Floating ground and ground faults - Theory questions

Discussion in 'OnBoard Electronics & Controls' started by DarylH, Oct 12, 2012.

  1. DarylH
    Joined: Oct 2012
    Posts: 2
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Canada

    DarylH New Member

    Hello,

    I've run into some issues with DC ground faults on an Aluminum hull jet boat and I'm a bit above my head when it comes to diagnosing ground faults.
    The boat design is such that the hull is to be completely isolated from the electrical system.

    I'm an Electronics Technologist working with a group responsible for Shore to Sea telecommunications, marine electronics, and data communications etc. I'm not a marine electrician and I'm certainly not a physicist that knows everything about electricity.

    We're occasionally called-out to this vessel when the crew notice a ground fault alarm. We're very careful with the ships electronics, keeping all chassis (typically grounded to the DC negative) isolated from the hull, wiring to do with ships electronic is done to very high standard etc. We've fixed faults that are usually related to moisture. For example the digital cell antenna lead, where the shield and connector are tied to the chassis of the transceiver (and DC negative) would trip the GND Fault alarm when sufficiently wet, as the moisture must pass enough current from the connector to the mast post (bonded to the hull) to trip the alarm. The solution in this case was using plastic mounts and insolators to keep the connector at a good distance from the post, and of course making extra precautions to wrap everything to keep things dry.

    While this alarm seems to trip when there is the slightest issue with the electronics, my suspicion is there are ground faults to do with engine electrical that has put this GND fault sensor close to it's threshold, and then a low current electronic equipment (latest was a wind sensor) with a high resistive connection to the hull (moisture) is enough to trip the alarm.

    The engine electrical system is not my responsibility, but it may be up to me to prove to the crew that there is an underlying problem and that they need hire an expert to diagnose.

    So, with all electronics off at the main breaker, when in the engine compartment, we looked at the voltage between the hull and the negative post on the 24v battery bank. We measured 15v. This doesn't seem right to me. In my opinion they have something quite wrong.
    Then I got to talking to a marine electrician, who told me that the isolated hull would sit at a voltage somewhere inbetween 0 and 24V and that ideally it should be 12.5 volts. He said a ground fault on the negative side would drag the hull voltage closer to 0 and that would be an indication of a fault.

    I don't agree. I figure the hull (if properly isolated) could not be referenced to the battery at all and that a good reading would be 0 volts. He says that if I measure 0 volts it would indicated the hull is grounded to the DC negative and indicate a fault. While I admit if the hull were grounded to the DC negative I would read 0 volts however I would also measure 0 volts if it were isolated, because there would be no circuit, and no current, and if no current, my meter could read no voltage.


    Can anyone clear this up for me? Also how best should I use my meter to prove there is fault. I could measure current between the DC positive and the hull, if there is current then the hull is finding a ground. That makes sense to me.

    Any insight on this would be a great help.
     
  2. jonr
    Joined: Sep 2008
    Posts: 721
    Likes: 11, Points: 0, Legacy Rep: 57
    Location: Great Lakes

    jonr Senior Member

    Ideally you would have 0V from the hull to either the positive or negative terminals of the battery (or anything else in the electrical system).

    When you are talking about very low currrents, it is easy to get induced voltages in anything made of metal. Yes, you can quantify this with a current measurement from the hull to any point in the electrical system. Also note that any of these things may be varying rapidly and may not show up if you are using DC measurements.
     
  3. Petros
    Joined: Oct 2007
    Posts: 2,934
    Likes: 149, Points: 63, Legacy Rep: 1593
    Location: Arlington, WA-USA

    Petros Senior Member

    I am no expert at marine wiring either, but my engineering experience would tell me if the hull is isolated you should not get voltage across it as you suspect.

    Can you measure continuity across the system to the hull using an ohm meter? If you can cut the electrical power and measure a closed circuit between between the hull and the electrical system than clearly the hull is not isolated. I would think that is a way to prove the there is a ground fault.
     
  4. erik818
    Joined: Feb 2007
    Posts: 237
    Likes: 21, Points: 0, Legacy Rep: 310
    Location: Sweden

    erik818 Senior Member

    Petros,
    I'm surprised if the power lines are truly floating from hull. What we would do at my company (I work with land vehicles so I'm extrapolating to boats) would be to make a floating electric system and then connect battery negative to the hull with a resistor at one point, typically at the engine. Any current through the resistor indicates a ground fault, so letting the ground fault sensor be the connection point between hull and power lines would be a smart move. It isn't necessary to connect hull to negative, you might as well connect it to the positive side or, by using two resistors, to any voltage between.

    The way to check such a system is to disconnect the grounding point and then measure the resistance between hull and power lines, often with a voltage according to specification applied.

    So, to understand what's happening you should start with understanding how the ground fault sensor works in your system and how/where the hull is connected to the power lines. I wouldn't be surprised if the problem is at the engine as you are guessing.

    Erik
     
  5. DarylH
    Joined: Oct 2012
    Posts: 2
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Canada

    DarylH New Member

    OK well the tips the marine electrician was giving me could make sense if the hull were tied to DC through a resisitor as erik is saying. Perhaps he had made that assumption. Good tips, thanks erik.

    Although I was told the vessel should have a completely isolated hull, I'll need to check my facts, but if it's true then its a different ball game.

    In regards to how a true isolated hull should measure I'm glad I haven't lost my mind.
    Thanks for the insight petros and jonr.

