Neutral Grounding Resistor (NGR)

NGR is a resistance used in system earthing inorder to reduce the magnitude of fault current.

Apart from limiting the fault current, it also ensures protection of the healthy phases against Overvoltage conditions, helps in the efficient operation of protective devices and protects the live conductors against thermal distress due to faults. Therefore, the design and selection of NGR is very much crucial to ensure equipment & personnel safety as well as supply continuity. Before discussing on the sizing of NGR, let’s try to understand the very basic nature of NGR and why it is so necessary.

As the name itself suggests, NGR or Neutral Grounding Resistor is a resistance connected between neutral and ground inorder to limit the fault current of the system. Any person with basic knowledge of physics would figure-out that a higher amount of resistance leads to lower levels of current in any circuit.

So, isn’t it better to not connect any NGR? i.e. an Ungrounded system, so that we have a theoretically infinite resistance (practically very high impedance) between the neutral and the ground.

True - by not connecting any NGR, the fault current is reduced to the least possible value in an Ungrounded system but this gives rise to a new problem i.e. Overvoltage.

Let us consider an Ungrounded Star connected system for our analysis. Although, there is no connection between neutral and ground, Ungrounded system can be considered as a system grounded through its Natural Capacitance, as shown in the figure above. The insulation (of various electrical equipment like Surge capacitors, Cables, Motors etc) behaves like a dielectric material between two different levels of voltage i.e. the system voltage and the ground reference. Hence, each system has its own naturally occuring capacitance due to its own equipment; and if it is an Ungrounded system, it can be understood as being grounded through the same capacitance.

In a healthy and balanced system, all the three phase voltages are equal & 120° apart and the neutral voltage is equal to zero. Since the three leakage currents are capacitive in nature, they are at quadrature i.e. at 90° leading to their respective voltages. These currents are also called Capacitive Charging currents. The equivalent system, its phasor diagram and the flow of currents can be observed from the figure below.

Suppose a Line-to-Ground(LG) fault occurs on phase-A. Voltage of phase-A falls down to zero, now this becomes the new reference point. Due to the absence of direct/indirect connection to the ground, the voltage of the neutral is not fixed and hence rises upto voltage -Va. Consequently, voltage of the other two healthy phases rises from Phase voltage to Line Voltage level i.e. 1.732 times than the nominal value; further, these voltages are now only 60° out of phase.

Since the capacitance on phase-A is shorted due to the ground fault, the current Ia now flows through the grounding path and it is equal to the vector sum of currents Ib and Ic. These currents, being capacitive in nature, are still at quadrature to their respective voltages but their magnitude is now 1.732 times their nominal value (due to the increase in their terminal voltages). The equivalent system, its phasor diagram and the flow of currents can be observed from the figure below. The voltage and current phasors of faulty system are superimposed over the phasors of healthy system (faded phasors) for comparative analysis.

Unless the ground fault is removed at voltage zero position, certain DC offset voltage is left on the neutral. Since there is no discharge path, this offset voltage shall remain on the neutral.

For example, if the fault is cleared at voltage maximum(+Vmax) position, voltage of the neutral shall remain DC Offset +Vmax and the voltage levels of all the phases shall now oscillate between 2Vmax to 0 i.e. all the three phases shall reach a maximum voltage of double the nominal value, easily damaging the insulation material leading to more faults. This can be clearly understood from the figure below.

This is not the whole problem but just the beginning of an even bigger problem.

What if the fault we were discussing till now was an Intermittent fault (i.e. an arcing, re-striking or vibrating type) instead of a solid fault?

In this case, the fault current path is through the Arc. Whenever the gap voltage rises above a certain threshold value, the arc strikes and when current passes through zero value, the arc extinguishes.

Since arcing is a transient phenomenon, the Line Inductance should also be taken into consideration. When Line-to-Ground(LG) fault happens on a phase, the voltage across the line inductance opposes the sudden change in the terminal voltages - thereby, the phase voltages do not change abruptly but in an oscillating waveform whose time constant depends on the magnitudes of line inductance and the natural capacitance. But since this is an Arc fault, the arc extinguishes when the current passes through zero value. This means the arc extinguishes when the oscillating voltage waveform is at one of its maxima points and the voltage of neutral at that instant of time shall remain as a DC offset thus shifting all the phase voltages by that amount.

