12V Wire Size Calculator
12V DC wire sizing is fundamentally a voltage-drop-driven calculation, not an ampacity-limited one. The core problem is arithmetic: at only 12V nominal, every milliohm of conductor...
Formula
Source: ABYC E-11, SAE J1128, NEC Table 310.16 (ampacity reference) | Last reviewed: July 26, 2026
Examples
20 Amps
= 0.001 AWG
- length = 25
- voltage_drop_pct = 3
20 A at 12V 25 ft → #6 AWG to maintain <3% VD
100 Amps
= 0.003 AWG
- length = 6
- voltage_drop_pct = 3
100 A inverter 6 ft → #2 AWG to maintain <3% VD
5 Amps
= 0.0005 AWG
- length = 40
- voltage_drop_pct = 3
5 A LED lights 40 ft → #10 AWG for <3% VD
Quick Reference Table
| Amps | 5 ft | 10 ft | 15 ft | 20 ft | 25 ft | 30 ft |
|---|---|---|---|---|---|---|
| 5 A | #16 | #14 | #12 | #12 | #10 | #10 |
| 10 A | #14 | #12 | #10 | #8 | #8 | #6 |
| 15 A | #12 | #10 | #8 | #6 | #6 | #4 |
| 20 A | #10 | #8 | #6 | #6 | #4 | #4 |
| 30 A | #10 | #6 | #4 | #4 | #2 | #2 |
| 50 A | #8 | #4 | #2 | #1 | #1/0 | #1/0 |
| 75 A | #6 | #2 | #1/0 | #1/0 | #2/0 | #2/0 |
| 100 A | #4 | #1 | #1/0 | #2/0 | #3/0 | #4/0 |
| 150 A | #2 | #2/0 | #3/0 | #4/0 | 250 kcmil | 350 kcmil |
| 200 A | #1 | #3/0 | #4/0 | 250 kcmil | 350 kcmil | 500 kcmil |
| AWG | Ampacity (A) | AWG | Ampacity (A) |
|---|---|---|---|
| #18 | 20 | #4 | 160 |
| #16 | 25 | #2 | 210 |
| #14 | 35 | #1 | 245 |
| #12 | 45 | #1/0 | 285 |
| #10 | 60 | #2/0 | 330 |
| #8 | 80 | #3/0 | 385 |
| #6 | 120 | #4/0 | 445 |
Where is this used?
(1) RV battery-to-inverter cabling — the single most critical 12V circuit.
A 2,000W inverter at 12V and 90% efficiency draws 2,000 ÷ (12 × 0.90) = 185 A at full load.
For a 3 ft one-way run at 3% VD: CM = (2 × 3 × 12.9 × 185) ÷ (12 × 0.03) = 14,319 ÷ 0.36 = 39,775 CM → #4 AWG (41,740 CM).
But check ampacity: #4 AWG at 105°C marine rating is 160 A — too low for 185 A.
The ampacity constraint pushes to #2 AWG (210 A marine, 66,360 CM) which also improves VD to 14,319 ÷ 66,360 = 0.216V (1.80%).
For a 3,000W inverter at 278 A: CM = 21,516 ÷ 0.36 = 59,767 CM → #2 AWG (66,360 CM) with VD = 0.324V (2.70%).
Most inverter manufacturers recommend 2/0 AWG minimum for 3,000W at 12V, reflecting the practical need for minimal VD at the inverter's surge rating (typically 2× continuous for 5-10 seconds, so 556 A for a 3,000W unit).
(2) Solar charge controller-to-battery connection — the voltage accuracy of this connection directly determines charging performance.
A 40 A MPPT controller at 7 ft with 1% VD target (0.12V): CM = (2 × 7 × 12.9 × 40) ÷ 0.12 = 7,224 ÷ 0.12 = 60,200 CM → #2 AWG (66,360 CM).
This seems large for 40 A, but the 1% target is unforgiving.
