Handing: The One-Letter Answer That Holds Up Orders

Handing describes which edge of the door the hinges are on and which way the leaf swings. It matters because a lever has to fall the right way, a latch bolt has a bevel that must face the frame, and a lock case machined for one hand will not close on the other. It is one letter, and it is the field most often left blank on an inquiry.

There are two conventions and they do not agree. The European method stands the observer on the side the door opens TOWARDS (the pull side) and calls it left-handed if the hinges are on the left. The American method (ANSI/BHMA A156.115) stands the observer OUTSIDE, on the side you would use a key from, and adds "reverse" for doors that swing away. The same physical door is DIN-left under one and right-hand-reverse under the other.

This is why "left-handed" written on its own is not an answer. It is an answer plus an assumption about which convention the reader uses, and the reader is on the other side of the world. The unambiguous way to say it is to describe the door: hinges on the left as you look at it from the corridor, opens away from you. Twenty words that cannot be misread beat one letter that can.

A useful test that needs no convention at all: stand where the key goes. If the hinge knuckles are on your left, it is left-handed in the American sense; if the door swings towards you from there, it is "reverse". Say which side you stood on and the sentence survives translation.

Some hardware does not care. A tubular latch with a symmetrical bolt, most euro-profile cylinders, hinges themselves, and any lever on a sprung rose are non-handed (universal, marked U) and the field can honestly say so. Reversible latches are a middle case: the bolt can be turned on site, which is a real convenience and also a real opportunity for it to be turned wrong and never checked.

What is genuinely handed, and where the cost lands: mortise cases with an asymmetric follower, rebated pairs, and anything with a deadlocking latch whose auxiliary bolt has to meet the keep. On those, a wrong hand is not adjustable with a file, it is a different part number, and on a container quantity it is a different shipment.

The practical rule for a schedule: give every opening a hand or the word universal, and never leave the column empty. An empty column reads as "not decided yet" to the person filling the order and as "universal" to the person in a hurry, and only one of those is safe.

The four North American terms, spelled out

The European convention gives two answers; the North American one gives four, because it adds a second axis. Stand outside the door. Note which side the hinges are on and whether the door swings away from you or toward you. The four combinations are the four terms:

"Reverse" does not mean the door is unusual or that anything has been flipped. It only means the leaf swings toward the person standing outside, which is normal on an outward-opening exit door, and therefore very common on exactly the doors that carry panic hardware.

DIN, and what a European product code is telling you

European codes carry the hand as DIN left or DIN right, and DIN follows the European convention: look at the door from the side it opens toward you, and name the hand by the hinge side. A code ending in a DIN designation is describing the lock case, not the door, which matters, because a DIN-left case fitted into a door somebody has described as right-handed in the American sense is not necessarily a mistake.

That is the trap in one sentence: a door correctly called left hand in one convention can be correctly called right hand in the other, and nobody has made an error. It is why "LH" written alone on an international order is a coin toss, and why the convention has to be named next to the hand every single time.

Reversible, universal and field-reversible are three different claims

These three words get used interchangeably in catalogs and they promise different things. The difference decides whether your fitter can solve a problem on site or has to order another part.

When a supplier says "reversible" and means the first row, everyone is happy. When they say it and mean the second row, a fitter on a ladder discovers it at the worst possible moment. Ask which of the three it is, in those words, and ask whether the conversion needs the lock removed.

What our own catalog actually says

We should be concrete about this rather than general, because the answer surprised us when we counted it.

184 of our products publish a handing row. Of the 188 handing rows across them, 187 say the product is fully reversible or non-handed. Exactly one names a handed left-or-right choice.

So for almost everything we sell, the honest answer to "what hand do you need?" is: we do not need one. The lever, the latch, the rim lock, the cylinder, they go either way.

That is not a reason to stop recording the hand, and this is the part worth saying plainly rather than quietly. The hand is not a property of the hardware; it is a property of the opening. It still decides which way the door is hung, which side the frame is rebated, where the closer arm goes, which leaf of a pair is active, and which way round the strike is cut. Reversible hardware removes one of those decisions and leaves the rest untouched.

Where handing still bites, even with reversible hardware

The frame and the strike: a keep is cut into one side of the frame. Reversible hardware does not move it.

The closer: a closer arm, a backcheck setting and a parallel-arm bracket are all handed by the opening even when the lock is not.

Pairs of doors: which leaf is active is a handing decision, and it determines where the astragal, the flush bolts and the coordinator go.

Anything with a thumbturn, an indicator, or a lettered face: symmetry ends where text or an icon begins, an occupied/vacant indicator read upside down is a defect, not an orientation.

Anything bought elsewhere: the ironmongery on the same door from another supplier may be genuinely handed, and it will be handed in whichever convention that supplier writes in.

So record the hand on the schedule, name the convention beside it, and then enjoy the fact that on our side of the order it usually will not change what we ship.

Why there are two conventions at all

Neither convention is wrong, and knowing why they differ makes them easier to keep straight than memorising them does. They were written by two different trades solving two different problems.

The European convention was written from the leaf. A joiner hanging a door cares which edge the hinges go on, and answers that standing where the door opens toward him. One axis, two answers, and it maps directly onto the work.

The North American convention was written from the lock. A locksmith cares where the key goes, so the observer is planted on the outside face, and once you are standing outside, the swing direction becomes a second thing you can see and therefore a second thing worth naming. One axis became two, and two terms became four.

So the conventions are not rivals; they are answers to "which way does this door work?" asked by a carpenter and by a locksmith. That is also why the disagreement is so stubborn: each one is the shortest correct answer to its own question.

When the order already says "LH" and nobody knows whose LH it is

This happens constantly on a re-order, where the original schedule is gone and all that survives is two letters. Do not guess, and do not assume the previous supplier used your convention.

Go back to the physical evidence in this order. Look at an installed door from the original batch, if one exists, and record the three plain observations, hinge side, swing direction, which face is outside. Those three survive any convention. Failing that, look at the part itself: a handed lock case shows its latch bevel, and the bevel always faces the direction the door closes in. Failing that, ask what the original supplier's catalog said, because the convention is usually printed in its front matter even when the order is not.

If none of the three is available, say so on the order rather than picking one. On our catalog it will almost certainly not matter (see the count above) and on a genuinely handed item, an honest "hand unknown, please confirm" costs one email. A guess costs a container.

Where a model is handed, our specification tables state it as a row rather than burying it in the description, and the guided selector asks for it before it offers a model number, because the answer changes which models can be offered at all, not just which variant of one.