Earth faults

I have installed an identical air conditioning system on two aluminium vessels.
The system consists of a combination of marine AC parts and automotive parts.

The compressor is an engine driven unit with a 24volt supply for the clutch but the return is via the compressor body therefor the unit has been "insulated" from the hull with nylon washers and spacers and the negitive is attached to the body of the compressor.

The vessels are fitted with isolation monitors set to 25k ohms, one vessel shows no fault but the other vessel shows a fault when the compressor is running.
We have established there is current leaking to earth from the compressor.
The boats are built to Lloyds standards, my question is, is there a standard which gives a value for leakage current which is acceptable?

 

My guess is that it will be in terms of millamps per square cm and will change with the hull material and the voltage. Doesn't nylon absorb water and then conduct?

 

Any difference between between the two as far as transmission and shafting? Anyway...with the power off and lead to the compressor detached, what does the resistance between ground measure for each compressor body and lead? You need to isolate the cabeling from the compressor to determine the leak path.

Then there is the flip side of the coin,.... has one of the isolation monitors failed.

 

The readings that have been taken indicate that with the cable disconnected from the compressor one boat reads 4.5 M ohms and the other 2.5 M ohms and iit is a stable reading.
When the compressors are running the reading is unstable an a reading cannot be taken.
I am intrested in what the prev post said about nylon absorbing water can anyone confirm this? could this account for the different reading?

 

Its not clear. You are only trying to isolate the compressor clutch ? Without the clutch you have no continuity to earth ?

 

For an aluminum hull, any current is probably too much if the flow is away from it. What is the voltage difference and the polarity of the charge? Is the whole engine isolated electrically from the hull? If it is, then the A/C should not be any different from the rest of the components. If the engine is not electrically isolated from the rest of the hull, then the A/C shouldn't be either.

 

According to this, moist nylon is terrible in terms of electrical conductivity. Replace it with some other plastic.

www.dtic.mil/dtic/tr/fulltext/u2/732378.pdf

 

>2 M ohms for the cable or the compressor housing? You need to check both to ensure the cable is not leaking. Often too tight a cable banding will cause a high ohm jacket leak.

And jonr is correct, nylon absorbs water (~3-6% by weight and an increase in volume by ~0.5-1%) and reduces its resistance, especially if it is sea water. FWIW, nylon is ~1000 times less effective than HDPE for electrical insulation and 10^6 times less effective than some of the dedicated plastic insulators.

http://www.professionalplastics.com/professionalplastics/ElectricalPropertiesofPlastics.pdf

 

http://www.professionalplastics.com/...ofPlastics.pdf

wow thats a pretty neat site !! never knew there was so much to know about plastics !!

 

If direction of current flow is an issue, you can then reverse polarity and the coil will still work fine.