How Much Ground Fault Is Too Much? 1

[mellomain]

I have an ungrounded 125VDC diesel generator start system that indicates a positive ground via (1) ground detection light illumination on the battery charger and (2)voltage to ground measurement. (+) to grd reads 75.4VDC while (-) to grd reads 56.7VDC.

The criticality of this level of grd on the system is often second-guessed (some want to let it go because it is too small to find or too time consuming to find).

How much of a ground is too much?

Thanks in advance for your response!!

 

[dpc]

Why do you think a higher voltage reading from + to ground indicate a positive ground? If the positive side of your battery was tie directly to ground, what would the voltage reading be?

 

[waross]

You may have a low resistance ground or you may have a fairly high resistance ground. Try a 120 volt, low wattage (15 watts or less) incandescent lamp from each terminal to ground.
If the lamp lights you probably have a low resistance ground.
If you can't get the lamp to light, you have a high resistance ground.
Either can be an indication of a failing component. A very high resistance ground path may be an indication of surface contamination which will respond to a good cleaning.
respectfully

 

[ScottyUK]

From the reasonably balanced voltages, you possibly have two ground paths on the two poles, thus creating a potential divider.

Waross' method is quite effective. A variation on the technique to prove a system is ground free is to connect a multimeter between one pole and ground. If there is very high resistance to ground the voltage measured by the multimeter should decay as system capacitance disharges through the multimeter. The effect should be similar on both poles. An old analogue multimeter like an AVO 8 is great for this because it has a relatively low input impedance compared to the modern digital types.

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

I don't suffer from insanity. I enjoy it...

 

[RRaghunath]

Mellomain,

Your 125V DC system seems healthy, to me. The system is ungrounded - true but still can show voltage with reference to earth. It happens this way - In ungrounded DC systems, A high value resistance is connected between (+) & (-) poles and the centre point of the resistor is connected to earth through the earth fault detection relay. Thus, strictly speaking, the system is not ungrounded.

The voltage to earth of each of the poles during healthy conditions, is 125/2 = 62.5V. In case of grounding of (+) pole, for example, the (+) to earth voltage would read zero and that of (-) to earth 125V.

In your case, the slight difference in measured volts could suggest minor leakage due to moisture or any other in the terminal box etc. involving the (-) pole (as it shows lesser voltage to earth).

Trust the above is helpful.

 

[waross]

I should have read your initial post more carefully.

(1) ground detection light illumination on the battery charger

This would indicate to me that the ground leakage is in excess of the level chosen by the original designer.
respectfully

 

[ScottyUK]

Yet the balanced voltages relative to ground would indicate that there is a similar level of current leakage from the other pole. If one or both faults becomes more solid, things could get exciting.

Or the GF detection circuit is kaput.


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

I don't suffer from insanity. I enjoy it...