Outdoor Hardware: IP44, IP65, IP67, Power, Enclosures, Cables, and Weatherproofing
Outdoor Hardware: IP44, IP65, IP67, Power, Enclosures, Cables, and Weatherproofing
A setup from Lesson 1 that ran flawlessly all summer can go silent forever after the first freeze and rain. This lesson is the one place in the course where the actual IP numbers, operating temperature ranges, and cable types decide whether your hardware survives winter, not just looks good in August.
You're not mounting anything permanently today. You're learning to read the markings on an enclosure, match a cable to its installation location, and understand why "waterproof" on a box doesn't always mean what you think it means.
Budget around 80 minutes. From here on I'll shorten Home Assistant to HA.
The core rule of this module
Outdoors, automation has to be more resilient than it is indoors.
Read and diagnose state first.
Then manual control and a Plan B.
Only then automation.
A Problem from Real Life: The "Outdoor" Sensor That Died After the First Rain
You bought a temperature sensor listed as "waterproof" on a budget marketplace and hung it under the eaves. After a week of wind-driven rain, the entity stopped reporting. Water vapor had condensed inside the plastic shell and corroded the board. The listing never gave an actual IP rating; "waterproof" in an auction description isn't a technical standard, it's marketing copy.
Second scenario: a Shelly relay in its stock plastic case (no extra enclosure) went into an irrigation valve pit. The pit flooded after a downpour and the module shorted out. A bare Shelly 1 or Plus UNI doesn't have anywhere near the IP rating needed for submersion: it needs an additional sealed mounting box.
The IP Code: What Those Two Digits Actually Mean
IP (Ingress Protection, sometimes International Protection) is two digits: the first covers protection against solids and dust, the second against liquids. A higher number doesn't automatically mean "better at everything": IP67 (submersion to about 3 ft) doesn't protect against a high-pressure jet of water the way IP69K does.
| Rating | Liquid protection | Where it makes sense |
|---|---|---|
| IP20 | No liquid protection | Indoors only, a dry panel or cabinet |
| IP44 | Splashing water from any direction | A covered patio, a carport, anywhere without direct rain |
| IP54 | Splashing plus partial dust protection | A driveway, a porch, somewhere with occasional rain exposure |
| IP65 | Water jets from a nozzle, full dust protection | Open ground, direct rain, an exterior wall |
| IP66 | Powerful water jets | Areas that get pressure-washed nearby |
| IP67 / IP68 | Temporary / permanent submersion | Drainage pits, tank level sensors, flood-prone installs |
IP says nothing about temperature or UV
An enclosure can be rated IP67 and still crack in a hard freeze around 5°F if the manufacturer never specified an operating temperature range, or fade and go brittle under UV exposure after two summers in direct sun. Check both separately: the stated operating temperature range (say, -4°F to 122°F) and UV resistance (a "UV resistant" label, or an ASA/PC material instead of plain ABS).
What's at Risk
Marketing instead of a spec: "waterproof" with no actual IP number on the label or the datasheet.
Condensation inside a sealed enclosure: a sudden temperature drop after a warm day can condense water vapor even inside a closed box with no vented membrane.
Indoor cable used outside: lamp cord or flat telecom cable stiffens and cracks in a freeze, and isn't UV-rated.
A cable gland with no seal: feeding a cable into an IP65 enclosure without a proper gland (a PG9 or PG13.5 gland, for example) voids the entire protection rating.
Batteries in the cold: alkaline cells and some lithium-ion cells lose a large share of their effective capacity below 14°F, and some sensors stop reporting long before the battery is actually dead.
