lw395
Well-known member
The objection to fig 2 is surely that it requires the anodes etc to be isolated from the boat 'earth' (which comes out of the consumer unit).
So why not just isolate them completely?
DIY Galvanic isolator.
- Thread starter Vara
- Start date
The objection to fig 2 is surely that it requires the anodes etc to be isolated from the boat 'earth' (which comes out of the consumer unit).
So why not just isolate them completely?
misterg
Active member
When afloat, you create an earth that's relative to the boat. see the following diagram from This IET 'Wiring Matters' document
When ashore, you either use an earth rod instead of the 'metallic points in contact with water', or link the 'not connected' earth in the shore power lead to the boat earthing system.
No GI needed. This *is* the was to install shore-power to the wiring regulations and avoid galvanic corrosion due to voltages on the shore power earth.
Still not sure where you're coming from with the fuses (doesn't matter), but (IMHO) running a double insulated charger, effectively on an extension lead has a lot going for it in terms of safety and freedom from galvanic action provided there's no connection from the DC side to the earth connector on the plug.
Andy
halcyon
Well-known member
In the US indication is required for non fail safe GI's, the new types do not need it.
Back in the old days they were in the bonding circuit, after all that is what you protecting.
Brian
Brian
nigelmercier
RIP
- Joined
- 20 Jun 2007
- Messages
- 16,234
- Location
- Live in Kent, boat in Canary Islands
... that it effectively bypasses the GI!
VicS
Well-known member
My reason for suggesting fig 2 is that faults on the AC installation would bypass the GI and therefore it is less likely to be blown by high currents.
the boats own ground, DC negative, anodes etc are still connected to the shorepower earth through it so an Ac short to that system would still be earthed by the shorepower earth conductor.
It still blocks galvanic currents that could cause rapid anode loss
Last edited:
JamieAngus
New member
Everybody, please note, ****Safety Issue****
There is a hidden safety problem when using a bridge rectifier. These are made using hybrid and not monolithic technology.
What this means is that the individual diodes in the bridge are not matched and therefore should not be used in parallel.
Why?
Because, unlike resistors, paralleling up 2 diodes does NOT divide the current equally between them. Instead the current mostly flows through the diode with the lowest voltage drop, which then gets more current as it heats up as this causes its voltage drop to reduce (by about 2mv per degree C). This is known as "Current Hogging" or "thermal runaway".
Therefore, if you want to make a galvanic isolator using a diode bridge then each individual diode must be capable of taking the FULL fault current. Only then will your GI be safe, and in this case the second parallel arm will provide some useful redundancy in the case of failure.
Be safe, and do not try to parallel diodes for extra current, even if it is cheaper.
Regards,
Jamie
dial-a-monkey
New member
I have wondered in the past why GIs are installed as they are.
If they were fitted not in the incoming earth conductor but in the bonding connection to the DC negative etc there would be less likelihood of them being subjected to these big fault currents. Even if they were and they failed they would not disable the safety earth connection, merely the bonding to the DC negative .. that in itself something that many people wont accept anyway. I think the diagram below illustrates what I am asking about. Fig 1 shows a normal installation, Fig 2 show the GI just in the bonding connection to the DC negative etc
I'm wondering if in the time since this thread was started has anyone discovered/realized a/any reason why the Galvanic Isolator is not in the AC to DC ground circuit as VicS has suggested - to me this seems like the most obvious place to put it?
Lon nan Gruagach
Active member
I have wondered in the past why GIs are installed as they are.
If they were fitted not in the incoming earth conductor but in the bonding connection to the DC negative etc there would be less likelihood of them being subjected to these big fault currents. Even if they were and they failed they would not disable the safety earth connection, merely the bonding to the DC negative .. that in itself something that many people wont accept anyway. I think the diagram below illustrates what I am asking about. Fig 1 shows a normal installation, Fig 2 show the GI just in the bonding connection to the DC negative etc
Can you explain what the objection to Fig 2 is .
I am a tad surprised that Vic didnt know this, maybe in the last 10years he ha learnt but forgot this rather dead thread...
If (and you wouldnt know about it) your onboard mains earth manages to connect to the 12V -ve your GI will be bypassed and your boat bits will rot.
How likely is that? Just think how much equipment has 12V and mains...
Inverter
Battery charger
Generator
Dual power fridge, TV,.....
All could develop a wiring fault to cause problems, a fault for which there is no commonly installed detection device. New kit would need fully testing too.
RichardS
N/A
But, surely, if your onboard mains earth wasn't connected to 12V -ve you would not have a GI fitted anyway? Although if there was no installed connection then I guess you're suggesting that a fault could short the mains earth to the 12V -ve. I've no idea how likely that is.
Richard
Last edited:
dial-a-monkey
New member
Thanks Lon,
If I understand this statement "your GI will be bypassed and your boat bits will rot" correctly - you are saying that
(1) IF a short occurs between the AC earth wire and the Negative/DC ground in a ?battery charger? for example it will render the GI ineffective IF placed on the DC ground side of the equipment. - that makes sense..and at that point your anodes will soak up other peoples laziness - which seems a common default.
But If I understand it correctly?
(2) The GI will still be there and still short fault currents to ground IF the - redundant - shore power AC ground cable is damaged? for those 10mS until the RCD trips or 30ms for the thermal.
