SORIVO · Cable Application Note · Thermal Rating · 2026-09

Cable Derating in a Heatwave: IEC 60287-3-1 and BS 7671 Appendix 4 Reference Conditions, Audited

A cable does not know what the weather is doing. It knows what its conductor temperature is, and the ampacity on the schedule was calculated from somebody’s assumption about the air and the soil around it. Here is that assumption, the correction arithmetic you can run yourself, and the six lines to put in a specification before the next hot July. If you came for the cross-section tables rather than the assumptions behind them, start from our cable ampacity chart by mm² (IEC 60364-5-52) and come back here for the derating layer.

Published 2026-09-03Updated 2026-09-03Reading ~11 minLevel: Spec / derating
Direct answer

Most European installation tables are built on 30 °C air and 20 °C ground, with a soil thermal resistivity around 2.5 K·m/W used as a conservative reference where no site data exists; AS/NZS practice is commonly built on 40 °C air and 25 °C ground instead. If your summer ambient is 40 °C, an XLPE 90 °C conductor keeps roughly 91 % of its tabulated ampacity before any grouping correction; at 50 °C it is about 82 %. A PVC 70 °C cable falls faster: roughly 87 % at 40 °C and 71 % at 50 °C.

For buried circuits the air number is the smaller half of the story. Published engineering references put the ampacity cost of soil drying from 1.0 to 2.5 K·m/W at somewhere between 15 % and 30 %, and the mechanism is self-reinforcing: the cable dries its own backfill, the backfill insulates it better, the conductor gets hotter, the soil dries further. That feedback is documented in peer-reviewed work; a specific public outage attributed to it is not something we could find. And no, we are not going to tell you the 28 April 2025 Iberian blackout was caused by heat — the final report says something else, and it is worth reading.

01What the number on your schedule is standing on

An ampacity is not a property of a cable. It is the result of a steady-state heat balance: I²R loss in the conductor, plus dielectric and armour losses, pushed out through insulation, bedding, sheath, and finally through whatever the cable is sitting in. The standard tables solve that balance once, for a fixed set of outside conditions, and print the answer.

The reference conditions behind the tables

Table 1 — Reference conditions the published installation tables are built on
InputIEC / BS system, as implemented in published toolsAS/NZS system, as implementedWhy it matters in a hot summer
Air ambient30 °C40 °CExposed runs, tray, ducts and risers in a roof space can exceed 30 °C for weeks, not hours
Ground ambient20 °C in the national implementations; IEC field reference conditions span 15–25 °C, and “standard UK conditions” uses 15 °C25 °CShallow buried circuits in a dry summer sit above every one of these numbers, which is the point
Soil thermal resistivityIEC 60287-3-1 tabulates 0.7 / 1.0 / 2.0 / 3.0 K·m/W by moisture state; 2.5 K·m/W is the conservative value when a site is unmeasured1.2 K·m/W commonly used as the design basisThe two systems are not comparable until this number is stated — and it usually is not
Maximum continuous conductor temperaturePVC 70 °C; XLPE and EPR 90 °Csame material limitsThis is the ceiling the whole calculation is built to reach, not a safety margin
Short-circuit limit (max 5 s)PVC 160 °C; XLPE/EPR 250 °Csame material limitsUnrelated to summer rating, but the number buyers most often confuse with it
Sources: the 30 °C air reference is stated in a German implementation of DIN VDE 0298-4 and in a BS 7671 Appendix 4 table guide. The ground value is not settled — national implementations use 20 °C, IEC field reference conditions span 15–25 °C, and “standard UK conditions” per BS 7769-3-1 uses 15 °C with 1.2 K·m/W — so both readings are shown rather than one being picked, because the CENELEC and BSI clause text is paywalled and we could not obtain it. The resistivity set is IEC 60287-3-1 as tabulated by an independent cable-calculation tool. Conductor and short-circuit temperature limits are corroborated by four independent cable-standard guides, match the figures already published on this site, and belong to IEC 60724 (1/3 kV) and IEC 60986 (6/30 kV) — not to IEC 60502-1, which is a product standard.
Where the argument usually diesTwo suppliers quoting different ampacities for the same cable are usually quoting different assumptions, not different cables. Before you compare a number, compare the ambient, the soil resistivity, the burial depth and the grouping the calculation used. Our derating guide breaks the correction factors down; this note is about the assumption underneath them.

