Fault 125VDC ground MarkVie

Good day,
I am new to a thermoelectric plant where they operate with General electric gas turbines. We use the MarkVIe, we have a ground fault at 125VDC. Well, the only way I have found to solve this fault is to disconnect the power supply from the JPD * one by one ... is there any other way to solve this fault?
 
abed...
Search this Forum's archive for VKGupta's Aug 19, 2015, query, and BLANK's reply. Blank is the name of the only person the forum's security system doesn't allow me to mention !!!
Regards, Phil Corso
 
aned77,

There are other ways, BUT they take too long to describe and if you’re not familiar with how the I/O is connected to the Mark* it’s very confusing.

The thing to remember is that Phil Corso could have provided the link to the thread, but he likes to make people WORK for the information.

The link is:

https://control.com/forums/threads/125vdc-both-legs-having-difference.44231/

UNFORTUNATELY, the codes in the pre-new Control.com system don’t translate well to the new Control.com system, so the “drawing” of the voltage divider network isn’t going to display properly.

But, if the rainy season has started at your site it’s highly likely the ground is caused by water in a junction box from a leak in the conduit to/from the junction box. So, something outside the enclosures (Accessory-or Turbine- or Load- or Collector Compartments.AND, it could be a device or circuit NOT directly connected to the Mark*. Because the Mark* monitors ALL the circuits and devices which are powered by the 125 VDC battery—NOT just devices connected to the Mark*.

It’s difficult to describe without drawings, but it’s not impossible to troubleshoot. The good news is you will get better at it—with time and experience.

Please write back to let us know how and where you found and resolved the ground!!! Best of luck!
 
Good day,
I am new to a thermoelectric plant where they operate with General electric gas turbines. We use the MarkVIe, we have a ground fault at 125VDC. Well, the only way I have found to solve this fault is to disconnect the power supply from the JPD * one by one ... is there any other way to solve this fault?
Nope. It takes a systematic approach and time. My experience is such that each wire going to its respective field device will have to be lifted or isolated one by one until the ground fault fault clears. Then its a matter of legwork to locate the device and start opening the junction boxes and covers to access the grounded switch or other device. A "megger" or insulation resistance instrument is essential in troubleshooting ground faults. Some manufacturers used slide links to open the circuit for troubleshooting that eliminates the need to remove terminal screws and lift wires. In my many years working in power plants, chasing DC control circuit grounds, I found they come in many forms. Water intrusion into J-Boxes and housings, aging and heat damaged wire, sloppy work where a wire gets pinched under a box cover and it takes a while for the insulation to displace and expose the conductor to ground, and conductive paths that form on or under terminal strips. Much of my work was in old and aging plants built in the 60's. In one case the cause is an overly sensitive ground detection circuit in the power supply, that would fault if you breathed on it.

Good Luck.
 
A systematic and logical approach to troubleshooting is always the best. As the original poster said, the design of the Mark VIe is such that groups of devices can be somewhat isolated by unplugging power supply cables to different I/O cards.

For example, most of the discrete (contact) inputs are powered from two or three parallel connections to the JPDx card, and each connection is connected to multiple input cards in a serial (“daisy chain”) method. By working from the end of the serial loop back towards the JCPx card quite often the problem can be traced to a group of inputs to one discrete input card, then from there it’s possible to start lifting wires on that input card to find the root cause. A similar thing is done with solenoid output circuits fed from one or Mose connections to serial loops of solenoid output cards. By beginning from the last card in the loop it’s often possible to isolate the problem to a group of solenoid outputs connected to a particular card, and then start lifting wires from that card until the root cause is identified.

But, in my experience it’s usually water ingress into a junction box, or a melted wire insulation from an abnormally high temperature around a conduit. Often when the root cause is identified it’s found to have been caused during a recent maintenance outage, or—again—a severe rainstorm or even water or condensate dripping on a broken conduit connection or improperly closed junction box. That’s why I usually ask what happened just before the problem started—because there’s usually some trigger for the problem such as rain, or a maintenance outage or an excessively high temperature in a compartment.

A little thought and analysis—amd some luck—can be very helpful to solving a problem like this. A colleague used to say, “The harder I work—the luckier I am. And, I’d rather be lucky than good!” Meaning, the more effort one puts into understanding the situation and configuration and what has or might have changed, the more likely it is that the problem will be resolved sooner and without a lot of unnecessary stress and physical work.
 
abed77...
Since you asked how to fix the fault you must first find it. Following is the method I used, eliminating the dangerous method of disconnection.

Using a DC clamp-on ammeter measure the current in each (+) (-) wire/lead connected to each terminal block !

Of course it would be better if you knew what the "expected" or "normal" current is ! So, in every plant or facility I worked in or visited, I would insist a Current-Table be provided for every AC and DC control panel !

Regards, Phil Corso
 
Phil Corso,

Discrete inputs to a Mark* normally draw micro amperes of DC current. Solenoids can draw 2 amps or more on inrush, but holding current is usually less than one amp.
 
I had an experience where a ground fault wasn't found despite meticulously isolating all of the inputs. It turned out that our Reuter-Stokes flame scanners run on 125VDC on another circuit. Ours had fuse so isolating as a group was easy. Once we saw the ground fault go away, we went scanner by scanner and ended up replacing one of the scanner cables.
 
Top