Sonel PQM analyzers case study #14: Transformer idling

Induction machines, i.e. motors and transformers, should not be operated with little or no load, as their efficiency is then low. Generally speaking, larger squirrel-cage induction motors show an efficiency of more than 80% in the load range of about 40...100%. Power transformers have the highest efficiency in the range of about 30...50% load. Low load operation causes adverse effects such as increased reactive power consumption and a decrease in power factor.

Description of the identified problem

A 2 MVA medium-voltage transformer supplies power to a production plant with a specific daily duty cycle. The transformer remains idle for most of the day. The effects and side effects of this way of powering the facility need to be analysed.A 2 MVA medium-voltage transformer supplies power to a production plant with a specific daily duty cycle. The transformer remains idle for most of the day. The effects and side effects of this way of powering the facility need to be analysed.

Measurement tools used:

• Sonel PQM-710 analyser
• Set of F-2A clamps (In=3000 A)
• Sonel Analysis software

 

Figure 1 Graph of apparent power showing one day of operation of the production plant

 

Figure 2 Graph of tg(φ) during transformer idling (marker no. 1 and no. 3) and active power load (marker no. 2)

 

Figure 3 Graphs of active energy (green) and reactive inductive energy (pink)

 

PRELIMINARY CONCLUSIONS:

1. An analysis of energy consumption indicates that the load is inductive throughout the operation of the production plant (Fig. 3).
2. During idling (marker no. 1 in Fig. 2), tg(φ) is between 0.7 and 1.4 causing a penalty to be charged for exceeding the 0.4 threshold of the Quality Regulation.
3. During operation at load, the reactive power is well compensated (marker no. 2 in Fig. 2).
4. A similar daily profile applies to all days of the week except for Sundays, when only the transformer idling condition occurs.

By analysing the values of active and reactive power, and hence tg(φ), one may attempt to calculate the capacity to compensate for inductive reactive power during transformer idling.

General formula for calculating the power of compensating capacitors:

QK = (tg(φ) - tg(φ)Z) * P

where:
QK – value of compensating reactive power
tg(φ) – actual value
tg(φ)Z – setpoint value, assumed level of 0.35
P – load power

The level of active power during idling is on average 13 kW at tg(φ)=1.2...1.4 and 11 kW at tg(φ)=0.7 (markers no. 1 and 3 on Fig. 2). Hence, the calculated compensating powers are 11...12.5 kvar and 2.5 kvar. The discrepancies indicate that the transformer is not idling but operating in the low load state. For this case, it is not possible to effectively compensate the operation with a fixed compensator.

FINAL CONCLUSIONS:

1. The occurrence of different reactive power values in the state without production processes in the plant clearly suggests that this is not typical idling, but low load operation.
2. The level of transformer under-compensation is not negligible and accumulates over a long period of time to a large value of reactive power, generating significant charges.
3. In order to maintain a tg(φ)<0.4 around the clock, regulated reactive power compensation is required; single-stage compensation cannot be used.

RECOMMENDATIONS:

1. The installation of a dynamic compensator or a graded capacitor battery should be considered.

 

Authors:
Krzysztof Lorek, Marcin Szkudniewski