How can a current probe avoid saturation?

Question

It is a buck converter, And I want to know how current probe can avoid saturation. Turns ratio is 1:1000. And DC makes transformer saturation. So... How can probe avoid saturation?

Answer 1

You cannot use a transformer (whether or not a current transformer or Rogowski coil) to sense or measure DC current. That being said, there is a trick you can use if your buck converter is operating in Discontinuous Conduction Mode (DCM). In DCM, the valley amplitude of the current ripple in the inductor is 0 amps. In that case, if you have a scaled and shifted replica of the current in the inductor, and you shift it (for example with a clamp circuit) so that the valley of this replica signal is at 0 amps, then this clamped/shifted signal is then a scaled replica of the original inductor current. Be forewarned that this ONLY WORKS if the buck converter is in DCM, and will not work if the converter is operating in Continuous Conduction Mode (CCM).

From this point on, I will assume that you are only interested in the ripple current amplitude.

I want to know how current probe can avoid saturation.

Typically saturation figures are given in terms of the magnetic B field (i.e. magnetic flux density). But for a current transformer, there is a very simple way to determine if a core will saturate using the H field (magnetic field strength).

Define \$H_{sat}\$ as the magnetic field strength at the point where the core enters saturation. (Note that because of hysteresis, the core may not leave saturation until H is much lower, or even reversed in polarity).

If you don't have a B-H chart for your core material, then you can get an approximate value for \$H_{sat}\$ from:

$$H_{sat} \approx \frac{B_{sat}}{\mu}$$

This relationship is only approximate because the B-H curve not perfectly linear below the saturation point.

The amplitude of \$H\$ in a transformer core is simply the net ampere-turns of current (which is the same as the magnetomotive force in the core) divided by the effective length of the core.

$$H = \frac{\mathscr{F}_{mag}}{\mathscr{l}}$$

The net ampere turns of current for a current fed primary, i.e. for a current transformer, is maximum when the secondary is open-circuit. Current flowing in the secondary can only reduce the net ampere turns. So, if a CT transformer core does not saturate when the secondary is open-circuit, it will not saturate when it has a burden resistor.

Now consider a small inductor, such as this one manufactured by Miller and distributed by Jameco. It is not clear whether this is an image of a 100 uH 2.4 A inductor, or another inductor from the same series. However, it is about 20 mm in diameter and about 8 mm thick. It appears to have about 24 turns.

If this indeed is the right inductor, it can tolerate 2.4 A DC without saturating. The core can tolerate 24 * 2.4 = 57.6 ampere-turns before saturation. If it were used as a current transformer with a single turn primary, it would handle 57.6 amps of instantaneous current before saturation. Thus, even 57.6 amps of DC current should not saturate the core. So, (at least some) current transformers with even relatively small toroids can sustain a relatively high instantaneous or sustained current without saturation.

If your CT is saturating, the \$H_{sat}\$ value for the core material or the effective core length is too small, and a larger diameter core, or a core with a larger \$H_{sat}\$ should solve the problem.

Answer 2

How can probe avoid saturation?

Yes, it can be a problem. For the AC part it isn't a problem because the ampere-turns on the primary are cancelled by the ampere-turns on the secondary. So, the only way to avoid saturation from DC levels is to choose a core that is big enough or, choose a core that has a low enough permeability to avoid saturation.

Or use a hall-effect probe that won't exhibit this problem. Then, you are faced with finding a hall-probe that has decent bandwidth and minimal problems from the voltage edges seen on the primary upsetting the signal output. There are a few coming out from Allegro that look really, really good but, I'm not sure they are widely available yet.

You can also find CTs that use a hall-effect sensor to measure the DC flux and utilize a 3rd winding (fed from a high-compliance current generator) to cancel out that flux. More expensive and, from what I've seen, they don't offer massive bandwidth improvements over hall-effect sensors.

Answer 3

As you seem to be interested only in the AC part of the current and saturation is your specific problem, changing the current transformer to a rogowski transducer sounds like the obvious replacement.