Split Core Current Transformers

What is a split-core current transformer? More specifically how does a split-core current transformer differ from the typical current transformer? Just like the typical current transformer, the split-core current transformer measures alternating current flowing through a conductor. The distinguishing feature of the split core current transformers is that their design permits them to be assembled around a buss bar without disconnecting the buss bar. The typical current transformer is usually a toroidal coil, which is slipped over the end of a buss bar, hence requires disconnecting the buss bar. "C" - cores and "U" core structures are commonly used for split-core current transformers because they are relatively easy to take apart and put back together around the buss bar. Some sort of bracketry or band clamps and holds the assembled pieces of the split-core current transformer together. Historically, this has not been as practical ( but is possible ) for toroidal coils. The bracketry is more complicated. Typically, the coil(s) must be sector wound on the toroid before cutting the core in half, whereas the “U” and “C” core structure of the typical split-core current transformer permit use of bobbin wound coils which can be wound independently of the core. There are now some flexible toroids, which permit the “split-core” feature of installing it around a buss bar.

The electrical performance of split-core current transformers is not as good as that of the continuous toroidal coil. The “circle” like ( or “ring” like ) shape of the toroid usually offers a shorter magnetic path length than other cores. Since the toroids are continuous, they do not add any air gap to the core structure. Split-core current transformers ( including toroidal split- cores ) add some air gap to the core structure. Consequently, the split-core current transformers will draw more magnetizing ( exciting ) current than a continuous toroidal current transformer made of the same core material ( assuming comparable size and/or weight. ). The toroidal shape provides better magnetic coupling and less leakage inductance than the “C” and “U” core structures commonly used in split-core current transformers.

Split-core current transformers for lower frequency applications ( power frequencies ) typically use grain oriented silicon steel or nickel alloys for the core material. There are some more exotic materials available. The material is cut into strips and then wound on an arbor ( mandrel ) to form a core. The core is then cut in half. These are known as “tape-wound” cores because their construction resembles a roll of tape. Strip thickness varies from 0.025” down to 0.0005”. The thinner strips have less core loss at higher frequencies hence they are used in higher frequency applications up to about 10 kilohertz. High accuracy current transformers require low core losses hence they either utilize the thinner strip thickness, the lower core loss materials such as the nickel alloys, or both. Ferrite materials are usually used for very high frequency designs, up to several megahertz. Some very specialized applications may require a core-less ( air-core ) coil. Some theory of current transformer operation is given further below

Butler Winding can make ( and has made ) split-core current transformers in a variety of shapes and sizes. The "U" and "C" cores structures are the most typical, but Butler Winding is capable of producing a variety of other custom designs. Butler Winding already works with various standard types of "core with bobbin" structures ( E, EP, EFD, PQ, POT, and others ), and does some custom bobbin wound designs. Usually, we can readily adapt our bobbin winding equipment to wind the split-core current transformer coils you need. Our upper limits are 40 pounds of weight and 2 kilowatts of power. We have experience with foil windings, litz wire windings, and perfect layering. For toroids, we can ( and have done ) sector winding, progressive winding, bank winding, and progressive bank winding. Butler winding has a variety of winding machines, bobbin/tube and toroid. That includes two programmable automated machines and a taping machine for toroids. Butler winding has vacuum chamber(s) for vacuum impregnation and can also encapsulate. To ensure quality, Butler Winding purchased two programmable automated testing machines. Most of our production is 100% tested on these machines. For more information on Butler Winding's capabilities, click on our "capabilities" link.

Current Transformer Design Specifications

The designer must either determine or be supplied with the information needed to design the current transformer. The needed information is listed below along with a brief description if needed. Add any additional items required by your particular application.

Describe Primary Current – State maximum current value and type of measurement ( r.m.s., average, peak, etc. ), Give type of waveform ( sine wave, square wave, triangular, etc. ). State either continuous current or describe the applicable duty cycle.

Give Number of Primary Turns – This is the number of times the primary conductor ( buss bar ) passes through the core window.

