SustainX

A (short) Story of SustainX

The short overview of my time at SustainX goes something like this:

I joined SustainX as employee #10 in Jan 2010 as a Heat Transfer Testing Intern, and they kept me on as a R&D Project Engineer and eventually I became R&D Manager. We grew to about 60 employees, and by Q3 of 2013 we successfully built a 1.5MW-1MWhr near-isothermal energy storage system in Seabrook, NH. The machine was an incredible technical success, and its “teething” period was one of the shortest I’ve ever seen: from mechanically complete to full power in just a few weeks.

But the cleantech investment scene had essentially completely died in 2011 (see Can Energy Startups Be Saved? or Venture capital and cleantech: The wrong model for clean energy innovation).

So in 2014, instead of finding investment, layoffs began. From 60 people to ~30, and some weeks later, down to ~15. Eventually, we were down to 6 people and to stay afloat for longer, we merged with our competitor, General Compression (who was also on the ropes), to form GCX. 2015 was another skeleton crew year, hoping against hope some new investment would come through. January 2016, we shut down for good. David Perkins and I were the last two SustainX people.

Primer

The technology that SustainX developed was essentially a large reciprocating piston compressor (2MW ~2700HP) to use electricity to compress air up to 3000 psi – this stores energy. When the energy is needed on the grid, the same machine is then used as a pneumatic engine (1.5MW), with the high pressure air providing the motive force to drive the engine to produce electricity. In 2013, I gave a talk to UNH Chemical Engineering students – this presentation is a crash course in energy storage, SustainX and its tech:

Isothermal Primer

A lot of my SustainX work focuses on mixtures of air & water – in the first years, a water spray in air, then in later years, foam. The “secret sauce” of SustainX was isothermal compression or expansion – maintaining a near-constant temperature by actively doing heat transfer inside the machine – and water spray or foam was the key to making that work at speed & scale. For foam, the water of the foam provides the heat sink or source for the air it envelopes, and that’s what keeps the temperature from changing drastically during a rapid compression or expansion.

Research Engineering

Tackling technical unknowns with test stands

The general approach we took was to identify risks/concerns/questions, see how far our current analysis could take us, then commit to building a ‘test stand’ – an apparatus for experimentation. As research engineer & manager, I led and executed the process from concept to conclusion: develop the impact of risk, what would the test stand look like to learn about & thereby reduce the risk, then research it, design it, build it, run it, analyze it, and provide conclusions on how to address the risk.

Heat Transfer Test Stand

The Heat Transfer Test Stand (HXTS) demonstrated that we could spray liquid water into high pressure air and that water would absorb the heat of compression / provide heat during expansion to keep the process near isothermal.

HXTS showed a water spray inside the cylinder during a 3 second compression or expansion over a 5 ft stroke could achieve 95% of the ideal isothermal process, which was pretty incredible.

This water spray heat transfer method was the design for the 40kW Pilot system. In the Pilot, the hydraulic drivetrain – an electric motor/generator connected to a hydraulic motor/pump – moved hydraulic cylinders which in moved the pneumatic cylinders via a common platen.

The water spray heat transfer method and its systems were patented:

Foam Origin Story

The foam story fits a couple archetypes: ‘innovation through accidental discovery’, and ‘If you can’t beat them, join them’.

As you’ll see in the Corrosion section below, air + water + steel is a recipe for rust. We trialed many different corrosion inhibitors added to the water. Almost anything dissolved in water tends to reduce its surface tension and increase its tendency to foam. We found certain inhibitors generated a bunch of foam when we emptied & reset the test stand, and those tests achieved amazing (98% isothermal) heat transfer results at the highest speeds we could achieve on the HXTS (1 second stroke). Foam was enabling 3x the speed AND even better efficiency.

The great heat transfer results with foam were, at first, a novelty we didn’t know what to do with, then an annoyance as foam made a mess of things and slowed testing down. So we tried many more corrosion inhibitors to try to keep away from foam, but we found ourselves locked into a trade-space: good corrosion prevention but too much foaming, or low foaming but inadequate corrosion prevention.

Roughly at the same time, we received news about the next-gen high-efficiency hydraulic drivetrain being developed by an industrial partner: the hydraulic machine was not achieving its efficiency targets. Efficiency of all steps is critical on energy storage as inefficiencies cut you twice: once on the way in & charging from the grid, and again on the way out & discharging to the grid.