Many installations use undersized #8 or #10 AWG here, resulting in 0.3-0.5V drop that causes chronic undercharging — the charge controller believes it's delivering 14.4V but the battery terminals see only 13.9-14.1V.
(3) Automotive starter cable sizing — the constraint is not voltage drop percentage but absolute voltage at the starter during cranking.
A V8 starter drawing 250 A at 12V through 4 ft of cable: the cable must deliver ≥ 9.6V at the starter motor (SAE cold cranking requirement of 7.2V per 12V divided across two batteries in series for 24V systems, but for 12V systems, ≥ 9.6V at 0°F is standard).
The cable voltage drop budget is 12.6V (fully charged battery at rest) − 9.6V = 3.0V.
CM = (2 × 4 × 12.9 × 250) ÷ 3.0 = 25,800 ÷ 3.0 = 8,600 CM → #10 AWG (10,380 CM).
But at −18°C cold cranking, the battery voltage sags to ~10V, reducing the budget to 10 − 9.6 = 0.4V, requiring CM = 25,800 ÷ 0.4 = 64,500 CM → #2 AWG (66,360 CM) — which is why automotive starter cables are typically #4, #2, or #1/0 AWG despite carrying 'only' 250 A.
(4) Marine navigation lights — ABYC E-11 §11.14.1 mandates ≤ 3% VD for navigation lights.
A masthead anchor light drawing 1.5 A at 12V on a 35 ft sailboat with 45 ft total circuit length (up the mast, through deck penetrations): CM = (2 × 22.5 × 12.9 × 1.5) ÷ (12 × 0.03) = 871 ÷ 0.36 = 2,419 CM → #16 AWG (2,580 CM) minimally, #14 AWG (4,110 CM) preferred for corrosion margin.
The corrosion issue is significant in marine environments: saltwater exposure corrodes copper, reducing effective cross-section over time; the ABYC recommendation is to up-size by one AWG step for circuits exposed to bilge or weather.
(5) RV DC distribution to appliances — a 12V compressor fridge (Danfoss/Secop BD35, 5 A running, ~7 A startup) at 20 ft: CM = (2 × 20 × 12.9 × 7) ÷ (12 × 0.03) = 3,612 ÷ 0.36 = 10,033 CM → #10 AWG (10,380 CM) for 3% VD.
If the same circuit uses #12 AWG (6,530 CM): VD = 3,612 ÷ 6,530 = 0.553V (4.6%), the compressor runs longer to achieve the same cooling because the reduced voltage causes it to spin ~5% slower and draw proportionally more current, compounding the VD problem.
(6) Off-road vehicle auxiliary lighting — a 300W LED light bar (25 A at 12V) on a 4×4 with the battery under the hood and lights on the roof rack (15 ft one-way routing): CM = (2 × 15 × 12.9 × 25) ÷ (12 × 0.03) = 9,675 ÷ 0.36 = 26,875 CM → #6 AWG (26,240 CM borderline) or #4 AWG (41,740 CM) to provide margin for hot engine-compartment routing where copper resistance increases ~10% at compartment temperatures.
(7) Electric trailer brake wiring — the 12V feed from the tow vehicle's 7-pin connector to the trailer brakes (typically 12-15 A combined, over 25-30 ft) must maintain voltage at the brake magnets to ensure proportional braking force.
A 10% VD at the brakes means 10% less braking force — a significant safety margin reduction.
(8) Portable power station solar input — compact 'solar generators' with built-in MPPT controllers often use 8mm or Anderson Powerpole DC inputs limited to 10-15 A.
Users commonly extend the included cables with SAE connectors and 16 AWG wire.
A 100W panel (5.5 A at 18V V_mp) over 30 ft of #16 AWG: VD = (2 × 30 × 12.9 × 5.5) ÷ 2,580 = 1.65V, dropping panel voltage from 18V to 16.35V — below the MPPT's minimum input voltage for some units, causing the controller to stop charging entirely even in full sun.
