Every car still hides a small lead-acid battery - the cheap, overlooked cell that cranks an engine or keeps an EV's electronics alive while parked. In China's electric-vehicle boom, that humble component has become an unlikely battleground. According to tech outlet Huxiu, automakers are now pursuing three distinct routes to replace it: swapping in lithium or sodium cells, raising the low-voltage network to 48V, or integrating the low-voltage system into the main high-voltage pack.
The lead-acid starter battery has reigned since Bosch adapted it for cars in 1927. Its virtues are extreme cost and reliability: a replacement costs a few hundred yuan and any roadside garage can swap one in ten minutes. But it is heavy, lasts only about two years, and carries a toxic lead footprint. GEP Research estimates global scrap lead-acid batteries have passed 68 million tonnes, with China contributing about 21 million tonnes, or 30.9 percent. Yet only about 62 percent of that is recycled through formal channels, leaving nearly 38 percent to leak into informal workshops that pollute soil and groundwater.
Regulation is tightening. The EU's new battery law requires 75 percent recycling efficiency for lead-acid units by the end of 2025, while China's updated standard GB/T37281-2019 and a policy push aim for a formal recovery rate above 60 percent by the end of 2026. On EVs, the mismatch is stark: a lead-acid cell typically weighs 15-20kg, against roughly one-third that for a lithium low-voltage pack; it needs replacement every two years; and it cannot share the warranty of the main traction battery.
Three routes, three risk profiles
The most conservative fix is a direct swap to lithium iron phosphate or sodium-ion cells at the same 12V voltage. BYD claims to be the first automaker to mass-produce a 12V LFP starter battery, saying it lasts as long as the vehicle itself. Geely, Li Auto and SAIC have adopted similar solutions on some models. Forvia Hella's fully integrated 12V lithium pack has won orders worth several hundred million euros, with a China-developed version due to start production there in summer 2027, the company told Huxiu. This route cuts weight by over 60 percent, extends cycle life beyond 2,000 charges, and keeps the existing repair ecosystem - but a replacement part costs roughly three times as much as lead-acid.
More ambitious is the 48V architecture championed by Tesla. Its Cybertruck already runs a 48V low-voltage network, slashing cable volume by about 75 percent and wiring weight by 70 percent. Chery and Bosch have signed a 48V vehicle-architecture cooperation deal, lifting onboard power to 15kW and cutting system weight by over 10kg. Bosch told Huxiu that more than 60 percent of new cars in China will adopt 48V within three years. The gain is efficiency: quadrupling voltage cuts current to a quarter, shrinking copper cross-sections - a 10-metre cable for a 1kW load drops from 896g to 134g of copper, an 85 percent cost saving. But the switch forces re-certification of every low-voltage component and a full supply-chain overhaul, so Tesla still keeps some 12V branches.
The boldest path is Leapmotor's CTC 3.0 system, which eliminates the external 12V battery entirely and integrates low-voltage modules into the high-voltage pack. Leapmotor's vice-president for batteries, Song Yining, told Huxiu the hardest part is not the technology but integration: traditional cars keep low- and high-voltage circuits separate, each with controllers, distribution boxes and wiring. Leapmotor's all-stack in-house development lets it merge the two into one system, managing thermal zones and an AI scheduler that allocates energy between circuits. The payoff, claims Leapmotor, is a seven-fold increase in low-voltage capacity and standby time - from about one month to seven months - and a five-fold longer auxiliary battery life. At a typical 400 yuan ($56) replacement every two years, that could save at least 2,000 yuan ($282) over a decade.
Leapmotor has open-sourced the technology, hoping rivals adopt it and drive down supply-chain costs. Huxiu notes the logic echoes Tesla's early patent open-sourcing: for a market chaser, sharing is the cheapest way to build ecosystem influence. If only Leapmotor uses it, costs stay high; if others follow, standardization rises and the pioneer reaps scale benefits.
The three routes are not about technical superiority, Huxiu argues, but about who bears the risk and who pockets the gain. Deeper changes bring bigger rewards - and heavier burdens for automakers and owners alike.
