Weather, cold and sun: what an outdoor smart lock has to survive

A lock set four feet back under a porch roof leads an easy life. It catches blown rain in a storm, an hour of low evening sun, and otherwise sits in still air at roughly the temperature of the house behind it. A lock on a side gate with nothing above it gets everything the sky produces, plus whatever the sprinkler throws at it twice a week.

Both are sold as exterior locks. The environmental part of a spec sheet is the section nobody reads, and on a sheltered door that is a defensible choice. On an opening that is genuinely exposed it becomes the whole decision, and the one rating that looks like it answers the question mostly does not.

Exposed is a smaller category than it sounds

Work out which door you have first, because the answer decides how much of the rest matters. A recessed entry, a covered porch, a door under a deep overhang, or an entry behind a storm door is sheltered. Rain reaches the hardware sideways in a gale and not otherwise, the temperature swing is damped, and a standard keypad deadbolt will be fine there for years.

Genuinely exposed openings make a shorter list: a yard or garden gate, the side door of a detached garage, an unsheltered entry on the windward wall of the house, an apartment breezeway that funnels wind and rain along its length. The crude test is to stand at the door in a downpour. If your shoulders get wet, so does the lock.

One case runs the other way. A storm door keeps rain off the entry door behind it and then bakes the pocket between the two on a sunny afternoon. Shelter and a comfortable operating temperature are not the same thing.

What an IP rating tells you, and what it leaves out

IP ratings come from IEC 60529 and carry two digits. The first covers solids, from nothing at 0 up to dust-tight at 6. The second covers water, running from dripping at 1, through splashing at 4, jets from a nozzle at 5, powerful jets at 6, and only then to temporary immersion at 7. An X in either position means that property was not tested or not claimed, which is not the same as being immune to it.

That makes the usual comparison misleading. IPX5 and IP65 offer identical water protection, because both were tested against jets from a nozzle. IP65 adds a dust-tight claim the other does not make. Both are laboratory results on a new sample using clean fresh water at room temperature.

What the rating never covers is most of what kills exterior hardware. Ultraviolet exposure is not in it, and neither is salt, freeze-thaw cycling, condensation forming inside a housing as a cold lock warms up in the morning, or five years of wind-driven rain working at a gasket that has gone hard. A lock can hold IP65 on the day it leaves the factory and let water past in its fourth winter.

Two practical notes. The rating is usually quoted for the exterior escutcheon, since the interior half lives indoors and is not rated at all. And it assumes the lock as designed, gasket included. Fitting the exterior half without the foam gasket, or over a ridge of old paint that holds it off the door face, throws the number away.

Cold: the operating window nobody reads

Any lock with electronics in it has a published operating temperature range, and manufacturers commonly state a lower limit somewhere around -4F. The figure lives in the manual rather than on the box. Check it against the coldest week your door actually sees rather than against the winter average, because this is a problem that only shows up in the outliers.

Below freezing the cells matter more than the electronics do. Alkaline capacity falls away in the cold, which is why a lock that has run happily since March gives up in a January snap. Lithium AA cells hold their output much lower down and would be the obvious answer, except that several lock manufacturers specify alkaline only and warn against lithium, because the different discharge behavior upsets their low-battery detection. Read the manual before upgrading anything, and understand why a lithium cell warns you so much later than an alkaline one before winter rather than during it.

Cold also thickens whatever grease is inside the lock body, so the motor works harder for the same throw, and it slows touchscreens.

Ice, swelling and the bolt that will not throw

Most winter lock complaints are not electrical at all. A door and its frame move with temperature and humidity, a wooden door swells through a wet autumn, and weatherstrip stiffens in the cold. The bolt then meets the lip of the strike plate instead of the middle of its hole, and a motor that had an easy job in September is driving a bolt against friction.

The lock’s own response disguises the cause. It stalls, retracts, beeps a jam alert and burns through cells several times faster than normal, so the owner fits new batteries, gets one good week, and concludes the lock is faulty.

The test takes a minute and needs no tools. Open the door and throw the bolt by hand with the thumbturn, then close the door and do the same. Smooth with the door open and stiff with it closed means alignment rather than the lock, and the strike plate is where to look. The other reasons a motorized deadbolt stalls before it finishes locking have their own walkthrough, including the lubrication question. That one matters here because the sprays people reach for in cold weather are the wrong ones for a lock with a motor in it.

Water freezing solid inside a bolt hole is rarer than the internet suggests. Gates are the exception, since a gate post moves with frost in the ground, and alignment that was right in October can be a quarter inch out in January.

