Why Mirrors Break in Transit — and What ISTA 3A Actually Tests

Short answer: mirrors do not usually break because someone dropped the carton. They break because of edge impact, panel flex, unsupported profile points, stack compression and sustained vibration — five mechanisms that a carton designed by eye will not survive and that ISTA 3A testing is designed to reproduce. If your supplier cannot describe which of the five their packaging addresses, the damage rate is a matter of luck.

Failure mode one: edge impact

Glass is strong in compression and weak at the edge. Almost every fracture in a shipped mirror starts at a chip or a micro-crack on the edge and runs inward. This is why edge work matters as much as packaging: a panel that has been cut but not properly seamed and arrised arrives at the packing line already carrying the origin of its own failure.

The packaging answer is a rigid corner and edge protection system — EPE profiles that hold the glass edge away from the carton wall by a margin, rather than foam sheet wrapped around a panel. If you can press the carton wall and feel the glass edge through it, the mirror is unprotected in the one place it matters.

Failure mode two: panel flex

A large thin panel behaves like a spring. Set it flat and stack weight on it, or set it upright and let it bow in a rolling container, and it will flex until it fails at the point of maximum curvature — usually the centre, which is why large mirrors sometimes arrive with a clean crack straight across.

The answer is honeycomb board or a rigid backer bonded across the panel face, which converts a flexible sheet into a stiff assembly. For anything above about 1200 mm this is not optional. For panels above 1800 mm, a vertical timber crate with interleaved separators is the only approach that reliably survives sea freight, because it constrains the panel in the axis it wants to flex in.

Failure mode three: unsupported profile points

This one is specific to shaped mirrors and it is the reason organic profiles have a higher damage rate than rectangles. A cloud, wave or pebble profile has concave nodes where the glass narrows and where there is no straight edge to bear against the carton. Under vibration those points move independently and fracture.

The answer is rigid inserts placed at each concave node, cut to the profile rather than generic. It makes the carton larger and the packing slower, and it is the difference between a 1.5% damage rate and a 6% one on the same product. When a supplier quotes a shaped mirror at the same packing cost as a rectangle, this is what has been left out.

Failure mode four: stack compression

Cartons at the bottom of a container stack carry the weight of everything above them for the length of the voyage, in humidity that softens corrugated board. A five-layer carton at 60% humidity has materially less compressive strength than the same carton dry, which is why a stack that was stable at the factory can be a slumped mass at the destination.

The answer is a carton specification stated as a burst or edge-crush value rather than as a layer count, plus a stacking limit marked on the carton and communicated to the loading team. A "five-layer carton" with no stated strength value tells you how many plies there are and nothing about whether they hold.

Failure mode five: sustained vibration

Road transit to and from the port produces continuous low-amplitude vibration at frequencies that can match the resonance of a packed panel. Over hours, resonance loosens fillers, works corner protectors out of position and abrades the mirror backing. Damage from vibration often does not appear as breakage at all — it appears as scuffing on the rear coating, which becomes black edge six months later.

The answer is a packing system with no free movement in any axis and a backing surface that is protected rather than left against bare board.

What ISTA 3A actually is

ISTA 3A is a performance test procedure for packaged products shipped through parcel and general freight systems. It subjects a packed carton to a sequence that includes atmospheric conditioning, drops from specified heights at specified faces and corners, random vibration on a shaker table, and compression loading representing stack weight.

Two points are worth understanding. First, it tests the package, not the product: passing means the packaging protected the item through that sequence. Second, it is a test procedure rather than a certification — there is no ISTA licence hanging on a factory wall. What exists is a test report for a specific packed configuration of a specific product, and that is the document to ask for.

A supplier saying "ISTA 3A packaging" without a report for the configuration you are buying is describing an intention. Ask which product was tested, in which carton, at what gross weight, and when.

What to specify, and what to accept

• Edge and corner protection stated as a material and a thickness, not as "foam".
• Face stiffening — honeycomb or rigid backer — on any panel above 1200 mm.
• Profile-cut inserts at concave nodes on shaped products.
• Carton strength stated as a burst or edge-crush value, with a marked stacking limit.
• A test report for the packed configuration you are buying.
• A spare-parts allowance — because on a category with genuine transit risk, an allowance is more honest than a zero-damage claim.

Summary

Mirror damage is not random. It comes from five identifiable mechanisms, each with a specific packaging answer, and ISTA 3A is the procedure that tells you whether the answers work. Mirrorwright packs every category to a tested ISTA 3A configuration and includes a 3% free spare-parts kit with every order. Ask us for the test report for the models you are considering.