Leakage Currents

[mica]

This query relates to the October 98 article in ST on measurement of leakage currents ('Stop that Rot')

Martin Cornfoot, the author, explained how to make leakage current and resistance meanssurements the ship's system isolated.

Values of resistance so measured were listed as follows:
10,000 Ohms - High Resistance - no problem
5,000 Ohms - Slight leakage, but not serious
1,000 Ohms - Significant leak, should be identified and eliminated
500 Ohms or less - serious and potentially catastrophic leak

I carried out resistance measurements on my own boat with a high impedence digital multirange test meter and the results were as follows:

With main switches off and ancilliary connections, Adverc sensor lead and shorepower leads, disconnected, measured resistance is 17,000 Ohms, so no problem. With the main fuse (100 amp) removed the resistance is 970,000 Ohms - even better. The problem arises when I measure the leakage current under these circuit conditions. The equivalent figures are 1) 8.5 milliamps and 2) 3.5 milliamps.
Now, applying Ohms Law which as every schoolboy and most sailors will know, relates to resistance by the formula E (voltage) = I (current) x R (resistance), the figures do not stack up.

In case 1) resistance 17Kohms is equivalent to a current of 12 volts divided by 17,000 ohms = 0.7 milliamps. The actual measured current is however much higher than this. The same applies to the second measurement. 12 volts divided by 970Kohms = 0.01mA.

To work it the other way round : 12 volts divided by 8.5mA = 1400 ohms, a worrying lower resistance figure. Even with the main fuse removed, the resistance by Ohms Law (12 volts/3.5mA) is about 3400 Ohms, not 970Kohms.

Which set of results should I believe, and why is there this discrepancy between measured current and measured resistance?

Can anyone explain or tell me where I am going wrong.

MICA

 

[HaraldS]

Your instrument measues with less than 1V. Any non-linear behaviour would cause that difference. You could use an adjustable power supply to slowly crank up the volatge and check the current flow at different levels.
My guess would be soon around 1V and upwards you might in fact see linear behaviour.
In such case it can some instrumentation like sense wires to regulators and chargers that you might find starting to draw some small current. You might have to disconnect wires to find the source. If it is one of those, then it is probably no problem.
Unfortunately electrolytical current flow isn't linaer either. Depending on the metal combination, you might have to overcome a certain voltage before current starts to flow. Not sure where your main switch sits, but if there is virtually nothing connected when it's open, you might indeed have an electrolytical problem.
Disconnecting things is probably the way to find out. This can be tedious. Good luck.

 

[Guest]

do you have any problems in practise. are your batteries going flat? prop shaft eroding? etc

put it another way, when did you last see a mag article along the lines of "dont bother...." If we followed everything they said, we'd have boats down to the gunwales with safety gear, never leaving harbour.

am not suggesting that you totally ignore the article, but that you stand back and ask yourself if you had a problem before you read the article, and if you have a problem in practise.

 

[LadyInBed]

I agree with Howard, if it aint broke etc.
But, as an academic exercise it is quite (maybe just a little bit) interesting.
First, your resistance measurements. Between what points are you taking them?
I would go about it the other way around. Disconnect the earth bond to the anode and measure the current down that path. Then you will know that paths resistance.
You will probably (certainly) have a path via your prop shaft, so do the same again.
Then, after reconnecting all bonds, disconnect your neg battery terminal (assuming that you have neg ‘ground’) and see what current is flowing. If all your other currents add up to this one, you’ve cracked it, if they do not, then there is some other leakage and its time to put your thinking cap on, but I would put any ‘spare’ current down to ‘earth loops’ then go to bed and RIP.

 

[VMALLOWS]

The slightest potential difference will make any resistance reading completely invalid. You must completely isolate the item you want to measure with a normal multimeter. ( I assume we're not getting into wheatstone bridges etc). Just try reversing the leads when you're measuring.... reading should stay the same unless you have a diode or some potential difference (an anode in the sea, for example).

In theory current measurements on a DVM (digital meter) should be OK, but I still prefer to trust my analogue AVO (or Simpson) in the mA range on the boat.