0
By clicking the button, I accept the Terms of Use of the service and its Privacy Policy, as well as consent to the processing of personal data.
Don’t have an account? Signup
Powered by :
The first reference point is March 2022. SECI issued what was then a landmark tender: 500 MW × 2 hours, or 1,000 MWh of standalone BESS in Rajasthan, with each project required to support two charge-discharge cycles a day. The 2-hour duration was the working assumption for the next two years. Almost every major early tender — from GUVNL’s early phases through NTPC’s first BESS rounds — was built around the 2-hour cycle, sometimes with the variation of two cycles per day to stretch the discharge envelope.
That has changed in the last 18 months. Look at what is actually being tendered now.
The current SECI 4,800 MWh tender is 1,200 MW × 4 hours, the second such issue from SECI — the previous 4.8 GWh round was awarded at Rs 6.27-6.28/kWh to Adyant Enersol, Serentica Renewables, AMPIN Energy, and ACME Solar. UPPCL allocated 1,500 MWh (375 MW × 4 hours) of standalone BESS at Rs 6.45-6.46/kWh in May 2026 under a 15-year contract. SECI’s 125 MW/500 MWh Odisha tender in early 2025 was 4-hour batteries. NTPC’s August 2025 thermal-co-located BESS tender deliberately split its 4,000 MWh allocation: 300 MW at 4-hour duration alongside 1,400 MW at 2-hour batteries, perhaps to ensure adequate and competitive price bids at a volatile period. Adani Green’s recently commissioned 3.37 GWh BESS at Khavda — the world’s largest single-location battery deployment outside China — is sized for multi-hour discharge profiles.
This is not to say that the 2-hour has, or will disappear. GUVNL’s Phase VIII tender in December 2025 was 335 MW × 2 hours. The 2 GW Phase VII GUVNL tender awarded in November 2025 was a 2-hour. State utilities’ and their standalone BESS tenders, where the use case is peak shaving and ancillary services, continue to lean on the 2-hour cycle. But the new FDRE, peak-supply, and ISTS-connected tenders — the categories that are growing fastest — have moved decisively to four hours. By tender volume, 4-hour is now the architecture being designed around, with 2-hour holding on at the state distribution end.
The shift is consistent with what the most authoritative independent modelling of India’s storage requirements predicts. The August 2025 report Strategic Pathways for Energy Storage in India through 2032, jointly published by the Power Foundation of India and the India Energy and Climate Center at UC Berkeley, identifies 2027 as the modelled inflection point — the year from which 4-hour batteries become predominant. The chart below, reproduced from that study’s Reference Case, shows the trajectory. That seems close to reality right now, considering the 12-18-month lead time for most large BESS storage projects.

Figure 1: India’s energy-storage build-out, 2023-2032 (Reference Case). 2-hour BESS plateaus at 20 GW from 2029 onward; 4-hour BESS rises from 4 GW in 2028 to 67 GW by 2032. Source: PFI/Berkeley IECC, August 2025 (Figure ES-2).
SIDEBAR  ·  THE TARGET TRIANGLE
What different authorities say India needs
India’s storage requirement is one of those numbers that depends on who you ask. Four official and quasi-official forecasts now sit on the table for 2030. They tell subtly different stories — and the differences explain a lot about why the duration architecture is shifting.
 
AUTHORITY
DATED
GW BY 2030
GWh
BESS / PSP SPLIT
CEA  ·  Optimal Generation Mix 2030
Feb 2026
60.63
—
41.65 BESS  ·  18.98 PSP
CEA  ·  National Electricity Plan
2023
~62
—
BESS-led; 411 GWh by 2032
PFI / Berkeley IECC  ·  Reference Case
Aug 2025
61
218
51 BESS  ·  9 PSP
CEEW  ·  600 GW high-RE pathway
Mar 2025
83
—
70 (4-hr) BESS  ·  13 PSP 
 
Where they agree.  On the headline GW count for 2030, three of the four authorities cluster between 60 and 62 GW. CEEW’s 83 GW figure sits higher because it models a more ambitious 600 GW non-fossil pathway rather than the 500 GW reference.
Where they diverge.  The duration mix. PFI/Berkeley specifies 31 GW of 4-hour batteries in 2030 versus 20 GW of 2-hour. CEEW makes the duration explicit — 70 GW of its 2030 BESS is 4 hours. CEA’s Optimal Generation Mix 2030 (60.63 GW) carries no duration breakdown, leaving the discom-vs-FDRE tilt as a procurement decision. The GW number has converged. The hours behind it haven’t.
Where the build is.  India’s installed storage at end-2025 is approximately 6 GW, almost all of it pumped hydro. Getting to 60 GW by 2030 requires roughly 12 GW of net additions per year for the next five years. IESA estimates 9.2 GWh of storage will be commissioned by the end of 2026— under 3 GW at a 3-hour weighted-average duration. The 2026 run-rate is a fraction of what every one of these target sets demands.
 