02Ampacity vs temperature: the correction arithmetic, done in the open

You do not need a licence to check the temperature-correction half of this. The form the standards use is a square-root ratio of temperature differences, and we publish the same one in our derating guide:

Kθ = √[ ( θmax − θambient ) / ( θmax − θreference ) ]

Take θreference = 30 °C, θmax = 90 °C for XLPE and 70 °C for PVC, and the numbers fall out. We ran them; the result also lands within a point or two of the published code correction factors for the same insulation classes, which is the cross-check we wanted before printing anything.

Kθ worked through, against the published code factors

Table 2 — Kθ from the formula above, with a 400 A tabulated circuit as the worked example
AmbientXLPE 90 °C: KθXLPE: usable ampsPVC 70 °C: KθPVC: usable ampsPublished code factors, 75 °C / 90 °C insulation
30 °C1.000400 A1.000400 A1.00 / 1.00
35 °C0.957383 A0.935374 A0.94 / 0.96
40 °C0.913365 A0.866346 A0.88 / 0.91
45 °C0.866346 A0.791316 A0.82 / 0.87
50 °C0.816327 A0.707283 A0.75 / 0.82
Our own arithmetic on the formula above, not a transcription from any standard: 400 A is a round tabulated figure used only to make the percentage visible. The last column is the code correction factor published for the same ambient band, taken from two independent published implementations; the agreement is the point — the formula is doing what the tables do. Grouping, duct-bank and burial-depth corrections multiply on top of this; they are not included here, and neither is the load factor. A cable that is 60 % loaded all day does not need the same thermal headroom as one that runs at 95 % through the evening peak — see why rated current and actual current diverge.
SHARE OF TABULATED AMPACITY RETAINED, BY AMBIENT TEMPERATURE 100% 80% 60% XLPE, 90 C rated conductor PVC, 70 C rated conductor 100 100 95.7 93.5 91.3 86.6 86.6 79.1 81.6 70.7 30 C 35 C 40 C 45 C 50 C ambient air temperature → values from Table 2; axis starts at 60% so the differences are visible, not at zero
Figure 1 — The same data as Table 2, drawn so the divergence is visible: XLPE and PVC start together at the reference ambient and separate as it rises, because the 70 °C cable has far less temperature difference left to dump heat into. Values are our calculation from the square-root correction formula in section 02, not transcribed standard tables.

03The buried half: the soil is the rating

Air temperature is the part buyers argue about. Soil is the part that quietly decides the answer, because for a buried circuit the ground is the heat sink, and its ability to move heat is a material property that changes with moisture.

Thermal resistivity is written ρ and measured in K·m/W — higher is worse. The standard tabulates it in four steps by moisture state, 0.7, 1.0, 2.0 and 3.0 K·m/W, and recommends taking the value from the reference operating conditions rather than from a soil label. In practice 2.5 K·m/W is the conservative number used when a site has not been measured, and UK practice for unmeasured ground often assumes 1.2 K·m/W. Our own soil resistivity note uses 1.0 K·m/W as the BS 7671 method-D reference and 2.5 as the no-data design value, and reports that specifying a good 1.2 K·m/W imported backfill buys back roughly 12–20 % of ampacity compared with leaving poor native soil in place. If you want the value measured rather than assumed, the guide for doing it is IEEE 442, IEEE Guide for Thermal Resistivity Measurements of Soils and Backfill Materials.

The published penalty, and what it is worthEngineering references agree on the direction and roughly on the size: doubling ρ costs about 15–25 % of rating, and moving from 1.0 to 2.5 K·m/W is quoted at about 25 % in one cable-calculation reference and 30 % in another. Treat those as tool-and-vendor figures, not standard values: we did not find a two-source percentage printed in the standards themselves, and the true number depends on depth, spacing, load factor and how much of the circuit sits in the dried zone. Their honest use is as a range to stress-test a design against, not as a line in a calculation note.

Then the part that turns a correction into a mechanism. Current heats the conductor; the conductor heats the soil immediately around it; water moves away from the hot zone; dry soil resists heat flow harder; the conductor gets hotter for the same current; the drying front moves further out. This is not folklore — it is the subject of an IEEE experimental study on drying-out of soil around underground cables, of an IET paper on cables in soils whose resistivity depends on temperature, and of work on thermal instability in buried cable backfills. Our own burial-spacing note describes the same loop and the 105–110 °C emergency overload band that an XLPE cable can sit in while it happens.

