Introducing Grid 2.0, the Open Protocol for the American Grid (Bonus episode)

Episode Summary

Executive Summary: The conversation argues that U.S. grid interconnection is too slow because it treats all capacity as firm, forcing expensive studies and upgrades for the 1% of worst-case hours while leaving the other 99% underused. Chris Shelton proposes "Grid 2.0": a protocol-driven, permission-based model that lets large loads and batteries access non-firm capacity in real time, potentially accelerating data center hookups, unlocking hidden megawatts, and lowering rates.

Main Topics: Why grid interconnection is slow (Priority: 5/5): Large new loads like data centers trigger long studies because utilities must prove the system can survive rare peak conditions across complex, multi-state networks. The process protects reliability but often delays projects for years. Battery storage as precedent (Priority: 5/5): Shelton recounts how AES helped prove batteries could be treated as capacity resources rather than backup, culminating in the Alamitos project. This earlier reframing of storage is used as a template for rethinking demand access. Grid 2.0 and connect-and-manage (Priority: 5/5): The proposed paradigm formalizes non-firm access through permission-based, near-real-time control. Instead of waiting for full firm upgrades, users could connect sooner and use available capacity when the grid is healthy. Data centers, hyperscalers, and flexible load (Priority: 4/5): The model is aimed at large AI/data center loads, which can potentially use diversified flexibility, site-level assets, chillers, batteries, or jointly managed capacity without disrupting core computing workloads. Buffers, BYOC, and network-directed storage (Priority: 4/5): Shelton argues the grid needs a new class of asset—buffers—between load and supply. These could be supplied by customers themselves via bring-your-own-capacity models to solve both network and supply constraints. Rates, revenue, and utility incentives (Priority: 4/5): Because the grid is already paid for during most hours, using its unused capacity could generate more revenue, spread fixed costs, and potentially lower rates while still supporting utility investment. Standards, governance, and the task force model (Priority: 3/5): The Grid 2 Task Force is framed like the IETF: an open, cross-sector standards effort with GitHub, WhatsApp, and public drafts to create interoperable protocols rather than one-off utility solutions.

Key Arguments: The grid is overstudied for rare 1% conditions, causing delays even though the system has excess capability in 99% of hours. A permission-based, near-real-time access standard can preserve reliability while unlocking earlier interconnection. Data centers do not necessarily need fully firm service for all functions; some site loads and support systems can be flexible. Batteries should be treated not only as supply assets but as buffers that solve network topology and time-shifting problems. Using existing assets more fully can create additional revenue that helps utilities recover costs and can reduce rates. A standardized protocol is needed so the industry does not create 2,000 different versions of the same solution across utilities. The model is intended to be collaborative and optional, not mandatory, especially for hyperscalers with non-interruptible workloads. Current interconnection logic hides "hidden megawatts" by failing to monetize capacity that is already built and paid for.

Data Points: Data center grid connection wait time: 3 to 7 years - Hyperscalers seeking new grid connections face long delays due to study and upgrade processes. Engineering study time: 22,000 engineering hours - Used to determine whether new data center load can interconnect safely and how it affects the grid. Battery project size threshold: First battery bigger than 100 megawatts - Alamitos was described as the first U.S. battery storage project exceeding 100 MW. Battery flexibility vs power plant flexibility: 100 MW battery = 200 MW of flexibility - Shelton contrasted batteries with thermal plants that often operate at partial output for efficiency/emissions reasons. Grid capacity usage: 99% of the hours - Claim that the grid has surplus capability most of the time, which could be monetized under Grid 2.0. Constraint frequency: Less than 1% of the time - The rare conditions that force expensive studies and drive most interconnection upgrades. Potential rate reduction: 5% to 10% - Cited as a possible downward pressure on effective rates if the 1% flexibility is used to access 99% of hours. Purdue study load reduction window: About 88 hours - Referenced as roughly 1% of the year where accepting load reductions could increase hosting capacity. Purdue study hosting capacity impact: 80% increase - Claim that load flexibility during rare peak periods can significantly raise system hosting capacity. PJM battery queue: 60,000 megawatts - Scale of battery capacity waiting in the interconnection queue in PJM. Historical battery startup scale: Two 1 megawatt systems - First lithium-ion battery system built by Shelton's team in 2008. Current battery penetration in states: 20% of power in Texas and California for several hours a day - Used to show how large batteries have become in peak supply in major markets. Utility count: 2,000 utilities - Reason given for needing a standardized national approach rather than many local variants. Industrial precedent: 7% of peak demand - Batteries now represent about 7% of peak demand across the U.S., according to the discussion.

Pivotal Quotes: "We have a lot of hidden megawatts." — Chris Shelton: Explaining that existing grid assets are underutilized and could be monetized through flexible access. "The grid shouldn't be a system that you can break by using it." — Chris Shelton: Summarizing the core philosophy behind Grid 2.0 and why protocols should make the grid more internet-like. "It's simple. It's just not easy." — Jigger Shaw: A concise framing of the challenge of turning Grid 2.0 into a working standard and policy framework.

Implications: If adopted, Grid 2.0 could speed data center interconnection, improve grid utilization, attract flexible investment, and potentially lower rates. It shifts the industry from firm-only planning to real-time, standards-based access.