 
Four reasons, none of which alone would have shifted the default. But together, they have, as we highlighted last year as well.  
The first is what the Indian grid is actually being asked to solve. Two-hour storage was designed for arbitrage and peak shaving — absorb the steepest spike of the evening ramp, return the energy in a narrow window. That was the right design for a grid where solar capacity was below 50 GW and the evening peak was a short, sharp event. Importantly, BESS technology itself was untried, untested and uncertain when it came to pricing.  Much has changed since then. Solar capacity is past 100 GW. The evening peak is no longer a spike — it stretches from roughly 5 PM to 11 PM in winter, 7 PM through midnight in summer. Two-hour storage covers the steepest moment of that window. Four-hour storage starts to cover the demand profile itself. The arithmetic of the duck curve has caught up with the tender design.
The second is the FDRE construct. Firm and Dispatchable Renewable Energy (FDRE) is increasingly in focus as large users demand committed, scheduled delivery during peak hours, not a vague quantum of energy delivered over a year. To bid into an FDRE auction with any confidence, a developer needs storage sized for the contracted dispatch window, not for the maximum-spike second. The math forces a 4-hour minimum, and increasingly a 4-hour-plus combined with a wind or hybrid generation tail.
The third is policy alignment. The Ministry of Power’s Viability Gap Funding scheme, in its second tranche of December 2024, formally included the 4-hour 1-cycle design — 6,300 cycles over the contract life — as an eligible architecture, alongside the older 2-hour 2-cycle structure. The VGF mechanism, sized at roughly Rs 16 lakh per MWh, applies to both — but the inclusion of the 4-hour pathway was the regulatory signal that the duration shift was officially in. The PFI/Berkeley study estimates that an expanded VGF scheme could unlock 50-100 GW of solar and 16-32 GW of storage capacity by 2027. The ISTS charge waiver for solar-plus-BESS projects extending through 2028 further tilts the financial economics towards co-located, longer-duration designs.
The fourth, and least discussed, is the cost curve. Falling lithium iron phosphate cell prices, particularly through 2025, have made the 4-hour design financially viable in a way it simply was not in 2022. Prices have also confounded those who predicted a sharp recovery in 2026 so far. The PFI/Berkeley report’s reverse-engineering of recent auction wins puts co-located battery storage capital cost at $100-120 per kWh, with standalone storage somewhat higher at $150-200 per kWh — co-location saves roughly 20% through balance-of-systems sharing. The report further projects that these costs could drop another 15-20% by 2030. At that level, doubling the storage hours from two to four moves the project tariff up by a manageable increment when set against the grid problem it solves.
This is where the architecture change actually moves the model.
Take SECI’s landmark December 2024 auction — a 2,000 MW solar project co-located with 1,000 MW × 4-hour battery storage, 4,000 MWh in total. The winning bid was ₹3.52 per kWh. The PFI/Berkeley analysis decomposes this number cleanly: assuming a solar LCOE of ₹2.5 per kWh, the evening-peak storage adder works out to about ₹1 per kWh. The project stores roughly 33% of daily DC solar generation in the batteries — enough to push power back into the grid during the four-hour evening peak.
Compare that with the SECI July 2024 auction (1,200 MW solar + 600 MW × 2-hour BESS, 1,200 MWh). The winning bid was ₹3.41 per kWh. About 20% of daily solar generation was stored. The storage adder reverse-engineers to ₹0.81 per kWh. Halving the storage hours saved roughly ₹0.20 per kWh — not nothing, but not a decision-driver either, once the buyer’s grid problem requires the longer discharge window.
The capex impact is the real story. Doubling the discharge duration at the same power rating roughly doubles the cell count — because the inverters, transformers, balance-of-plant and land footprint scale primarily with MW, while the cells scale with MWh. A 1 GW × 2-hour BESS is a 2 GWh project. A 1 GW × 4-hour BESS is a 4 GWh project. The cell-procurement tonnage doubles. Every gigawatt of new 4-hour standalone storage now pulls roughly twice the lithium iron phosphate demand into the supply chain that an equivalent 2-hour facility would.
Layer the ALCM regime that took effect on June 1, 2026, and the implications for the domestic cell industry become hard to overstate. India’s ALMM-aligned cell capacity is on a trajectory from roughly 13 GW in July 2025 to about 100 GW by December 2027, per ICRA. That looks like a lot of capacity until you multiply the PFI/Berkeley projections through it: 51 GW of battery storage (20 GW × 2-hour + 31 GW × 4-hour) cost-effective by 2030, rising to 87 GW (20 GW × 2-hour + 67 GW × 4-hour) by 2032. Convert those numbers into MWh — 164 GWh by 2030, 308 GWh by 2032  and the cell demand pull becomes a structural shift instead of incremental.
