Ecmweb 8437 Motor Facts Pr
Ecmweb 8437 Motor Facts Pr
Ecmweb 8437 Motor Facts Pr
Ecmweb 8437 Motor Facts Pr
Ecmweb 8437 Motor Facts Pr

Motor Protection Pitfalls

Sept. 21, 2016
Limitations of residual ground-fault elements in medium-voltage motor protection schemes

Medium-voltage motors are critical to a facility’s power system. Not only are they expensive, but their failure can have crippling, long-lasting impacts on a plant’s operation. As such, properly protecting these motors is of great importance. Many devices exist to help accomplish this task, but protective relaying is a primary contributor. Through connections to instrument transformers, discrete indicating devices, and even resistance temperature detectors (RTDs), relaying can provide several important protection functions.

This zero-sequence CT is used to detect ground current in a 3-phase circuit.

A function of particular interest is ground-fault protection because it addresses a relatively common abnormality: faults involving ground. These comprise the vast majority of faults in a power system. Many begin low in magnitude, but ground faults within motors can cause serious winding and stator damage. The severity of this damage is largely dependent upon fault duration, which can be limited by properly designed protective relaying schemes. Various ground-fault protection methods exist, each with unique limitations. Specifically, this article discusses three primary methods (ground return, residual, and flux-summation), presents the basic concepts involved, shows how the methods apply to medium-voltage motors, and evaluates a real-world example.

Examining the methods

Ground-fault current magnitudes are limited, in part, by zero-sequence impedance. In medium-voltage industrial power systems, this impedance often includes a neutral grounding resistor (NGR). This resistor is inserted between the grounding electrode and the system neutral point. Circumstances exist in which ground-fault magnitudes are limited to values lower than a circuit’s load current. When this occurs, faults can persist undetected by conventional phase-overcurrent protection schemes because these schemes must be desensitized to load. Therefore, supplementary protection techniques must be applied to quickly isolate ground faults.

Kirchoff’s Current Law shows us that the ground-fault current equals the vector sum of the three phase currents and the neutral current. Medium-voltage industrial power systems used to power large motors are typically 3-wire systems so neutral current will be excluded from further discussion. Under normal operating conditions, phase currents should sum to near zero. During a ground fault, however, phase currents become unbalanced because some current takes abnormal paths through the earth or the grounding electrode. Therefore, ground faults can be detected by sensing the magnitude of this imbalance. Various methods exist to detect phase-current imbalance. Three commonly used are: the ground return method, the residual method, and the flux summation method (Fig. 1).

Fig. 1. Three methods can be used to detect phase current imbalance.

Ground return method — The ground return method detects imbalanced current in the three phase conductors by measuring current in the system’s neutral connection to the grounding electrode. This can be done with a window-type current transformer (CT) around the connection. Alternatively, a bar-type CT can be placed in series with the connection. This method is limited to zones in the system that contain a neutral-ground connection, such as the secondary winding of power transformers. It is atypical, however, for distribution equipment such as medium-voltage switchgear to contain an accessible system neutral conductor, so other methods must be used in these zones.

Residual method — The residual method measures current imbalance amongst separate CT secondary circuits. This can be done either numerically by summing the phase current vectors within a relay, or directly by placing a sensor in the neutral circuit.

Flux-summation method — The flux summation method utilizes a core balance CT, also called a zero-sequence CT, which encompasses all phase conductors. These special instrument transformers have enlarged openings adequate for multiple cables and can either be rectangular or toroidal (Photo). Often, they have low turns ratios and do not require very high accuracy ratings because their primary currents are typically limited by neutral impedances as discussed above. A 50:5A, C10 CT is commonly used in this application. Principally, flux is only produced in the core balance CT when a non-zero net magnetic field results from a current imbalance in the encompassed conductors. This flux will produce CT secondary current.

As a side note, supplementary techniques such as flux-balancing-differential relaying, can provide very effective fault protection. These schemes detect abnormalities by evaluating the difference between the incoming and outgoing current on a given motor winding. Flux-balancing differential schemes are more costly because they require an additional set of CTs located at the motor necessitating additional physical space and associated wiring. For these reasons, they are typically found on larger medium-voltage motors and are excluded from the discussion of the more basic methods in this article. See page 49 of IEEE Std. C37.96-2012, “Guide for AC Motor Protection” for further information.

Which method should you use?

Out of three basic methods described, two are useful (in practice) for ground protection in medium-voltage motor circuits: the residual method and the flux summation method. They will be further discussed in turn.

Theoretically, residual current is proportional to current imbalance in the primary system. However, since residual current is the net secondary current from three separate phase CTs, unequal CT performance can disrupt this proportionality. When each of the three phase CTs inaccurately transforms primary current to varying degrees, false-residual current arises. This artificially inflates the perceived system imbalance.

