GE turbine MARK V control


Thread Starter


We had three instances when all the three MARK V processors of GE Frame V turbine had rebooted during lightning. It was observed that lightining struck one of the CVTs in the 132 KV yard adjacent to the captive power plant and at the same time all the processors rebooted, leading to the tripping of the turbine.

The earthing of the MARK V is connected to 2 separate earth pits which are nearer to the yards. Also at the same time other Frame V machine of similar capacity and with a similar earthing status (except that they are having separate earth pits) did not have any effect.
Grounding systems are not created equal, even on the same site. Soils can vary even on the same site. Moisture and soil content can greatly affect ground grids even on the same site.

Most ground grids are not very well constructed, and some are "inadvertently" damaged during construction and never properly repaired.

Have you had your ground systems tested lately? You might be (un)pleasantly surprised; it can be an eye-opening experience.

It takes accurate equipment, proper planning, and requires experienced analysis to accurately decipher the results and plan any remedial action (which can be very costly and which are never usually implemented).
I am responding with very little advice in hope for a lot of education.

The Mark V is usually powered by DC voltage and it's own set of batteries and grounded via a capacitor. Computerized systems usually require "special" grounding. How is your Mark V powered and grounded? Did C core reboot? If not, any process and diagnostic alarms would be appreciated.

I would look to my Mark V manuals and investigate to ascertain if the Mark V is properly grounded.

Lighting, however, can cause problems that are impossible to troubleshoot because it is the nature of lighting to do pretty much anything it likes.
Do you run Arcnet between buildings? If so, is it fiber or copper? If it is copper, lightning can definitely cause this kind of problem if it strikes near the network cable. It can even take out the network. Normally, Arcnet runs that go between buildings are fiber for this reason.

The Mark V does indeed run on 125 VDC. Some people have added DC-to-AC converters called <DACA>s, which are usually powered by AC from inverters, so that means that AC is converted to DC to be inverted back to AC to be converted to DC. Makes a lot of sense to me (not). (NOTE: Many steam turbine control panels historically ran on AC, and when the Mark V is used for steam turbine control and there is no 125 VDC battery available <DACA>s are used to power the Mark V.)

There is a capacitor in parallel with the ground lead.

This is a huge controversy: whether or not to ground industrial control systems separately from "protective" ground, or earth. Sometime the separate control system ground is referred to as "functional earth." The Mark V was not *designed* for such a grounding system, though it has been modified to make it compatible. In my estimation, this is really difficult for any control system unless all field devices are also capable of being separately grounded, and we've heard from at least one site where they were having problems "frying" components if the instrument cases were gounded. I've been to several sites where there were several volts differential between the functional earth and the protective, and this was proven to cause problems.

Lightning is a problem for many control systems be they electronic, digital, or electromechanical. As some guy says, when copper coaxial cable was used to connect Mark Vs to operator interfaces in remote control rooms in lightning-prone locations there has been a history of issues, particularly when the coaxial cable was run in overhead cable trays.