The standalone tariff data confirms the same direction. UPPCL’s 4-hour standalone BESS allocation at ₹6.45-6.46 per kWh in May 2026 sits well above the ₹1.8-1.9 lakh per MW per month range that GUVNL was clearing for 2-hour standalone in late 2025. SECI’s 4-hour peak-power tender awarded earlier at ₹6.27 per kWh sits in similar territory. The market is paying for the duration, and the discoms, FDRE buyers, and SECI’s procurement desks have decided the duration is worth paying for. Especially with power prices hovering around the Rs 10/kWh ceiling on exchanges during peak, non-solar hours.
The winners’ list begins with the obvious. Domestic cell manufacturers — Reliance, through its ACC PLI award, Amara Raja, Exide, the Adani-Karur Vysya consortium, the Ola Cell line, and the integrated cell-to-pack see their addressable market roughly double per MW of project awarded. Large integrated developers with bankable balance sheets to fund the higher per-MW capex — Adani Green (which has used Khavda to prove operational capability at 3 GWh-plus scale), ReNew, JSW Energy, Tata Power Renewable, Greenko, Serentica, Acme — are best placed to win 4-hour tenders. 
The harder side of the ledger is for mid-sized developers and project finance arrangers. A 4-hour project at the same MW rating is roughly twice the cell capex and has a longer payback profile. Financial closure becomes a different conversation. The “tender to commissioning” gap that India has struggled with — 130-plus GWh of tenders in 2025 against just 9.2 GWh expected to commission in 2026 — is partly a function of exactly this. Bidders win at aggressive tariffs, then discover the financing math does not close cleanly when the cell sourcing, land, and debt syndication all come together at the higher per-MWh asset value. The longer the duration, the more the gap can widen.
Existing 2-hour assets do not get stranded. Their PPAs run for 12-15 years; their economics still work. The PFI/Berkeley Reference Case keeps 2-hour BESS at a roughly stable 20 GW from 2028 onwards- a meaningful share of the system, just not the growth engine. The 4-hour asset class is where the next 60-plus GW of new capacity will sit. A discom looking to procure peak supply in 2027 will increasingly find 4-hour quotes that deliver more grid flexibility than a 2-hour comparable at not-quite-double the tariff. The 2-hour asset is fine as long as it stays in its lane. It cannot bid into the FDRE.
A few signals will tell us which way the architecture continues to move.
The first is whether 4-hour standalone BESS tariffs hold at the current ₹6.27-6.45 per kWh band, or drift down as cell prices continue to fall through 2026. If the band breaks below ₹5.50 per kWh, the case for 4-hour displacing 2-hour as the default everywhere — including at the state utility level — becomes overwhelming. The big risk here is rising module prices due to the tight market conditions for DCR modules now.
The second is whether the next tender wave adopts a duration beyond four hours. NTPC’s August 2025 hedge between 2-hour and 4-hour storage looks, on reflection, like a transitional design. The PFI/Berkeley study finds that batteries remain more economical than pumped hydro up to roughly 6-8 hours of daily discharge — and that pumped hydro becomes competitive beyond that boundary. Pumped storage tenders totalled 74 GWh in 2025. Long-duration energy storage projects, such as the Triveni Turbine awarded by NTPC, are being trialled. The 8-hour configuration is the next logical battlefield.
The headline that “SECI has issued a 4.8 GWh tender” reads like business as usual in a busy storage market. The substance underneath is that India’s storage tendering has moved on from peak-shaving to grid-shaping — from absorbing the spike to covering the demand window — and the duration parameter is where that move is showing up first. Every gigawatt of 4-hour storage that gets built is functionally two gigawatts of 2-hour storage in cell-procurement terms, and India’s cell manufacturing build-out is being scaled, financed, and policy-supported on that arithmetic.
For developers, the shift selects for balance sheet strength and integrated sourcing. For cell makers, it doubles the demand pull. For lenders, it raises the per-MWh exposure and the duration of the underwriting. For the grid, it starts to bend the duck curve into something the discoms can actually live with, allowing for sustained additions of solar into the grid. 
The next 18 months will decide whether the architecture sticks at four hours or pushes onward into the longer-duration territory where pumped storage and emerging chemistries are already lining up to compete.
We are India’s leading B2B media house, reporting full-time on solar energy, wind, battery storage, solar inverters, and electric vehicle (EV)
Quick Links
© 2025 Saur Energy. All Rights Reserved.

source