Unequal performance can be attributed to a number of factors including varying levels of remanent flux, unequal secondary burdens, and manufacturing variances. These factors lead to unequal saturation, which is most prevalent in CT performance-intensive situations when current magnitudes are elevated. Examples of these situations are motor starts or line-to-line faults. Ground-fault element operation is undesirable in both of these cases.

Fig. 2. These current waveforms were captured during an actual medium-voltage motor start.

Figure 2 shows current waveforms captured during an actual medium-voltage motor start. The top waveform shows the phase currents, the middle waveform shows the residual current (residual method), and the bottom waveform shows the core-balance CT current (flux summation method). The oscillography was captured by a motor protection relay, and the graphic was produced by the manufacturer’s software program. The motor characteristics in this example were as follows:

• Power rating: 5,800 hp

• Full load rotational speed: 713 rpm

• Rated voltage: 4,000V

• Full load current: 808A

• Locked rotor current at 80% voltage: 3,510A

• Locked rotor current at 100% voltage: 4,528A

Additional relaying application data includes:

• Three 1,200:5A, C50 phase CTs

• One 50:5A, C10 core-balance CT

The middle trace, IG(A) in Fig. 2, represents residual current. The relay numerically derives this waveform by summing the three phase current input channels. The absence of a ground fault during the event suggests that the current shown is false-residual current. Notice the magnitude of the current reaches nearly 700A.

The motor starting current in the example was only around 300% of the CTs rating. Thus, it is unlikely that the amount of false-residual current shown can be attributed to the elevated current magnitude alone. Instead, the asymmetry (DC offset) of the primary current may have caused enough saturation to produce the residual current observed. The varying level of asymmetry in each phase further exacerbates the problem. Note that the DC offsets have been filtered out of the waveforms in Fig. 2 so asymmetry will not appear graphically.

Notice that all of the causes of false-residual currents listed above are due to unequal performance of different phase CTs and associated circuitry. Since flux summation only requires one CT, it isn’t subject to these same pitfalls. The IN(A) trace in Fig. 2 represents the current on the input channel connected to a core-balance CT. The current on this channel is near zero, as we expect in the absence of a ground fault.

IEEE Std. 242, “Recommended Practice for Protection and Coordination of Industrial and Commercial Power Systems,” recommends instantaneous ground-fault protection on medium-voltage motors. If residual overcurrent is used to provide this protection, the pickup setting must exceed the maximum expected false residual current during motor starts. Otherwise, you would risk nuisance tripping. Recall that the false residual current in the example was nearly 700A. This is a high lower-limit on the pickup setting, especially when you consider that maximum available fault current magnitudes on low-impedance grounded systems typically range from 100A to 1,200A. Pickup settings greater than the available fault current would of course render the protection useless.

To go one step further, IEEE Std. 242 cites a 10A to 30A primary instantaneous ground-fault setting. Residual settings this low are rarely achievable with any significant amount of false residual current. Therefore, the flux summation ground-fault protection method offers a distinct advantage over residual methods in medium-voltage motor applications.

Miscellaneous items for consideration

There are a variety of other things to consider when designing these protection systems. For example, surge arresters can be another source of residual current. IEEE 242 states that discharge current through surge protection can cause ground-fault protection elements to misoperate. Another problem commonly found in the field related to these schemes is the connection of cable shields. When shielded cable passes through a core balance CT, the shield conductor must be brought back through the CT. This cancels out any shield current. If this is not done properly, ground-fault current may flow undetected. This is described in more detail in IEEE Std. 242. Be careful when applying protection to fire pumps. Ground-fault protection is often excluded or limited in circuits supplying fire pumps. See the National Electrical Code for additional information. Lastly, it is worth noting that negative-sequence elements are also susceptible to false residual currents.       

Thornam, P.E., is a senior electrical engineer with Stanley Consultants in Centennial, Colo. He can be reached at [email protected].

About the Author

Joseph Thornam, P.E. | Senior Electrical Engineer

Voice your opinion!

To join the conversation, and become an exclusive member of EC&M, create an account today!

Sponsored Recommendations

Electrical Conduit Comparison Chart

CHAMPION FIBERGLASS electrical conduit is a lightweight, durable option that provides lasting savings when compared to other materials. Compare electrical conduit types including...

Fiberglass Electrical Conduit Chemical Resistance Chart

This information is provided solely as a guide since it is impossible to anticipate all individual site conditions. For specific applications which are not covered in this guide...

Considerations for Direct Burial Conduit

Installation type plays a key role in the type of conduit selected for electrical systems in industrial construction projects. Above ground, below ground, direct buried, encased...

How to Calculate Labor Costs

Most important to accurately estimating labor costs is knowing the approximate hours required for project completion. Learn how to calculate electrical labor cost.