Photovoltaic inverter shutdown due to voltage variation

A problem that affects many photovoltaic systems is the sudden shutdown of the inverter when generation reaches high values. The phenomenon occurs mainly on sunny days, when everything seems to be perfect for the photovoltaic system to perform at its best. The root of the problem lies in the quality of the electrical network to which the inverter is connected. Installing a photovoltaic system requires care not only in the components and connections of the inverter, modules and the string box. The designer also needs to be careful with what exists after the inverter. The energy generated must leave the inverter and be injected at the connection point without causing disturbances to the electrical network, both within the property and in the distribution network. The main disturbance to the electrical grid caused by photovoltaic systems is the increase in voltage. This variation is normal and in many cases it will not even be noticed. However, when we have fragile or very precarious, poorly sized electrical networks, the voltage variation can exceed the voltage and operating limits programmed in the inverter, causing it to shut down.

Understanding the problem

Electrical conductors have electrical resistance. When an electric current passes through, this resistance causes a voltage drop. Due to the voltage drop, in a home or any electrical installation, the voltage obtained at the distribution board will be lower than the voltage delivered by the concessionaire at the input pole, as we see in the following figure. When the installation's consumption increases, the voltage drop intensifies and the electrical voltage available in the distribution board may drop even further depending on how the installation was sized.

In electrical installations, in general, the voltage at the point of consumption will be lower than the voltage at the input pole due to the voltage drop. In very poorly sized installations (which are not uncommon) the voltage drop will be more pronounced. However, this is a problem that existed even before the arrival of photovoltaic systems. What changed then? In weak electrical networks, due to poor cable sizing, the voltage available for internal loads will always be reduced with the circulation of current through the circuit, as shown in Figure 1. The more current is circulating through the installation, the more voltage drop there will be. to occur. With the injection of energy by the inverter, the electrical current present in the circuit can decrease in intensity or even be reset to zero (when solar generation is equal to internal consumption), causing the voltage drop to decrease, which causes the voltage at the point to rise. connection (where the inverter is connected), as shown in the graph in Figure 2. When solar generation is very large (situation illustrated in Figures 2 and 3), the amount of energy injected by the inverter can cause a current in the opposite direction, when energy is exported to the concessionaire. In this case, the voltage at the injection point will be higher than the input voltage supplied by the utility, and in some cases may exceed the operating limit for which the inverter is programmed.

A little theory: resistance and impedance

Conductive materials have an internal resistance that depends on the metal used (copper or aluminum), the cross section (cable gauge) and the length of the cable.

The longer the cable, the greater its resistance. The smaller the cable gauge, the greater the resistance. A very thin conductor makes it difficult for electric current to pass through, so its resistance is greater.

And each material used to manufacture the cable has its own characteristic called resistivity, which affects the cable's resistance. 

A Resistance of an electrical conductor can be calculated using the formula:

In the formula above, ρ is the resistivity of the material, L is the length of the cable and S is its cross section (area).

In short electrical circuits, resistance is the main cause of voltage drops. In longer circuits there are also voltage drops caused by reactances, which are the result of the capacitive and inductive effects present in electrical circuits.

A more complete model of a cable's characteristics involves factors such as the cable's capacitance and inductance. Capacitance and inductance are related to the distance between the cables, the voltage between them and the medium in which they are surrounded.

The sum of the effects of resistance, capacitance and inductance is called impedance. In cables that carry direct current, the effects of capacitance and inductance are null due to the very nature of electromagnetic effects.

ANDIn very long alternating current circuits, capacitance and inductance cannot be neglected, as their effects alter the voltage drops and conduction losses of the cable. Impedance can also be significant in electrical connections. Poorly made or inappropriate connections have significant impedance, which can even cause a loss of energy in the form of heating. The following figure shows the overheating of a cable and an electrical contact. The cause of heating is the dissipation of energy over the electrical resistance of the contact. In addition to the resistances and impedances present in the cables, poor contacts in the connections are also an important cause of voltage drops. 

Voltage drop

When a current passes through an impedance in a circuit, there is a voltage drop between the positive and negative terminals of the impedance.

When taking into account that a real cable has a non-zero impedance, we can say that the passage of current in a cable also causes a voltage drop between its terminals. The voltage drop in a circuit caused by a real cable is determined below:

As we are analyzing a circuit with two conductors, the voltage loss in the resistance of cable 2 must also be taken into account. Therefore:

In the above formula we have The= 2 for single-phase or two-phase circuits and The=1.73 for three-phase circuits.

