The losses of solar inverters are the core factors affecting their conversion efficiency (usually measured by MPPT efficiency and total conversion efficiency), mainly due to the physical characteristics of power electronic devices, circuit topology design, and operating conditions. They can be divided into the following six categories, and the generation mechanisms, key influencing factors, and characteristics of each type of loss are as follows:
1、 Switching loss: the core loss of high-frequency switching action
Switching losses are the main source of losses in high-power operating conditions of inverters, originating from the non ideal states of power semiconductor devices (such as IGBT, MOSFET) during the "on" and "off" processes - when the device transitions from "off" to "on" or "on" to "off", there is a brief overlap between voltage and current, resulting in instantaneous power losses.
Generation mechanism:
Turn-on loss: When the device is turned on, the current rise speed is slower than the voltage drop speed, and the two overlap to form a success rate loss (especially IGBT needs to overcome junction capacitance charging);
Turn off loss: When the device is turned off, the voltage rise rate is slower than the current drop rate, resulting in overlapping loss again (IGBT still has "tail current", further increasing the loss);
Reverse recovery loss: If the device has reverse conduction characteristics (such as the freewheeling diode of IGBT), a spike voltage will be generated when the reverse current disappears, resulting in additional losses.
Influencing factors:
Switching frequency: The higher the frequency (such as 20-50kHz commonly seen in string inverters), the more switching times per unit time, and the loss increases linearly;
Device characteristics: The switching speed and reverse recovery time (trr) of IGBT directly determine the size of losses (the switching losses of new SiC and GaN devices are only 1/5-1/10 of traditional IGBT);
Working voltage/current: The higher the input voltage and output current, the greater the instantaneous power in the overlapping interval, and the higher the loss.
2、 Conduction loss: Ohmic loss during device conduction
Conduction loss is the fundamental loss of an inverter under full power conditions, originating from the inherent resistance of power semiconductor devices when conducting (such as Rds (on) of MOSFETs and Vce (sat) of IGBTs). Essentially, it is the Joule heating loss generated by the current flowing through the resistance (according to Ohm's law P=I ² R or P=V × I).
Generation mechanism:
When MOSFET is turned on, there is a conduction resistance Rds (on) in the channel, and a loss P=I ² × Rds (on) occurs when current I flows through;
When IGBT is turned on, there is a fixed on state voltage drop Vce (sat) (usually 0.8-1.5V), and the loss P=Vce (sat) × I (the proportion is higher when the current is small).
Influencing factors:
Output current: The loss is proportional to the square of the current (doubling the current increases the loss by 4 times), and is the main loss under low power factor or overload conditions;
Device selection: MOSFETs with low Rds (on) and IGBTs with low Vce (sat) can significantly reduce losses;
Conduction time: The inverter adopts PWM (pulse width modulation) control, and the higher the conduction duty cycle of the device, the greater the conduction loss.
3、 Core Loss: Hysteresis and Eddy Current Loss of Inductors/Transformers
The filtering circuit of inverters (such as output LC filtering and high-frequency transformers of isolated inverters) requires the use of magnetic core components (such as ferrite and silicon steel sheets). The magnetic core will produce hysteresis loss and eddy current loss in alternating magnetic fields, collectively known as magnetic core loss (also known as iron loss).
Generation mechanism:
Hysteresis loss: The alternating magnetic field causes the magnetic domains inside the magnetic core to repeatedly flip, overcoming the frictional resistance between the domains and generating energy loss. The magnitude of the loss is proportional to the hysteresis loop area of the magnetic core material;
Eddy current loss: An alternating magnetic field induces a closed current (eddy current) inside the magnetic core. The current flows through the resistance of the magnetic core, generating Joule heat. The loss is proportional to the square of the magnetic core thickness and the square of the magnetic field frequency.
Influencing factors:
Working frequency: The higher the frequency (such as 10-100kHz for high-frequency transformers), the significantly increased hysteresis and eddy current losses;
Magnetic core material: Ferrite is suitable for high frequencies (low loss at high frequencies), while silicon steel sheets are suitable for low frequencies (such as 50/60Hz power frequency transformers);
Magnetic core structure: adopting a "layered stacking" (silicon steel sheet) or "small air gap" design, which can reduce eddy current and hysteresis losses;
Magnetic flux density: The higher the magnetic field strength in the magnetic core (such as during overload), the more nonlinear the hysteresis loss increases.
