The construction of new solar photovoltaic generation plants in Latin America has brought great challenges regarding the design and construction of grounding systems (SPAT) due to their large size.

Engineering evolves, as do the computational tools used to model and solve these challenges. In general, there are three (3) ways to carry out SPAT studies, each with its own limitations:

  • Use of equations or Excel spreadsheets with the simplified formulas of the IEEE 80 of 2013.
  • Use of specialized software that considers finite element models based on an equipotential or DC hypothesis (GSA, ETAP, CYMGRD, MALT, etc.).
  • Use of specialized software with complete PEEC electromagnetic models in AC (GSA_FD).

This article compares the results of a real case, a large SPAT, using two resolution models: DC vs AC. The objective is to demonstrate the difference that exists when using models in SPAT designs without considering the limitations they have:

The soil model to be considered is four (4) layers:

The results of the GPR and mesh resistance are presented below:

It could be observed that the resistance of the mesh using the equipotential hypothesis is much lower than the resulting impedance using an AC model (an error of more than 400%). This is because the equipotential model considers only the real part of the impedance, that is, it neglects the effect of the self/mutual inductance and capacitance of the buried conductor.

It is worth mentioning that when using an AC model, the impedance of the system will depend on the injection point of the fault current and the topology of the mesh. Therefore, it is necessary to simulate different injection points in the design stage.

Below, the ground potential is presented in the same mesh using both models. These figures help clarify the behavior of an equipotential SPAT and the effects that the own and mutual impedances can have:

In an equipotential model the value remains relatively constant along the surface of the mesh, unlike the AC model where you can observe how the ground potential decreases as you move away from the point of failure. This greater potential difference between point and point is what generates risks in the facilities.

The results of the step and contact voltages are presented below. To facilitate its interpretation, the following should be considered:

  • Parkland: They are the safe zones, corresponding to the area where the step and contact voltages are below the permitted limits.
  • Yellow zones: They are the areas where the step voltages are met, but not the contact voltages. This area may be yellow if there are no metal elements that can be touched.
  • Red zones: They are unsafe areas, where both passage and contact limits are not met.

The previous graphs allow you to quickly see the areas where there are problems in the SPAT. The differences that can be generated in the use of one or another model are evident, while the equipotential model guarantees that there are no risks of contact voltages for people in the solar park area, a more rigorous evaluation considering the impedance of the mesh highlights situations where (depending on the location of the fault), the SPAT is risky for plant personnel.

According to the results, it is possible to affirm that the equipotential condition (DC) generates non-rigorous results, that is, it underestimates the passage and contact potentials of large installations, generating possible risks derived from the design stage.