent off the sensor reading. The gap between reaching a condition and holding it for weeks is where cheap chambers and serious ones part company, since anything can hit a number for a minute and only a well-built box keeps it through a fortnight of door openings, load changes, and quiet nights. Meeting the standard's tolerance means a mapped, uniform space, a steady supply of pure water, and a control loop fine enough to trim moisture and heat in small steps, all proven before the parts ever go in.
Where the parts get applied
The same family surfaces across industries that share nothing else. A vaccine maker leans on an ultra low temperature stability run. A cosmetics house runs a fragrance and cosmetic shelf life soak. A data centre qualifies hardware in a chamber for liquid cooled AI servers. One standard family, three products, the same handful of part numbers.
Maintenance the standard quietly assumes
A 60068 run assumes the box holds its numbers, and the long damp heat severities expose any humidifier that does not. Scale cuts the steam output over a 56 day soak, so the electrode replacement schedule for the steam humidifier earns a fixed slot on the maintenance calendar.
Inside a single Db cycle
One Db cycle is 24 hours, and the shape inside it is fixed.
The air rises from 25 degrees to the upper limit over three hours, with the humidity held at 95 percent on the way up so moisture drives into the surface rather than beading on it. The upper limit then holds for nine to twelve hours, long enough for the warmth to push water deep into seams, under coatings, and along the lead frames where it does its slow work. The fall back to 25 degrees takes three to six hours, and the standard requires condensation to form during that descent, because the wet surface is the entire point of the run: as the part cools faster than the air around it, dew settles on it the way it beads on a cold glass carried into a warm room, and that film is what drives the corrosion and the ion migration the test is built to find. Skip the condensation and the cycle becomes a mild humidity soak that misses the mechanism it was written to expose.
The remaining hours dwell at the low point before the next cycle starts.
Part 2-30 names two upper limits.
Variant one tops out at 40 degrees, variant two at 55. The 55 degree run is the harsher of the pair and the one written into a great many automotive and aerospace contracts. A chamber that holds 40 comfortably can stall trying to keep 93 percent humidity as it climbs past 50.
Rate of change, the hidden severity
Severity is not only the end points. It is how fast the chamber moves between them.
The cyclic parts cap the ramp, typically near one kelvin per minute averaged across the transition, and a profile that ramps faster or slower than the window is out of spec even when the end points read perfectly. This is the line a budget chamber crosses first, since a slow refrigeration stage cannot follow the descent on a loaded shelf. The rate limit, buried in an annex, decides more purchases than the headline temperature range.
The class codes under 60721
IEC 60721-3 writes the field as a code, and the code repays a close read.
A heated, humidity controlled indoor space sits at class 3K3. A space with no climate control reads 3K5. A sheltered outdoor location climbs to 3K6 or 3K7, and the codes carry the temperature and humidity extremes the location reaches. The value is the bridge they build. A product sold into a 3K6 environment maps to specific 60068 severities, and the map removes the guesswork from the test plan.
Match the wrong class and the test drifts from reality.
A part qualified to 3K3 and shipped into a 3K6 site fails in the field, and the lab record offers no cover because it tested the wrong climate from the start.
One letter, one climate
The method letter does the heavy lifting. Change it and the whole character of the test changes, from a steady soak to a swinging cycle.
Three documents, one family
The series splits into layers that confuse first time readers. IEC 60068-1 holds the general rules and the standard atmosphere for measurement. IEC 60068-2 holds the test methods, the parts with the part numbers and the variant letters. IEC 60068-3 holds the background and the guidance behind the numbers. A spec that cites a severity without naming the 60068-2 part it came from leaves the lab guessing which method to run.
Tolerance the parts name outright
The damp heat parts do not leave tolerance to the chamber maker.
Part 2-78 fixes the air at 40 degrees within two kelvin and 93 percent humidity within three points, held at every point in the working space across the full soak. Part 2-30 carries the same temperature band through the cycle and asks for 93 percent or higher at the top, dropping toward saturation on the descent. A chamber that meets the band at the control sensor and misses it in a corner has failed the part, and only a loaded uniformity survey catches it before the qualification leans on the result. The band is the promise the standard makes to the next lab that repeats the run, and a box that cannot keep it turns a qualification into a figure the next lab cannot reproduce.
The cheapest mistake
Reading a severity number wrong is the cheapest mistake to make and the dearest to find, since the wrong test quietly passes the wrong part.
Picking the part number
The sequence is short. Name the field class under 60721. Pull the 60068 parts that class demands. Read the variant letter so Db and Cab never swap. Fix the severity from the product class, then list the duration and the recovery window. Only then size the chamber and choose the configuration the part calls for.
A buyer who works this order reads a data sheet as a checklist and prices it line by line. The part number outlives the box on the floor, since the chamber gets replaced every decade while the severity it was bought to satisfy stays fixed, and a spec written against the standard still reads true when the next chamber arrives. The part number is the one line on the order that survives the hardware, the supplier and the decade. It earns the first reading and the last word.