Extremely Efficient Energy Storage Based On Three-Level Silicon Carbide Power Module

SiC makes the difference – both consumers and energy providers benefit from high-efficiency energy conversion between decentralized energy storage and the main power grid. The attraction for cost-effective and environmentally sustainable energy storage peaker plants is unlimited.

Decentralized battery energy storage systems (BESS) reduce the reliance on fossil fuels and allow a wide range of other powerful benefits. Power grids need sufficient extra capacity to work properly during peak demand periods and to satisfy reliability requirements. This disadvantage is strengthened by the fact that the energy sources the energy providers prefer to use are only available with different time lags. In order to counteract large immediate peak demands, energy providers implemented pricing structures that led to a common use of combustion turbines (CT) in the past providing instantaneous energy. The benefit-side of the overall cost-benefit analysis will make energy storage more attractive to customers. If a proper analysis is done, not only acquisition costs, but also operating costs must be taken into consideration. The price for battery cells plus peripherals and the inverter competes against the acquisition and operating costs of CT plants.

 

Flexibility of CTs versus energy storage

The well-known output characteristics, acquisition and operating costs of CTs led to high acceptance in the world for adding peak power capacity to the main power grid. Furthermore, they were historically proven to be very reliable. However, when discussing the good points of CTs, the benefit of meeting shorter-duration peaking capacity requirements compared to coal or nuclear plants needs to be moved from the benefits of CTs to the ones of energy storage: Compared to energy storage, CTs are turtles in start and ramp-up phases. In addition, distributed energy storage facilitates the integration of variable wind and solar resources. The California Public Utilities Commission (CPUC) reported the negative impacts on grid stability occurred primarily due to solar PV. The potential of energy storage has been discussed in “Guide to procurement of flexible peaking capacity: Energy storage or combustion turbines” (by Chet Lyons, Energy Strategy Group, 2014): Flattening system load with energy storage synergistically reduces the need for all major categories of utility asset investment, including generation, transmission and distribution.

 

Table 1: Projected costs (price) for a 1-MW, 4-hour redox flow battery (Source: ViZn Energy, used by Chet Lyons)

 

Cost-benefit analysis

CTs are often chosen to be the most cost-effective solution due mainly to their huge acceptance. At the same time, temperature, air pressure and humidity have a huge impact on the efficiency of those air-polluting machines. The CPUC defined use cases for storage in the California energy market that are described by Chet Lyons in more detail. In this paper, a 1-MW 4-hours redox flow battery is compared to a conventional simple cycle CT with a Capex of $1,390 per kW assuming mid-range CT costs since there are so many on the market. By 2017, storage can be roughly competitive with many conventional simple cycle CTs. More advantages can significantly increase the cost-effectiveness. The modular architecture of energy storage improves asset reliability and system resiliency through redundancy. The Electric Power Research Institute (EPRI) found out that distributed storage has higher value than central station storage mainly due to added distribution upgrade deferral and circuit stability control. Simply deferring capacity investments, regardless of type, lowers capital investment risks and improves total return on assets. The cost trend of energy storage as environmentally sustainable zero emissions energy resources that is dropping confirms the added value.

 

Figure 1: Exemplary E3 power module for a 60-kW solar inverter on IGBT basis (footprint: 62 mm x 122 mm)

 

High operational savings with three-level topology

Three-level advantages with compact NPC2 topology lead to high-efficiency inverter solutions due to smaller output voltage steps, reduced switching losses and doubly effective switching frequency. The design with the newest SiC technology allows a very high switching frequency of greater than 50 kHz that minimizes the required output filter dimensions. In addition, only SiC-type gate drivers are needed for the inverter. Due to its physical nature, the use of MOSFETs offers further advantages in the partial load area compared to IGBT solutions and in a reduced size package by more than 35% from E3 to E2 as shown in Figure 1—2. In conclusion, the E2 power module in Figure 2 and its circuit in Figure 3 is the first battery storage power module using SiC technology. The great amount of energy that is transferred through the inverter has a high impact on the operational cost savings resulting from the high efficiency. This benefit should be kept in mind to perform a proper costs-benefit analysis of power modules.

 

Figure 2: Exemplary E2 power module for a 60-kW energy storage inverter with SiC MOSFETs (footprint: 45 mm x 107.5 mm)

Figure 3: One NPC2 leg for a 60-kW three-phase energy storage inverter with SiC MOSFETs and internal resistors to adjust switching characteristics

 

Customer-specific pin-out

The E2 module design has high flexibility to adjust switching characteristics due to a great amount of internal surface-mounted components. Therefore, MOSFETs can be easily replaced, and the power can slightly be up- or downscaled by minor changes to optimize system costs. Low temperature operation through improved thermal stack by using an AlN substrate leads to cost-efficient bonding and joining technologies that can be applied while still meeting the high lifetime requirements!

 

Outlook

The use of SiC power modules might only be beneficial if the battery stack costs (2017: approx. $350/kWh) are high compared to the inverter costs. Although the price for battery cells is soon expected to drop below $100/kWh, especially due to SiC components, high inverter efficiency is beneficial for the cost-benefit analysis to lower the attraction of CTs as peak capacity is added. Furthermore, as mentioned by Chet Lyons, simply overlaying storage on a central station basis won’t maximize grid performance or cost reduction. In addition, storage can increase the importance of decentralized independently operating micro-grids that are disconnected from the main grid in emergency situations. Beside energy storage, this E2-SiC module can additionally be used for compact designs of high-efficiency solar inverters.

 

About the Authors

Alexander Streibel works at Danfoss Silicon Power as an Application Engineer. He is responsible for automotive and industrial customer's requests; sales support including design, layout and simulation of power module concepts and lifetime requirements; evaluation of upcoming semiconductor technologies and supplier portfolios; investigation about new cooling concepts; silicon carbide power MOSFET packaging; and transfer of knowledge. He is particularly skilled in power electronics, FPGA, and programming. He earned his Bachelor's Degree in Electrical Engineering at Technical University Carolo-Wilhelmina of Braunschweig. He then acquired his Master's Degree in Mechatronics at the University of Southern Denmark located in Odense, Denmark.

Roger Cooper works as the Sales Manager at Danfoss Silicon Power, where he is responsible for driving sales of Danfoss customer-specified power modules in the US, Canada and Mexico to power conversion OEMs serving the Automotive, Industrial and Renewable spaces. He is also skilled in the filled of electronics, semiconductors as well as in sales. He earned his Bachelor's Degree in Electrical Engineering Technology at Texas A&M University located in Texas, USA.

Ole Mühlfeld is the Director of Application Engineering at Danfoss Silicon Power. He has worked at Danfoss since 2012, beginning as an application engineer in automotive. He graduated with his degree in physics from Christian-Albrechts-Univerität zu Kiel in 2007 and stayed on as research staff until 2011.

 

This article originally appeared in the Bodo’s Power Systems magazine.