The original technology roadmap had SustainX transition to a crankshaft drivetrain, but after bringing the hydraulic drivetrain to scale & market. In light of the hydraulic dead-end and the promise of foam heat transfer, SustainX changed its plans dramatically. We added a few years to the product development timeline (which causes all kinds of heartache for investors), and began to work on a ‘power unit’ (the compressor/expander) 38X more powerful than what we just built, based on foam which knew next to nothing about and a drivetrain that was completely new to us. We also had to bring together many other critical items simultaneously: development of innovative hydraulically-actuated valves, high-inlet-pressure water pumps, and an Industry 4.0 control system; and procuring the long lead items (like a marine diesel engine and wind turbine permanent magnet generators) and manage the major plant construction project. We had to discover the physics/design/buy/build the plane as we were flying it.

And we built it in under 3 years.

High Speed Heat Transfer Test Stand

We wanted to see if foam could work at the speed of a 120RPM crankshaft engine – a 1/4 second, 5ft stroke. I designed the High Speed Heat Transfer Test Stand as a modification to the original HXTS such that the same hydraulic flow rates would produce the 1/4 second stroke. Tests on this sounded like a cannon.

We successfully demonstrated 98% isothermal efficiency on a 1/4 second stroke – 12X faster than water spray and more efficient.

We proved to ourselves the promise of foam was real – fast, high efficiency heat transfer – and now we ‘just’ needed to figure out everything else about foam.

The Racetrack

The Racetrack was our first attempt at foam at scale: 16″ diameter ductwork in a loop, with a inline centrifugal fan driving the air, centrifugal pump for spraying the water, and an axial fan converted into a “weed whacker” to break down the foam.

We learned:

• Spraying horizontally makes it difficult to make good foam –> new design rule: spray water downward onto the screen, with air also flowing in the same direction
• We needed to have more ability to try many different screen configurations
• The Weed Whacker broke the light foam down, but whipped it into a wet, dense foam. Ultimately, you just have to wait for the foam to breakdown on its own (measured in hours). Foam Separation would become its own topic and challenge.

BenchTop Foam Test Stand: the workhorse

With the learnings from the Racetrack, we built something smaller (3″ diameter pipe) to be easier to handle & modify, air & water flowing downwards, and more modular. We would end up doing hundreds upon hundreds of tests with this apparatus.

With our learnings from all the foam test stands, we successfully jumped from foam-making at 10 gallons per minute in a 3″ pipe diameter to 10,000 gpm in a 48″ diameter duct in the full-scale system.

MPV: Mid Pressure Vessel

We wanted/needed to demonstrate how to make foam flowing upwards and at scale, but at atmospheric pressure & not at process pressure (250psi) so we could outfit the test stand with observation windows.

(videos start to work after about 8 seconds in – it’s a website bug I haven’t figured out…)

Foam Patents

Foam Rheology

‘Expansion Ratio’ or ‘ER’ is a term from firefighting foam – it’s the ratio of the volume of the foam to the volume of the liquid in the foam.

Thermodynamics and Data Analysis

Corrosion & Coatings & Inhibitors

Air, water and steel may make a SustainX machine, but it’s the perfect recipe for rust. We tried different avenues to prevent corrosion (alloy composition, chemical inhibitors, and coatings), and many test stands. Our first corrosion test stands were just 6 Home Depot buckets with aquarium pumps splashing metal coupons with a variety of corrosion inhibitors – that was in fact a great & informative start! We got more sophisticated, as needed, from there.

Sunapee – The 1.5MW-1MWhr Isothermal Compressed Air Energy Storage System

Sunapee, a mountain (and lake) in New Hampshire, was the project name for this 1.5MW-1MWhr Isothermal Compressed Air Energy Storage System built in Seabrook, NH.

Reflections

The Uniqueness of SustainX

Paraphrases, as I don’t have direct quotes:

• Mike Schaefer, controls engineering lead: SustainX is a once-in-a-lifetime job
• Jeff Modderno, fluid systems director & Rich Bradshaw, valve development manager: only Fluid Power Products (FPP) compares to SustainX’s humble but effective technical culture
  • FPP was modeled after Sun Hydraulics, a HBS case study company that re-invented itself with a radically flat organization structure and became renowned for its innovative designs, its quality products, and its highly ethical standards for business dealings.

This is the best description I’ve found to date about SustainX: