Two first-of-a-kind projects in Pennsylvania are trying to replace power generation with power reduction to balance the grid. If they can turn lots of equipment down with the accuracy and speed necessary to match the big natural gas-fired turbines that do the job today, they stand to make a good deal of money at it, too.

Startups Viridity Energy and Enbala Power Networks are behind the two projects, which Mid-Atlantic grid operator PJM announced Tuesday morning. They’re the first under a new PJM program that lets small-scale (100 kilowatts to 500 kilowatts) projects sell their power into frequency regulation markets, which balance the ups and downs of grid frequency on a second-by-second basis.

That’s a lot faster than demand response -- that is, turning down power use to ease grid constraints -- happens today. Most of it is done manually via phone or email alerts. Even automated demand response (AutoADR) usually takes at least 10 seconds -- and usually minutes or longer -- to execute the complicated commands needed to downshift electricity usage by manufacturing lines, air conditioners, lighting banks and other commercial and industrial power loads.

But Viridity and Enbala say they’ve got it down to the four-second intervals required by PJM to compete against “peaker” plants, the natural gas-fired plants that manage the vast majority of the regulation market. In fact, both projects have already sold their services on the market that serves PJM’s 15-state territory.

How does it work? In Enbala’s case, the Toronto-based startup has linked up water pumps at an American Water wastewater treatment plant in Washington County, Pa. Those usually run at less than full-bore capacity, which means they can turn up or turn down to match the grid’s needs, CEO Ron Dizy said in a Tuesday morning conference call. That’s kind of like how turbines can slow down to restrain frequency shifts, as well as speed up to increase them.

Enbala’s technology connects individual pumps and central plant power controls, but Dizy envisions a network of lots of different plants in different parts of the region that could provide a grid-wide set of assets to call on. In a way, it’s using the stored energy of water, and how fast you choose to move it, as a way of mimicking the way a battery or a pumped hydro facility could provide stored power.

In Viridity’s case, it’s connecting an actual battery to the grid. Lead-carbon battery maker Axion Power is providing the battery, as well as the power-shifting capacity of its New Castle, Pa. manufacturing plant, to back it up.

It isn’t the first project to bid energy storage of some kind into U.S. frequency regulation markets. Bankrupt flywheel maker Beacon Power has had 20 megawatts of kinetic energy storage selling into the New York ISO since this summer. But where Beacon’s costs were all tied up in its expensive-per-kilowatt energy storage systems, the Viridity-Axion setup is using its battery more like an interface for the plant’s potential power, Viridity CEO Audrey Zibelman said. As the battery acts to inject grid power, it’s being recharged by the facility, which in turn is being adjusted on an ongoing basis by Viridity’s software.  

Pennsylvania-based Viridity raised $14 million in January, and is balancing and storing power at several university campus microgrids, Philadelphia’s SEPTA train system, a wind power storage management project in Pennsylvania, though CEO Audrey Zibelman didn’t say if any of its other PJM projects might start bidding in frequency regulation markets soon.

Indeed, hooking up a plant to support split-second grid support entails a lot of investment into intelligent building sensors and controls, which then lends itself to general energy efficiency improvements. Add energy savings to the revenues that can come from frequency regulation and you’ve just turned an expense into an asset.

The neat thing about doing grid balancing this way is that you don’t have to keep building new peaker plants, Zibelman noted. The vast majority of the capital cost -- the sites that are using all that adjustable power -- is already in the ground. All you have to do is hook it up properly.

The returns could be significant. Enbala already claims revenue potential of $100,000 to $200,000 per year for an upfront investment of $50,000 or so for each water treatment plant, and Dizy said the company was lining up more customers, though he wouldn’t name names. Zibelman noted that average prices for regulation service in PJM are enough to pay $160,000 to $240,000 per year for each megawatt of storage delivered, which would work out to $8,000 to $12,000 for a 500-kilowatt system, to bring it back to the scale of these projects.

Under a Federal Energy Regulatory Commission ruling last month, frequency regulation resources will be getting more money for faster and more reliable performance, which should make demand-side resources that much more lucrative. That means frequency regulation could help pay for all the other benefits of this kind of fine-tuned, fast-acting power control, like absorbing intermittent wind and solar power into the grid, Zibelman noted.

Of course, it’s hard to turn a building, factory, campus or real estate portfolio into a “closed-loop, non-disruptive” power shifting entity, as Zibelman put it, to serve power markets like a generator does. Viridity and Enbala aren’t saying how much it cost to set up their projects. No doubt the industry will be watching to see if the projects can do what they say they can -- and at what price.