2
\$\begingroup\$

We have a complex network of earth wires and earth pits. These earth wires are connected to power panels and CNC machines. However, the earthing system is very old and poorly maintained.

Is there a way to check if the panel/ machine is properly earthed or not?

Obviously, visual inspection and line tracing is an option to trace the connectivity of earth wire, but I am looking for a more efficient solution.

For example - Is there any specific method to test if the body of earthed machine/ panel is at earth potential and will remain as such?

\$\endgroup\$
  • 1
    \$\begingroup\$ An earth impedance/integrity tester springs to mind. \$\endgroup\$ – Andy aka Aug 25 '17 at 16:09
2
\$\begingroup\$

In the USA this is called a Ground Bond Test. There is specialized equipment that sends a high current through the grounding/bonding conductors, measures the resultant voltage, then calculates the resistance to ground.

In a well-grounded system, this voltage will be very low. If the voltage exceeds a certain threshold (meaning that the resistance to ground is too high) the test is automatically halted.

In your case, you would first find (or create) a known-good earthing point. Then you would perform the ground-bond test from this point to the ground connections of individual machines/panels.

Ground-bond testing equipment is expensive; you may wish to find a testing agency that can rent it to you, or which will perform the tests for you.

\$\endgroup\$
  • \$\begingroup\$ Hiring the test equipment is almost impossible (at least in my country) because the license is binded with calibration certificate and the operator course/exam. So nobody is so stupid to rent an instrument and to loose the money for inspection. \$\endgroup\$ – Marko Buršič Aug 25 '17 at 17:45
  • \$\begingroup\$ @MarkoBuršič: Interesting! Around here it is pretty standard, at least in my experience. ElectroRent, for example, sells new equipment for a few thousand USD or rents it for a few hundred USD per month. (This isn't a recommendation; I haven't done business with them). It's true that you can't certify your results if you do it yourself, but this may not matter to the OP. \$\endgroup\$ – bitsmack Aug 25 '17 at 18:11
  • 1
    \$\begingroup\$ @SagarUpadhyay Well, a megger is opposite of what you need. It provides a high voltage into a (nominally) open circuit. You need something that can drive a high current into a (nominal) short circuit. If you have an appropriate power supply you could create a bunch of current and measure the resultant voltage separately. But most power supplies can't feed a short circuit. They either limit themselves or simply fail :) Also, the measured voltage should be in the mV range, so it's tricky... \$\endgroup\$ – bitsmack Aug 25 '17 at 19:35
  • 1
    \$\begingroup\$ @SagarUpadhyay And a reason to not simply use an Ohmmeter to check for resistance is because they only source a few mA. An iffy ground connection could sink a few mA successfully and still be dangerous in the case of higher amperage fault. \$\endgroup\$ – bitsmack Aug 25 '17 at 19:42
  • 1
    \$\begingroup\$ Sorry, @SagarUpadhyay, but this is the only good way I can think of. You could, of course, run all new grounding wires to the utility ground, and hope that it's sufficient. But even then, you'd probably want to perform a ground-bond test on the new connections. Perhaps someone else here will post a better idea! \$\endgroup\$ – bitsmack Aug 26 '17 at 17:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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