IRON CORE TYPES OF POWER TRANSFORMERS BASIC INFORMATION

Oriented (anisotropic) silicon-steel laminations.

The iron cores of conventional transformers consist of anisotropic silicon-steel laminations with lamination thickness ranging from 0.1 mm to 0.4 mm. In a transformer, the flux travels mostly within the limbs in the with-grain direction, and in the cross-grain direction only near the corners and lamination joints of transformer cores; thus oriented steel sheets are used.

The with- and cross-grain structure of oriented steel is determined by the rolling direction of the sheets during manufacture. Each side of a lamination is coated with insulating material so that no eddy currents can flow between laminations.

The coating does not significantly interfere with the passage of flux. The magnetic resistance, or reluctance, is only slightly increased and is taken into account via the iron-core stacking factor ϕFe = #(iron cross section of all laminations of core)/(cross section of entire core including insulation between laminations).

The stacking factor is in the range of 0.93 ≤ ϕFe ≤ 0.97 for 60 Hz units. For anisotropic electrical silicon steel the relative permeability is larger (and thus the magnetization required is smaller) in the with-grain direction (direction of rolling) than in the cross-grain direction. Similarly, the core losses are small in the with-grain direction and relatively large in the cross-grain direction.

Amorphous (glass-type) cores.

Amorphous magnetic materials either are obtained by quenching the molten material at high cooling rates or are manufactured by deposition techniques in a vacuum. The quenching process does not permit the forming of a crystalline structure, and therefore amorphous magnetic materials have a structure similar to glass.

The cores of transformers with amorphous alloy (AMTs) can be fabricated in the same manner as those made of oriented-silicon-steel. METGLAS (trademark of the Allied Signal Co) cores are 30% heavier than comparable oriented silicon-steel cores, but the no-load losses in amorphous alloy wound cores are only 30% of those in comparable oriented-siliconsteel wound cores.

However, the rated power efficiencies of present-day designs of AMTs and silicon-steel pole transformers with wound cores are about the same. For example, the rated power efficiencies of 20 kVA and 50 kVA wound-core AMTs at unity power factor are ηpower = 98.26% and 98.59%, respectively, while that of a 25 kVA oriented-silicon-steel wound core (10) at unity power factor is ηpower = 98.31%. The fabrication cost for AMTs with wound cores is higher than that for oriented silicon-steel wound cores.

HIGH PERMEABILITY STEEL USED IN TRANSFORMER CORE BASIC INFORMATION

Use of cold-rolled grain-oriented steel as described above continued with only steady refinement and improvement in the production process until the late 1960s.

However, in 1965 the Japanese Nippon Steel Corporation announced a step-change in the quality of their electrical steel: high-permeability grainoriented silicon steel.

Production is simplified by the elimination of one of the coldrolling stages because of the introduction of around 0.025% of aluminium to the melt and the resulting use of aluminium nitride as a growth inhibitor.

The final product has a better orientation than cold-rolled grain-oriented steel (in this context, generally termed ‘conventional’ steel), with most grains aligned within 3° of the ideal, but the grain size, average 1 cm diameter, was very large compared to the 0.3 mm average diameter of conventional material.

At flux densities of 1.7 T and higher, its permeability was three times higher than that of the best conventional steel, and the stress sensitivity of loss and magnetostriction were lower because of the improved orientation and the presence of a high tensile stress introduced by the so-called stress coating.

The stress coating imparts a tensile stress to the material which helps to reduce eddy-current loss which would otherwise be high in a large-grain material.

The total loss is further offset by some reduction in hysteresis loss due to the improved coating. However, the low losses of high-permeability steels are mainly due to a reduction of 30 40% in hysteresis brought about by the improved grain orientation.

The Nippon Steel Corporation product became commercially available in 1968, and it was later followed by\ high-permeability materials based MnSe plus Sb (Kawasaki Steel, 1973) and Boron (Allegheny Ludlum Steel Corporation, 1975).

TRANSFORMER CIRCUIT MAGNETIZING REACTANCE

For an ideal transformer, the magnetizing current is assumed to be negligible. For a real transformer, some magnetizing current must flow when voltage is applied to the winding in order to establish a flux in the core.

The voltage induced in the winding by the flux restrains the magnetizing current. It was shown earlier that the magnetizing current is not really sinusoidal, but contains many odd harmonics in addition to the fundamental frequency.

If we neglect the harmonics and concentrate on the fundamental frequency, the magnetizing current in the winding lags the applied voltage by 90°. In a two-winding transformer, this is equivalent to placing a reactance Xm, called the magnetizing reactance, in parallel with the transformer terminals.

