Galvanic Isolators. Isolation Transformers. SMPSU. Anodes.

[CaptainBob]

No response from the engineers yet. But I tried something else.

I removed one connection from the GI and put my ammeter in-line in the circuit, plugged everything back in and checked again for AC or DC currents in the earth line. Both read 0.000A, not even a milli-amp flowing.

I'm wondering this...

If the GI is working correctly then unless a voltage across the GI of over 1.2V or therabouts exists, it will not conduct.

As my voltage is 0.76, even though this is perhaps the RMS of the peak voltage - that still means it's peaking at only just over 1V and so conduction is not occurring.

Further, as the earth in my boat has AC currents nearby, is it not possible that all I'm measuring is an induced voltage in the boat side earth wire, which is floating there almost like a static charge because of the GI stopping it being locked down to earth?

 

[Billjratt]

It's the current that matters - or lack of it!
If you don't have a current flowing it won't gobble up anodes or skin fittings- and you say you don't, but something is still nibbling at your anodes.
It's not neccessarily your load/usage that causes the problem, as a fault on a neighbouring boat can inject current into the water, which your correctly earthed boat provides a safe path for, but sacrifices in the process.
You can always add to your anode armoury without lifting out, by dangling one/some over the side, connected to the same place as the existing ones. I did that as a matter of course when leaving the boat in a marina.
I agree with your views on the voltage measured across the isolator, <1.2V is in my comfort zone.

 

[VicS]

I dont disagree violently with Bill J Rat but the fact that the volts disappear when you switch off your shorepower main switch indicates that the source of the reading is your own installation rather than elsewhere. (The earth connection is not broken when you just switch off the main switch)

The lack of any measurable current is good, at least nothing is eating at your anodes via the shorepower connection .... at the moment. BUT you need to have good margin between the AC reading you observe and the volts at which the GI will conduct because any galvanic source there may be will add to it. OK it'll only conduct on parts of alternate half cycles so to begin with the effect is small. (That's my understanding of the situation from the info I've read anyway) 1.2 volts is not so comfortable if you add another 0.4 or more to it from galvanic sources.

To put things into perspective I calculated that a current of 1mA continuously for a year will remove about 11g of zinc from an anode. So we are looking at measurable currents to cause problems not immeasurable fractions of milliamps. Please check my calculations though. Bit rusty on Faraday's constant and all that stuff.

Despite all of this got to avoid fixing something that isn't broken I think

 

[CaptainBob]

Thanks for your help Vic - again /forums/images/graemlins/smile.gif