The Desired Current Ratio – This is simply the desired secondary current value ( at a specified value of primary current ) divided by the primary current value that generates said value of secondary current. Alternatively, a turns ratio could be specified. but don’t expect the current ratio to exactly equal the turns ratio.

Define the Output Burden ( Load Resistor ) – Specify the value and type of the intended secondary load. The type of load is usually resistive ( a resistor ), but could be inductive or capacitive ( which complicates things ). Alternatively, the desired output voltage per unit of primary current can be specified. The value of the load resistor can then be calculated.

Required Accuracy – This is usually expressed as either a maximum percentage or maximum absolute change over the entire primary current range. It includes both measurement tolerances and variations over the operating range(s). It may be expressed over a portion of the operating range or at specific operating points.

Minimum Inside Window Dimensions – This is the primary conductor ( buss bar ) dimensions plus any additional distance needed to clear any obstacles encountered during installation of the current transformer..

Dimensional Constraints – Overall width, length, thickness.

Termination – Describe how you want the secondary terminated. Some possible examples are: terminal block, lead wires ( with or without terminal lugs), or header
( with p.c.b. pins or pads ). If leads, what length, insulation type, voltage rating, etc..

Mounting -- Describe how you expect it to be mounted. Will it be supported by the primary conductor ( hang on the buss bar ), or will the current transformer support the primary conductor.

Voltage Isolation Requirements – In many applications, the current transformer’s secondary winding rests on the primary conductor ( buss bar ), hence it must be adequately insulated according the expected conductor voltage potential and/or the required equipment voltage classification for the intended application.

Corona Requirements, if applicable – Give test criteria: maximum test voltage, minimum voltage ramping time, minimum voltage inception value, minimum voltage extinguish value.

Maximum Temperatures – Specify the maximum ambient temperature and the maximum expected temperature of the primary ( buss bar ) conductor. If applicable, state the maximum allowed temperature rise.

Application Standards -- Application standards may exclude use of some materials and require use of some materials . Some examples of such standards are minimum temperature ratings ( regardless if actual is less ), flame retardancy, vibration, out-gassing, and required labeling.

Environmental Restrictions – Examples are: poor cooling due to confined space, corrosive environment, water spray, ultra-violet light, and vibration.

Current Transformer Theory of Operation.

In the typical current transformer application, the primary winding consists of one to a few turns of wire. The primary wire size is much larger than the secondary wire size. The number of secondary winding turns is a selected multiple of the primary turns. Figure 1 gives a circuit schematic of a current transformer application. The current transformer shown represents an ideal transformer. The ideal transformer has infinite no-load input impedance, 100% magnetic coupling between transformer windings ( hence no leakage inductance), zero winding resistance, zero core losses, and no capacitance. ( Capacitance, leakage inductance, winding resistance, and core losses are considered to be parasitic components. ) The output voltage is exactly proportional to the primary voltage times the turns' ratio. There is no regulation drop. There are no losses. Since there are no parasitic components the ideal current transformer is 100% accurate. The conservation of energy requires that the output power equals the input power, hence Vp x Ip must equal Vs x Is. Since Vs = Vp x Ns / Np, it can be shown that Is = Ip x Np / Ns. Is = Vs / RL, hence Ip = Ns x Vs / ( RL x Np ). With an ideal current transformer there is no phase shift ( except 180 degrees depending on the choice of output connections ).

The ideal transformer’s secondary resistive load consumes power equal to Is x Is x RL. This same amount of power must be consumed at the primary terminals. The secondary load RL can be replaced ( commonly referred to as “reflected” ) with a resistor across the primary terminals, RLr. By applying the conservation of energy, one can show that RLr equals Np x Np x RL / ( Ns x Ns ), OR RLr equals RL times the turns ratio squared ( where turns ratio = Np / Ns ). If Np / Ns is small, then the RLr is very small. The primary voltage drop is Ip x RLr. A very small value for RLr means that the current transformer presents a low insertion loss to the primary current and a low primary voltage drop.

