What’s a startup with just under $7 million in venture capital going to do with $125 million worth of Tesla’s grid batteries?

Last week, Advanced Microgrid Solutions announced plans to install up to 500 megawatt-hours of Tesla’s Powerpack battery systems over the coming years. That would make it the biggest single customer yet for the electric automaker’s new push into the energy storage field.

It’s also the smallest and least well-known company on a list of Tesla battery customers that includes utilities Oncor and Southern California Edison, demand response provider EnerNOC, data-center giant Amazon Web Services, and the country’s biggest utility-scale storage project developer, AES Energy Storage.

Advanced Microgrid Solutions, by contrast, has kept very quiet on many of the details of its business, including its funding to date. It's in the midst of raising an $18.8 million funding round with some 14 participating investors, according to a filing with the U.S. Securities and Exchange Commission.

But the San Francisco-based startup does have a contract with Southern California Edison to deploy 50 megawatts of batteries at sites across the West Los Angeles Basin region, as part of the utility’s Local Capacity Requirement procurement last year. Those storage systems need to be capable of providing four hours of capacity per day, which means that AMS will need 200 megawatt-hours of batteries -- two-fifths of the total it plans to get from Tesla -- to meet the terms of that agreement.

As for how AMS will line up financing for its SCE project and those yet to be announced, “We have a financing plan in place that we’re not talking about now, but we will be talking about it in the near term,” Jackalyne Pfannenstiel, AMS co-founder and board member, said in a Friday interview. “We expect this to be equity-financed.”

Pfannenstiel, a former chairwoman of the California Energy Commission and assistant secretary of the U.S. Navy for energy and environment issues, declined to say how much financing AMS was seeking to raise, or how much it expected its 500-megawatt battery order to cost. But Tesla has said its 100-kilowatt-hour Powerpack units will cost $25,000 each, or $250 per kilowatt-hour.

That means that 200 megawatts would call for 2,000 units, or $50 million, and 500 megawatts would call for 5,000 units, or $125 million, at Tesla’s currently stated price. That’s a rough calculation, and doesn’t include the significant share of additional costs that go into a working energy storage system. On the other hand, it also doesn't include the potential for prices to fall over time, since AMS won’t be buying all its batteries right away.

But perhaps more important than battery price is what value AMS is able to capture from them. In that respect, its contracts with Southern California Edison are expected to provide a solid, dependable revenue stream for projects that will give the company a chance to prove it can manage its fleet of distributed batteries as a grid resource, while also providing benefits to the buildings where they’re installed, she said.

“We are operating on both sides of the meter, initially, as a virtual power plant for utilities, but we’ll also be working with the individual customers,” she said -- a combination the company has dubbed a "hybrid-electric building." AMS plans to install its first 10-megawatt project for SCE in Irvine, Calif. next year, and expects to install 200-kilowatt to 600-kilowatt blocks of batteries at individual sites.

Those systems’ primary role is to provide the utility with an aggregated resource that can be dispatched to inject power during times when it’s needed to meet steep upward ramps in energy demand. In Southern California, those most often come during the late afternoons or early evenings, and are normally covered by natural gas-fired peaker plants.

SCE’s West LA Basin area is facing the future prospect of significant power shortages and difficulty delivering stable and reliable power to congested areas of the grid, due to the closure of the San Onofre nuclear power plant and the planned closure of gas-fired power plants that use seawater for “once-through” cooling.

Batteries can store power at night or during peak solar production hours, then discharge it to meet those system peak needs, without burning natural gas and emitting carbon dioxide or other pollutants. They can also be installed in much smaller increments than a single new power plant, and can be located at specific trouble points on the grid. These advantages could outweigh the fact that they still cost more than gas-fired power plants to build and operate, don’t actually generate any power on their own, and don’t last as long.

Under the terms of its contract with SCE, “We are a power plant, in essence, a virtual power plant, selling power to the utility,” she said. At the same time, “there are other applications on the customer side, behind the meter, that this helps, and I think we’ll have conversations with other customers as we go.”

This isn’t a new concept. Companies like Stem, Green Charge Networks and Coda Energy, and Tesla and partner SolarCity are installing batteries in buildings to reduce demand charges and, potentially, serve as grid-responsive energy assets. While AMS hasn’t specified just how it plans to work with customers that are hosting its batteries, “whether it’s their own demand response plans, or other behind-the-meter uses, we have the software to help them,” she said.

AMS is working on “a combination of proprietary software we will develop, and pieces we will buy from existing providers.” There are a number of energy storage software providers out there -- Greensmith, Younicos, Geli and 1Energy are some of the notable companies working with multiple battery types and project developers -- but Pfannenstiel didn’t say which ones AMS might be working with on this goal.

Picking the right batteries and software for particular grid tasks is an important part of the energy storage value proposition. So is building in flexibility for how those assets may perform different revenue-generating tasks over their working lifespan. Pfannenstiel didn’t go into details on which potential grid services or customer-facing applications the startup was exploring for its future projects.

“You need to have some real-life experience with this stuff working,” she said. “I’d say that over the next five years, you’re probably going to see a big change in the perceived risk of battery and storage technology generally.”

That should help prove the business case for future contracts, whether to meet California utility procurements under the state’s 1.3-gigawatts-by-2022 energy storage mandate, or in other parts of the country, she said.

“We’re starting in California, but we think -- in fact, we’re sure -- that this business is applicable throughout the country,” she said. “Anyplace where a utility will need additional capacity, this is a good way of them getting that capacity.”

Tesla’s lithium-ion batteries won out in AMS’ most recent RFP in terms of cost, performance and scalability, but the startup isn’t limiting itself to one battery technology, she added. “We’ve said often -- and we believe deeply -- that we need to be technology-agnostic.”