Learnings

Outcomes

From the literature review, high frequency earthing electrodes have been recommended by various standards, but quantification of safety and impedance improvement needs to be carried out. The project therefore investigated the performance of various earthing arrangements including a small-scale 5m×5m, 25 mesh grid in isolation, with HF rod enhancements and with a new alternative method using insulated horizontal conductors. The test scenarios were then modelled to further enhance the understanding of the test results.

Test results:

A scaled earth grid was installed at Cardiff University. A vertical rod was connected to either the grid only or the grid with the above insulated conductor. The rod (up to 4.8m) was installed at the centre of the grid, with a small separation between the rod and grid to allow tests on each electrode to be performed separately.

1. DC Earth Resistance:

The DC earth resistance measurements for the rod, grid and grid bonded with rod was carried out experimentally.  The results showed that as the length of the earth rod increases, both the earth resistance and its rate of fall-off decrease. Also, the changes in DC earth resistance appear similar for all the configurations and are attributed to variations in soil resistivity influenced by moisture content and temperature. Furthermore, a significant reduction was obtained when the grid bonded with the high frequency electrode.

2. Variable frequency tests for vertical electrodes

Frequency-dependent earthing impedance was measured for different lengths of rods in a two-layered soil. The results showed that the impedance magnitude for a 4.8m vertical rod is lower than that for shorter lengths of rod, up to a particular frequency which depends on the soil resistivity value. Generally, at these lengths, the effect of capacitance is more pronounced with sharper downturns occurring at a high frequency (100 kHz). Following the first minimum (at 600kHz), the impedance increases dramatically due to the inductive effect, after which the impedances follow an oscillatory pattern with frequency.

3. Variable frequency tests for scaled earth grid

The frequency response of the earth grid electrode was determined experimentally. Variable frequency current was injected at the centre of the grid. The results showed that the impedance magnitude results at low frequency are very close to the DC earth resistance. However, at high frequency, where inductive effects start to appear, a sharp upturn is seen.

4. Frequency response for earth grid bonded with vertical electrode

The results of the variable frequency tests have shown that bonding of high frequency electrode to the grid had a beneficial effect on measured earth impedance and can reduce the earth resistance by approximately 23% compared with grid only. In general, not only is there a considerable reduction in earth resistance/impedance over the low-frequency range, but there is also a reduction at higher frequency.

5. Impact of moisture and temperature

It was found that the impedance at low frequency increases by around 15.7% in the Winter season compared to that in the Summer. For high frequency, the impedance significantly increased in the Winter compared to Summer season, and this could be attributed to a particularly dry winter period. This also indicates how an earthing system buried in the soil layer near the surface is more exposed to seasonal variation than deeper earthing systems.

6. Frequency response for earth grid with enhancements

The new proposal suggested by Cardiff University uses horizontally laid, insulated conductors to be bonded at the point of injection at one end and positions further out on the earth grid at the other, with the main purpose being an increase in the effective area/length over which the current is dissipated, and to reduce the inductive effect at high frequency. Three different arrangements were studied.

The results show that the addition of a parallel insulated conductor has no major effect on the impedance of the earth grid at low frequency. However, at high frequencies, a reduction in impedance becomes apparent. It is thought that the drop in earth impedance occurs due to a reduction in electrode inductance compared to that of the bare conductor alone, and an increase in the effective area of the electrode such that the current density is more evenly distributed across the electrode-soil interface.

7. Frequency response tests at Swansea North substation

High frequency and impulse simulations of the earthing system at Swansea North substation demonstrated the advantages of connecting high frequency electrodes with surge arresters. The results show that the reduction of earth impedance at high frequency is greatest with the electrodes embedded in the lowest resistivity soil, and the earth impedance of a grid-bonded surge arrester may be reduced when enhanced with vertical rods. Impulse simulation results also show that increasing the length of the high frequency electrode can reduce the peak of the transient earth potential rise (TEPR).

8. Impulse test results

A low voltage impulse generator was used to inject an impulse current of variable magnitude and shape into the grid, with and without the bonding of the 4.8m rod under test. The results show that the addition of the 4.8m rod to the earth grid can reduce the earth potential rise (EPR) by approximately 48.4% and 32% compared to the 4.8m rod and grid only respectively. Also, the impulse characteristics of the grid buried in a non-uniform soil at an outdoor test site with enhancement conductors were tested. It was shown that a reduction in earth potential rise can be achieved. The results showed that the performance of one of the enhanced arrangements is very close to that of the grid with bonded rod

Simulation results

1. Simulation results for earth grid and vertical electrode

In order to improve the understanding of the test results the test setup was simulated using CDEGS-HIFREQ software. For the frequency response, at low frequency injections, the measured results show differences between the measured and simulated results and could be attributed to the soil resistivity model used in the CDEGS software. At high frequency, the measured results indicate capacitive response up to 700kHz, while the simulated results presented inductive effect over the same frequency range. In addition, the comparison for the transient response showed that the computed transient voltage impulse has a slightly higher peak magnitude compared with the measured value. Overall, there is a good agreement between computed and measured values, and the differences may be due to the uniformity of the stratified soil model used in the simulation model.

2. Simulation results for Swansea North Substation

The frequency response for the earthing system at Swansea North Substation was investigated experimentally and compared with the simulation using CDEGS.  There is good agreement between measured and simulated results at low frequency. However, at high frequency there are noticeable differences between measurement and simulation.  These might be due to mutual coupling errors.

Geometry of down-lead conductors

The effect of bending of down-lead conductors was tested under impulse current injection, and it was found that a straight downlead conductor is ideally suited for the best-case scenario regarding reliability and safety compared with curved downlead conductors. The bending of down-lead conductors has a significant effect on the voltage at the bend which indicates that the electric field in the bent area of copper is much higher than the straight downlead conductor.

Recommendations for further work

Further work in this area will be particularly useful to investigate GIS earthing transients, coupling with other systems on the same earthing mat and the development of engineering guidance for ground resistance calculations.

Lessons Learnt

From the literature review, it was clear that all standards (BS, IEC and IEEE) have specifications for high frequency earthing practice. However, there is a need to evaluate the benefits of each approach and quantify their impact on both reducing earth impedance and safety voltages. This project has explored a further method, first proposed by the research group at Cardiff University, in which an insulated horizontal conductor/s is used to fulfil the high frequency shortfall.

The main point to emphasise is that the national and international standards require a high frequency electrode but very little experimental evidence had been available to date to support this requirement. High frequency and impulse performances of earthing systems are strongly affected by inductive or capacitive effects, and these need to be taken into account during the testing. Tests on scaled grids indicate a complex response which requires detailed study.

Dissemination
A dissemination webinar has been held on 06/07/2021 where the findings from the project have been disseminated to key stakeholders from other network licensees, suppliers and interest groups.