By Allan Ludbrook

Sags are the most frequent cause of downtime on industrial production systems. A common cause of sags or faults on utility system lines. Utilities have evolved complex protection coordination schemes intended to minimize the inconvenience to their customers when faults occur. But these schemes do not always benefit the user of modern computer controlled equipment and variable frequency drives. Some options for users and utilities are discussed.

Definitions

The following definitions were taken from the IEEE Dictionary and the Westinghouse Electric Utility Distri-bution Systems Engineering Reference Book.

Reclosure: The automatic closing of a circuit interrupting device following automatic tripping.

Bulk Power: Receives power from the transmission system and transforms it to the subtransmission voltage.

Subtransmission: The function of transmitting power from a bulk power source to Distribution Substations.

Distribution: Receives power from the subtransmission circuits and transforms it to the primary feeder voltage.

Primary Feeders: Circuits emanating from distribution substations providing a path for power flow to the distribution transformers.

Distribution: The function of transmitting power from the distribution substation to the Distribution Transformers

Distribution Transformers: Trans-forms the primary feeder voltage to the consumer utilization voltage.

Figure 1 is a on line diagram showing the basic components of a distribution system.

The Utility System

Over the last 10-15 years utilities have been eliminating systems with distribution voltages below 25kV, e.g. 4160 Volt and 13,800 Volt systems. The motivation for this move was to save energy, capital costs, and real estate. Previously, power was transmitted from Bulk Power Substations at voltages of 25 kV and above (subtransmission voltages) to Distribution Substations (sometimes called Municipal; Substations) where the voltage was stepped down to levels below 25kV (distribution voltages). On the secondary of this Distribution Substation transformer, circuit breakers served a number of primary feeders.

Capital, real estate, and energy costs were reduced by eliminating the Distribution Substation, Energy was saved by increasing the voltage of the primary feeders.

The functional features sacrificed for these cost saving benefits were:

the loss of the buffering action of the Distribution Substation transformer impedance leading to an increase in the number and severity of sags experienced by every customer on the system.
loss of the sectionalizing capability provided by the Distribution Substation circuit breakers.
the ability to adjust the steady state voltage levels of small groups of Primary Circuits.

These factors relate to most significant industrial plants with sensitive loads fed from the same Bulk Power Station as large areas of residence loads.

In fairness to the utilities, when the decisions were made to alter the distribution system, the information age had not yet arrived. Therefore sags and the like were not recognized as a reliability issue.

Sag Severity

Figure 2 shows one line diagrams of fault impedances of the new and old distribution systems. The impedances values shown are chosen for simplicity and relative significance. Note that the Distribution Substation transformer has a significantly higher impedance relative to the lines and the Bulk Power Substation transformer. Exhibit 3 serves to compare the sag performance of the new and old systems. It shows:

the variation of severity of sags with fault location and measuring point location.
the comparative severity of sags with fault location and measuring point location.
the duration of the sag that just meets the IEEE Emerald Book Power Quality Envelope (PQE).

Sag Incidence

Figures 4 and 5 are the examples of one line diagrams for the old and new distribution systems respectively. The susceptibility to sags, also shown in Figure 3, is calculated on the basis that:

each unit length of subtransmission and primary circuit line has the same exposure to faults.

the subtransmission line and the primary circuits are, for simplicity of calculation, all the same length.

there are 4 subtransmission lines or equivalent per bulk power station.
there are 4 distribution substations per subtransmission line (old system).
there are 3 primary circuits per distribution substation (old system)
there are 12 primary circuits in the new system for each subtransmission line in the old system.

The susceptibility to sags is also shown in Figure 3. Note that the faults at nodes 3 and 4 have the highest susceptibility level because they include the greatest length of line. Also note that in the old system, node 3 and 4 faults can last for 1 second or more without falling below the PQE, while with the new system any fault lasting for more than 0.1 seconds falls below the PQE. The results are calculated for bolted faults. Less severe faults will permit longer duration sags while still meeting the PQE.