Real-World Usage Scenarios
Off-road winch stalls halfway up a recovery pull
A Jeep owner installed a 12,000 lb winch (rated 400 A at full load) on a front bumper with the battery in the stock engine location, using the winch manufacturer's included #2 AWG cables (6 ft total, 3 ft positive + 3 ft ground). On a steep recovery pull requiring near-full winch power, the winch stalled after 20 seconds. Investigation found the positive cable had been extended 2 ft with a #4 AWG splice by the installer to reach a relocated battery terminal. The total positive path now had 2 ft of #4 AWG (41,740 CM) + 3 ft of #2 AWG (66,360 CM), and the ground path was 3 ft of #2 AWG through the frame. At 400 A: VD in the #4 section = (2 × 2 × 12.9 × 400) ÷ 41,740 = 0.495V. VD in the #2 positive section = (2 × 3 × 12.9 × 400) ÷ 66,360 = 0.467V. VD in the ground path (assuming frame return: 2 × 3 × 12.9 × 400 ÷ 66,360 = 0.467V). Total VD = 1.43V. But the real issue was the splice: a 5/16-inch bolt through ring terminals, which added approximately 0.5 mΩ of contact resistance — at 400 A, that's an additional 0.2V drop across the splice alone. Total system drop = 1.6V, and the battery under 400 A load sagged to 10.5V. Winch motor terminal voltage = 10.5 − 1.6 = 8.9V — below the motor's 9.0V minimum operating threshold. The fix: replace the entire positive cable with a single continuous #1/0 AWG (105,600 CM) run, eliminating the splice. VD at 400 A: (2 × 5 × 12.9 × 400) ÷ 105,600 = 0.489V. With battery sag to 10.5V, motor terminal voltage = 10.0V — comfortably above 9.0V.
RV lithium upgrade triggers inverter low-voltage shutdown at 80% battery
An RV owner upgraded from two 100 Ah lead-acid batteries to a 300 Ah LiFePO₄ battery bank, retaining the existing 2,000W inverter and its #2 AWG cables (3 ft round-trip, 1.5 ft one-way). The lithium battery's discharge curve is much flatter than lead-acid: at 80% depth of discharge, lead-acid would be at ~11.6V, triggering the inverter's 10.5V low-voltage disconnect. The lithium battery maintained 12.8V at 80% DoD — seemingly a huge improvement. But under the microwave's 150 A surge (1,800W ÷ 12V, plus inverter efficiency): VD = (2 × 1.5 × 12.9 × 150) ÷ 66,360 = 0.087V. The math checked out — this tiny VD couldn't be the problem. Deeper investigation revealed the owner had used the same #2 AWG but had extended the negative cable by 2 ft to reach a new shunt-based battery monitor, using a brass busbar with four stacked ring terminals. The busbar contact resistance measured 1.8 mΩ across the stack. At 150 A, that added 0.27V. But the real issue was the LiFePO₄ battery's BMS: the internal MOSFETs had an R_ds(on) of approximately 2 mΩ total, adding another 0.3V internal drop at 150 A. Total system VD = 0.087 (cable) + 0.27 (busbar) + 0.30 (BMS) = 0.66V. Battery terminal voltage at 80% DoD under load = 12.8 − 0.66 = 12.14V — still above 10.5V disconnect. The intermittent shutdown occurred when the microwave cycled on for the third time during a 5-minute cooking period, because the BMS FETs heated up and R_ds(on) increased to ~3 mΩ, pushing total VD to 0.81V and terminal voltage to 11.99V — not the shutdown cause on its own but combined with the cable inductance (L × dI/dt) generating an additional 0.3V transient dip at each magnetron cycle start, momentarily dipping below 10.5V and latching the inverter off. The solution: a dedicated 2/0 AWG connection directly from battery terminals to inverter, bypassing the busbar for the high-current path, and adding a 50,000 µF electrolytic capacitor bank at the inverter DC input to supply the magnetron's 120 Hz inrush pulses locally.
