How Does Your Capacitor Work?

Surge Protection Capacitors

Surges and Protection against surges:

Electrical networks experience surges wherein a voltage or a current rises rapidly to unsafe values and destroys the dielectric insulation. These, along with partial discharges which these start, are blamed for the major portion of failures of electrical equipment of all types.

As per modern thinking, most of the surges are current – sourced as against the normal voltage sourced electric power supply. An amount of let off energy, determines this current which flows to ground – irrespective of the circuit resistance. If a contact of a lighting conductor stripe is bad, it creates dangerous voltages – rather than reducing the current.

This rapid rate of rise of current is responded by a magnetic circuit (of all types of transformers) with an equally rapidly rising flux, a back EMF and a very high induced voltage. This voltage causes breakdowns, flash overs, partial discharges and so on.

This surge has two or three parameters which lead to electrical break – down –

Any capacitor can not be charged to a full surge voltage instantly. It will take our indefinite amount of current to do so. Thus, it takes time to get charged. This time slopes down the almost vertically advancing surge were – though not substantially. Even a small reduction in di/dt reduces the magnetically induced voltages from an infinite value to a finite value and this is how surge capacitors help.

Surge capacitors by themselves are protective on small voltage spikes – with limited involved energies. They have to be supplemented with lighting arrestors which can ground large amounts of surge energies.

Let us see what is involved in designing and building up a surge capacitor so that it gives a desired performance.

Typical Calculations:

  1. Performance of a 0.25mfd Surge capacitor against an impulse type surge:
    2.

  1. We take the following figure for our illustration:

Wave is travelling at a speed of 200KV per microsecond.

The transmission line has a surge impedance of 400 ohms.

The transmission line has a flash over voltage of 180KV.

The surge capacitor rated at 18KV, 0.25mfd is fixed at point A.

Then without the capacitor, the surge will require 0.9 microseconds to reach 180KV.

The current will rise at a rate of 200/400 KA per microsexcondi.e. 500 Amps/ microsecond.

The magnitude of the current at 180KV point will be 450 Amps.

The energy on the wave, in this 0.9 microseconds, will be –

0° t I x V x dt = 0° t 500 x 106 x 200 x 103 x 106 x t2dt

= 0° t [10 x 1019 ] [t2]dt

= [10 x 1019 ] [t3/3]0 0.9 x 10-6

= 24.3 Joules.

Under the worst possible conditions of coincidence of surge peak with system voltage peak, we will have peak voltage on surge capacitor = 18 x 1.414KV i.e. 25.45KV. Plus voltage peak due to surge.

Energy stored in the capacitor will be = ½ CV2

= ½ x 0.25 x 10-6 x (182 x 2) x 106 Joules.

= 81 Joules under voltage peak conditions.

This capacitor will be superimposed with additional joules of 24.3 taking its total to 105.3 joules.

Its voltage will rise momentarily to

105.3 = ½ x 0.25 x 10-6 x [KV]2 x 106

i.e. KV = 29.02 or by a more 3.568KV in 0.9 microseconds.

Thus this rate of rise of voltage has been reduced to 3.568/0.9 KV/microsecond i.e. 3.964 KV/microsecond as against 200 KV/microsecond.

Two things will happen:

  1. The voltage will climb up at this rate till the associated lightning arrestor fires at its rated voltage or

  2. The energy in the spikes is so small that the capacitor holds the system to a safe enough small value.

  1. Suppose that the surge energy is very large – say 5000 joules and we want to restrict the peak voltage build up to 30KV only.

    2. Then 5000 = ½ C [302 – 25.45]2

    i.e. C = 39.63mfd i.e. almost 4MVAR, 18KV, per phase.

    This is too large.

  2. Today’s ZnO type lightning arrestor also slope down a surge slightly before firing. However, if the speed of surge is high, the voltage at which they fire also becomes higher and beyond the safe zone. A surge protection capacitor – in parallel, helps.

  1. How is the value of 0.25mfd arrived at?

    2.

    Let us take another example.

    A 13.8KV motor is connected across a bus which has a surge impedance of 100 ohms and the connected network constants and circuit breaker parameter can develop a, surge capable of reaching 120KV peak. [This can be verified at site with a storage oscilloscope.]

    This bus has a lightning arrestor capable of firing at 39KV.

    We want to design a surge capacitor which will clamp the surge voltage rise down to 28KV peak in one microsecond.

    The peak surge current = 120(KV)/100 KA i.e. I = 1200 Amps.

    I/C = rate of rise of volatge = 28000 volts/microsecond.

    = 1200/C x 10-6

    i.e. C = 1200/28000 x 10-6

    = 0.043 mfd.

    A surge capacitor, standardised at 0.25mfd, is more than adequate.

    Normally, when more than one H.T. motors operate on a common bus, it is recommended to employ individual surge capacitors and one group lightning arrestor.

  2. Inductance and Resistance in series with a Surge Capacitor:

    3.

Any inductance or resistance will have a high voltage drop across it on account of a high surge current. Thus, a surge will charge the system vertically to this voltage drop – before being sloped down by the surge capacitor. This vertical rise is detrimental to transformer windings in the system and defeats to an extent, the very purpose of employing surge capacitors.

It is essential:

  1. To have a specially designed surge capacitor with minimum self inductance and with minimum internal series resistance &

  2. To connect it as nearer to the equipment, to be protected, as possible,

To keep the connecting links shortest possible and

To have more than adequate, conducting cross section on the leads.

  1. Rated voltage of a Surge capacitor:

    2.

    The surge capacitors are normally, single terminal, body grounded type. If these get connected across a system with ungrounded neutral, there is a possibility of the line terminal getting full line voltage – instead of a phase voltage, should one of the phases get shorted (under a surge). Besides they are subjected to high rate of charging when they cater to surges. As such they are rated at the line voltage or slightly higher – even though normally they will operate at phase voltage.

  2. Specific Failure modes:

    3.

High speed Circuit Breakers on generator bus ducts and on industrial arc furnaces have given instances of surge capacitor failures.

The overall bus capacitances are low in the region of a few picofarads – which give a natural resonance frequency in multiples of kilo-hertz. When a 0.25mfd capacitor per phase is added to this system, it lowers the resonance frequency substantially. Should this coincide with any harmonic of 50 Hz, then there will be instant build up of high voltages and capacitor failures. This will be

prominent with Vacuum Circuit Breakers which are liable to pre-arcing or restriking. The best remedy for this would be to add another capacitor in order to detune under away from a resonance.

On arc furnaces it is recommended to use a R.C. combination, rather than a surge capacitor. Since the VCB operation is very frequent, it is desirable to step up the capacitor voltage rating.

The surge impedance figures for H.T. buses, transmission lines etc, are given in standard Reference Handbooks. Actual surges can be pictured and held on todays digital storage oscilloscopes. These, together, can pinpoint the trouble mode, against which suitable remedial measures can be adopted.

The surge capacitors have fairly high built in safety factor. The failures are very rare in actual practice.

Back