Large AC to DC has high current in ground wires

[KWRON]

There is a large 23MW rectifier which has up to 300A on ground wires attached to the transformer ground pads. the building has a ground grid and attaches to the building columns. The transformer ground pads are attached to this underground ground grid.
The grounds attached to the columns also have a large amount of current.
These currents are continuous. The transformer secondary has 6 phase bus (WYE and DELTA) connected to the rectifier. The bus support for the secondary used to get so hot that it glowed red. Several ground wires have been installed to this bus to remove this energy. Also, prior to attaching these ground wires, arcs would periodically jump from the transformer tank to building steel (about 8 inches apart). While the arcs have no longer been seen and the steel support does not glow red, the support is still over 300 degrees F.
This is a 230VDC, 80,000A rectifier. I know there has to be some serious magnetic fields (I've seen metal pins stand up), but could it create hundreds of amps?
Could it be ground loops or leakage current?
The transformer and rectifier have been tested and no problems found. With the transformer secondary being 230VAC, 6 phases and 23MW, I could see that 300A on ground wires would not be much compared to the phase currents
Any ideas would be appreciated

 

[Skogsgurra]

With 80 kA DC, there is usually quite a lot of 600 (or, if on a 60 Hz grid, 720 Hz) ripple current. Ten - twenty percent ripple current is not unheard of if your load is inductive and a lot more if you have electrolysis or electrophoresis with minimum (or no) DC reactors.

Such a high ripple current at 600 or 720 Hz easily induces lots os circulating currents in nearby conductors. Your PE/ground wire is one of the most likely conductors to be nearby. So, it is a natural thing that you get high induced currents.

Balancing the DC conductors in order to minimize stray flux is usually not a practical way of solving the problem. Introducing some inductivity in the ground loop has helped, but you will then get potential differences between parts of your plant and that may do nasty things to your instrumentation and control circuits.

Going the opposite way; increasing the ground wire area to accommodate the high current is also possible. But then you have to make sure that all parts och the circuit can take the high current without getting hot.

A DC reactor is often the best solution. Not cheap, but may be good for your process as well. Especially if your plant is for electrophoresis.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ScottyUK]

At that current I am surprised that the secondary conductors are not enclosed in an isolated phase bus just to control the external magnetic field. Possibly an expensive solution to retro-fit, but probably the way it should have been built in the first place.

Look for any closed loops of steel enclosing the conductors. Acceptable loops would enclose all six phases or all three phases of a group. Any other loops will get hot. Don't forget that a buried earthing mat may form part of the enclosing loop.


----------------------------------

If we learn from our mistakes I'm getting a great education!