Sailboat anchor windlass wiring fire narrowly averted
A 42 ft cruising catamaran had its 1,200W anchor windlass (100 A at 12V) wired with #4 AWG copper (41,740 CM) from the house battery bank amidships to the bow locker, a one-way run of 28 ft through conduit behind cabinetry. The wire was sized for 10% VD at 100 A: VD = (2 × 28 × 12.9 × 100) ÷ 41,740 = 1.73V (14.4%). The windlass worked adequately in shallow anchorages (30-50 ft of chain) but when retrieving from 90 ft depth with 200 ft of chain deployed, the gearbox required additional torque and the motor drew its stall current of 280 A briefly. At 280 A: VD = (2 × 28 × 12.9 × 280) ÷ 41,740 = 4.85V (40%). Motor terminal voltage dropped to 12.6 − 4.85 = 7.75V — insufficient to turn the gearbox, so the motor stalled with full current flowing. The locked-rotor condition persisted for 8 seconds until the owner released the up button. In those 8 seconds, power dissipation in the #4 AWG cable was I²R: the cable resistance R = (2 × 28 × 12.9) ÷ 41,740 = 0.0173 Ω, so P = 280² × 0.0173 = 1,356 W — over a kilowatt of heat in the cable for 8 seconds. The conductor temperature rose from 25°C to approximately 25 + (1,356 × 8) ÷ (0.1 × mass_of_copper), and the insulation (PVC, rated 105°C on marine-grade cable) reached ~90°C — hot enough to soften but not melt. The owner, unaware of the thermal stress, repeated this cycle three times. On the third attempt, the insulation had degraded sufficiently that a minor arc occurred at a chafe point where the cable passed through an un-grommeted bulkhead penetration. The arc was small and self-extinguished when the windlass button was released, but the charred insulation was discovered during a subsequent survey. The surveyor mandated: (1) re-wire with #1/0 AWG (105,600 CM) for a VD of (2 × 28 × 12.9 × 280) ÷ 105,600 = 1.92V (16%) at stall, (2) install a 150 A thermal-magnetic circuit breaker within 7 inches of the battery connection per ABYC, (3) add a dedicated battery in the bow locker with a DC-DC charger from the house bank, reducing the main windlass power run to 5 ft (#2 AWG adequate) and providing redundancy if the long cable fails under load.
Common Mistakes to Avoid
Sizing for ampacity instead of voltage drop
A #10 AWG marine-grade copper wire at 105°C insulation rating carries 60 A per ABYC ampacity tables. A designer might select #10 AWG for a 50 A windlass circuit because 'the ampacity table says it's fine.' The voltage drop over 30 ft tells a different story: VD = (2 × 30 × 12.9 × 50) ÷ 10,380 = 3.73V (31.1%). The windlass motor receives 8.3V instead of 12V — it may not even overcome the gearbox's static friction. The rule for 12V systems: voltage drop ALWAYS governs conductor selection, never ampacity. The conductor that satisfies VD will automatically be large enough for ampacity. The only exception is very short runs (< 3 ft) at moderate currents (< 30 A), where ampacity and VD converge on the same gauge. This error is so common that ABYC E-11 explicitly states: 'The conductor size shall be such that the voltage drop does not exceed 3 percent for panelboard main feeders, navigation lights, bilge blowers, electronic equipment, and other circuits where voltage drop must be kept to a minimum, and 10 percent for other conductors.' If your 12V wire sizing process doesn't start with VD, you're doing it wrong.
