On paper, the Champlain Hudson Power Express was a triumph of vision and patience. Fifteen years of planning. A decade of permitting. Four years of construction snaking 546 kilometers (339 miles) underground and underwater from the Canadian border to the heart of Queens. A price tag of $6 billion, privately financed by Blackstone and Hydro-Quebec. When it switched on this May, it became the longest underground transmission line in North America, capable of delivering up to 20 percent of New York City’s electricity from Quebec hydropower.
A month and a half later, the line has been offline for nearly three weeks.
On July 1, a converter station tripped on the Canadian side, briefly knocking the line out. It was restored, then failed again on July 4 when a damaged cable section was discovered on the U.S. side. As of July 22, the CHPE was still dark. Hydro-Quebec spokesperson Lynn St-Laurent told reporters that repairs should be complete by the weekend of July 25-26. But for a project that was supposed to mark a turning point in the region’s clean energy transformation, the timing could not be more uncomfortable.
The CHPE is an extraordinary piece of engineering. It consists of two high-voltage direct current (HVDC) cables, each roughly 12.7 centimeters (5 inches) in diameter, buried beneath highways, streets, and the muddy bottom of the Hudson River. HVDC technology allows electricity to travel long distances with far lower losses than alternating current, which is why it is the backbone of nearly every major long-haul transmission project in the world. The CHPE’s route crosses 150 bodies of water, threads through two states, and terminates at a converter station in Queens that transforms the direct current back into the alternating current that powers New York’s subway trains, apartment lights, and office towers.
But building the line was only half the challenge. Operating it, it turns out, is the other half.
Two strikes in two weeks
The July 1 outage was caused by what grid operators call a converter trip, a protective shutdown on the Hydro-Quebec side of the system. The second failure, on July 4, was more concerning: a physical cable fault in the U.S. section of the line that required excavation and repair crews to mobilize. As of this week, the line is still down.
Startup glitches are common with first-of-a-kind infrastructure, and the CHPE is pushing boundaries even by international standards. “Any new system presents startup challenges,” said Normand Mousseau, a physics professor at the University of Montreal who studies energy systems. “What matters is how quickly problems are identified and resolved. These are not fundamental flaws.”
Mousseau is right to be measured. HVDC systems are notoriously finicky in their first months of operation. The converter stations that switch between AC and DC are among the most complex pieces of electrical equipment ever built. They contain thousands of power electronics modules that must fire in perfect synchrony, and even a minor control software bug can take weeks to identify and patch. A physical cable fault requiring excavation (the July 4 failure) is a different category of problem, but also one that the industry knows how to fix. The repair timeline of roughly three weeks from fault to fix is within normal range.
Still, the outages point to a deeper reality that the clean energy industry is only beginning to confront. Generating renewable electricity has become remarkably cheap and abundant. Solar panels cost a fraction of what they did a decade ago. Wind turbines grow larger and more efficient every year. But getting that electricity to where people actually live, work, and cool their apartments on a sweltering July afternoon is turning out to be the harder problem.
The drought dimension
Add to this a complication that no cable repair crew can fix. Quebec has been in the grip of an intense drought for three years, drawing down the water reserves behind the dams that power Hydro-Quebec’s vast hydroelectric system. Even when the cable is fully operational, the hydropower flowing through it depends on water that is increasingly scarce.
The CHPE was conceived in an era when Quebec’s hydropower seemed inexhaustible. The province generates more electricity than it can use, and for decades it has sold the surplus to neighboring provinces and U.S. states. But climate change does not respect engineering assumptions. The drought has reduced reservoir levels, and while Hydro-Quebec has not publicly said it will be unable to meet its export commitments, the risk is real and growing.
This creates a strange paradox. The clean energy transition needs more transmission (lots of it) to connect renewable generation centers to population centers. The sunniest and windiest places are rarely the most populated ones. But every new transmission line is a bet that the engineering will hold, that the weather will cooperate, and that the politics of routing power across multiple jurisdictions will not dissolve into lawsuits and delay.
How New York saved itself by planning for the worst
The New York Independent System Operator (NYISO), which manages the state’s grid, did something that in hindsight looks very prudent. In its planning studies for summer 2026, it did not assume the CHPE would be available. The model treated the line as a bonus, not a necessity.
That conservative approach meant that when the CHPE went down, the grid did not buckle. New York City did not face rolling blackouts. The contingency plans (older gas plants kept on standby, demand-response programs that pay businesses to reduce consumption on peak days) absorbed the shock. It was not elegant, but it worked.
Compare this to what might have happened if planners had bet the farm on the CHPE’s timely arrival. The difference between a grid that survives a major transmission outage and one that collapses is exactly this kind of margin. Infrastructure optimists often argue that building more capacity is always the answer. The CHPE’s first two months suggest that reliability margins matter just as much.
Maine’s parallel experience
The CHPE is not alone in its growing pains. The New England Clean Energy Connect (NECEC), another Quebec-to-U.S. HVDC line that began operating in January 2026, has also suffered outages and delivered only minimal additional power to the New England grid in its first months. The pattern is the same: planning and construction were hard, but operations are proving harder still.
Two major international HVDC lines, both built to bring Canadian hydropower to U.S. cities, both experiencing startup struggles within months of each other. This is not a coincidence. It is a signal that the industry’s learning curve for long-distance HVDC transmission is steeper than many anticipated. The technology works. It has been deployed successfully in Europe and China for years. But the pace at which North America is building HVDC capacity may be outstripping the continent’s operational experience.
The bottleneck shifts
For years, the conventional wisdom about the clean energy transition has been that generation is the hard part. Build enough solar and wind farms, the thinking went, and the rest will follow. The CHPE’s rocky debut tells a different story. The bottleneck is shifting from how we make electricity to how we move it.
The physics of transmission are unforgiving. A solar farm can be built in phases, with panels coming online as fast as they can be installed. A single cable fault can take down an entire intercontinental power line for weeks. A drought can silently drain the value of a multibillion-dollar interconnection before anyone notices. Generation is modular and increasingly cheap. Transmission is monolithic, expensive, and exposed to a wide range of failure modes that no amount of planning can fully anticipate.
The CHPE will almost certainly be repaired by the end of July. It will likely run smoothly for years after its initial kinks are worked out. The projects that follow will learn from these early stumbles. But the lesson of the Champlain Hudson Power Express is not about one cable or one converter station. It is about the nature of the challenge ahead.
The clean energy transition has plenty of electrons. The question is whether we can reliably deliver them to where they need to go.