In a 40-product launch, the standout item is a 1200V silicon carbide MOSFET, not a compute chip. India's first semiconductor path is power devices.
On September 18, L&T's semiconductor subsidiary LTSCT unveiled 40 products at an Indian semiconductor exhibition. The list includes secure identity platforms, vision camera SoCs, connectivity modules, and USB-C gallium nitride (GaN) charger solutions, but the headline item is a 1200V silicon carbide (SiC) MOSFET platform.
An Indian newcomer leads with power devices. This choice warrants scrutiny: it is both a result of capability constraints and a reflection of market structure.
1200V
SiC MOSFET voltage rating
Number of products launched simultaneously
4 Regions
Operational coverage
The three figures correspond to the voltage rating of the power device, the number of products launched simultaneously, and the company's operational presence across the Americas, Europe, Japan, and India. The last point is notable: it indicates this is not a company selling only in the local market, as its R&D and sales reach extends overseas.
What device was chosen
According to LTSCT, this 1200V SiC MOSFET platform targets next-generation power conversion: EV fast charging, microgrids, solid-state transformers, and traction inverters. These scenarios share high voltage, high power, and sensitivity to efficiency.
Silicon carbide replaces silicon-based IGBTs here. Its value lies not in vague claims of being 'faster,' but in two specific physical properties: lower switching losses at the same voltage rating and the ability to operate at higher temperatures. At the vehicle level, higher switching frequencies allow smaller passive components, reducing inverter volume and weight; every unit of energy saved in conversion reflects in range, though these gains are ultimately shared across the battery, motor, and thermal management systems.
The 1200V voltage rating was deliberately selected. It corresponds to 800V high-voltage vehicle architectures, which are becoming a prerequisite for fast-charging capability. Choosing a specific voltage tier means the product's market positioning is clear, rather than developing the product first and finding customers later.
Why not start with logic chips
If the goal is to rapidly build commercializable chip capabilities, advanced logic is almost always the first option on new entrants' lists—and the hardest. It requires not just one fab line, but an entire chain of process, EDA, IP, and customer adoption; missing any link stalls progress, and the capital barrier is in the tens of billions of dollars.
Power semiconductors have a different structural profile. The devices are discrete components with low dependence on advanced nodes, requiring process maturity primarily in substrates, epitaxy, and back-end packaging. LTSCT adopts a fabless model, outsourcing manufacturing to partners while concentrating resources on design and product definition. This aligns with India's relative advantage: an ample supply of engineers, though capital for production lines and mass-production experience require time to accumulate.
Market structure also drives this choice. Demand for power devices is dispersed across automotive, industrial, energy, and home appliance sectors, unlike advanced logic chips which are concentrated among a few customers. India's local markets for electric two-wheelers, home appliances, and grid upgrades are still growing nearby markets. The startup costs for power devices are lower, and the path to cash recovery is shorter.
There is an upstream constraint on this path. The bulk of silicon carbide (SiC) device costs lies in substrates and epitaxy, segments currently dominated by a few suppliers. Fabless power device companies face equally challenging procurement in these areas. Design capabilities can be built quickly, but material supply cannot.
A further move occurred concurrently: LTSCT and Tata Electronics announced a strategic partnership, linking the former's chip design capabilities with the latter's packaging and wafer manufacturing capabilities. One handles design, the other manufacturing; the intent to create a local closed loop is clear. Whether this loop truly closes depends on when manufacturing capabilities are ready.
A testable judgment follows: India's first scaled export category in semiconductors will be power devices, with logic chips following later. The test is direct: if India has a mass-production scale logic wafer line supplying external markets by 2028, this judgment fails. Conversely, if India's exports remain concentrated in power, analog, and sensing chips at that time, the path is set.
Entering via wide-bandgap devices bypasses the hardest segment but relinquishes control over advanced process nodes. The final landscape of this path will likely depend on whether the first automotive customers are willing to place orders with a company that only begins shipping in 2026.