The current rating printed in the tables is not the current your cable can carry. It is the current the cable could carry under reference conditions your installation almost never matches: 30 degrees ambient, one circuit alone, no insulation touching it, standard soil, standard depth.
Every departure from those conditions attracts a correction factor, and the factors multiply. A cable can lose half its tabulated capacity without a single dramatic condition, just several ordinary ones stacking. Here are the seven that matter under BS 7671:2018+A4:2026, with the actual numbers.
Ca: ambient temperature, Tables 4B1 and 4B2
Ratings assume 30 degrees in air. A 70 degree thermoplastic cable in a 40 degree plant room keeps 0.87 of its rating. At 50 degrees it keeps 0.71. Thermoset cable survives heat better, holding 0.91 at 40 degrees, which is one of the quiet arguments for specifying it in hot spaces.
Buried cables reference 20 degrees ground temperature instead, via Table 4B2, and warm ground bites harder than most designers expect: 30 degree soil takes a thermoplastic cable down to 0.89.
Cg: grouping, Tables 4C1 to 4C6
The steepest factor on the list. Circuits bunched together heat each other, and Table 4C1 prices it: two circuits bunched keep 0.80, four keep 0.65, nine keep 0.50, twelve keep 0.45. A single layer on a perforated tray is kinder, four circuits keeping 0.77, and spacing cables apart recovers capacity, which is why containment layout is a design decision rather than an installer's preference.
Ci: thermal insulation, Table 52.2 and Regulation 523.9
A cable surrounded by thermal insulation for 0.5 m or more keeps exactly half its rating. Shorter contact lengths derate on a sliding scale from Table 52.2: 0.88 at 50 mm, 0.78 at 100 mm, 0.63 at 200 mm. In a modern insulated building this factor applies far more often than it gets applied.
Cs and Cd: soil resistivity and burial depth, Tables 4B3 and 4B4
The tables assume soil at 2.5 K.m/W and burial at 0.7 m. Real ground varies enormously, and the correction runs both directions: wet ground at 1.0 K.m/W uprates a direct buried cable by 1.5, while dry ground at 3.0 derates it to 0.9. Depth matters less but measurably, 0.92 at 2 m. Amendment 4's split of Method D into D1 and D2 made these corrections asymmetric between ducted and direct buried cables, which I covered in detail in the Method D article.
Cc: the buried overload factor everyone forgets
Where a buried circuit relies on its protective device for overload protection, a further 0.9 applies. It lives in the Appendix 4 preliminary text rather than a table, which is exactly why it is the most commonly missed factor on this list.
Harmonic rating factor, Table 4Aa
Third harmonic currents from electronic loads sum in the neutral instead of cancelling. Between 15 and 33 percent third harmonic content, the cable keeps 0.86 of its rating. Above 33 percent, the neutral becomes the sizing conductor, which is a different article in itself.
How they combine, and what that does in practice
The factors multiply. Take an unremarkable circuit: 70 degree thermoplastic cable in conduit through a 40 degree plant room, bunched with three other circuits.
Ca 0.87 × Cg 0.65 = 0.566.
The cable delivers 57 percent of its tabulated rating. A cable tabulated at 24 A is actually good for 13.6 A, and if the designer sized it against 24 A, the circuit is overloaded on paper the day it is energised, with every conductor running hotter than its insulation was rated to survive long term.
That is the whole reason the factors exist. They are not pessimism, they are the physics of heat finding its way out of a conductor, priced per condition.
A detail from implementing this
While validating CurrentProof's buried cable calculations, we found that the soil and depth factors, Cs and Cd, were sitting correctly in our database and being ignored by the code that should have read them. The values were right the whole time. The query never asked for them. Every buried result was silently uncorrected until the audit caught it.
I mention it because the same failure exists in spreadsheets everywhere: a corrections tab that exists, is correct, and is not referenced by the cell doing the sizing. The dangerous factor is never the one you got wrong, it is the one that never entered the calculation at all.