3
\$\begingroup\$

I am trying to determine the best option for monitoring the current in a specific application but am having difficulties deciding between a current transformer, Rogowski coil or Hall effect sensor.

The main purpose of the sensor is to determine whether there is any fault or unbalanced loads in the circuit. Under normal operating conditions the current flow will be fairly low in the 0-20 A range. For unbalanced loads the current can go up to 60-70 A. The main concern are jacket faults which can lead to currents up to 400 A for an extended period of time.

The aim of the current sensor is to detect when these currents rise so that action can be taken before any significant damage is done to the cable.

The main concern is accuracy, however there is some leeway there as an issue with the cable will usually cause a fairly significant increase in current.

Another major concern is price.

The current transformer seems like a viable choice since it has a large current range and is fairly accurate. However, while the Rogowski coil is typically used as a method for temporary current measurements, I have also seen that it can be used as a low cost alternative to a CT. From my understanding, many of the issues with accuracy for a Rogowski coil are due to positioning, but this would be installed in a solid box so the coil would always be in the same predetermined position. I do not know if this is enough to achieve a higher accuracy though.

The Hall-effect sensor I do not think makes much sense since we do not need to know DC currents and they are more difficult than CT's to engineer for high accuracy. Also, from what I have seen they tend to be higher cost but they do seem to come in a fairly wide range of prices. I did not want to rule them out altogether at this point.

Thank you for any help or guidance you can offer.

\$\endgroup\$
2
  • \$\begingroup\$ Take a look at the LEM 'UltraStab' CTs (that are actually linearised flux gates) less than 0.02% from DC to 800kHz \$\endgroup\$
    – Jay M
    Dec 13, 2019 at 16:10
  • \$\begingroup\$ Peripheral: note that Hall devices with integral current shunts MUST be safe at the highest possible current that will be attained. If the device melts (literally) under overload then isolation will be lost with 'potentially' severe consequences. This may also be true for Czt & RC but more robust protection against such failure is usually much easier. \$\endgroup\$
    – Russell McMahon
    Apr 9, 2021 at 15:03

3 Answers 3

3
\$\begingroup\$

You are correct in that a hall sensor is not as useful for AC currents and if you have no DC offset that is probably not the best choice.

The issue with a current transformer is core saturation. As long as you can use one with the current range you need that will be your best bet.

The Rogowski coil will work but you will need an integrator to generate the current waveform which usually offsets the additional cost of the current transformer.

\$\endgroup\$
4
\$\begingroup\$

CT everyday of the week.

You are building a device for protective function not for measuring/monitoring. As others have said watch saturation but a CT rated for the highest normal loads would be the best bet. eg 70A in your example so 100A CT. That will protect for small overcurrents and for larger overcurrents (you will experience a degree of saturation depending on the CT characteristics) although the rate of rise (current) should be detected by the device and trio quickly.

That said 400A is not a huge amount so you could get a larger CT but this is unusual in protection schemes.

The other alternative would be to look into sensitive fault protection eg earth fault and residual, which uses a much smaller CT (say 10% of your main CT) to detect differences in current between conductors.

Not sure what you are doing but I would imagine 100A CT would be fine.

Sorry didn’t realise this post was so old! Was going to offer additional help via email if needed but I’d imagine you have sussed it by now.

Sounds like you are protecting cable from a generator or maybe susbsea equipment?

\$\endgroup\$
1
\$\begingroup\$

My 2¢ on this:

  1. Rogowski coil must be positioned to capture the field it is supposed to capture, and be protected from / designed to cancel out external fields which may generate false readings. There are ways to do this, and if the coil is inside the PCB, the ways may be easier to implement
  2. To protect the coil from undesirable external fields, you have to either simulate the system or make educated assumptions about external fields, under all possible circumstances (normal operation, transients, imbalanced loads, fault conditions) and design accordingly
\$\endgroup\$
1
  • \$\begingroup\$ Take a look at the Slimsensor (by Gridkey/Lucy Electric) this is immune to external magnetic fields to a level that exceeds requirements of industry tests in that area. \$\endgroup\$
    – Jay M
    Dec 13, 2019 at 16:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.