Using chassis/frame as the sole ground return without verifying the path resistance
In vehicles and boats, the metal chassis or bonding system is often used as the DC negative return path, eliminating one conductor. This is legitimate and saves weight and cost — but ONLY if the return path resistance is verified. A common scenario: the battery's negative cable bolts to the engine block (excellent connection, thick cast iron/aluminum). The winch's negative cable bolts to the frame rail (also good). But the engine block connects to the frame through rubber engine mounts — the electrical path from engine to frame is via a braided ground strap, which may be corroded, loose, or undersized. A ground strap with 5 mΩ resistance carrying 300 A winch current drops 1.5V — instantly consuming 40% of the VD budget before the positive cable is even considered. The diagnostic test: measure voltage between the battery negative post and the winch motor case while the winch is running. Anything above 0.2V indicates a ground path resistance problem. The fix: add a dedicated negative cable from the battery to the winch (full two-wire circuit) OR add a heavy-gauge redundant ground strap between engine and frame. ABYC E-11 requires that if the hull or bonding system is used as a current-carrying conductor (which is discouraged), the voltage drop across the return path must be included in the total circuit VD calculation.
Ignoring the voltage sag of the battery itself under load
A 12V battery is not a constant voltage source. Under heavy load, its terminal voltage sags due to internal resistance (R_int). For a fully charged 100 Ah lead-acid battery, R_int ≈ 10-15 mΩ at 25°C. At 200 A load, the internal voltage drop is 200 × 0.012 = 2.4V. The battery terminal voltage (open-circuit 12.6V) drops to 10.2V under 200 A load — a 19% reduction BEFORE the external wiring even enters the calculation. When a designer calculates wire size using 12V as the source voltage, they neglect this sag and underestimate the percentage drop. The more accurate approach: use 10.5-11.0V as the effective source voltage for high-current circuits (> 100 A) when calculating VD percentage. For LiFePO₄ batteries (R_int ≈ 3-5 mΩ per 100 Ah), the sag is less severe but the flat discharge curve means the BMS's low-voltage cutoff may be triggered by the combined cable + BMS drop even at moderate state of charge. For critical high-current circuits (inverters, winches, bow thrusters), measure the actual battery terminal voltage under a load test and use that measured value — not 12.0V — in the voltage drop calculation.
Industry Standards Referenced
Frequently Asked Questions
What size wire for a 1000W inverter at 12V?
1000W ÷ 12V = 83.3 A continuous DC input (with ~90% efficiency, closer to 93 A actual). For a typical 3 ft one-way run (inverter near battery) at 3% VD: CM = (2 × 3 × 12.9 × 93) ÷ (12 × 0.03) = 7,196 ÷ 0.36 = 19,990 CM → #6 AWG (26,240 CM) provides VD = 0.274V (2.29%) with margin. However, most inverter manufacturers specify #4 AWG or larger for 1,000W units because the surge rating (typically 2,000W for 5-10 seconds) doubles the current to ~186 A. At 186 A surge with #6 AWG: VD = (2 × 3 × 12.9 × 186) ÷ 26,240 = 0.55V (4.6%) — still functional but approaching the inverter's low-voltage warning threshold during surge. Using #4 AWG (41,740 CM): surge VD = 0.345V (2.87%) — much safer. For installations where the battery-to-inverter distance exceeds 3 ft (common in RVs where the battery bank is in a compartment and the inverter is in the living space), wire size escalates rapidly: at 8 ft with #2 AWG (66,360 CM): VD at 186 A surge = 0.58V (4.8%). The practical recommendation: keep the battery-to-inverter run as short as physically possible (ideally < 3 ft), use the heaviest gauge the inverter terminals will accept (typically #2 AWG or 2/0 AWG), and fuse the positive cable within 7 inches of the battery terminal with a Class-T or ANL fuse rated at 125-150% of the inverter's continuous rating.
Why is 48V better than 12V for solar battery systems?
