Earth return ?

[by'eck]

I know this is reaching unlikely circumstances but are your batteries bonded to the hull and is your exhaust welded to the hull , if so you have a return path from the starter motor case. You yourself mentioned the prop shaft.

 

I think Nick concluded that as long as the starter motor and alternator negatives are cabled directly to their batteries with the hull negative bond at the starter motor, the much preferred path will be via the cables. If for example the hull bond was closer to the battery negative, the chance of a parallel path via the hull would be greater. I know its splitting hairs but it does make sense.

 

There will always be secondary negative return paths even via salt water to outside of hull from raw water pump on sea going vessel, but as long as they are relatively high resistance to the preferred path they can be ignored.

[nicknorman]

I believe the engine "earth strap" is to avoid the current finding a return path that perhaps wouldn't be obvious ( such as those mentioned above ). I remember years ago it would be the choke cable on a car engine that would get hot. So I am now thinking I should remove the -ve from the alternator to the hull and run it to the battery -ve directly. Previous posts asking about the potential return path, what about DC bonding to the hull ? I am unsure how the mains earth, 12v -ve, and GI combine to connect to the hull.

But as I keep saying, there is no return path through the hull unless you have two connections, one from engine to hull (which you probably do have) and have, AND another one from the battery -ve to somewhere else in the hull, which you shouldn't have.

[ditchcrawler]

No, my battery -ve is definitely not bonded to the hull other than via the cable to the engine, so there is no circuit. The hull is connected to the engine, but one connection doesn't make a circuit!

I see. Ta

[Nickhlx]

Well, I understand, and agree with, your view, Nick...

 

(another) Nick

[MtB]

Why would current through the hull matter anyway? Sounds like red herring to me.

 

Surely electrolytic erosion relies on the electroplating effect where material migrates from the anode to the cathode in a circuit using an electrolyte, so we need to prevent the hull taking on the role of anode in such a circuit.

 

I don't quite see how having two negative connections to the hull might cause this. Or have I misunderstood the problem?

[Dave Clinton]

But as I keep saying, there is no return path through the hull unless you have two connections, one from engine to hull (which you probably do have) and have, AND another one from the battery -ve to somewhere else in the hull, which you shouldn't have.

Despite advice to have a 0V bond to the hull adjacent to the 240V PE as a safety measure recommended, amongst others, on the smartgauge site?

[Chewbacka]

The theory is that there will be a voltage difference caused by the current flow. So if you had say 0.1 ohm resistance in the Steel hull below the water line between the bows where the tunnel light is and the battery earth point at the stern then using ohms law you would have -

 

Assume a 35w tunnel light so current about 4 amps, using V=IR we get

V=4 * 0.1 = 0.4 V

 

A volt difference in water of more than about 200mV will encourage corrosion, and in this example we have 400mV. However they are a long way apart which will diminish the effect.

 

So in theory this is bad, in reality I am not so sure if only little currents, but with large currents running for long periods then it must be avoided. By the way, I have no idea what the resistance is from one end of the boat to the other, it may be higher than the 0.1 ohm I used above, which would make the problem much worse.

[Chewbacka]

Despite advice to have a 0V bond to the hull adjacent to the 240V PE as a safety measure recommended, amongst others, on the smartgauge site?

If I remember correctly then for a new boat with 240v ac power system it is mandatory that the battery negative is bonded to the hull - safety reasons. Therefore it is very important that there are no other bits of kit such as a horn or tunnel light using the hull as an earth return.

[MtB]

The theory is that there will be a voltage difference caused by the current flow. So if you had say 0.1 ohm resistance in the Steel hull below the water line between the bows where the tunnel light is and the battery earth point at the stern then using ohms law you would have -

 

Assume a 35w tunnel light so current about 4 amps, using V=IR we get

V=4 * 0.1 = 0.4 V

 

A volt difference in water of more than about 200mV will encourage corrosion, and in this example we have 400mV. However they are a long way apart which will diminish the effect.