    Petros, continuity was what I first thought, and it would help if there was low resistance closed circuit but the fault isn't so simple. In the spot's I've checked, resistance is in the mega ohms with a small fluke meter and I wouldn't go near the hull with a megometer.
     
  6. Ike
    Joined: Apr 2006
    Posts: 2,704
    Likes: 496, Points: 83, Legacy Rep: 1669
    Location: Washington

    Ike Senior Member

    Well I am very familiar with marine electrical systems. The hull should be completely isolated from the DC system. There should be 0V on the hull. If you are reading voltage on the hull then you have leakage somewhere. This can result in severe stray current corrosion on metal parts, especially aluminum. If it is reading 15 volts then you have some voltage drop due to resistance. the resistance could be any number of things including a resistor, but I serious doubt that. I think you have a high resistance connection to the hull probably through some piece of equipment. Engines are always good suspects. But with all that electronics on board I would suspect someone grounded the case of an electronic device to the hull (maybe through a metal shelf or table) or the device has an internal ground connection to the case and the case is resting or touching metal.

    It's going to be a ***** to find.
     
  7. jonr
    Joined: Sep 2008
    Posts: 721
    Likes: 11, Points: 0, Legacy Rep: 57
    Location: Great Lakes

    jonr Senior Member

    If all your wires run back to one place, then you may be able to switch off circuits (power, neutral and ground) one by one to find roughly where the problem is.
     
  8. CDK
    Joined: Aug 2007
    Posts: 3,324
    Likes: 148, Points: 63, Legacy Rep: 1819
    Location: Adriatic sea

    CDK retired engineer

    The Fluke meter has too high an impedance to tell you the truth in any floating circuit.
    With an antique needle gauge (10 K-ohms/volt) you won't measure 15 V.
     
  9. jonr
    Joined: Sep 2008
    Posts: 721
    Likes: 11, Points: 0, Legacy Rep: 57
    Location: Great Lakes

    jonr Senior Member

    To word that a little differently, voltages at very low currents are inconsequential. And there might not be any solid material to find that is causing it (voltages can travel through the air).
     
    Last edited: Oct 15, 2012
  10. gonzo
    Joined: Aug 2002
    Posts: 17,094
    Likes: 1,857, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    Boat are systems that are, as you know with the antenna problem, wet. This causes electrical connections at random places. Usually you will find several electrical paths that add up to the threshold that trips the ground fault. Oftentimes, there are pieces of equipment with ground leaks that add up to the problem. It looks like an intermittent problem cause by accumulated leaks. You may need a systematic approach to troubleshooting each section for ground leaks.
     
  11. Bglad
    Joined: May 2010
    Posts: 175
    Likes: 5, Points: 18, Legacy Rep: 67
    Location: Jacksonville, Florida

    Bglad Senior Member

    Integral tanks connected with wire reinforced hose to engines could make a connection too.
     
  12. brianb00
    Joined: Sep 2012
    Posts: 24
    Likes: 1, Points: 0, Legacy Rep: 25
    Location: San Francisco

    brianb00 Junior Member

    What CDK said, only you can still use the fluke and take about 10K ohm resistance and place across the fluke leads when making the voltage measurement. This will lower the fluke impedance and discharge any odd voltages that would only be measureable at high impedance. If you still get significant voltages, like above a volt, you likely have some kind of low R path. You can try lowering the parallel resistance to 1K, then 100 ohms to determine the significance of a leakage path. If you are convinced you do have a leakage path one approach you might try is to connect a device for tracing wires in home walls. This injects a low level AC current in the audio range. Then a sniffer is used to follow the path of the AC current. You can likely follow this signal to the errant leaky component/path. These devices are sold at places like home depot.
    Good luck.
     
    1 person likes this.
  13. CDK
    Joined: Aug 2007
    Posts: 3,324
    Likes: 148, Points: 63, Legacy Rep: 1819
    Location: Adriatic sea

    CDK retired engineer

    This story once more illustrates how difficult it is to keep DC wiring isolated from a metal hull. The engine wiring may have one or more sensors with a low path to ground, electric motors with isolated leads may have carbon deposits or a noise suppressor does not only contain capacitors and coils but also a built in resistor.

    I like to know what kind of ground fault interrupter is triggered by such low currents.
     
  14. Frosty

    Frosty Previous Member

    It depends what engines are fitted too . Some are proper marine with 2 wire senders and 2 wire starters if not then the engine block will be 12v negative. All it needs from there is a wet and oily engine mounting and there you go.

    Not only that control cables can easily transfer leaks. I think I have something similar from induction from 240v wires lying next to engine control cables.
     

  15. powerabout
    Joined: Nov 2007
    Posts: 2,944
    Likes: 67, Points: 48, Legacy Rep: 719
    Location: Melbourne/Singapore/Italy

    powerabout Senior Member

    is all the dc equipment IMO approved as isolated earth?
     
Loading...
Similar Threads
  1. yodani
    Replies:
    3
    Views:
    1,855
  2. Frosty
    Replies:
    48
    Views:
    8,581
  3. sdowney717
    Replies:
    11
    Views:
    295
  4. Mick Rogers
    Replies:
    9
    Views:
    3,976
  5. sdowney717
    Replies:
    10
    Views:
    8,006
  6. fallguy
    Replies:
    8
    Views:
    2,114
  7. sdowney717
    Replies:
    12
    Views:
    5,059
  8. fireman_bob
    Replies:
    21
    Views:
    10,977
  9. cgoodwin
    Replies:
    38
    Views:
    11,302
  10. sdowney717
    Replies:
    18
    Views:
    8,424
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.