Energy Vault raised another $100mln, and I still don’t get it
Crazy storage tower Energy Vault picked by World Economic Forum
If a successful investor throws $ 110 million of its assets on an idea, is it a good idea? If the World Economic Forum (WEF) honors you as ‘Technology Pioneer’, is your idea sound? Energy Vault has a good chance of being a million-dollar hype.
This article was first published in Dutch on August 17th, 2019. The selection of Energy Vault as one of WEF’s Technology Pioneers 2020 motivated me to translate this piece. This version has been slightly expanded with proceedings since August 17, 2019.
The Vision Fund of Japanese telecom giant Softbank has a track record of venture investments in large tech companies. Uber, Slack, and Wework are some of the startups that Softbank enabled to grow.
With the $ 110 million investment in Energy Vault, Softbank in August 2019 took its first step in the growing energy storage market. Softbanks participation in the Swiss startup really took off in business newspapers and on tech blogs. ‘Finally, an investor that recognizes the importance of energy storage and dares to think beyond lithium batteries.’
Storage of wind and solar power in gravity
The Energy Vault concept is dead simple and consists of blocks (35 tonne a piece) of concrete and six lifting arms. When wind is available and power is cheap, the lifting arms stack the blocks into a tower. As the demand for electricity increases, the lifting arms lower the concrete blocks and the lifting motors work in reverse as a generator. The movements of the six arms are tuned so that the tower can absorb or deliver power continuously.
The claim that lured Softbank, previous investors such as Tata Power and Cemex and media is that Energy Vault provides much cheaper storage than lithium batteries and does not require the specific geologic features (mountains) that enable traditional energy storage in gravity (pumped hydro power). The Energy Vault consists of nothing more than cheap concrete, simple lifting arms and a bit of software. Furthermore, 90% round trip efficiency, a service life of 30 years and (once charged) zero energy loss have been the kind of promises that convinced investors to participate.
Rapid growth in attention and growth capital
Energy Vault ‘demonstrated’ its technology in August 2018, see second video. According to Quartz that got the scoop at the time, it took ‘only $ 2 mln and nine months’ to realize this demonstration. Still quite some time and money, only to lift some oil drums filled with concrete. At that time I ignored the hype. A few months later, Indian company Tata Power ordered a full scale tower, able to store 35 megawatt hours and deliver 5 megawatts of gravity power. I barely managed to ignore the hype.
The tower ordered by Tata should have been operational in 2019, somewhere in Italy. According to Energy Vault it would be the first off many. Every subsequent tower should be constructed cheaper than the last.
On the occasion of this translation I did some research as to exactly where in Italy the tower has been, or should have been, build. A 130 meter high stack of concrete blocks is hard to miss and should get the attention of at least local press. The intended location seems to be the Swiss town Arbedo-Castione, near the Italian border. Good enough. The project has been scaled down to be 60 meters high. Furthermore the prototype will not generate power but only serves to test the software. Fair enough. Unfortunately, I was not able to find a new (intended) delivery date for this scaled down prototype.
Simplicity is great, but a simple thought is not an energy storage system
In the past years, I have observed the progress and growing media coverage for Energy Vault with increasing amazement. When I saw the video of the first demonstration in 2018 and read Quartz’s report, I thought of a single, over-enthusiastic journalist. The hype has developed without much criticism and now this startup is $ 110 mln and several awards richer. I don’t understand how this has happened. But I can’t ignore it any longer.
Let’s risk making a complete fool of myself and question the skills of an investor with $97 bln under management. A few points of interest:
- Lifting speed. Energy Vault promises up to 8 megawatts of power. With two blocks of 35 metric tons moving with or against gravity, this means that the blocks must have a speed of almost 12 meters per second (43 km/h). That is a large step compared with the oil drum demo (video above). If we assume that the oil drums are 90 centimeters high and it takes 6 seconds to put the highest drum on the ground, the demo only reaches 1 meter per second. All the more worrying that the playback speed of the video has been tweaked, by how much is unknown to me.
- Profile of power delivered. Concrete blocks will not last for 30 years if you let them crash into the ground at 43 km/h several times a day. A good part of the 120 meter height will have to be used to get the blocks up to speed and slow down again. Supplying stable power at 8 megawatt with just six lifting arms suddenly is far less simple than it seemed. Especially considering the height of the tower changes continuously during charging and discharging.
