[CSP] plants being built,’ he says. ‘They don’t produce anything until they are 100% complete. You need to install all 200,000 mirrors and build a seven-storey tower and have every last pipe and heat exchanger and water cooling tower built before they can produce 1 kWh.
‘Our units are very small (100 kWe). After one unit [is installed] we can produce power; this also means we can build five units at a time, which will each come online one day after they’re finished. We’ve calculated the initial systems, the controls are hard-wired, there are wireless controls to give instructions to the mirrors to move, and this cuts down on construction time. With 10 crews in the field, we can build 10 units every three months. The only real limiting factor is availability of the components.’
How it works
The Tulip system uses sunlight concentrated by an 800-square-metre field of sun-tracking mirrors, or heliostats, onto a solar receiver, heating air which drives an Ansaldo Turbines gas-fired turbine. The air (at 88 cubic metres per minute) passes through the receiver in 1.4 seconds and is heated from the intake temperature of 600 degrees C to around 1000 degrees C, ‘which is the temperature we need in order to yield the 100 kW nominal output from turbine,’ says Rosenzweig. A recuperator is installed to reclaim heat from the exhaust before its release to the atmosphere at 850 degrees C. ‘We recoup 600 degrees of it, more or less,’ says Rosenzweig.
Because the air coming out of the solar receiver is not hot enough to drive the turbine to its maximum capacity, he says, if the temperature is less than 950-1000 degrees C, the system will automatically turn on the combustion chamber and start the fuel pump to provide the missing heat which keeps the air temperature steady. ‘This occurs without any need for human intervention, so we can set the system to run either at a set output level of 65 kW, 70 kW, 100 kW,’ he says.
Aora’s standard plant features 50 heliostats, but the exact number depends on its location. The Israeli plant has 50 mirrors while the Spanish plant has 52 ‘because the location is somewhat further north,’ says Rosenzweig, and more mirrors are needed to achieve the turbine’s nominal capacity year-round.
Each Tulip plant is scaled to its location, Rosenzweig explains. While the land area needed will be similar – around 2500-2700 square metres – the area used for each plant ‘will depend on where it is, because at higher latitudes the sun spends more time lower on the horizon so we have to space the heliostats further apart to avoid blocking and shading’. The Israeli plant is 2500 square metres, while the Spanish plant is 2800.
And the heliostat field is configured differently for each plant, based on a program developed by Aora to optimize the mirrors’ locations in relation to the tower in order to maximize annual production. ‘To model the heliostat field we chose 100 different locations,’ says Rosenzweig, ‘and modelled the transit of the sun every day for a 365-day cycle to see which location yields the highest predicted reflection of heat and power. We can optimize for June 21, July 21 or March 21.’
The tulip-shaped towers are between 30 and 35 metres tall, with a focal height of between 32 and 37 metres. ‘Our modelling shows that if height is an issue we can cut as much as five metres without affecting the total output,’ Rosenzweig notes, ‘and add a heliostat if the height of the tower becomes an issue.’ For example, the Israeli plant ‘is on the flight path into an airport, so [the tower is] below the maximum height allowed.’
‘Especially when dealing with remote locations where you have a single generator as the only source of electricity,’ Rosenzweig says, ‘it’s very important to keep the output steady, otherwise you have brownouts, refrigeration and cooling outages, and it can cause harm to the compressor if the power levels keep shifting. At night the issue is simply that there is no solar heat, therefore all power generation is taken over by the alternative fuel.’
For this fuel, he says that because the Tulip system is aimed at villages in remote locations, ‘we have to use processes that make use of whatever biomass matter is available in the location. We can’t pre-specify the raw material or the biogas conversion process.’
For the firm’s two Ethiopian projects, it aims to use waste by-products from the country’s sugar industry – ‘basically molasses’, Rosezweig says. This material ‘has a lot of energy but the process of turning it to biogas is difficult’, and the company is working with an Ethiopian university to develop an efficient way to process this waste into biogas.
The Spanish installation uses diesel as a backup fuel, ‘which is not exactly a green fuel,’ Rosenzweig acknowledges, and which also presents some problems for the system. ‘The components of diesel fuel are not very forgiving on combustor equipment. We can run the system, but we need to disassemble the combustor and clean it on a regular basis. We’re using it to demonstrate the concept of hybrid operation but it’s not something that we imagine to be the final configuration installed.’