Outdoor Cabling: The Right Wire for the Job
| Application | Cable type | Note |
|---|---|---|
| 120/240V power outdoors, buried or exposed | UF-B direct-burial cable underground, outdoor-rated THWN/THHN in conduit above ground | Selection and installation: a licensed electrician |
| Low-voltage signal (sensors, reed switches) | Outdoor-rated, direct-burial low-voltage cable, shielded where noise is a concern | Avoid flat telecom cable: it isn't UV-rated |
| Ethernet / PoE to a camera | Cat 5e/6 in an outdoor-rated jacket (gel-filled, UV-stabilized) | Ordinary indoor patch cable cracks after one season in the sun |
| Low-voltage DC power (strips, 12/24V sensors) | Two-conductor cable with PVC insulation rated to -13°F | Check gauge against run length and voltage drop |
Every place a cable passes through a wall or enclosure needs a seal: a cable gland, silicone, or a purpose-built grommet. The point where a cable enters a box is the single most common failure point for "waterproof" hardware: the enclosure itself can carry a high IP rating, but a badly sealed cable entry lets water in regardless.
Power: Battery, PoE, or a Transformer
You'll run into three power paths most often in this module:
Battery: simple to mount, but needs replacing and loses capacity in the cold. Good for low-draw sensors (reed switches, temperature sensors), bad for anything driving a motor or a pump.
PoE (Power over Ethernet): one cable carries both power and data. The standard for outdoor cameras: it eliminates a separate power supply at the device, but it needs a PoE switch or injector and a well-planned cable run.
A transformer or low-voltage power supply: for LED lighting, 24V AC solenoid valves, and gate operators. The transformer itself mounts in a dry, ventilated spot: not in a drainage pit.
PoE and cold weather
PoE cameras run on thin cable over long runs (past 200 to 230 ft) can see voltage sag at the camera end in cold weather, when draw increases (a lens heater, IR illumination). Check the PoE spec (802.3af vs. 802.3at) and the switch's real power budget, not just the sum of the cameras' nominal draw.
Add-On Enclosures: When the Device Itself Isn't Enough
Modules like the Shelly 1, Shelly Plus UNI, or typical Zigbee relays are designed to mount inside a junction box or a light fixture: the bare electronics usually carry a low IP rating (IP20-IP40). For outdoor use they need:
A sealed IP65/IP66 mounting box with cable glands on every cable entry and exit.
Glands mounted facing down (never pointing up), so water runs down the cable instead of into the box.
Clearance off the ground: a box sitting on the ground collects moisture from dew and melting snow even without direct rain.
Hardware Selection Card for Each Mounting Location
# Outdoor hardware selection card
Mounting location: ____________________
Direct rain exposure: yes / no
Possible submersion (pit, tank): yes / no
Minimum required IP rating: ____
Winter temperature range at this spot: ____ to ____ °F
Power: battery / PoE / transformer / 120V mains
Cable run and length: ____ ft
Cable type: ____
Extra enclosure needed: yes / no, model: ____
Cable glands: count and size ____Junction Boxes and Cable Glands: The Detail That Determines Durability
Even a device with a stated IP65 rating loses that protection if a cable enters through a plain hole with no gland (a cable gland). A gland is a threaded rubber or plastic fitting that seals the point where a wire passes through an enclosure wall: without one, water running down the cable flows straight into the box, even if the enclosure itself is otherwise sealed.
An IP65/IP67 junction box: a small, separate enclosure mounted at a wire-splice point, for example where power reaches an in-ground sensor or a wall-mounted device
A cable gland sized to the cable's diameter: too small and it won't tighten down, too large and it won't seal: check the manufacturer's stated diameter range
A drip loop: the cable routed with a slight downward sag before it enters the enclosure, so water running along the wire drips to the ground instead of running inside
These three details together, not the IP number on the label alone, decide whether an install survives its first rainy season.
Enclosure Materials: Polycarbonate, ABS, and the Difference in Durability
Not every enclosure marketed as weather-resistant is equal: polycarbonate (PC) holds up to UV radiation far better than cheaper ABS, which turns brittle and cracks over time under sunlight, especially in spots that get a lot of direct sun. Enclosures marketed as UV-stabilized typically last 5 to 10 years in the sun with no visible degradation, while plain ABS can start cracking after just 2 to 3 seasons in full sun.
Enclosure color has a practical effect too: dark enclosures heat up more in the sun, which for battery-powered devices (the sensors from Lesson 11, for instance) speeds up battery drain and can skew temperature readings if the sensor shares its housing with the electronics.
