MISO Caps New Large-Load Ramp Rates at 30 MW Per Minute, Forcing Data Centers to Build On-Site Flexibility

MISO will impose a 30 MW per minute ramping limit on new large load interconnections, primarily data centers, after frequency excursions during winter 2026 forced the grid operator into emergency dispatch. Stakeholders confirmed the cap at the April 26 reliability subcommittee meeting, according to RTO Insider. The constraint applies to future hyperscale connections rather than existing campuses, but it changes the engineering precondition for getting on the grid in the largest US RTO by footprint.

A 30 MW per minute ceiling is not a generation rule. It is a load rule. The customer is being told how fast it is allowed to consume.

The frequency event that produced the rule. MISO’s reliability subcommittee traced the winter 2026 emergency dispatch to load steps from a small number of large customers whose ramp behavior was incompatible with the operator’s reserve response time. The grid operator could not procure replacement reserves fast enough to cover the gap between a hyperscale customer demanding incremental megawatts and committed generation responding. MISO declared an emergency and dispatched out-of-merit units to hold frequency.

That is a sequence the operator can survive once. It cannot underwrite a queue full of data center applications on the assumption that it survives every time. The 30 MW per minute cap is the structural answer.

What the cap does to interconnection economics. A hyperscale campus targeting 600 MW of total load can no longer commission its full draw inside an hour. It must either step up over twenty minutes, oversize on-site flexibility to absorb the difference, or accept a slower commissioning curve that pushes revenue dates to the right. Each of those options has a cost. The first stretches construction milestones and tenant move-in. The second requires capital deployed at the customer site. The third defers the entire project’s financial model.

The cleanest engineering answer to a ramp-rate ceiling is a battery sized to absorb the delta between the customer’s instantaneous draw and the rate the grid will accept. The battery does not generate power. It buffers the rate of change. Diesel generators can do similar work but introduce permitting friction in jurisdictions already hostile to data center backup fleets, and they do not address ramp-down behavior the way a four-quadrant battery does.

The policy pattern across RTOs. PJM’s 14.9 GW conditional load backstop, approved in concept and now awaiting a FERC decision in June, already contemplates storage as an eligible flexibility resource for data centers seeking accelerated interconnection. Michigan’s megawatt-for-megawatt rule, signed in March, requires data centers to procure matched storage capacity as a condition of grid access. ERCOT has been studying equivalent ramp constraints inside its large-load study group. The MISO action is the third RTO in a calendar quarter to translate data center load growth into a flexibility requirement priced to the customer rather than absorbed by the grid.

The direction is consistent. Operators are no longer willing to socialize the cost of accommodating hyperscale ramp behavior across the rate base. They are pricing it back to the customer either as a procurement obligation, a queue-position penalty, or, in MISO’s case, an outright operational ceiling.

The commercial read. The MISO rule is written for hyperscale, but the precedent does not stop at hyperscale. Once an RTO has established that ramp rate is a permitted basis for interconnection conditions, the threshold can move down. A 30 MW per minute cap on new large loads in MISO today is a regulatory artifact that any state commission or distribution utility can cite when a large commercial campus seeks an upgraded service in a constrained substation tomorrow. The mechanism, customer-side flexibility as a condition of grid access, is now codified at the wholesale level in three of the four largest US RTOs.

For commercial campuses already on the grid, the immediate effect is indirect. Their existing service is grandfathered. The second-order effect is that the utility’s marginal cost to upgrade their substation rises, because the operator now has formal grounds to demand offsetting flexibility from any new large load sharing that infrastructure. Substation upgrade costs flow through demand charges. Demand charges flow through commercial bills.

The supply-side complication. The same April 26 reliability discussion that produced the ramp cap also surfaced a separate problem: domestic battery cell supply is tightening, not loosening. RTO Insider reported the same day that announced US cleantech manufacturing investment continued falling through Q1 2026, with battery manufacturing among the hardest-hit categories. The 82.4% Section 301 tariff on Chinese LFP cells and the FEOC 55% non-prohibited foreign entity threshold, both active in 2026, narrow the addressable pool of cells that can be procured by a US developer building under ITC bankability constraints.

A grid operator that imposes a flexibility requirement on new large loads at the same moment that domestic flexibility hardware is structurally constrained has created a sequencing problem. The customer is being told it must bring storage to the interconnection table. The supplier base for that storage is shrinking. The lead times that follow are not optional; they are physics and trade policy stacked on top of each other.

Where the rule lands by 2027. MISO’s 30 MW per minute cap will not stay at 30 MW per minute. Operators tend to ratchet ramp constraints downward as load volatility data accumulates, and the winter 2026 event will not be the last frequency excursion attributable to large-load behavior. PJM’s eventual FERC ruling in June will either ratify or modify the conditional-load template, and whichever direction it takes, the precedent will spread. The salient question is no longer whether RTOs will price ramp behavior back to large customers. It is how steep the curve gets, and how quickly commercial customers below the hyperscale threshold inherit the same treatment.

A grid built for slow, predictable load growth is being asked to accommodate fast, lumpy load growth. The first regulatory response was capacity charges. The second was interconnection queue reform. The third, now visible in MISO, PJM, and Michigan, is operational rate limits enforced at the customer’s meter.

Sources