The peak value of the magnetizing current is determined from the B-H curve of the core, which we have seen is very nonlinear. Therefore, the magnetizing reactance is not a constant but is voltage dependent; however, if the peak flux density is kept well below the point of saturation, Xm can be approximated by a constant reactance in most engineering calculations.

It is generally desirable to maximize Xm in order to minimize the magnetizing current. We saw earlier that inductance is inversely proportional to the reluctance of the core along the flux path and the reluctance of an air gap is several thousand times the reluctance of the same distance through the steel.

Therefore, even tiny air gaps in the flux path can drastically increase the core’s reluctance and decrease Xm. A proper core design must therefore eliminate all air gaps in the flux path.

Since any flux that is diverted must flow between the laminations through their surfaces, it is vital that these surfaces lie perfectly flat against each other. All ripples or waves must be eliminated by compressing the core laminations together tightly.

This also points out why the oxide layers on the lamination surfaces must be extremely thin: since these layers have essentially the same permeability as air and since the flux that is diverted from the air gaps must then travel through these oxide layers, the core’s reluctance would greatly increase if these layers were not kept extremely thin.

TRADE-OFF BETWEEN STEEL AND COPPER IN THE DESIGN OF A TRANSFORMER

The previous section illustrated the fact that reducing a winding conductor’s length enables a corresponding reduction in the conductor’s cross-sectional area to maintain the same total I 2R losses.

Therefore, to maintain constant losses, the required volume of copper is proportional to the square of the conductor length L.

Vcu directly proportional to L^2

The required number of turns in a winding N is inversely proportional to the volts per turn generated by the core. The volts per turn are proportional to the total magnetic flux, and the flux is proportional to the cross-sectional area of the core AFe for a given allowable peak flux density, expressed as volts per turn.

N = inverselt proportional to Afe

From simple geometry, we know that the conductor’s length is equal to the number of turns times the circumference of the coil. If the cross section of the core is nearly circular and the winding is placed directly over the core, the circumference of the coil is roughly proportional to the square root of the core’s cross-sectional area.

Assuming that the core’s volume is roughly proportional to the core’s cross sectional area, The relationships given indicate that the volume of copper required to limit I 2R losses is inversely proportional to the volume of the core for a given KVA rating, winding configuration, and applied voltage.

In other words, adding 25% more core steel should permit a 25% reduction in the quantity of copper used in a transformer. This results in a 1:1 trade-off in copper volume vs. core volume.

However, that the total core losses are proportional to the core volume for a given flux density. For example, if we decide to reduce the volume of copper by 25% by increasing the volume of core steel by 25%, the core losses will increase by 25% even though the conductor losses remain constant.

In order to maintain the same core losses, the flux density must be reduced by increasing the cross-sectional area of the core, meaning that additional iron must be added. Therefore, the 1:1 trade off in copper volume vs. core volume is only a very rough approximation.

There are also other practical physical limitations in selecting the dimensions of the core and windings; however, this exercise does illustrate the kinds of trade-offs that a transformer design engineer can use to optimize economy.

TRANSFORMER NO LOAD LOSSES BASICS AND TUTORIALS

TRANSFORMER NO LOAD LOSSES BASIC INFORMATION
What Are The Transformer No Load Losses?


Alternating magnetic flux produces both hysteresis losses and eddy-current losses in the steel. As we have seen, hysteresis losses depend on several factors including the frequency, the peak flux density, the type of core steel used, and the orientation of the flux with respect to the ‘‘grain’’ of the steel.

All of the above factors, except the frequency, are under the control of the transformer designer. Core losses are sometimes referred to as iron losses and are commonly referred to as no load losses, because core losses do not require load currents.

Decreasing the induced voltage per turn can reduce the peak flux density. This obviously involves increasing the numbers of turns in both the primary and secondary windings in order to maintain the same transformer turns ratio.

The disadvantage of adding more turns is that this increases the length of conductor and increases the conductor resistance. More cross sectional area is required in order to keep the resistance constant.

Doubling the number of turns requires about four times the volume of copper. Another way of reducing core losses is to use various types of low-loss core steels that are now available, including ‘‘amorphous’’ core materials, which have extremely low losses and superior magnetic properties.

Unfortunately, amorphous core materials have ceramic-like properties, so fabricating transformer cores with these materials is much more difficult than with laminated steel cores.

