IP65, IP67 and drop tests: what a rugged handheld terminal actually certifies

Three numbers do most of the selling on a rugged Android handheld: an IP code, a drop height, and the phrase "MIL-STD-810 tested". All three are real engineering artefacts, and all three get quoted in a way that means something different from what the buyer needs to know. The gap is mostly shorthand rather than dishonesty — a rating certifies a test, a test certifies conditions, and your depot is not those conditions.

This decodes those claims: what each digit of IEC 60529 is tested with, what the water digit assumes, what ingress protection never looks at, what has to be specified before "1.5 m drop" means anything, and why MIL-STD-810 is a menu rather than a badge — plus the wording to put into your RFQ, because the cheapest test of a claim is its conditions in writing.

The two digits of IEC 60529 do different jobs

IEC 60529 classifies the protection an enclosure gives against solid objects and against water: two digits, rated independently. A device listed IP65 and one listed IP67 share the solid-particle digit and differ only in the liquid test.

The first digit's top two levels are different apparatus, not steps on one scale. IP5X permits dust to enter provided the deposit does not interfere with safe operation; IP6X uses a sealed chamber with a pressure differential across the enclosure walls and admits no harmful deposit. In a mill, a cement plant or a foundry, permitted interior deposit ends up in a key switch, a vent or a charging interface, and it stays there.

The second digit escalates through dripping, spraying, jetting, immersion and a 9K high-pressure class that some datasheets borrow from road-vehicle practice rather than from the base IEC table. Every one of these is run on a device in a defined state: covers closed, ports terminated as specified, normal operating orientation, dry air at room temperature, still fresh water. That sentence is where most field surprises come from.

What the water digit assumes that a depot does not

IP65 is a jet test: a defined nozzle, a defined flow, a target roughly 2.5 to 3 m away, and no sustained pressure on one seam. IP67 is an immersion test — in the standard, a depth in the region of 1 m for 30 minutes. Still, fresh, shallow, brief, clean.

The field adds mechanisms the test does not contain. A depot pressure washer puts a narrow, high-pressure stream at distances and angles the jet test never applies, and the hot, detergent version is a different physical problem: detergent lowers surface tension and reaches passages water would not, and heat changes how a gasket sits under compression. Immersion while warm is another — a housing heated by a shift cools after being dropped in a puddle, and the contracting air inside draws water through paths a room-temperature test never demonstrated. Salt water and melting snow are not the test medium either. What the rating does not describe is washing the device.

What ingress protection never looks at

Dust already inside. IP6X says nothing about contamination present at assembly. Build-line particulate, machining swarf or omitted desiccant ends up inside a nominally dust-tight housing, where it bridges contacts and scratches a lens indefinitely. Ingress protection is not cleanliness.

Charging contacts. Exposed pogo pins and a cradle interface sit outside the protected volume, and their wet-field failure is electrochemical rather than hydraulic: water bridging energised pins migrates metal across the gap, and a connector that was merely dirty becomes conductive and stays that way. The instruction "do not charge a wet device" is a behaviour problem, not a rating problem.

The seal after one service opening. Gaskets are compressed to a designed fit by a defined torque pattern, and they are consumables: they take a compression set, they are nicked by a screwdriver, and replacement packs are rarely stocked. A device opened at month nine for a battery or a port is an IP67 device only in the memory of the report. So ask which joints are meant to be reopened, whether the seal is reusable, and what the classification is expected to be afterwards.

Thermal cycling. Repeated expansion, UV on a shell held outdoors, dropped covers and overtorqued fasteners all spend a seal's margin. The rating is measured on units from a fresh lot; in year three you are paying for whatever the geometry has become. The honest question is not "is it IP67" but "what is it after a year of openings, and what do you sell me to put that back".

Drop testing: the height is the least informative word

A drop claim without conditions describes one event that happened to a good sample. The parts that decide whether your device survives your floor:

softest is a different product.

and corners concentrate load into the display assembly where the failures are.

not. Repeated after thermal soaking, or done once on a new unit.

a lost session are six outcomes. If the pass criterion was "it boots", nothing about the scanner window or the seal was proven, and the case you intend to issue changes the load path again.