When the Arc re-striking happens, voltage of the neutral shall be sum of that particular phase voltage (AC) + Capacitor Offset voltage (DC). After clearing, it leads to even higher phase voltages than the voltage before the fault. In this manner, the Intermittent fault can cause the phase voltage to rise upto 6 or 8 times the phase voltage, leading to a breakdown of insulation on one of the Non-faulted phases and the development of a Line-to-Ground-to-Line(LLG) fault.

To understand this phenomenon more clearly, simulation of intermittent fault on an Ungrounded Star connected system was done using SPICE simulator, as shown in the figure here. To simulate arcing, a switch is designed to turn ON basing on the gap voltage and turn OFF when current passes through zero, thus mimicking a real time arcing phenomenon.

In our SPICE model, Line-to-Ground Arc fault occurred at Point ① as shown in the figure below. This causes the voltage to discharge through the capacitances, drop to zero and then towards opposite polarity at the rate defined by the resonating frequency established by the capacitances and inductance of the circuit. This continues until the current reaches zero value at Point ② and the arc extinguishes. If it were a solid fault, the voltage of the faulty phase shall oscillate for a transient period and finally converge to zero, depicted by the faded red waveform in the figure.

The Arc re-strikes again between point ③ to point ④ and point ⑤ to point ⑥. Finally, the phase voltage oscillates between 1120 V and 1840 V which is 5 times the nominal phase voltage. It is to be observed here that the magnitude of current for every re-striking fault is larger than the previous one and the voltage also increases multiple fold. This process repeats until some device fails, typically a lower BIL (Basic Impulse Level) device such as a control power transformer, a motor starter coil, a motor winding or a dry-type transformer etc.

Therefore, it is established that unless connecting the Neutral to Ground, heavy overvoltages are experienced. Neutral cannot be left open.

Since we answered the “WHY” and the "WHAT", the next logical question is the “HOW”. How to do the sizing of an NGR?

While sizing NGR, contrary to popular belief, it is not the resistance which is the deciding factor. An optimum value of NGR Let-through current is determined first. The value of resistance is then derived from this current and system voltage. To decide on the optimum value of NGR Let-through current, the following pointers should be followed.

Points to consider in the sizing of NGR

1. An NGR should be so sized that it limits the fault current to value which can be withstood by the equipment being protected. This sets the maximum limit for NGR Let-through current.

2. The magnitude of the NGR Let-through current should not be so low that it cannot be detected by the protection system.

3. The NGR let through current should be atleast more than or equal to 3 times the system charging current. System charging current is the current flowing through the natural capacitances in a healthy and balanced system (discussed previously). The line-to-ground capacitance associated with system components determines the magnitude of zero-sequence charging current. The charging current of a system can be calculated by summing the zero-sequence capacitance or determining capacitive reactance of all the cable and equipment connected to the system. Annexure-1 discusses the charging current values of various equipment at different voltage levels and also calculation of system charging current. This condition i.e. I ≥ 3 x Charging current sets the maximum limit of resistance.

4. NGRs installed at different voltage levels should be coordinated w.r.t to their magnitudes such that highest voltage level has the largest resistance NGR and so on.

5. Also, nowadays Manufacturers produce NGRs with standardised magnitudes for each voltage level. Industrial consumers can select from the available options whose value should lie between the maximum and minimum limits discussed above.

Considering the above 5 points, the Let-through Current rating of NGR is finalised. Finally, the resistance of NGR can be derived by dividing the system voltage (Vph) by the current rating of NGR.

Apart from the Voltage rating, Current rating and Resistance rating, other parameters in the design of NGR are Rated Time, Rated Continuous Current, Rated Time Temperature Rise.

Rated Time is the time during which the rated current passes through the equipment under the specified conditions, taking into account the thermal limits in accordance with the standards. The recommended rated times are 10 seconds, 1 minute, 10 minutes and extended time. (Rated times outside of the mentioned values can also be used). In practice, due to the adjustment of protective equipment, the duration of the fault passing through the NGR is much less than its rated time. In determining the rated time, possible occurrence of consecutive short-circuit faults should also be considered.

Rated Continuous Current is the amount of current that can be continuously passed through the equipment, taking into account the thermal limits in accordance with the standard. As a result of this current flow, the temperature of the element will reach steady state temperature.

Rated Time Temperature Rise is the maximum temperature increase over the steady state temperature caused by the rated current flow over the rated time. As per IEEE 32, NGRs are designed for a maximum temperature rise of 720° C. Resistance should be so designed that the variation over temperature range should be below ± 10%.

Kindly leave your corrections/suggestions & feedback in the comments section.

Annexure reference: iGard