 

So in theory this is bad, in reality I am not so sure if only little currents, but with large currents running for long periods then it must be avoided. By the way, I have no idea what the resistance is from one end of the boat to the other, it may be higher than the 0.1 ohm I used above, which would make the problem much worse.

 

 

Thanks for the explanation. This is the bit that diverges from my understanding.

 

How/why does a voltage difference between two parts of the hull encourage corrosion? (Notwithstanding the fact that 0.1 Ohms seems a rather high resistance for a cross sectional area of several square inches if steel!)

 

It can't be from electrolysis because this requires two different metals of different nobility (AIUI).

[Chewbacka]

It's the same thing, you can either generate a voltage difference by different metals in contact, the electrochemical series defines the voltage difference that will be created or you can have a single bit of metal and by forcing a current along it a voltage will appear between the 2 ends. The metal ions will then move from one region of steel to the other.

 

As an aside, if the metal is not uniform in it's composition then different regions of metal can be at slightly different potentials. You can sometimes see this on old concrete rebar where when the concrete is removed some areas are rusty and others shiny

[Chewbacka]

Another example - if you did chemistry at school you may have had 2 wires in copper sulphate solution and nothing much happens until you connect a battery and copper moves from the copper wire to the other wire. The applied voltage difference causing the metal loss from the copper wire. Though in this experiment we were more interested in the copper arriving on the other surface.

[MtB]

It's the same thing, you can either generate a voltage difference by different metals in contact, the electrochemical series defines the voltage difference that will be created or you can have a single bit of metal and by forcing a current along it a voltage will appear between the 2 ends. The metal ions will then move from one region of steel to the other.

 

 

I'm still not understanding. Two metals in contact will have no voltage difference surely. As I understand it from skool the two metals have to be separated by an electrolyte for a voltage difference to happen.

 

I understand that once a voltage difference is created along a length of steel a current can flow (provided there is a circuit), but I still don't understand how a current in that metal causes corrosion. I accept that corrosion happens, obviously but attributing that corrosion to electrolysis troubles me.

[Chewbacka]

When you join the 2 metals you need an electrolyte to join them for the metal ions and electrons to move about in. So steel and copper or copper and aluminium are fine joined together until they get wet and the circuit is complete. Then metal migration will happen. But you must have the electrolyte for the ions to be able to move in.

 

A battery is the other way around, the metal plates are in the electrolyte (acid in a lead acid battery) and nothing much happens until you join the metal plates together. Normally through a load like a light etc.

 

Added - as the battery discharges the anode will lose metal ions, however when we recharge the battery the voltage must be high enough to stop and reverse the natural flow of ions and electrons.

 

When we don't want metal migration to happen we call it corrosion, but when we do it's called battery discharge or electroplating etc.

Edited September 2, 2015 by Chewbacka

[Giant]

Let's put some numbers on this.

 

Take, say, a 10mm (0.01m) baseplate on a 20m boat, 2m wide.

 

Resistance R = ρ × L / A

 

where ρ is the resistivity (1.43×10⁻⁷ for carbon steel), L is the length (20), A is the area (0.01× 2).

 

R =1.43×10⁻⁷× (20 / (0.01× 2)) = 0.000143 Ω

 

To get a 200mV voltage drop from one end to the other you would need:

 

I = V / R = 0.2 / 0.000143 = 1398 AMPS!!!

 

Even if the baseplate is 3mm, you need 419 A down the entire length of the boat to create that sort of voltage difference.

 

And that's not even accounting for all the extra conductance through the hull sides and the cabin walls/roof.

Edited September 2, 2015 by Giant

[Chewbacka]

I wasn't trying to justify the rules, just giving an opinion as to why they came about, and the theory behind it all.

[MtB]

Let's put some numbers on this.

 

Take, say, a 10mm (0.01m) baseplate on a 20m boat, 2m wide.

 

Resistance R = ρ × L / A

 

where ρ is the resistivity (1.43×10⁻⁷ for carbon steel), L is the length (20), A is the area (0.01× 2).

 

R =1.43×10⁻⁷× (20 / (0.01× 2)) = 0.000143 Ω

 

To get a 200mV voltage drop from one end to the other you would need:

 

I = V / R = 0.2 / 0.000143 = 1398 AMPS!!!