• Surface. Concrete is the honest one; steel plate, epoxy floors and a bench are progressively softer, and a claim taken from the
• Orientations and count. Six faces plus corners and edges is the meaningful set. A single face-drop is one event, not a specification,
• Device state. Powered and running during the sequence, or not. Tested cold, where a polymer embrittles, or at 25 °C where it does
• Failure definition. Cosmetic scuffing, a cracked window, an unseated battery, an intermittent charging port, a scanner that lost its aim and

"1.5 m to concrete" is therefore not a specification; it is the middle of a sentence you should require in full.

MIL-STD-810 is a catalogue of methods, not a pass

MIL-STD-810 is a United States compilation of test methods for defence equipment: shock, vibration, rain, icing, salt fog, sand and dust, humidity, temperature, altitude and more, each with procedures, categories and conditions selected by a test plan. Passing "810" is not a thing that exists; there is no single result to quote. Read the report for:

drop-shock procedure at one height is not your floor.

fresh sample.

1. The method, procedure and category names actually run — a transport vibration profile says nothing about a vehicle cradle, and a
2. Whether conditions were tailored to a commercial environment or left at defaults, and whose numbers the tailoring used.
3. What the equipment under test was: a working handheld, or a housing with ballast.
4. Whether one unit ran the whole sequence, so combined damage — drop, then dust, then rain — got answered, or whether each method got a
5. The laboratory and report identifier, which is the difference between a test and a claim.

Packaged transport is a separate question, answered by an ISTA carton test rather than by an IP code.

What each claim covers, and what it does not

Sentences to paste into an RFQ

Require an answer per line, referencing a report:

ports terminated.

powered and running.

IP class afterwards.

1. State the IEC 60529 classification tested, the report number, the laboratory, and the sample state at test — battery fitted, covers closed,
2. Confirm whether the liquid digit was verified by jet or immersion, and give depth, duration and water temperature.
3. Give drop height, surface material, the orientation and corner sequence, number of drops, sample size, and whether the device was
4. Define failure separately for housing, display, scanner window, charging interface, battery retention, and the ability to still meet its
5. List the MIL-STD-810 methods, procedures, categories and tailored conditions run, and identify the equipment under test.
6. Identify every seal intended to be opened in service, whether it is reusable, and the part number for a replacement.
7. Confirm drop and ingress performance with the case and screen protector you intend to issue.
8. State spare-part and service-manual availability for the model's lifecycle, with the price basis for screens and charging contacts.

The answers either exist as documents or they do not, and one reply tells you which.

What this looks like on a spec sheet

EDOO lists several handhelds and tablets with an IP classification inside the model description — M93-UHF and M86 as IP67 rugged Android terminals, P1020 as an IP67 rugged tablet, P105G as IP68 — and the fixed-mount conveyor reader ED5000 as IP67, while its sibling ED3000 is listed by use case (fast-moving Data Matrix) rather than by an IP digit. Those are the classifications to ask about, and the eight questions above are how you find out whether the rating describes your site. Published commercial terms are OEM quantities from 100 units, 15-day samples and 30-day production from the ISO9001 Shenzhen facility.

FAQ

What is the difference between IP65 and IP67 on a handheld terminal? Both share the same solid-particle digit — dust-tight under an IEC 60529 chamber test — and differ only in the liquid test: IP65 certifies protection against a water jet from a defined nozzle at a defined distance, while IP67 certifies short immersion in still fresh water, in the region of 1 m for 30 minutes as the standard specifies.

Is an IP67 rugged PDA safe to wash in a pressure-washer bay? No — an IP67 rating on a rugged handheld is earned with still, fresh water at room temperature under controlled depth and duration, whereas a pressure washer adds a narrow high-pressure stream and usually hot water and detergent, which reach passages the standard never stressed. Only the higher-pressure road-vehicle class addresses anything similar, and only under its own conditions.

What makes a drop-test claim useful rather than decorative? The conditions around the height: surface material, orientations and corner hits, number of drops, sample size, whether the device was powered and running, whether it was tested cold, whether it wore the issued case, and what counted as failure beyond "it still boots".

Is "MIL-STD-810 tested" a certification for a rugged handheld? No. It is a catalogue of methods whose procedures, categories and severities are chosen per test plan, so the claim becomes information only when a report names the methods run, the tailoring applied, whether the equipment under test was a working device, and whether one unit survived the whole sequence.

Related EDOO resources

services · All EDOO solutions · Android PDA and POS terminals · IP67 product page · IP65 product page · EDOO product datasheet · FAQ · ODM / OEM services · Product documentation and SDKs · Contact EDOO