What we could not findWe searched three times for a documented, publicly attributed network outage caused specifically by soil drying around a buried circuit, and did not find one. The mechanism is well documented in the engineering literature; a named incident is not. Anyone quoting a blackout as proof of the drying effect is borrowing authority they have not earned, including, on the evidence, some very confident articles.

04What the 2025 reports actually said

Because this is where the easy version of this story goes wrong. The 28 April 2025 loss of the Spanish and Portuguese systems produced a 472-page ENTSO-E expert panel final report, published on 20 March 2026. Its finding is a systemic failure of voltage management, not of thermal capacity: most renewable plant in Spain was running in fixed power-factor mode rather than voltage-control mode; conventional units missed their reactive-power references; shunt reactors were switched manually; a national derogation let 400 kV networks sit up to 435 kV, above the 380–420 kV band in the system-operations guideline; and a significant number of generator overvoltage protection settings did not comply with the requirements for grid connection. Over 700 MW of solar and thermal generation tripped on overvoltage in the first seconds, and the voltage climb became self-reinforcing.

The report is blunter still about the weather. In its own description of the day it records that the incident happened on a typical spring day in Spain, with mild temperatures and mostly sunny weather. Heat is not the culprit in that account, and neither is thermal capacity — the report does not turn on cables running hot. That matters for a buyer, because the fixes it points at (voltage-control mode instead of fixed power factor, automatic control of shunt reactors, audits of overvoltage protection settings, wider PMU coverage, periodic black-start testing) are not the same purchases as a bigger conductor.

We also dropped an event from this article during the checking. The trade press described a 24 July 2025 trip on the French–Spanish interconnection as heat-related, and we could not establish an official causal finding for it from a primary source, in several attempts, with the sources we could reach disagreeing with each other. So it is not in here. Heat affects the rating of a specific circuit, and that we can calculate. Heat caused a specific major outage is a claim we are not going to make.

Why this section is in a cable articleBecause the same confusion shows up in specifications. We are regularly asked for a cable that will “handle the heatwave conditions of the design year”, which is not a defined requirement, next to a request for a circuit whose ambient, burial depth and soil resistivity are stated — which is. The second one we can quote against.

05Dynamic line rating: what exists for buried cable, and what we could not find

“Dynamic line rating” means calculating the real ampacity from real weather instead of a conservative seasonal assumption. For overhead lines this is a mature, standardised field: IEEE 738, current edition 738-2023, is titled IEEE Standard for Calculating the Current-Temperature Relationship of Bare Overhead Conductors and gives the numerical method relating conductor core and surface temperature to current and weather conditions — with the standard itself noting that it does not recommend which weather data or conductor parameters to use. CIGRE’s overhead-line work covers the same ground (a technical brochure on the thermal behaviour of overhead conductors, and a 2025 ELECTRA article on forecasting dynamic thermal ratings).

For buried cables the picture is different. The steady-state method is IEC 60287-1-1, which is built on a 100 % load factor, and the standard that covers time-dependent loading of buried circuits is IEC 60853 — cyclic and emergency ratings derived from a load profile, not from live weather and soil data. We searched for a standard doing for buried cable what IEEE 738 does for a span of conductor and did not find one; what exists on underground dynamic thermal rating is research literature, not a normative method. That is a statement about our search rather than a claim that nothing exists — but it is the practical reason a buried circuit’s “dynamic rating” is usually sold as instrumentation plus an engineering study rather than as a standard you can cite in a tender.

Where the numbers are documented — all overhead, and all worth reading as ranges rather than promises. The US Department of Energy has reported dynamic rating increasing transfer capability by 10–40 % on monitored circuits, and separately that some lines can carry around 50 % more than their labelled limits. CIGRE’s own framing is more conservative: roughly 10 % above static rating for most of the time. At project level, Belgium’s TSO Elia has reported taking a monitored corridor from 200 to 600 MVA, and in May 2026 National Grid reported a deployment across 585 km of routes with an expected uplift of about 8 %. Those last two come from a single report each, so we treat them as leads, not benchmarks, and no number from this paragraph appears in any table on this page.