In commercial installations there are not many options for conductive materials, leaving the designer to choose between copper and aluminum. Circuit size also cannot always be adjusted freely, as there are physical and allocation demands on the devices.

The most practical way to control the voltage drop in a circuit is to choose the cross section of the conductor. The standard for low voltage electrical installations – NBR 5410 – provides for a maximum limit on the voltage drop in a circuit, as shown below.

Grid-connected inverter

When connected to the electrical grid, the photovoltaic inverter receives the voltage available at the connection point at its terminals. If the inverter is connected directly to the consumer's input standard, the voltage at the connection point will be more stable (as it will only depend on the voltage drops existing in the electricity utility's distribution network circuit. When the inverter is connected to a At the innermost point of the installation, the voltage at the inverter connection point may vary according to the voltage drops caused by the electrical currents present in the installation. The voltage at the connection point may also vary depending on the intensity of the current injected by the inverter itself. as we discussed at the beginning of the article. If the inverter injects too much current, to the point of reversing the energy flow, the voltage at the connection point will rise beyond the value found in the consumer input standard. However, there is a limit to the voltage. voltage at which the inverter can operate, which is defined by the NBR 16149 standard, as shown in the table below. The inverter voltage cannot be lower than 80% of the utility's nominal supply voltage. The inverter voltage also cannot be greater than 110% of the nominal supply voltage. Each concessionaire has a nominal supply voltage and the inverter must be parameterized according to this voltage. The voltage at which the inverter is authorized to operate is in the range between 80% and 110% of the nominal voltage of the distribution network, as shown in the table.

Table 1. Export voltage limits of grid-connected inverters

Therefore, if the inverter injects a lot of energy and if there are significant resistances (or impedances) in the installation, the voltage at the connection point may increase and exceed 110% of the nominal grid voltage, causing the inverter to automatically shut down.

This is a very common situation on big generation days. The inverter is switched off for this reason always occurs on days and times when solar radiation is most intense.

The photovoltaic system is disconnected at the moment when it should generate more energy – a very unpleasant situation that can be avoided with the correct design of the electrical installations (avoiding voltage drops) or with the parameterization of the inverter (changing its minimum and maximum operating limits , without violating the limits of standard NBR 16149). 

Transformer impedance and voltage drop

Larger solar photovoltaic systems are typically connected to the grid through a coupling transformer that raises the system's nominal voltage to the same utility level.

Transformers also have an internal impedance that can cause a voltage drop from the inverter's point of view. The equivalent circuit of a real transformer shown in the figure below indicates the internal impedances of this device. For a photovoltaic solar system to inject energy, the voltage on the side of the transformer facing the internal grid (secondary), represented by Us, must be within the operating range of the inverter, as explained in the previous section (i.e., between 80% and 110% of the concessionaire's nominal supply voltage).

Inverter shutdown due to voltage drop

When an inverter is installed in a high-impedance circuit, i.e., a cable with an inadequate cross-section, poorly made connections or high-impedance transformers, the voltage at the connection point can be significantly high due to the impedance of the cable, poor contacts and to the presence of the transformer. If the voltage increase is considerable, exceeding 110% of the nominal grid voltage (if the inverter parameterization is correct), the inverter will automatically shut down. This shutdown is a protection function present in inverters that in some cases causes unwanted shutdown of the equipment. The inverter shutdown using the mechanism described above typically occurs at times of maximum generation – close to noon. These inverter shutdowns can be identified in the daily generation curves of the affected device, as shown in the example below.

Measures to overcome the problem

The first step to solving the problem is to check whether the voltage configured on the inverter is the same as the utility supply voltage, as the inverter must only operate within the limits established in Table 1.

The phenomenon described in this article is closely related to the impedance that the inverter sees on the AC side and, as shown previously, the sources of impedance seen by the inverter are: cables, connections and transformer.

Therefore, cables sized taking into account the voltage drop criteria described in the NBR 5410 standard minimize the risk of inverter shutdown due to voltage rise. As for connections, they must be well made, following current standards and periodically checked to ensure there is no looseness in the tightening.

If the internal impedance of the transformer or the cables that connect the photovoltaic plant's transformer to the utility's transformer are very high, the same reasoning applies as for the voltage drop in the cables.