CT Saturation: What It Is and How to Detect It

What CT saturation is

A current transformer (CT) reproduces in its secondary the primary current, scaled down to a standard value. To do this, its magnetic core needs to build up a flux. While the core has capacity to spare, the ratio holds and the relay receives a faithful replica of the current.

The core is made of a magnetic material that has a limit: when the required flux exceeds it, the material saturates and stops coupling the primary to the secondary. From that moment the secondary delivers less current than it should, and with a distorted shape. This is known as CT saturation.

Saturation is not a defect of the CT. It is a physical limit that the design must respect. What matters is whether it happens during a fault, before the protection has been able to decide. For the full picture of the device, see the guide on the current transformer. If you already have your CT data, you can estimate its margin with the CT saturation tool.

The physics: what the flux depends on

The flux the core must sustain is proportional to the voltage the CT has to generate in its secondary, integrated over time. That voltage is the secondary current multiplied by the total impedance of the circuit: the resistance of the winding itself plus the connected burden (leads, relay, terminals). For this reason, three factors decide whether the CT saturates:

• The fault current. The more current, the more voltage is needed.
• The secondary burden. Long or small cross-section leads, and relays with higher consumption, increase the impedance the CT must drive.
• The DC offset of the fault. This is the hardest factor to see and the one that surprises most.

The effect of DC offset

A fault close to a transformer or a generator does not start as a symmetrical wave. Depending on the instant at which it occurs and the ratio of reactance to resistance of the circuit (X/R), the current carries a DC component that decays slowly. That component is not transformed like the AC: it builds up in the core as flux.

In the worst case, the required flux can be several times that of a symmetrical fault of the same value. The commonly used reference, the IEEE C37.110 guide, deals with it using the «1 + X/R» factor: the higher the X/R ratio, the more flux the fault demands and the less time the CT takes to saturate. A CT can reproduce a symmetrical fault well and saturate with the same current when the fault comes with DC offset.

The excitation curve and the knee point

The relationship between the applied voltage and the current the core absorbs is called the excitation curve. At first the excitation current is very small. As the core approaches saturation it grows rapidly. The transition zone is the knee point.

IEC 61869-2 defines the knee-point voltage as the voltage that, increased by 10 %, makes the excitation current grow by 50 %. That definition separates healthy core behavior from saturated behavior. The same standard uses the rated knee-point voltage to specify class PX and PXR CTs. For classes P and PR it defines a limiting electromotive force: the accuracy limit factor multiplied by the rated secondary current and by the total secondary impedance.

Remanence

After a fault, the core can retain part of the flux, like a weak magnet. That remanent flux subtracts margin: the next fault can saturate the CT much sooner. IEC 61869-2 defines the remanence factor as the ratio of remanent flux to saturation flux, and limits its value to 10 % in low-remanence classes, which use air gaps in the core.

Consequences for protection

The effect depends on the function that uses the current:

• Overcurrent. The measured current is lower than the real one. The protection may take longer to operate or, if saturation is severe, never reach its pickup.
• Differential. If two CTs measure the same external fault and one saturates, the relay sees a difference that does not exist in the primary and may operate with no fault in its zone. For this reason busbar differential protection and transformer differential protection include restraint and saturation detection. High-impedance differential is designed to remain stable even with a saturated CT.
• Distance and directional. A distorted current alters the impedance or angle measurement, with a risk of incorrect reach.

A saturated CT can also hide the magnitude of a fault in the record. Readings after an event should therefore be cross-checked against other sources.

For a transformer differential, Industrial transformer protection settings shows how the protection is set to tolerate that error, and for the feeders of a plant it is covered in Industrial feeder protection.

How saturation is detected

In the fault record. Oscillographic records in COMTRADE format (see the guide on COMTRADE files) show saturation directly:

• the secondary wave is flattened at the peaks or loses its sinusoidal shape;
• the current is lower than that measured by other CTs in the same event;
• the distortion starts after some time, not from the first instant, because the CT needs a few milliseconds to reach the limit;
• harmonic content appears that was not there before.

In the CT. The excitation curve measured from the secondary lets you compare the core with its factory curve, see the real knee point and detect residual magnetism. Instruments such as the CT Analyzer perform this test without injecting primary current.

In the relay. Numerical relays for saturation-sensitive protections, such as the differential, recognize it from the harmonic content of the current or from the time the wave stays undistorted at the start of each cycle. The relay uses that healthy interval to decide.

The record of a real fault is interpreted with the method in Fault analysis with oscillographic records.

How the risk is reduced

The risk is controlled from the design stage, with general criteria:

• Choose the CT class according to the function. Differential protection CTs and those on lines with reclosing need more margin than simple overcurrent ones.
• Keep the secondary burden low: short or larger cross-section leads and low-burden relays.
• Include the DC offset of the fault in the calculation, not only the symmetrical current.
• In critical applications, use gapped or transient-response classes, which limit remanence and have more margin.

The detail of each calculation and its verification depends on the equipment and the faults of the system. It is done in the short-circuit study and in protection coordination.

Calculate it here

To estimate whether a CT saturates with its burden and fault current, use our free CT burden and saturation tool.

We can help with your current transformers

At Tiza Group we review the sizing of protection CTs, analyze fault records where saturation is suspected and set the protections with that criterion. See our protection coordination service and our testing and commissioning service.

Related guides

• Current transformer (CT): parts, functions and tests
• Busbar differential protection (87B)
• Fault analysis with oscillographic records
• Industrial feeder protection
• Industrial transformer protection settings

References

• IEC 61869-2, instrument transformers, additional requirements for current transformers: definitions of knee-point voltage, secondary limiting electromotive force and remanence factor.
• IEEE C37.110, guide for the application of current transformers used for protective relaying purposes.

Frequently asked questions