The reflected load impedance acts in parallel to the transformers own input impedance. The ideal current transformer has infinite input impedance. This infinite impedance would correlate to an infinite inductance inserted in series into the path of the primary conductor. Without the load ( or burden ) the current transformer acts like an inductor and would completely block the primary current flow. Any constant value of alternating current would, in theory, produce an infinite primary voltage drop. In reality the current transformer’s input inductance ( hence also impedance ) cannot be infinity. The current transformer has an inductance value which acts in parallel to the reflected load. The core has losses which can be represented as a resistor in parallel with the reflected load and the transformer’s self inductance ( no load inductance ). Without the load resistor the inductance and core loss will place an upper limit on the primary voltage, but this voltage could still be substantial. Core saturation is also a possibility. A turns ratio step-up would result in even higher secondary voltage. Any circuitry beyond the secondary load resistor could be subjected to high voltage, possibly resulting in circuit damage. Because of this potential high voltage, the load resistor should never be removed from the secondary when the current transformer is being powered. Figure 2A shows an equivalent circuit schematic for a current transformer with load RL. The ideal ( induced ) secondary voltage is now denoted as Vsi and Vs now denotes the voltage at the secondary terminals. Notice that the schematic contains the ideal current transformer and load as before plus transformer mutual inductance Lm, secondary winding resistance Rs, core loss resistor Rc, secondary leakage inductance Lks, and primary leakage inductance Lkp. Just like for the load resistor, the other secondary circuit components can be reflected to the primary side of the transformer. This is illustrated in Figure 2C.

The parasitic components, Rs, Lkp, and Lks, all act to lower the output voltage across RL, hence the output voltage, Vout, will not equal the induced secondary voltage Vsi. Rs and Lks act in series with RL and are reflected to the primary side along with Rs. Their presence presents added impedance to the primary current hence an increase in primary voltage in proportion to the impedance. Consequently, RL still has the same voltage drop and current flow as it did without Lks and Rs even though Vs does not equal Vout. The phase shift associated with Lks will cause some slight deviation from the ideal current ratio ( equals the turns ratio ).

The current transformer’s self ( no-load ) inductance Lm and the core loss Rc shunt current away from the reflected load and reflected parasitic components. Their impedances act in parallel to the reflected impedances, consequently lowering the impedance seen by the primary current and the resulting primary voltage. Less primary voltage means less output voltage and less secondary current. Consequently Lm and Rc also cause deviation from the ideal current ratio.

As long as Rc, Lm, Lkp, Lks, and Rs are constant in value, The actual current ratio will be some fixed ratio times the ideal ( or desired ) current ratio. One can compensate for the deviation from the desired current ratio by appropriate choice of secondary turns. The number of turns will be a little lower than that for the associated ideal turns ratio. For constant values accuracy could be 100% except for any turn resolution limitations ( full turns versus fractional turns ).

Accuracy concerns arise from non-constant values for Rc, Lm, and to a lesser degree from Lkp and Lks. These values usually vary with core induction levels, hence they vary over the range of primary current being measured. ( Air core transformers are stable but magnetic coupling is relatively poor hence relatively large leakage inductances. ) Since Rc and Lm impedances act in parallel to the reflected load, higher Rc and Lm values have a smaller effect and consequently increase accuracy. Cores materials with high permeability and low core loss are preferred for high accuracy applications.

At higher frequencies winding capacitance becomes a concern. Figure 3 gives an equivalent circuit schematic, which includes winding capacitance. Leakage inductance and winding capacitance are actually distributed components, but are shown as lumped approximate equivalent components. Like Lm, winding capacitances shunt current around the reflected load. The inductances and capacitances can interact and consequently may produce spurious oscillations. it is also possible to develop “parallel resonance”. High frequency coil designs seek to minimize winding capacitances.

If you need assistance with your current transformer design, please contact Butler Winding and ask for Engineering.