Why is it important to keep reasonable switching losses?
Switching frequency is a decisive design parameter of a VFD – especially in medium voltage drives where the semiconductor capability may be more restricted. But switching losses have much bigger impact on the system. The end user shall understand it properly as it is a matter of his operational expenditures in the next years. In this short post we would like to emphasize the importance of keeping reasonable switching losses and the consequence when not doing so.
Remark: In medium voltage drives, unlike low voltage drives, it is important to be a bit more precise when talking about switching frequency in absolute figures. Medium voltage drives often utilize multilevel inverter topologies and there is a big difference between switching frequency per device (IGBT/IGCT/IEGT semiconductor) and total effective switching frequency seen in the output voltage waveform [1]
Technical capabilities
Modern power semiconductors allow operation at increasing switching frequencies. There are other trends in parallel that aim to reduce the switching losses or to allow operation at higher junction temperature. Motivation for reduced switching losses is an improved energy efficiency. On the other hand, operation at higher temperatures is motivated by volume optimization (“miniaturization” of power electronics) leading to a more compact design with reduced footprint.
Currently used semiconductor devices can principally switch at considerably higher frequencies than what the application actually needs. Medium voltage VFDs with multilevel inverter topology can achieve high quality of the output power while using moderate device switching frequency. Although the inverter could operate at significantly higher switching frequency, it is simply not necessary.
Life cycle cost
The cost of energy consumed by the drive system throughout its lifetime is multiple times higher than the initial purchase price of the equipment. Thus, even a small increase of system losses has a significant impact on the life cycle cost. Vice versa, reduction of losses by as little as several kW leads to attractive savings over the lifetime.
In this view it is not wise to utilize switching frequency to the maximum if not required. Instead, a balance shall be found between power quality, dynamic performance and energy efficiency.
Consequences of high switching losses
Excessive switching losses impact the drive system in several ways. Let’s look at the consequences.
- Increased losses / reduced efficiency
- Power derating
- Increased cooling demand
- Larger footprint
i. Increased VFD losses
Switching losses represent significant portion of the total VFD losses. They make some 40-60% of the semiconductor losses which are the most dominant part of the overall VFD losses. Increasing the switching frequency directly impacts the switching losses.
Increased losses have direct impact on the operating cost (higher energy bill). More losses can mean larger sizing of air conditioning in the room etc.
ii. Power derating
The equipment shall be operated within its temperature class to guarantee the design lifetime. Specifically for semiconductors the junction temperature shall not exceed certain value (e.g. 140°C) otherwise a fast degradation or a direct failure appear. To stay within those boundaries, the increased switching losses may need to be compensated by reduced conduction losses. This is achieved by reducing the current and thus going into power derating. In practice it means that VFD with higher switching frequency may have lower rated output current.
iii. Increased cooling demand
Higher switching losses require larger cooling capacity to evacuate the heat away from the source and maintain stable operating temperature. Increased cooling demand means larger cooling system (bigger fans or pumps), larger auxiliary power consumption and increased footprint of the cooling system. Moreover, the re-cooling equipment may also become more bulky.
iv. Larger footprint
Increased switching frequency and corresponding higher losses affect the overall footprint of the drive system. The cooling system gets larger due to higher cooling capacity. The VFD frame size may jump to next bigger one due to mentioned power derating.
Switching frequency shall not be too low either
Of course, too low switching frequency is not good either. It may just move the problem somewhere else. For example, too low switching frequency can result in very distorted output current which is then causing additional losses inside the motor. In case of asynchronous PWM modulation, small ratio between carrier frequency and fundamental output frequency leads to undesired side bands. Too low switching frequency generally impacts the stability of the control system.
Selection of optimal switching frequency depends on multiple factors, such as:
- VFD topology (e.g. number of voltage levels, type of output filter)
- VFD performance (dynamic requirements)
- Motor base frequency and operating range
- Type of modulation
Regarding the last point, direct torque control (DTC) is particularly strong with respect to optimum balance between switching losses and control of machine’s torque and flux [2]. DTC uses the lowest switching frequency to satisfy the requirements on the torque and flux control thus leading to lowest switching losses.
Would you like to learn more about VFDs and drive systems? Purchase one of our premium plans or participate in a dedicated webinar.
References
[1] VFD switching frequency, https://mb-drive-services.com/vfd-switching-frequency/
[2] How does DTC contribute to superior VFD energy efficiency? https://mb-drive-services.com/how-does-dtc-contribute-to-superior-vfd-energy-efficiency/
