Measuring DC with a CT 3

[LiteYear]

A CT is traditionally used to measure AC currents. Technically however, a CT is sensitive to DC as well but typically the core will become saturated quickly and indication will be lost.

Are there any established methods for enabling an ordinary CT to accurately measure, say, 0 to 90Hz? I can vaguely imagine it being possible by putting a second secondary coil (a tertiary coil?) around the CT which regularly resets the flux to avoid saturation. The measurement circuit could then maybe compensate for this offset.

We've been lead to believe that normal CTs are being used in the field to measure 0-90Hz, including DC. Any ideas on how this might be being done beyond the realm of theory?

 

[VEBill]

Hall effect sensor works at DC.

50 / 60 Hz is good old CT.

 

[LiteYear]

Yes, Hall Effect sensors are one way to do it, but they require a gap in the core to fit the sensor. They also have a bunch of drawbacks in fixed installations, such as sensitivity drift and requiring de-gaussing.

There appear to be two other patented techniques:

1) Orthogonal fluxgates. See US Patent No. 6,885,183. Effectively uses orthogonal coils, one of which is alternately driven into saturation. By measuring the effect on the other coil, you can determine the DC component passing through the primary of the inner coil. The AC component is then detected by adding a Rogowski coil.

2) Commutating windings. See US Patent Application Number: 20120063055. Effectively the primary current is alternately passed in "windings" of opposite direction. Simultaneously the secondary current is drawn from two windings of opposite direction. The commutation prevents saturation and also allows measurement of both DC and AC components up to several MHz.

I don't know if either design is used in practice. The latter in particular only got filed a couple of months ago.

Still, all solutions require modifying the CT to some extent, even if it's just to wind a second secondary or pass the primary through both ways.

However, there is at least one prominent manufacturer (see

http://www.littelfuse.com/acdc.html

and the linked datasheets) that claims it can measure DC to 90Hz with a standard earth leakage CT. Their "theory of operation" is to apply a known oscillating signal to the secondary and measure the effect of the primary on that signal.

Is anyone familiar with this method? Any idea on susceptibility to saturation or accuracy in practice?

 

[Skogsgurra]

There are DC transformers that work with magamp components, aka transductors (no, not transducers, transductors). That seems to be the technique used by Littelfuse.

Fluxgates are common and sensitive. Can be had commercially from DC to something like 5 MHz bandwidth. I use them.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[VEBill]

'DC Clamp' meters are not at all uncommon. It's my understanding the most of them at the lower price end of the market use Hall effect sensors. Having used them and seen them used many times to measure DC currents from 100s of mA to 200A.

Are you looking to simply buy and employ one, or are you looking to design and build one using some new and different technology?

 

[OperaHouse]

I remember seeing that saturation circuit more than 25 years ago to make an electronic compas so it certainly can be sensitive. Not that many components either. For some reason it never made it into industry.

 

[cranky108]

Optical CT's can be used over the full range of DC to 90 Hz. Still somewhat of an oddity, because of the amplifiers used to convert the optical signal to current.

They also can be used over an extended range because they don't saturate like wire CT's.

 

[LiteYear]

Thanks for all the feedback. All very enlightening.

My primary interest is in understanding the pros and cons of the various techniques of accurately measuring low amplitude, low frequency currents. The use of a CT of course, is one option. I want to be able to assess commercial products and also pick technologies for our own products.

Skogsgurra, the mention of current transductors was particularly revealing. I agree it seems to be the basis for the Littelfuse technique. It was very useful to read about magnetic amplifiers and saturable reactors. Thanks.