For the same power, a 48V system draws 1/4 the current of a 12V system (P = V × I, so I = P/V). This means 1/16 the I²R losses (since resistive losses scale with I²). A 4,800W load at 12V requires 400 A; at 48V it requires 100 A. For a 10 ft run at 3% VD: 12V requires CM = (2 × 10 × 12.9 × 400) ÷ (12 × 0.03) = 286,667 CM → 300 kcmil, a cable roughly 0.55 inches in diameter. 48V requires CM = (2 × 10 × 12.9 × 100) ÷ (48 × 0.03) = 17,917 CM → #8 AWG (16,510 CM) or #6 AWG (26,240 CM) — a cable 10× smaller by area. The cost differential is dramatic: 10 ft of 300 kcmil copper costs approximately $80-100, while 10 ft of #6 AWG costs $6-8. Multiply by every high-current circuit in the system (inverter, charge controller, battery interconnects, DC distribution) and the savings from 48V architecture can exceed $1,000 in copper alone on a mid-size off-grid system. Additionally, the lower current reduces connector and switch contact heating, reduces fuse size and cost, and allows standard DIN-rail circuit breakers instead of expensive Class-T fuses. This is why all modern residential energy storage systems (Tesla Powerwall, Enphase IQ Battery, FranklinWH, EG4) use internal DC bus voltages of 48-400V. The only reason to use 12V today is compatibility with existing 12V appliances and the massive 12V ecosystem in automotive and RV markets — a legacy constraint, not an engineering advantage.
Can I use household Romex for 12V DC wiring?
Technically yes for ampacity — NM-B (Romex) uses copper conductors sized per AWG standards, and 12V is well within its 600V insulation rating. However, there are three strong reasons not to: (1) NM-B is solid copper, which work-hardens and eventually fractures under the vibration and thermal cycling present in vehicles and boats. Stranded copper wire distributes bending stress across multiple filaments, dramatically increasing fatigue life. (2) NM-B's PVC insulation is not rated for the wet, oily, and high-temperature environments common in engine compartments, bilges, and battery boxes. Marine-grade wire uses cross-linked or high-temperature PVC rated at 105°C and tinned copper strands to resist corrosion. (3) NM-B's outer jacket is not abrasion-resistant enough for mobile installations — chafing against metal edges, vibration against zip ties, and repeated flexing will wear through the jacket. For stationary installations (off-grid cabins, solar sheds), NM-B in conduit or where protected from physical damage is acceptable. For mobile installations (RVs, boats, vehicles), use stranded, tinned copper wire meeting ABYC E-11 (marine) or SAE J1128 (automotive) standards. Specifically: Ancor Marine Grade or equivalent for marine, GXL or SXL cross-linked primary wire for automotive underhood, and fine-stranded welding cable (class K or M stranding) for high-flex battery and inverter cables where vibration isolation requires extreme flexibility.
What size wire do I need for a 12V solar panel to charge controller?
This depends on the panel's maximum power current (I_mp) and the one-way distance. For a single 100W 12V panel (I_mp ≈ 5.5 A, V_mp ≈ 18V) at 30 ft with 2% VD (recommended for solar source circuits to minimize MPPT tracking loss): CM = (2 × 30 × 12.9 × 5.5) ÷ (18 × 0.02) = 4,257 ÷ 0.36 = 11,825 CM → #10 AWG (10,380 CM) would give VD = 0.41V (2.3%), marginally acceptable. Moving to #8 AWG (16,510 CM) gives VD = 0.26V (1.43%) — preferred. But if you have multiple panels in parallel (3 × 100W = 16.5 A at 18V), the same 30 ft run at 2% VD requires: CM = (2 × 30 × 12.9 × 16.5) ÷ 0.36 = 35,475 CM → #4 AWG (41,740 CM). This illustrates why parallel panel configurations become wiring-impractical beyond 2-3 panels — series connection (3 panels at 54V V_mp, still 5.5 A) requires the same #10 AWG as a single panel. Series-parallel configurations (e.g., 2S2P for 200W: 36V at 11 A) are a compromise. The NEC 690.8 calculation for PV source circuits uses I_sc × 1.56 (1.25 for continuous × 1.25 for irradiance above STC), so the design current is higher: for a panel with I_sc = 6.0 A, I_design = 6.0 × 1.56 = 9.36 A. At this current and 2% VD (0.36V at 18V V_mp), CM = (2 × 30 × 12.9 × 9.36) ÷ 0.36 = 20,124 CM → #8 AWG (16,510 CM too small, #6 AWG at 26,240 CM required). The NEC-required safety margin pushes the wire one AWG size larger than the simpler I_mp calculation suggests.