Sag Duration

Sag duration is controlled by the circuit breaker reclosing relays and the operating time of the circuit breaker itself. Short Operating times result from relays set to trip on overcurrent instantaneously. Longer times result from relays set to trip on Òinverse timeÓ. Reclosing relays are usually set to trip instantaneously on the first fault. On reclosing, the relay trips inverse time is used to allow a downstream circuit breaker or fuse to clear the first and so minimize the number of customers affected by the fault. In the new system, the downstream circuit breaker has been eliminated and replacement for its fault interruption and sectionalizing function is rarely provided. Therefore, the additional trip time yields no benefit whatever; it only lengthens the duration of the sag imposed on customers fed from unfaulted feeders.

Conclusions on Utility Related Factors

Sags of magnitude exceeding the PQE are much more likely to occur from faults anywhere on the distribution system than they previously were. Therefore, a high level of preventive maintenance is now more important and also more difficult because all the voltage levels are also higher than before.

The duration of sags is a function of the protective relay settings, especially the settings prevailing after reclosure. Therefore, minimizing sag duration requires reducing the trip time on reclosure.

CASE STUDY

The following case study illustrates some of the principles discussed above. This case study involved a manufacturing plant served by a short feeder directly from the 27.6 kV bus of a Bulk Power Substation. Because production interruptions due to sags were excessive, the 27.6 kV input was monitored for a period of months.

Some typical sag records are described below:

the circuit breaker on a faulted feeder de-energized the faulted circuit and remained open. (Trace 1)
the circuit breaker on the faulted feeder de-energized the faulted circuit, the fault cleared and the circuit breaker reclosed. (Trace 2)
the circuit breaker de-energized the faulted circuit, the fault remained, the circuit breaker reclosed automatically and then tripped again. (Traces 3 and 4).
Interesting features of these recordings are:
The depth and duration of the initial fault.
The slight rise in the prevailing voltage after a fault reflecting the reduction of load due to the loss of the feeder.
The small sag after reclosure due to the re-energizing inrush.
The greater depth and longer duration of the sag following circuit breaker reclosure.
The increase in sag depth on reclosure.

The sags shown in Traces 1 and 2 caused no production interruption. The sags shown Traces 3 and 4 shut down all the production lines at the plant. Exhibit 6 shows the effect of utility faults on the operations of the plant over a number of months related to the PQE. With the exception of the one sag at 47 per cent for 1 second, there is only a small margin between faults that do and do not cause a process interruption. This result suggests that a combination of action by the utility in reducing the duration of sags on unsuccessful reclosures and action by the user in increasing the equipment tolerance to sags would eliminate most of the process interruptions.

Possible Solutions

The solutions fall into a number of categories:

1) Reduce the sag sensitivity of the plant machinery .

Back up the power with UPS -- But this option is too expensive.

Back up control with UPS and modify drives to restart automatically after a short sag -- This process is currently being designed.

2) Reduce the incidence of sags.

Feed the plant from the transmission voltage (115kV) -- Again, this is too expensive.

Run the Bulk Power Station with the tie breaker open -- Doing so can lead to a variety of operational problems.

Improve preventative maintenance on the 27.6 kV distribution system.

3) Reduce the amplitude of sags.

Install reactors in the main feeders -- this process it too expensive.

4) Reduce the duration of sags

Speed up protective relay operation on reclosures -- A process which is still under evaluation.

Chosen Solutions

1) Back up control with UPS and modify the drives to start automatically after short sags.

2) Investigate speeding up relay operations on reclosure.

3) Improve preventative maintenance on all the whole 27.6 kV system.

Neither of the above options is a trivial matter. Redesigning the control system of a coordinated drive using relays is difficult because the circuitry may be poorly documented and require wiring changes. System using PLCs with their self documenting features allow design changes to be made off line and without wiring changes. Changes to protective relay settings is not a matter undertaken lightly by a utility. The whole protection coordination design for the Bulk Power Station has to be thoroughly reviewed.

Allan Ludbrook is President and Principal Engineer with Ludbrook and Associates, an electrical engineering consulting firm located in Dundas, Ontario.