Gates, garages and doors with nothing behind them

An exposed opening is usually a remote one as well, which brings a second set of problems that have nothing to do with weather. A gate at the end of a yard can be two walls and forty feet from the router, a long way for the 2.4 GHz radio in a battery-powered lock, and a detached garage often has no usable signal at all.

The ways around it are a mesh radio with a node close enough to reach, or an access point out at the building. Failing both, the lock becomes a keypad-only device with no remote anything, which is not a failure. A gate that opens on a code and reports to nobody is a perfectly reasonable gate.

The hardware has its own catches. Most residential deadbolts assume a door between 1-3/8 and 1-3/4 inches thick with a 2-1/8 inch bore, and gate stiles are frequently thinner, thicker or too narrow for the backset. A gate has no jamb either, only a post that shifts with the season.

Sun, heat and what three summers do to a finish

A dark door on a south or west elevation reaches surface temperatures well above the air temperature on a clear afternoon. A stated upper limit near 120F is closer than it sounds on a black steel door in Phoenix, and heat shortens battery life while it softens adhesive gaskets and lets plastic trim creep out of shape.

Ultraviolet light does the slower damage. It chalks painted and powder-coated finishes and hardens the rubber seals and O-rings that were keeping water out. The seal is usually what fails first, and it fails invisibly, which is why an exposed lock often starts misbehaving in year three or four rather than in its first storm.

Finish quality, unlike weatherproofing in general, is measurable. BHMA A156.18 sets out finish tests including salt spray and ultraviolet exposure hours, and a lock quoting a higher finish code has been through more of both. What the ANSI and BHMA grades actually measure is worth reading before paying a premium, because the grade covering the bolt and the code covering the coating are separate things. In broad terms a physical vapor deposition finish outlasts plating, and plating outlasts paint.

Near salt water, assume less. Salt attacks fasteners and plated finishes quickly, and manufacturers commonly exclude coastal installations from their finish warranties, so read the warranty rather than the marketing copy.

What to check before buying for an exposed opening

Six things settle it, and all six come from the manufacturer’s documentation rather than the retail listing.

• The IP rating, and whether it is quoted for the exterior escutcheon or for the whole lock.
• The operating temperature range at both ends, checked against the worst week your door sees.
• Which battery chemistry the manufacturer permits, since several forbid the lithium cells that would otherwise be the obvious cold-weather answer.
• Whether a gasket is supplied for the exterior half, and whether the instructions require it.
• A mechanical keyway, or some other way in for the morning the electronics have had enough.
• The finish, its warranty, and whether that warranty excludes coastal addresses.

One structural answer is worth keeping in mind. An interior retrofit unit puts every piece of electronics on the warm dry side of the door and leaves the weather-facing side as a plain mechanical cylinder. Of the categories sold as keyless entry, it is the one that sidesteps most of this article, at the cost of having no keypad outside.

Common questions

Do smart locks work in cold weather?

Down to the manufacturer’s stated limit, which is commonly around -4F, yes, with the caveat that battery performance drops off before the electronics do. Below that limit the behavior is undefined rather than guaranteed to fail. Most winter complaints trace back to weak cells or a door that has moved.

Is IP65 waterproof?

No. IP65 means dust-tight and tested against water jets from a nozzle. Immersion is a separate test at 7 and above, and no door lock needs it. What matters over years is ultraviolet light, salt and freeze-thaw, none of which the rating covers.

My lock gets slow and eats batteries every winter. Is it failing?

Probably not. Cold thickens the grease and the door and frame move, so the motor is working harder against a bolt that no longer lines up with its hole. Check the throw by hand with the door open and then closed. If it is stiff only when closed, adjust the strike before buying anything.

Can I put a smart lock on a garden gate?

Sometimes. The obstacles are usually a stile too narrow or the wrong thickness for a standard bore, a post that moves seasonally so the bolt fights the keeper, and no useful wireless signal that far from the house. Measure the stile and the backset before shopping, and expect a keypad-only lock rather than a connected one.

The short version

Decide first whether the door is genuinely exposed, because most are not, and a sheltered porch lets you ignore nearly all of this. If it is, read the manual rather than the box: the IP rating covers dust and water jets but never ultraviolet light, salt or freeze-thaw, and the operating temperature window and the permitted battery chemistry are the two figures that decide how the lock behaves in February. Expect the gasket and the seals to fail some years in, rather than the electronics. Keep a mechanical way in, because an exposed lock has more ways to have a bad morning than a sheltered one.