What is the difference between AWG, SAE, and metric wire sizes for 12V DC?
Three gauge systems are commonly encountered in 12V DC work and they are NOT interchangeable. AWG (American Wire Gauge, also called Brown & Sharpe) is the North American standard, with a logarithmic progression where each 6-gauge step doubles the diameter. SAE (Society of Automotive Engineers) gauge is similar to AWG but approximately 6-12% smaller in cross-sectional area for the same gauge number — SAE #10 is roughly equivalent to AWG #11 in conductor area. This matters because '10 gauge' automotive primary wire from an auto parts store may be SAE standard, not AWG, and will have higher resistance (and thus higher voltage drop) than an AWG #10 wire. Always check the packaging: if it says 'SAE J1128' and '10 GA,' it's SAE gauge. If it says 'AWG 10' or just '10 AWG,' it's AWG. Metric wire sizes (used in European and Asian vehicles and marine equipment) are specified by cross-sectional area in mm². Common equivalents: 1.5 mm² ≈ #16 AWG, 2.5 mm² ≈ #14 AWG, 4 mm² ≈ #12 AWG, 6 mm² ≈ #10 AWG, 10 mm² ≈ #8 AWG, 16 mm² ≈ #6 AWG, 25 mm² ≈ #4 AWG, 35 mm² ≈ #2 AWG, 50 mm² ≈ #1/0 AWG, 70 mm² ≈ #2/0 AWG. When working on European-built boats (Beneteau, Jeanneau, Bavaria) or Japanese vehicles with metric wiring, use the mm² to AWG conversion table for replacements and upgrades. For critical VD calculations with metric wire, use the actual mm² area: CM = mm² × 1973.5 (since 1 mm² = 1973.5 circular mils).
How do I account for temperature when sizing 12V wires in engine compartments?
Engine compartments routinely reach 60-80°C during operation, and copper resistance increases by 0.393% per °C above 25°C. At 75°C (50°C rise), resistance is 19.7% higher. The practical correction: multiply your calculated CM by the temperature factor TF = 1 + 0.00393 × (T_engine − 25). For a common underhood temperature of 70°C (measured with a thermocouple during a hot-soak test after highway driving): TF = 1 + 0.00393 × 45 = 1.177. So a circuit that requires 20,000 CM at room temperature needs 20,000 × 1.177 = 23,540 CM in the engine compartment — one AWG size larger. Additionally, SAE J1128 specifies insulation temperature ratings: Type GPT (general purpose thermoplastic) is rated 85°C; Type GXL (cross-linked thin-wall) is rated 125°C; Type SXL (cross-linked standard-wall) is rated 125°C. For underhood wiring, use GXL or SXL exclusively — GPT will embrittle and crack within a few years of engine heat exposure. The ampacity of the conductor must also be derated for elevated ambient: a #10 AWG SAE wire carrying 40 A in free air at 25°C may only carry 25-30 A in a 70°C engine compartment before the insulation temperature limit is reached. For marine engine rooms, ABYC E-11 Table VI provides separate ampacity columns for 'inside engine spaces' (derated by approximately 15-20%) and 'outside engine spaces.' Always use the inside-engine-space column for engine room wiring.
Reviewed for accuracy
Cross-referenced against ABYC E-11 marine wiring standards, SAE J1128 automotive primary cable specifications, and industry practice for RV/off-grid 12V installations · Last reviewed: July 26, 2026
All calculations are for reference only. Always verify with manufacturer data and a qualified engineer for critical applications. Learn about our editorial process.