The Israeli plant also originally used diesel as its backup fuel, but natural gas availability has since increased and Rosenzweig says Aora has applied for a permit to build an underground gas storage tank, with a plan for fuel to be delivered by tanker trucks.
‘The gas turbine is agnostic to which fuel you’re using,’ he says, although ‘you obviously have to make changes to the combustor. You have to take into account the volatility of different fuels, and make sure you don’t have combustion occurring before you want it to happen.
‘Turbine suppliers are quite aware in terms of adjusting the combustion chamber in terms of type of fuel,’ he adds. Changing from one fuel to another requires around two to three hours to disassemble the combustor and swap out components to make sure they’re compatible with the fuel type, he says, but this is ‘not like when you build a plant and are stuck with the alternative fuel you chose on day one. With our system it can be changed with relatively little effort.’
The problem with storage
There is no energy storage component with the Aora system; instead the system runs on solar power during the day and switches over to burning its alternative fuel at night.
‘We think storage is inherently inefficient,’ Rosenzweig says, ‘because if you’re generating additional power in order to store it, you have to increase the capacity of your initial generation system. A PV installation would be effective six hours a day, so if you had to store 18 hours’ worth of power you’d have to use four installations side by side, one to provide power during the solar day, and the other three units to provide power during the 18 hours you have to use stored power.
‘Given the cost of a storage system, say batteries, you would need 18 MW of batteries operating with an initial generation capacity about 4.5 times the size of the rated required efficiency of the batteries – less than 100% – so you can’t just put in three times the additional generation capacity. You have to put in about 3.5 times that, because you won’t be able to put everything you generate into batteries and, once in, you won’t be able to take it out.
‘Even if we get the cost down, storage is not a very efficient system. Everybody forgets about the additional cost of surplus capacity,’ he says.
‘Even with heat, you have to size the system so the heat you have left is enough [for nighttime hours], which precludes the use of that heat for gainful purpose during the day. We are looking to use our heat for absorption chilling, and if we were storing that heat for backup power or running on non-solar times that option would not be available.’
A desert flower
An Israeli architect designed the tower to ‘look like a desert flower,’ Rosenzweig says. ‘We thought if we were providing something environmentally friendly we wanted to enhance the environment rather than create an eyesore.
‘We modified the original design, kept the concept and have made the flower on top a bit smaller. The original one was larger because it had a lot of monitoring instrumentation in it. Now that we’re confident that we understand the airflow inside the receiver and temperature losses between the receiver and turbine etc, we’ve modified it a bit so it looks like tulip before the flower opens.
‘Studies in terms of cost vs a standard vertical-type structure show that there’s not all that much difference. We decided to stick with the Tulip and rather work on optimizing the design so we get the most out for the cost. We changed the design so the tower itself, the stem, is enough to support the tulip on top with guy wires which helps lower the footprint and allows us to put units closer together.’
The only limiting factor for rapid development of multiple Tulip plants is component availability, Rosenzweig says, while the ‘real constraint there is availability of the turbines’.
‘My vision for the next five years would be installing 250-300 units per year in hopefully five different locations or countries. That will be constrained by the ability of the turbine supplier to supply the turbines. We would hope to negotiate delivery availability that would grow as we grow. This would mean the turbine supplier going back and negotiating with his supply chain to have all components available.
‘I’m confident that this five-year vision is quite achievable once we have our first two to three major installations,’ he adds, noting that the company is already working with Italian developers to install ‘a large number of units’ at a location in Sicily, and that these will be biogas-fuelled.
While the company is involved in ‘other initiatives in other African countries’, Rosenzweig terms Aora’s Ethiopian project the company’s ‘first foray into what we see as our niche market, which is providing power in remote locations in off-grid areas’. The company has signed a memorandum of understanding (MoU) with Ethiopia’s government for the development of three research units initially, although Rosenzweig sees the potential for 350 to 500 Tulip plants in the country.
However, there are some potential challenges. For one, there is the higher cost of CSP as opposed to PV. ‘In Ethiopia, one of the ways we’re able to deal with that is they have a reasonably developed industrial infrastructure and we will be teaming with a local industrial company to manufacture everything except the turbine, receiver and some of the very sophisticated ceramic insulation and other components inside the receiver,’ says Rosenzweig. ‘But basically the whole balance of plant will be manufactured locally. This helps us keep costs in check, and we’re looking at expanding the scope of the Ethiopian agreement to cover other countries given manufacturing in Ethiopia.’
By Tildy Bayar