Battery Life in Winter Conditions: Why Sensors "Die" in the Cold
Alkaline batteries lose capacity in cold weather
Standard alkaline CR2032 or AA cells can lose 20-30% of their effective capacity below freezing, which often shows up as a sensor suddenly vanishing from the Zigbee network in the middle of winter, despite running fine all summer. For sensors in cold locations, favor lithium cells (not alkaline) rated for low temperatures, or plan on more frequent battery swaps during winter.
Testing Enclosure Sealing Before Final Installation
Before you mount a new outdoor device in its final spot, it's worth testing its seal under controlled conditions: a few minutes under a running garden hose (not a pressure washer, which exceeds the test parameters of even IP65-rated hardware) will catch a bad seal before the failure shows up only after the first real downpour, when a fix is much harder to reach.
Choosing Corrosion-Resistant Screws and Mounting Anchors
A small, often-overlooked detail: plain carbon-steel screws rust outdoors within a season or two, leaving ugly streaks and weakening the mount. Stainless steel screws (grade A2 or A4) cost slightly more but eliminate the problem for years. The same goes for mounting anchors on siding or masonry: check before buying whether the manufacturer rates the anchor itself for outdoor exposure, not just the device it holds.
Testing Hardware Under Extreme Conditions Before Full Rollout
Manufacturers state IP ratings and operating temperature ranges based on lab tests that don't always perfectly match real conditions on your specific property: all-day direct sun at the height of summer, a freeze colder than the stated minimum on an unusually cold night, or condensation from sudden temperature swings. Before buying the same model in bulk for multiple locations, it's worth testing a single unit through a full season in the most demanding spot on your property, rather than trusting the manufacturer's spec sheet alone.
Labeling and Documenting Field Hardware
Once you have more than a few devices scattered around the property, it's easy to lose track of which physical device maps to which entity in Home Assistant. A simple, durable label (weather-rated) with a short identifier on each device, matching an entry in your smart home notebook, saves a lot of time when servicing hardware months or years after installation, once the installation details have long since faded from memory.
Real-World Lifespan of Outdoor Hardware: When to Expect the First Failures
| Device type | Typical trouble-free run |
|---|---|
| Battery sensor in a good IP65+ enclosure | 2-4 years to the first battery swap; the electronics itself usually longer |
| Cheap ABS enclosure with no UV stabilization | 1-3 seasons before visible cracking in the sun |
| PoE device in a solid enclosure | 5+ years, limited mainly by the electronics itself |
| Irrigation solenoid valve | 3-7 years, depending on water quality and cycle frequency |
Knowing these rough timeframes helps you plan a maintenance budget ahead of time, instead of treating every failure as an unexpected surprise.
How to Test This Lesson
1. Check the datasheets of two devices you already own or plan to buy: find the actual IP number, not just the word "waterproof."
2. Fill out the hardware selection card for three spots from your map in Lesson 1.
3. Measure or estimate the cable run to your farthest mounting point.
Common Mistakes
"Waterproof" with no IP number: buying on a word instead of a spec.
A gland pointing upward: water runs straight into the enclosure.
Indoor cable used outside: cracks after the first freeze, goes brittle from UV within a year.
Ignoring the operating temperature range: an IP67 enclosure cracks in a freeze because the manufacturer never tested it below 32°F.
Practical Task
☐ Fill out a hardware selection card for every mounting point on your map from Lesson 1.
☐ For each point, settle on a minimum IP rating and check whether you already own a matching enclosure.
☐ Plan your cable runs together with the cable type: note where you'll need glands and grommets.
Key Takeaways
IP is two digits: solids and liquids, separately: check both.
IP says nothing about temperature or UV: check those separately on the datasheet.
Glands and grommets are the most common failure point in "waterproof" hardware.
Shelly modules and relays usually need an extra IP65+ enclosure to go outdoors.
What's Next
Next lesson: Wi-Fi, Zigbee, LoRa, and PoE Range Outdoors: The Garden, the Garage, the Gate, and Sensors Far From the House. You now know which hardware survives the weather: next you'll check whether you can even connect to it.