4、 Copper Loss: Resistance loss of inductance/transformer windings
Copper loss is the Ohmic loss of winding wires in magnetic core components, similar to the principle of conduction loss, but occurring in copper (or aluminum) windings of inductors and transformers. Essentially, it is the Joule heating (P=I ² R) generated by the current flowing through the winding resistance.
Generation mechanism:
DC copper loss: The loss caused by the DC current flowing through the winding DC resistance Rdc, which is only related to the current and Rdc;
AC copper loss: Under high-frequency AC current, the effective resistance of the winding increases due to the "skin effect" (current concentration on the surface of the wire) and the "proximity effect" (interference between adjacent wire magnetic fields), resulting in a loss that is 30% -100% higher than DC copper loss.
Influencing factors:
Winding material: The resistivity of copper wire (1.72 × 10 ⁻⁸Ω· m) is much lower than that of aluminum wire (2.82 × 10 ⁻⁸Ω· m), and the copper loss of copper winding is lower;
Wire structure: Using "multi strand twisted wire" (Litz wire) can weaken the skin effect and reduce high-frequency AC copper loss;
Current size: Loss is proportional to the square of the current, and copper loss increases sharply when overloaded;
Winding turns: The more turns there are, the longer the wire and the greater the resistance, resulting in a corresponding increase in copper loss.
5、 Control Circuit Loss: Static loss of auxiliary systems
The control circuit of the inverter (such as MPPT controller, PWM drive circuit, sampling circuit, and cooling fan) requires a small amount of electrical energy to maintain operation, which belongs to static losses (regardless of whether the inverter is loaded or not, as long as it is powered on). The proportion of total losses is usually low (1% -3%), but it will increase under low-power conditions.
Main components:
MPPT and main control chips: operating power consumption of MCU and DSP (usually a few watts);
Driver circuit: a chip that provides driving signals for IGBT/MOSFET (such as optocouplers and driver ICs), which consumes driving current;
Auxiliary power supply: converts the DC bus voltage into the low voltage (such as 12V, 5V) required by the control circuit, with conversion losses;
Cooling system: The operating power consumption (usually 1-5W) of the cooling fan (if active cooling).
Influencing factors:
Auxiliary power efficiency: Efficient auxiliary power sources (such as flyback and LLC resonant) can reduce conversion losses;
Fan start stop strategy: Intelligent temperature controlled fans (stop running at low loads) are more energy-efficient than regular running fans.
6、 Other losses: Non core but non negligible losses
In addition to the five main types of losses mentioned above, there are also a small amount of dispersion losses, usually accounting for less than 2%, but may increase under extreme working conditions:
Busbar capacitance loss: The loss (P=I ² × ESR) caused by the equivalent series resistance (ESR) of electrolytic capacitors or film capacitors on DC busbars. At high frequencies, as ESR increases, the loss also increases;
Parasitic parameter loss: Parasitic inductance (such as wire inductance) and parasitic capacitance (such as distributed capacitance between devices) in the circuit generate peak voltage/current during high-frequency switching, resulting in additional losses;
Wiring and contact losses: Losses caused by contact resistance (such as oxidation or looseness) of inverter input/output terminals and terminals, which may increase due to poor contact after long-term operation.
Core technology direction for reducing losses
In response to the aforementioned losses, the industry typically optimizes inverter efficiency through the following technologies:
Device upgrade: Adopting SiC (silicon carbide) and GaN (gallium nitride) devices to reduce switching and conduction losses;
Topology optimization: Adopting LLC resonant topology (auxiliary power supply) and three-level topology (high-power inverter) to reduce switching losses;
Magnetic core optimization: High frequency and low loss magnetic cores (such as PC44 ferrite) are selected, and Litz wire windings are used to reduce magnetic core losses and copper losses;
Control strategy: dynamically adjust the switching frequency (light load frequency reduction), optimize the MPPT algorithm (reduce frequent switching), and reduce switching losses;
Heat dissipation design: Adopting liquid cooling or high-efficiency air cooling to reduce device junction temperature (an increase in junction temperature can lead to increased conduction loss).
Through the above optimization, the maximum conversion efficiency of mainstream string inverters has reached 98.5% -99.5%, and the efficiency of centralized inverters generally exceeds 99%.