 

Even if the baseplate is 3mm, you need 419 A down the entire length of the boat to create that sort of voltage difference.

 

And that's not even accounting for all the extra conductance through the hull sides and the cabin walls/roof.

 

 

 

Thanks for those numbers, they confirm my instinctive assessment that the volt drop along a hull carrying a bit of current would be so low as to be discountable.

 

This adds validity to my question "how does having multiple negative (earth) connections to a hull promote corrosion?" Chewbacka offered an answer suggesting once a 200mV drop along the hull occurs, corrosion will happen, but such a voltage has now been shown to be impossible under normal circumstances.

 

So, can anyone else say why multiple hull connections matter? Or explain how they might cause corrosion?

 

 

 

(Punctuation edit.)

Edited September 2, 2015 by Mike the Boilerman

[nicknorman]

Despite advice to have a 0V bond to the hull adjacent to the 240V PE as a safety measure recommended, amongst others, on the smartgauge site?

The 0v bond is recommended not as a safety measure in terms of electric shock protection, but so that if there is an inadvertent short of +12v to hull, it will blow the fuse. Gibbo et al explains why having the hull connected to +12v is a bad thing from a corrosion point of view. So I guess it depends on what you mean by a "safety measure".

 

But anyway, I am not advocating not having a 0v bond, I am just saying that you probably already have one (prop shaft, exhaust etc) and so if you want to add a dedicated 0v bond for "tidiness" then it should be along the same circuit path ie from the engine to the hull (so there will be no voltage difference around the circuit path, and hence no current) as opposed to from the battery -ve to the hull.

 

With our 175A alternator working hard we have a couple of hundred mV dropped in the -ve path and if I had a wire from battery -ve to hull some of that 175A would take that path through the hull, via say the exhaust, to the engine.

 

Now whether that hull current is actually a problem (viz debate between chewbaka and MtB) I wouldn't like to say, but when there are two ways to achieve the desired outcome, and one way does not involve any hull current, I would choose that way. Why not?

Thanks for those numbers, they confirm my instinctive assessment that the volt drop along a hull carrying a bit of current would be so low as to be discountable.

 

This adds validity to my question "how does having multiple negative (earth) connections to a hull promote corrosion?" Chewbacka offered an answer suggesting once a 200mV drop along the hull occurs, corrosion will happen, but such a voltage has now been shown to be impossible under normal circumstances.

 

So, can anyone else say why multiple hull connections matter? Or explain how they might cause corrosion?

 

 

 

(Punctuation edit.)

I can't offer any science but so far you have only contemplated the potential difference effects. Does the magnetic field so created have any bearing?

 

It does seem to be universally disallowed to use the hull for -ve return (the DC wiring ISO for example) so I think it is more likely that we haven't identified the corrosion mechanism, rather than it doesn't exist.

 

Edited September 2, 2015 by nicknorman

[by'eck]

Reasons for single ground point to hull for DC negative, ideally via engine block. Note this doesn't in any way remove the need for a dedicated negative cable return for DC devices, since using the hull as such will result in errosion of cable or hull if positive grounded! (some may remember issues with old postive grounded cars of the 50's & 60's)

1) Safety - any short from live AC to battery negative will go to hull/ground & trip RCD

2) Many engine ancillaries such as starter motor, alternator or senders rely on engine block being at DC negative. Even if they have a dedicated negative terminal its often connected internally to body of ancillary. Since block is bolted to or otherwise in contact with hull it is also at DC negative anyway.

3) Following on from 2), you know there will be no stray paths from DC negative to hull which may otherwise be through a thin cable or sensitive electronics.

4) Provides a common ground for stray currents induced into cable runs which may otherwise cause RF interference or hum on audio systems - tenuous point I agree.

5) A short from DC positive to hull without a ground strap will cause a current flow through any of possibly many stray paths. This may not blow the fuse or trip the breaker but could damage any item creating that stray path in a greater or lesser time.

6) Also following on from 2), many car electronics have their mounting chassis at DC negative, so a connection or short to hull cannot easily be avoided with such, although the path of least resistance would be via a dedicated return cable.