With grain-oriented steel, the direction of the core flux must be kept more or less parallel to the grain of the steel by mitering the corners of the laminations where the flux changes direction by 90°. Since the flux will cross the grain at about a 45° angle at the mitered edges, the hysteresis losses will increase somewhat in these places.

These additional localized core losses must be factored into the calculation of the total core losses. Building up the core with thin laminated strips controls eddy losses in the core, each strip having an oxide film applied to the surface.

The oxide film is extremely thin and it is more like a high-resistance film than true electrical insulation; but since the potential differences between adjacent laminations is quite small, the resistance of the oxide film is very effective in breaking up the eddy current paths.

During the manufacture of the core, the core cutting machine must not be allowed to get dull; otherwise, ‘‘burrs’’ will form along the edges of the laminations. Burrs are imperfections that form electrical bridges between the laminations and create paths for eddy currents and increased losses.

Sometimes the eddy currents near a burr can be large enough to cause localized overheating that can actually cause core damage. Core losses are approximately proportional to the square of the excitation voltage E applied to the transformer.

Therefore, placing an equivalent linear conductance Gm across the transformer terminals can approximate transformer core losses. The core losses are expressed by Wm = E^2Gm

TRANSFORMER CORE EARTHING BASICS AND TUTORIALS

TRANSFORMER CORE EARTHING BASIC INFORMATION
What Is Transformer Core Earthing?


Core earthing
Before concluding the description of core construction, mention should be made of the subject of core earthing. Any conducting metal parts of a transformer, unless solidly bonded to earth, will acquire a potential in operation which depends on their location relative to the electric field within which they lie.

In theory, the designer could insulate them from earthed metal but, in practice, it is easier and more convenient to bond them to earth. However, in adopting this alternative, there are two important requirements:

ž The bonding must ensure good electrical contact and remain secure throughout the transformer life.

ž No conducting loops must be formed, otherwise circulating currents will result, creating increased losses and/or localised overheating.

Metalwork which becomes inadequately bonded, possibly due to shrinkage or vibration, creates arcing which will cause breakdown of insulation and oil and will produce gases which may lead to Buchholz relay operation, where fitted, or cause confusion of routine gas-in-oil monitoring results y masking other more serious internal faults, and can thus be very troublesome in service.

The core and its framework represent the largest bulk of metalwork requiring to be bonded to earth. On large, important transformers, connections to core and frames can be individually brought outside the tank via 3.3 kV bushings and then connected to earth externally.

This enables the earth connection to be readily accessed at the time of initial installation on site and during subsequent maintenance without lowering the oil level for removal of inspection covers so that core insulation resistance checks can be carried out.

In order to comply with the above requirement to avoid circulating currents, the core and frames will need to be effectively insulated from the tank and from each other, nevertheless it is necessary for the core to be very positively located within the tank particularly so as to avoid movement and possible damage during transport.

It is usual to incorporate location brackets within the base of the tank in order to meet this requirement. Because of the large weight of the core and windings these locating devices and the insulation between them and the core and frames will need to be physically very substantial, although the relevant test voltage may be modest.

POWER TRANSFORMERS CORE IMPROVEMENT BASIC AND TUTORIALS

POWER TRANSFORMERS CORE IMPROVEMENT BASIC INFORMATION
What Are The Transformer Core Improvements?


The major improvement in core materials was the introduction of silicon steel in 1932. Over the years, the performance of electrical steels has been improved by grain orientation (1933) and continued improvement in the steel chemistry and insulating properties of surface coatings.

The thinner and more effective the insulating coatings are, the more efficient a particular core material will be. The thinner the laminations of electrical steel, the lower the losses in the core due to circulating currents. Mass production of distribution transformers has made it feasible to replace stacked cores with wound cores.

C-cores were first used in distribution transformers around 1940. A C-core is made from a continuous strip of steel, wrapped and formed into a rectangular shape, then annealed and bonded together.

The core is then sawn in half to form two C-shaped sections that are machine-faced and reassembled around the coil.

In the mid 1950s, various manufacturers developed wound cores that were die-formed into a rectangular shape and then annealed to relieve their mechanical stresses. The cores of most distribution transformers made today are made with wound cores.

Typically, the individual layers are cut, with each turn slightly lapping over itself. This allows the core to be disassembled and put back together around the coil structures while allowing a minimum of energy loss in the completed core.

Electrical steel manufacturers now produce stock for wound cores that is from 0.35 to 0.18 mm thick in various grades.

In the early 1980s, rapid increases in the cost of energy prompted the introduction of amorphous core steel. Amorphous metal is cooled down from the liquid state so rapidly that there is no time to organize into a crystalline structure.

