# How To Control Reactive Power In Larger Electrical Plants With Multiple Incomers

Home / Technical Articles / How To Control Reactive Power In Larger Electrical Plants With Multiple Incomers

## Controlling reactive power

Most compensation banks are controlled stepwise. For this purpose it is essential to ‘know’ when it is allowed to (de)activate a capacitor step by the power factor relay (controller).

The so-called C/k value is calculated by the step size C divided by the ratio k of the current transformer. It is clear that a capacitor with, for instance, 50 kvar may not be switched in if the power factor relay measures a deviation of just 10 kvar reactive power with regard to the preadjusted power factor target.

This technical article explains two ways of controlling reactive power in plants with with multiple incomers:

### 1. Measuring by Means of Summation Current Transformer

Larger electrical plants have multiple feed-ins with two or more power transformers usually working in parallel. Regarding the control of reactive power, there are two solutions possible.

The first solution is, according to Figure 1, to measure the load via three current transformers of 1500 A/5 A with each incoming supply fed by one power transformer of 1000 kVA. The three current paths are summed in a summation current transformer with three input paths, of 5 A each, and one output path of 5 A too.

This output is wired to the current path of the power factor relay controlling a 12-step central compensation bank of 600 kvar.

The method has a big disadvantage: for proper control of reactive power it is necessary to keep the two coupling circuit breakers 1 and 2 closed all the time! In case of any short circuit, all three power transformers generate a very high power as a rule.

Suppose coupling circuit breaker 1 is opened – then the power factor relay is not able to compensate the reactive power of consumers connected to the busbar to the left. However, the relay notes a higher request for capacitors to be switched in.

This could lead to overcompensation at transformers 2 and 3 – it means that the capacitive reactive power goes up to the busbar of the MV level and via transformer 1 to the consumers to be compensated. This transmission of reactive power causes additional active power losses along the cables and in the transformers of course.

The third disadvantage is that the power factor relay is not able to ‘see’ in which area reactive power arises, due to the summation current transformer and the central-type compensation connected to the busbar in the middle. Despite this, it is essential to discuss how to calculate the C/k value for correct adjustment.

For this purpose Equation 1 is used:

Factor k is to be determined separately:

The first term symbolizes the total ratio of the three current transformers, and the second, symbolizing the summation transformer to be multiplied, results in a very high total ratio of k = 900. It is then necessary to check whether the relay’s minimum sensibility of 1% will not be undersized.

Finally the C/k value is to be calculated according to Equation 1:

This value is adjustable at most power factor relays by manual C/k setting or at relays with half-automatic adaption. The market offers power factor relays to set the parameters ‘current transformer ratio’ and (smallest) ‘step size’ digitally. Wrong settings may lead to ‘hunting’ a capacitor step!

Suppose a fourth incoming supply with an additional power transformer of 1000 kVA increases the total ratio referring to four current transformers to be summarized up to k = 1200 and the C/k value decreases down to 0.038 approximately.

This value decreases the 1% level distinctly.

There would be only the possibility of varying the factor of 0.65 up to 0.85 or, in another calculation, to show at which percentage of the step size (50 kvar) a power factor relay with fully automatic C/k adaption would start to control:

The relay would re-energize the capacitor at a level of 85% referring to 50 kvar, or 42.5 kvar only. Proper control of the capacitors is no longer guaranteed due to the tolerances of the relay and the capacitor as well.

Regarding factor k, it does not matter whether all power transformers are in operation or not. Factor k is just a characterizing constant of the entire electrical plant.

This method of controlling reactive power has some disadvantages, as noted above. To install individual compensation banks for each incoming supply is much more suitable, as described in the following paragraphs.

Go back to contents ↑

### 2. Parallel Operation of Compensation Banks for Each Incoming Supply

Referring to Figure 2, the advantages can be seen at once. Control of reactive power runs individually for each incoming supply with the help of its own compensation bank, each controlled by an automatic reactive power controller.

It does not matter if the position of the coupling circuit breakers 1 and 2 is open or closed. If they are open, the compensation banks operate individually. If they are closed, one encounters the parallel operation, to be discussed next. All three power factor relays working in parallel are preset to the identical power factor target between each other.

Regarding the switching time delay per step, there is no need to preset the same value, just an approximate one, for example in the range between 35 and 40 s per step.

The C/k calculation from Equation 1 again is simple as no summation current transformer must be considered.

Thus the ratio of the current transformer amounts to:

This C/k value is to be preset at each relay with manual or half-automatic C/k adaption identically in the case of open coupling circuit breakers. But what is the situation with closed coupling circuit breakers?

The step size of 50 kvar is only one-third measured by the current transformers, supposing the same impedance voltage at the power transformers operating in parallel.

This would mean the preset C/k value depends on the position of the coupling circuit breakers and is to be corrected all of the time. This is indeed a major disadvantage if using power factor relays with manual or half-automatic C/k adjustment.

Thus the best solution would be to use power factor relays with the feature of ‘full automatic C/k adaption’.

They always register the so-called compensation effect for each step independently of the position of the coupling circuit breakers 1 and 2. Even if breaker 2 is open and 1 is closed, the current transformer of compensation bank C registers the full size of the capacitor, but the current transformers of compensation banks A and B only half the size.

It is not necessary to discuss the disadvantages of power factor relays with manual or half-automatic C/k adaption further.

During the installation of large electrical plants with currents of more than 1000 A approximately occurring attention should strictly be paid that, for example, cables from the power transformers to the busbars are of the same length in order to ensure symmetrical load distribution.

Another situation it is essential to mention is the case when one power transformer, for example no. 3, is out of operation due to maintenance and, provided that the coupling circuit breakers are closed of course, that the power factor relay of compensation C is energized but does not get any signal from the current transformer.

With regard to Figure 3 there is no vector existing that exceeds the C/k threshold level. Older relays would ‘stand by’ with the number of capacitors switched in as before. They would not be able to establish control until maintenance is finished.

Not all brands of power factor relays have the capability to disconnect the capacitor(s) after a defined measurement time of ‘no current’ or ‘I = 0’ is displayed. Thus it is essential to get information on this from the manufacturer.

Go back to contents ↑

### To sum up

Summarizing, it is best to focus on the described method in preference to the method of Section 11.6.1 as there are several advantages:

1. Control of reactive power independently of the position of the coupling circuit breaker.
2. Possibility to limit the power in case of a short circuit.
3. De-centralized compensation enabling location much closer to consumers of reactive power.

Go back to contents ↑

Reference // Reactive power compensation by Wolfgang Hofmann, Jurgen Schlabbach and Wolfgang Just (Purchase hardcover from Amazon)