Undersized conductors: how short cross-sections happen

What undersized really means

Buyers usually think of conductor size as a physical cross-section: 2.5 mm2, 25 mm2, 95 mm2. IEC 60228, the standard most of the world uses for conductors, does not define size that way in practice. It sets a maximum DC resistance at 20 C for each nominal size and class. A conductor is acceptable if its resistance is at or below that limit, whatever its exact geometric area. An undersized conductor is therefore one whose resistance is above the limit for the size printed on the sheath.

That definition matters because it tells you how to test. Measuring strand diameters with a micrometer is useful, but the decisive test is resistance, corrected to 20 C, against the table. It also explains why undersizing is attractive to a dishonest maker: copper and aluminium are roughly 80% of a typical cable's cost according to listed manufacturers' disclosures, so every gram of metal removed is margin, and nothing on the outside changes. Our conductor materials page covers the classes and metals in more detail.

How short cross-sections happen

Few undersized conductors are accidents. The common routes are:

• Fewer or thinner wires. A stranded Class 2 conductor built with one or two wires fewer than the design, or with wire drawn slightly fine. The conductor still looks round and full under the insulation.
• Poor metal. Recycled copper of low conductivity needs more area to reach the same resistance; using it at the nominal area produces a conductor that fails resistance even if its size is right.
• Substitution. Copper-clad aluminium, or steel wires mixed into a copper conductor, presented as copper. BASEC has reported finding steel wires, copper-clad aluminium and poor-quality recycled copper in cables that should have been copper.
• Price pressure. When copper prices rise, the temptation to use less copper rises with them. LME three-month copper reached an intraday record of USD 14,875 per tonne on 10 September 2026 (Fastmarkets), the kind of market in which buyers should check resistance records more, not less.
• Loose specifications. A purchase order that says "4x25 copper cable" with no standard and no resistance requirement gives a supplier room to build to its own idea of 25.

A warning from the UK market

The UK certification body BASEC issued a public warning about a multicore XLPE/SWA/PVC armoured cable marked as made to BS 5467 but not BASEC approved. Testing found the conductor resistance and the armour resistance above the limits; the effective cross-section was up to two sizes below the marking. The two consequences are different and both serious. An undersized conductor can be overloaded at a load the design assumed was safe. Armour with excessive resistance may not carry fault current well enough to work as the circuit protective conductor, so earthing can fail. Neither problem is visible on site. BASEC's advice to buyers is to look for its name printed on the cable, because a standard number alone is not evidence of approval.

The resistance table is your ruler

Keep the IEC 60228 limits next to every test report you receive. The values below are the maximum DC resistance at 20 C for plain copper and aluminium conductors.

A conductor one size smaller sits well above the next limit: a real 16 mm2 copper conductor has a limit of 1.15 ohm/km, far above the 0.727 allowed for 25 mm2. That gap is why a correctly measured resistance test catches undersizing reliably, and why a report that only states "pass" without the measured value is not good enough.

Two details trip buyers up when reading reports. First, use the right column: flexible Class 5 conductors have higher limits than Class 1 and 2 of the same size, and tinned Class 5 conductors higher again; the tinned 4 mm2 conductor in an H1Z2Z2-K solar cable is allowed 5.09 ohm/km against 4.95 for plain Class 5. Comparing a flexible cord's result with the Class 2 column will make a good cable look bad, and the reverse will hide a short one. Second, check the temperature correction. Resistance rises with temperature, so a value measured in a warm test room and not corrected to 20 C overstates the resistance, while a careless correction can understate it. Ask for the measured value, the measuring temperature and the corrected value on each record.

Checks that catch it early

1. Specify the standard and class. Write "annealed copper conductor, Class 2 to IEC 60228" and the size on the spec sheet, so resistance limits are contractual.
2. Ask for measured values per drum. Resistance in ohm/km, temperature-corrected, against the limit. Values sitting exactly at the limit on every drum deserve a second look.
3. Weigh a sample. Weight per metre compared with the datasheet is a quick screen. The reference 0.6/1 kV Cu/XLPE/PVC 4x25 mm2 construction on our LV power cable page is about 1,292 kg/km per the partner catalogue.
4. Cut and count. At inspection, strip a sample, count strands and measure wire diameters against the plant's own design sheet.
5. Test independently when stakes are high. A third-party lab resistance test on samples from random drums costs little next to a re-pull.

These checks belong in the pre-shipment inspection, and the desk builds them into the inspection plan by default. The broader list of failure modes is in the cable problems guide.

What it costs when it is missed

Cable sizing is done by the project engineer for a given installation method under IEC 60364-5-52, BS 7671 or the NEC as applicable. Every sizing calculation assumes the conductor meets its resistance limit. An undersized conductor quietly breaks that assumption: more voltage drop, more heat in the same duct or tray, and less margin under fault. Once cable is installed and covered, replacing it costs far more than the copper ever saved. That is why resistance records, not unit price, should be the first thing compared between two offers. To compare offers on that basis, send both constructions through the quote form and the desk will line them up layer by layer.