- Wind loads. In all ‘artist impressions’ provided, Energy Vault presents the storage tower in the middle of a wind farm. Logically, wind farms are located in places where the wind blows well and often. Wind and cranes are no friends. A 35-ton block of concrete has a volume of 15 cubic meters and will capture wind loads, especially while charging with excess wind power. A block of concrete on a cable of 120 meters will swing. Stabilizing blocks with guide ropes will greatly increase complexity. Picking up the blocks from the ground with an ‘empty’ cable will be even more exciting.
- Not in my backyard. In the above-mentioned artist impressions, Energy Vault always presents one tower per wind farm. But one such tower (the largest 80 MWh version) only stores the daily electricity production of one modern wind turbine. If Energy Vault is to be the breakthrough that makes storage really cheap, think about one concrete tower for every turbine installed. That will probably raise even more opposition than the turbines themselves. In addition, Energy Vault then blocks the wind from the turbines.
- Useless blocks. Ideally, with this concept you would put all blocks directly on the ground. Then however the lifting arms would have to be unrealistically long. As the energy tower discharges and shrinks, a ring is created around the tower. About halfway through the conversion, the tower is the same height as the ring around it. The stored energy is then used up. All concrete blocks that are still in the inner tower are only there for the height. They do not contribute to the storage capacity. The average height at which the blocks go up and down is therefore not 120, but a bit more than 60 meters.
- Foundation work. The largest Energy Vault offered stores up to 80 megawatt hours. With an effective height difference of 60 meters this Energy Vault has to have a useful mass of 480,000 tons. The useless bottom half of the tower adds another 480,000 tons. Including the steel upright and lifting arms and a smooth floor of concrete or asphalt to be able to stack neatly, think in excess of 1,000,000 tons per 80MWh tower. That is twice the weight of the 830 meter high Burj Khalifa in Dubai, but on a much smaller diameter. Good luck with foundation works.
Thousands of single points of failure
Purely on practical aspects, the Energy Vault proposal seems daring to say the least. More arguments might not be necessary to depreciate the concept and thus the investment of Softbank. But there are more arguments.
- Wear and tear. Energy Vault claims the towers will last 30+ years and operate unmanned. Concrete blocks that you pick up thousands of times over that lifespan and put down somewhere else, in windy conditions, wear out quick. Steel cables that continuously roll over pulleys and are subject to 35 tons of load wear out quick. Just like the pulleys, the motors and the lifting arms themselves will be subject to various dynamic loads. An unmanned service life of 30+ years would be much more than amazing, Mike.
- Expect hiccups. These towers consist of several tens of thousands of blocks. That means tens of thousands of movements a day where something may go wrong, especially if the wear increases. If a block gets entangled or falls on its side during lifting, one-sixth of the tower is out of order. At least one-sixth.
- Catastrophic failure. It is even more tricky when no one notices that a block in the (unmanned) tower has defects. If a bad block ends up low in the tower after a few times of shuffling and collapses underneath his colleague blocks atop, this results in a major chain reaction that can permanently disable the entire tower (and it’s expensive foundation).
A very risky ‘breakthrough’ for climate policy
The basic idea of the Energy Vault is to store surpluses of renewable energy for use in moments of shortages. In practice, the tower will mainly compete with gas-fired power stations. It is inconvenient that the towers consist almost entirely of steel and concrete. Two of the most CO2-intensive materials imaginable. Let’s consider the carbon balance per block:
- CO2 emissions during production. The concrete blocks will have to be strong because they bear the weight of all the blocks above. They probably should consist of roughly 15% cement and 4% reinforcing steel – or 5 tons of cement and 1.5 steel per block. For each block produced about 3 tons of CO2 will be emitted just for steel and cement.
- CO2 reduction in use. The Energy Vault stores an average of 5 kilowatt hours (kWh) per block (35 tons, 60 meters, 90% efficiency) per cycle. If we assume that the direct competitor is indeed a gas plant (± 400gr CO2/kWh), Energy Vault saves 2 kilos of CO2 per storage cycle per block.
- Carbon ‘payback time’. Assuming that the towers capacity is fully utilized 200 times a year, it takes about 7.5 years for a block’s production emissions to be offset by saved CO2 emissions from gas plants.
The above concerns only the moving (useful) blocks. With the CO2 emissions related to useless bottom halve, the steel hoisting construction, the foundation, the transport of all materials and the construction itself, you will quickly reach a carbon payback time of over 15 years.
It gets even worse when Energy Vault customers use the towers not for short-cycle but (as advertised) for long-term storage. If the tower is not stacked at least 100 times a year for 30 years straight, this tower will effectively increase carbon concentration in the atmosphere.
In short. This is another innovation that I hope I didn’t understand at all. What do I overlook? The comments are open.
Bron: Energy Vault, Financial Times, Quartz / Imagecredit: Energy Vault