Policy, briefly, because buyers ask — and because it is easy to get wrong. The Commission’s European Grids Package (COM(2025) 1005 final, 10 December 2025) does talk about grid-enhancing technologies and puts a number on them: their use can increase overall network capacity in Europe by 20 % to 40 % by 2040, with cost reductions of 35 %. What it does not do is name dynamic line rating — the term appears nowhere in that document, in its press release IP/25/2945 or in its Q&A QANDA/25/2946. The words appear one document over: the proposed TEN-E revision (COM(2025) 1006 final) carries a recital saying non-wire solutions such as dynamic line and transformer rating should play a greater role, and defines non-wire solutions by reference to grid-enhancing technologies. The older Directive (EU) 2019/944 already told operators to use dynamic line rating where appropriate. A cable buyer’s conclusion is narrow: policy is pushing the study and the instrumentation, not the copper.

06Six lines to put in the specification

  1. Design ambient, stated with a basis. Not “hot”: an air temperature for exposed and duct runs, a ground temperature for buried runs, and the reference the tables were built on, so the correction is unambiguous.
  2. Soil resistivity, measured or declared. Ask for a site ρ value with its test method, or accept a stated design value of 2.5 K·m/W and size accordingly. A buried circuit designed on 1.0 K·m/W without site data is a cable that will run hot.
  3. Backfill, not just trench. Specify the imported backfill’s target resistivity and compaction. This is the cheapest ampacity on the market — our burial-spacing note quantifies what a 1.2 K·m/W surround buys back compared with native soil left in place.
  4. Installation method and grouping. The correction only means something against the method table it came from: free air, duct, direct buried, tray, and the number of current-carrying conductors together. See the comparison by installation method.
  5. Termination and joint temperature class. The connection is usually the first thing to get hot and the last thing to be measured. Silicone cold-shrink bodies are rated over a wide temperature range and survive thermal cycling better than a taped joint that was never torqued to a schedule.
  6. What gets monitored. If the circuit will run near its thermal limit, decide now whether you want ammeter/temperature instrumentation, a rated-temperature margin, or a larger cross-section. Those are three different purchase orders, and the cheapest one is not always the first.

For the sizing arithmetic itself, the engineering tools page collects the calculators we use, and the IEC 60287 explainer covers where the losses in the heat balance come from. For the extreme-heat selection side — jacket chemistry, UV and the 51 °C question — our heatwave cable selection guide is the companion piece.

Quote against your numbers, not oursSend the design ambient, burial depth, soil resistivity (or say plainly that you do not have it) and the method table you are working from, and we will quote armoured LV cable and MV terminations against that basis — with the correction assumptions printed on the quotation: request a quote.
Related product · what Sorivo publishes for hot, buried and exposed runs
CU/XLPE/SWA/PVC 0.6/1 kV armoured LV power cable
ApplicationDistribution feeders, direct-burial and duct runs, industrial and renewable sites
StandardsIEC 60502-1; conductor to IEC 60228 Class 2 (as published on the product page)
ConductorAnnealed copper, Class 2 stranded compacted, 1.5–1000 mm²
InsulationXLPE; 90 °C continuous, 105–110 °C emergency overload, 250 °C short-circuit (5 s)
Voltage Rating0.6/1 kV (Um = 1.2 kV); 3.5 kV / 5 min high-voltage test
Temperature90 °C conductor rating; SWA armour with PVC ST2 outer sheath
Silicone cold shrink termination and joint, 10 kV–35 kV
ApplicationIndoor and outdoor terminations and straight joints on MV cable — the thermal bottleneck in most buried circuits
StandardsIEC 60502; GB/T 12706.4; IEEE 48/404 (as published on the product page)
ConductorFor conductors 25–1000 mm²
InsulationLiquid silicone rubber (LSR); dielectric strength ≥26 kV/mm
Voltage Rating8.7/10, 12/20, 26/35 kV classes
Temperature−50 °C to +130 °C operating range; IP68; UV and ozone resistant
Specifying for a hot site? Send the ambient, burial depth, soil resistivity (or say you do not have it) and the method table you are working from, and we will quote against your numbers, not ours: request a quote.
Q1What ambient temperature are cable ampacity tables based on?

In the IEC and BS system, the published installation tables are built on 30 °C air and 20 °C ground, with soil thermal resistivity around 2.5 K·m/W used as a conservative reference where a site has not been measured. AS/NZS practice commonly uses 40 °C air and 25 °C ground. Because the reference conditions differ, ampacity figures from the two systems are not directly comparable — and neither is either of them valid for a roof space that sits at 45 °C in August.

Q2How much ampacity do I lose at 40 °C ambient?

For an XLPE cable rated for a 90 °C conductor, the square-root temperature correction gives about 0.91 — roughly 9 % of the tabulated ampacity gone before any grouping correction. For a PVC cable rated at 70 °C, the same ambient costs about 13 %, and at 50 °C it costs about 29 % because the cable has so little temperature difference left to reject heat into. Grouping, burial depth and load factor all multiply on top of that.

Q3Does soil really matter more than air temperature?

For a buried circuit, yes. The ground is the heat sink, and published engineering references put the ampacity cost of soil drying out from 1.0 to 2.5 K·m/W at roughly 15–30 % — comparable to or larger than a very hot summer day, and it persists for weeks. Worse, the drying is self-reinforcing: the cable dries its own backfill, which insulates it further. That mechanism is documented in peer-reviewed studies of buried-cable thermal instability; a specific public outage attributed to it is not something we could find.

Q4Is dynamic line rating a thing for underground cables?

For overhead lines it is standardised: IEEE 738 (current edition 738-2023) calculates the current-temperature relationship of bare overhead conductors from current and weather data. For buried circuits we could not find an equivalent dynamic-rating standard, even though real-time thermal monitoring of underground cable is installed and sold. Practically, that means a buried circuit’s rating is a static calculation you either trust with margin or instrument — not something you can re-rate on the fly the way an overhead span can.

Q5Did the 2025 Iberian blackout happen because of heat?

No, on the evidence of the final report. ENTSO-E’s expert panel report of 20 March 2026 attributes the 28 April 2025 loss of the Spanish and Portuguese systems to a systemic failure of voltage management — fixed power-factor operation of renewables, missed reactive-power references, manually switched shunt reactors, 400 kV networks permitted up to 435 kV, and non-compliant generator overvoltage protection settings. Heat and conductor thermal ratings are not the mechanism in that account, and we would rather say so plainly than borrow drama from an event that does not support it.

Q6Should I size the cable from the worst day or the average day?

From the worst day you are willing to design for, and then check the average day for cost. A conductor sized on a 30 °C assumption in a 45 °C climate is running above its design temperature for weeks, and thermal ageing is cumulative — the familiar “every 8–10 °C halves the life” rule of thumb comes from oil-impregnated paper, not from XLPE, so treat it as a warning rather than a formula. What we ask suppliers for instead is thermal endurance data generated to the IEC 60216 method, and a stated design ambient we can both sign up to.

Quote against your site’s real numbers

Send the design ambient, burial depth, soil resistivity or backfill spec, installation method and circuit loading, and we will quote armoured LV cable and MV terminations against that basis — with the calculation assumptions written on the quotation.

Ask a Quote Book a Sample Kit
Senior cable application engineer at Sorivo
Reviewed by Luo Qiang — Senior Cable Application Engineer, Sorivo
15+ years in industrial and renewable energy cable specification. Experienced in cable specification aligned with IEC standards.

Sources checked for this article

  • Reference ambient and soil conditions, and soil resistivity ranges: a multi-standard cable-sizing comparison, a published BS 7671 cable-factor calculator, and a cable-engineering note on soil thermal resistivity. The CENELEC and BSI clause text is paywalled and was not obtained.
  • Thermal limits for PVC and XLPE: three independent cable-standard explainers, cross-checked against our own published derating notes and IEC 60287. The time-dependent loading method referenced in section 05 is IEC 60853.
  • Soil drying and thermal instability: IEEE Experimental study for drying-out of soil around underground power cables; IET Thermal analysis of power cables in soils of temperature-responsive thermal resistivity; IEEE work on thermal instability in buried cable backfills.
  • Dynamic rating: IEEE 738-2023 (IEEE Standards Association); CIGRE ELECTRA N°341 (August 2025) on forecasting dynamic thermal line ratings. Reported uplift figures come from single sources and are attributed as such in the text.
  • 28 April 2025 incident: ENTSO-E expert panel final report published 20 March 2026, plus two independent summaries of its findings.
  • EU policy: Commission European Grids Package (COM(2025) 1005 final), press release IP/25/2945, Q&A QANDA/25/2946, guidance C/2025/8473, EPRS legislative briefing, ENTSO-E package page.
  • Soil resistivity measurement method referenced in section 03: IEEE 442, IEEE Guide for Thermal Resistivity Measurements of Soils and Backfill Materials.

Related reading

Engineering figures in this article are our own calculation from the stated formula and assumptions, and are for specification discussion, not a substitute for a design calculation sealed by a responsible engineer. Standards, tables and incident findings change; the sources above were checked on 2026-09-03.