CT theory

It rather depends on the core size and the winding materials. The other problem is that in the CT the current is more or less fixed by the load side, and as many volts per turn are available from the supply to 'make it so'. In a voltage transformer the voltage is fixed, and the current varies with load.

As current flows in the 1 turn primary, unless and equal and opposite number of amp-turns flow in the secondary, the magnetic field builds up much like a choke ~(which is like a transformer with an unloaded secondary) At some point the core will saturate, when all the magnetic domains (imagine little molecular magnets) have rotated to align. At this point the effective inductance (and so the voltage drop per amp) drops. This results in a clipping of the induced voltage. However, to give good linearity and to fit round the bus bar,  CT cores are oversized, and the  turns ratio can be very high, especially the sort clamped to 100 amp bus bars. The pracitical up shot is that a turns ratio of 1000 to 1 or even 10,000:1 is realistic, so when the 1 turn measuring current is only dropping maybe a tenth of a volt to 1V , the secondary side may be enough to do serious injury.

In normal operation the primary is dropping a few millivolts.

As the metering wiring is intended  to handle a few tens of volts at most, it may well fail. It would be possible to wind a CT with suitable HV winding like a car ignition coil so that it did not flash over, but it is far easier to leave something connected to limit the voltage.


Personally I use CTs in a pulse power world, and there I like to see anti-series zener diodes or gas discharge devices, directly on the winding, or cascades of two transformer cores e.g.  two of 33:1 rather than 1 of 1000:1 with limiters on the transfer winding.

Mike.

It rather depends on the core size and the winding materials. The other problem is that in the CT the current is more or less fixed by the load side, and as many volts per turn are available from the supply to 'make it so'. In a voltage transformer the voltage is fixed, and the current varies with load.

As current flows in the 1 turn primary, unless and equal and opposite number of amp-turns flow in the secondary, the magnetic field builds up much like a choke ~(which is like a transformer with an unloaded secondary) At some point the core will saturate, when all the magnetic domains (imagine little molecular magnets) have rotated to align. At this point the effective inductance (and so the voltage drop per amp) drops. This results in a clipping of the induced voltage. However, to give good linearity and to fit round the bus bar,  CT cores are oversized, and the  turns ratio can be very high, especially the sort clamped to 100 amp bus bars. The pracitical up shot is that a turns ratio of 1000 to 1 or even 10,000:1 is realistic, so when the 1 turn measuring current is only dropping maybe a tenth of a volt to 1V , the secondary side may be enough to do serious injury.

In normal operation the primary is dropping a few millivolts.

As the metering wiring is intended  to handle a few tens of volts at most, it may well fail. It would be possible to wind a CT with suitable HV winding like a car ignition coil so that it did not flash over, but it is far easier to leave something connected to limit the voltage.


Personally I use CTs in a pulse power world, and there I like to see anti-series zener diodes or gas discharge devices, directly on the winding, or cascades of two transformer cores e.g.  two of 33:1 rather than 1 of 1000:1 with limiters on the transfer winding.

Mike.