Thus it forms the metal equivalent of glass and is often referred to as metal glass or “met-glass.” Amorphous core steel is usually 0.025 mm thick and offers another choice in the marketplace for transformer users that have very high energy costs.

TRANSFORMER CORE DESIGN AND CONSTRUCTION BASICS AND TUTORIALS

TRANSFORMER CORE DESIGN AND CONSTRUCTION BASIC INFORMATION
Transformer Core Design and Construction: A Tutorial 


Air gaps in a magnetic core will add considerable reluctance to the magnetic circuit. Remembering that the inductance of a coil and the magnetic reluctance are inversely proportional, air gaps reduce the inductance of the coil and increase the magnitude of magnetizing currents. In practical transformers, we want to reduce magnetizing currents to almost negligible levels; it is therefore important to eliminate all air gaps if possible.

One approach would be to make the core from a solid block of material. This is impractical from the standpoint of fabricating the transformer, since the coils would have to be wound through the core window.


Also, since metallic core materials conduct electric current as well as magnetic flux, the induced voltages would produce large circulating currents in a solid core. The circulating currents would oppose the changing flux and effectively ‘‘short out’’ the transformer.


A practical solution is to fabricate the core from thin laminated steel sheets that are stacked together and to coat the surfaces of the laminations with a thin film that electrically insulates the sheets from each other. Steel not only has excellent magnetic properties but is also relatively inexpensive and easy to fabricate into thin sheets.


In a modern transformer plant, steel ribbon is cut into sections by a cutting/punching machine commonly called a Georg machine. The sizes and shapes of the sections are determined by the core design of the individual transformer.

The thickness of the sheets varies somewhat; core laminations operating at 60 Hz are between 0.010 and 0.020 in. thick, with 0.012 in. being the most common thickness in use today.


Different methods of stacking core steel have been used in the past. One such method is called the butt lap method using rectangular core sections and is illustrated in Figure 1.11

Even if the edges of the segments do not butt together perfectly, as shown in the exaggerated edge view at the bottom of the figure, the alternating even and odd layers assure that the magnetic flux has a continuous path across the surfaces of the adjacent layers.

DISTRIBUTION TRANSFORMER COOLANTS BASIC AND TUTORIALS

DISTRIBUTION TRANSFORMER COOLANTS BASIC INFORMATION
What Are The Different Distribution Transformer Coolants?

Mineral Oil

Mineral oil surrounding a transformer core-coil assembly enhances the dielectric strength of the winding and prevents oxidation of the core. Dielectric improvement occurs because oil has a greater electrical withstand than air and because the dielectric constant of oil is closer to that of the insulation.

As a result, the stress on the insulation is lessened when oil replaces air in a dielectric system. Oil also picks up heat while it is in contact with the conductors and carries the heat out to the tank surface by self convection. Thus a transformer immersed in oil can have smaller electrical clearances and smaller conductors for the same voltage and kVA ratings.

Askarels

Beginning about 1932, a class of liquids called askarels or polychlorinated biphenyls (PCB) was used as a substitute for mineral oil where flammability was a major concern. Askarel-filled transformers could be placed inside or next to a building where only dry types were used previously.

Although these coolants were considered nonflammable, as used in electrical equipment they could decompose when exposed to electric arcs or fires to form hydrochloric acid and toxic furans and dioxins.

The compounds were further undesirable because of their persistence in the environment and their ability to accumulate in higher animals, including humans. Testing by the U.S. Environmental Protection

Agency has shown that PCBs can cause cancer in animals and cause other noncancer health effects. Studies in humans provide supportive evidence for potential carcinogenic and noncarcinogenic effects of PCBs (http://www.epa.gov). The use of askarels in new transformers was outlawed in 1977 (Claiborne, 1999).

Work still continues to retire and properly dispose of transformers containing askarels or askarel-contaminated mineral oil. Current ANSI/IEEE standards require transformer manufacturers to state on the nameplate that new equipment left the factory with less than 2 ppm PCBs in the oil (IEEE, 2000).

High-Temperature Hydrocarbons

Among the coolants used to take the place of askarels in distribution transformers are high-temperature hydrocarbons (HTHC), also called high-molecular-weight hydrocarbons.

These coolants are classified by the National Electric Code as “less flammable” if they have a fire point above 300˚C.

The disadvantages of HTHCs include increased cost and a diminished cooling capacity from the higher viscosity that accompanies the higher molecular weight.