If I understand this statement "your GI will be bypassed and your boat bits will rot" correctly - you are saying that
(1) IF a short occurs between the AC earth wire and the Negative/DC ground in a ?battery charger? for example it will render the GI ineffective IF placed on the DC ground side of the equipment. - that makes sense..and at that point your anodes will soak up other peoples laziness - which seems a common default.
But If I understand it correctly?
(2) The GI will still be there and still short fault currents to ground IF the - redundant - shore power AC ground cable is damaged? for those 10mS until the RCD trips or 30ms for the thermal.
Last edited:
Lon nan Gruagach
Active member
If it was a house it would have to be. I think it ought to be, apparently some think otherwise.
halcyon
Well-known member
When we first had them they were in the bonding circuit. We first case across this in 1990 with Sealine, lost a prop, we finally found that mains ground on the pontoon were the boat was, had a 2 volt drop between it and mains earth spike at marina. The pontoon pile was in sea bed as a earth spike 2 volt lower, hence prop at 2 volt higher than piling, thus corrosion.
Also had leakage due to a screw going though a bulkhead that attached battery charger, case at ains earth, screw shorting on 12 v dc cable, again fault on boat. The GI was then fitted to protect low volt leakage, up to around 2 volt, but allow allow high voltage leakage for safety, do understand the new system my self.
Brian
Lon nan Gruagach
Active member
Yes.
If you have a defective Earth wire in your lead then a L or N to E fault would (if possible) cause all the fault current to flow through the water. Hopefully there will be a RCD at the pontoon.
If your shore power earth is OK then some (if possible) of the fault current will flow through the water .
If wired as per Vic's drawing #1 then none of the fault current will flow though the GI
If wired as per Vic's drawing #2 then the part of the current that flows thought the water will also flow through the GI
I say "if possible" since it depends if the anodes and skin fittings are bonded to the 12v -ve. again, if it were a house then it should/must be since all exposed metal service connections must be bonded. Some consider that since its not a house (plastic hulls are pretty good insulation), combined with the danger of fault currents in water where there might be swimmers, that bonding a protective earth to the skin fittings/anodes is a bad idea.
halcyon
Well-known member
What is the danger to the swimmer ? other than if he has one foot in water and one on land, bit like a bird sitting on a mains overhead power distribution cable.
Just a passing thought.
Brian
dial-a-monkey
New member
I guess any fault current in the water would be arrested pretty quickly by the RCD - i think my tested at 8ms.
dial-a-monkey
New member
It seems most GI's only have a voltage drop of 1.2V to 1.4 Volts, which sadly would not have saved the "2 volt 1990 Sealine problem" that Halcyon - Brian has mentioned.
Perhaps I need to read over the replys again: but from my understanding of the posts it still seems ....
(1) A better idea to put the GI in the AC - DC bonding circuit? due to less risk to life than a possible failure ( from a complete short @ 4000 Amp surge) of the GI in the main primary AC ground of the boat shore-power to marina AC ground circuit?
(2) The secondary lower safety consequence risk mentioned by Lon where an AC ground- DC negative short will rendering the GI ineffective - but fault current will still be passed to ground - either through one or the other AC or DC grounds ..Hmmm.
Perhaps I need to read over the replys again: but from my understanding of the posts it still seems ....
(1) A better idea to put the GI in the AC - DC bonding circuit? due to less risk to life than a possible failure ( from a complete short @ 4000 Amp surge) of the GI in the main primary AC ground of the boat shore-power to marina AC ground circuit?
(2) The secondary lower safety consequence risk mentioned by Lon where an AC ground- DC negative short will rendering the GI ineffective - but fault current will still be passed to ground - either through one or the other AC or DC grounds ..Hmmm.
Last edited:
Lon nan Gruagach
Active member
Electric shock drowning.
If the circuit has faulty RCD(s) or even the fault current through a functioning RCD (believe me, it flippin hurts) can be plenty to cause drowning if not direct electrocution.
halcyon
Well-known member
But if swimming you do not have a circuit, the swimmer needs a earth to produce a shock, as said a bird can sit on a high voltage mains power line, as long as he does not touch another.
Brian
VicS
Well-known member
No I have not forgotten about this
The suggestion was originally made "for discussion" It was discussed 9 years ago and IIRC the only valid objection related to metal hulled craft. It may be in this thread or it may have been in another.
The point now raised about an AC system PE short to the DC negative bypassing the GI seems to be another valid objection.
For convenience below is the image illustrating my suggestion ( fig 2).
However today is Wednesday and this mornings priorities are a shit, shave and shampoo, to do any essential chores and be out by 1130 in order to get to the Age UK lunch club in time for lunch.
lw395
Well-known member
The danger to the swimmer is that if the water is not very salty, it is not very conductive. Therefore you can have sufficient electric field to kill or incapacitate you, because your various extremities can be at different voltages.
It is a particular problem in fresh water.
Say the boat is supplied by an isolating transformer. The metal parts of the boat can drift towards half the supply voltage if there is no bonding to shore ground at all. The prop for example, then has an electric field around it.
The impedance might be quite high, but enough 'volts to jolt', which can interfere with your ability to swim.
A separate issue is if any metal part of boat or pontoon is at a few volts different from the water or other metal, with a low impedance, you can get enough current to kill.