2
\$\begingroup\$

I'm working somewhere that uses an unearthed 3.3kV network for safety critical drives. I've been learning about ground fault detection on ungrounded systems, using VT's and measuring voltage across a broken delta arrangement to detect ground fault:

Ground Fault Detection

(Source)

I'm confused as surely the star point on the VT's effectively ground's the ungrounded delta connections, negating the purpose of having an ungrounded system in the first place? Each phase will now have a reference to ground, allowing for phase-ground faults?

Am I thinking about this the wrong way?

\$\endgroup\$
  • \$\begingroup\$ Your correct in that something is wrong. The 4800/120vac transformer acts as a ground reference, so corner-grounding a delta feed would create a short circuit. You can do one or the other-but not both. \$\endgroup\$ – Sparky256 Aug 2 '16 at 21:58
  • \$\begingroup\$ I'm not familiar with high-voltage systems, but could you insert a large resistance between the line and the VT used for measurement? \$\endgroup\$ – immibis Aug 2 '16 at 23:19
0
\$\begingroup\$

It may seen like that but it is the only way to detect a phase to ground leak/fault in the 'isolated' system.

If you did not have that ground connection then no current would flow into the ground fault and you would not be able to detect it. With this circuit you will have no current in the earth point if you have a balanced load (and supply).

If you have a fully isolated system you cannot detect a faulty to ground from any phase (or neutral) as no current flows. All that happens is that your system gains a real ground reference at some point (perhaps not the best point) but no ground/fault current will flow until some point of your system connects to a point with a potential difference, this will cause a current to flow.

Having no way to detect the single fault is risky especially with high voltages or current. If your circuit must be floating it could float at any voltage and possibly exceed insulation capabilities.

On consideration I think it may be the best way to manage the 'isolation' and it will reference your circuit to ground at the virtual neutral point. If your circuit must be isolated so it can float to the load potential then you should reference the ground to a suitable neutral point there.

\$\endgroup\$
0
\$\begingroup\$

After discussing this in work I've realised I was thinking about this wrong.

As the secondary of the transformer is not grounded, it does not matter that the star point of the VT's grounds the phases, the windings of the transformer are not being shorted out.

When an earth fault does occur on one phase, as the VT's are grounded this provides a path for current flow, allowing the fault to be detected. This current flow is additional to the current already present in the phase and the system continues to operate as normal.

Fault Condition

The imbalance between the 3 phases is picked up by the relay, which will indicate the fault.

\$\endgroup\$

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.