So to sum up, in an ideal situation the DC negative hull connection will never carry any current, but in the event of a fault or unforeseen situation, will provide a single safe path to ground, quickly tripping the circuit protection in the case of a DC positive short to hull. You will also be turning a tenuous situation into a known one and so create clear conditions for safety trips and fuses to work quicky and effectively. Stray current galvanic errosion on the hull is not an issue as long as its DC negative that's grounded, since any leakage will leave the hull as the most negative with any errosion plating hull with tiny amount of copper from cable.

Note that an AC ground point to hull is also required for the RCD safety device(s) to be effective. This should be close but separate to avoid the two straps becoming detached but remaining together and thus allowing the admittedly distant possibility of the DC system becoming live to AC mains.

Edited September 2, 2015 by by'eck

[steve7a3]

Sorry to open up an old thread but it seemed appropriate.

I have a Victron Phoenix Multiplus which appears to have no chassis earth connected, which I've been told is the probable cause of my indicated AC earth leakage, so I've been told I need to earth the Victron chassis to the hull. I'm assuming, after reading this thread, that it's likely that the battery negative will also be connected to the hull. If this is the case would it be reasonable to connect the Victron chassis to the battery negative or to the same connection (presumable on the engine block) that the battery negative is connected to or should I find a separate connection point on the hull?

 

 

[Tony Brooks]

1 hour ago, steve7a3 said:

Sorry to open up an old thread but it seemed appropriate.

I have a Victron Phoenix Multiplus which appears to have no chassis earth connected, which I've been told is the probable cause of my indicated AC earth leakage, so I've been told I need to earth the Victron chassis to the hull. I'm assuming, after reading this thread, that it's likely that the battery negative will also be connected to the hull. If this is the case would it be reasonable to connect the Victron chassis to the battery negative or to the same connection (presumable on the engine block) that the battery negative is connected to or should I find a separate connection point on the hull?

 

 

This topic seems to be about the 12 or 24 volt DC circuits, not mains and shorelines. I am going to askva  mod to move it into Equipment as a stand alone topic.

[WotEver]

1 hour ago, steve7a3 said:

... I've been told is the probable cause of my indicated AC earth leakage...

Indicated by what? Told by whom? 

It sounds like BS to me but we'll need loads more info yet. 

In the mean time, here's some reading on AC earthing: http://smartgauge.co.uk/earthing.html

[steve7a3]

I was told by Ed Watt from The Galvanic Isolator Company after I checked with him that I'd installed my new GI Status Monitor correctly as it was showing two lights.

His comment was:

"Many people have problems with earth leakage when using specifically Victron chargers. It's usually due to the charger not being adequately bonded to the boat's hull. You need to "Earth" the case of the charger to the boat's hull/ground using a short, thick cable, min 2.5mm cross sectional area or preferably greater.

 

In all the cases I have come across so far, this has resolved the problem."

 

I assume he knows what he's talking about.

[Tony Brooks]

3 hours ago, steve7a3 said:

Sorry to open up an old thread but it seemed appropriate.

I have a Victron Phoenix Multiplus which appears to have no chassis earth connected, which I've been told is the probable cause of my indicated AC earth leakage, so I've been told I need to earth the Victron chassis to the hull. I'm assuming, after reading this thread, that it's likely that the battery negative will also be connected to the hull. If this is the case would it be reasonable to connect the Victron chassis to the battery negative or to the same connection (presumable on the engine block) that the battery negative is connected to or should I find a separate connection point on the hull?

 

10 minutes ago, steve7a3 said:

I was told by Ed Watt from The Galvanic Isolator Company after I checked with him that I'd installed my new GI Status Monitor correctly as it was showing two lights.

His comment was:

"Many people have problems with earth leakage when using specifically Victron chargers. It's usually due to the charger not being adequately bonded to the boat's hull. You need to "Earth" the case of the charger to the boat's hull/ground using a short, thick cable, min 2.5mm cross sectional area or preferably greater.

 

In all the cases I have come across so far, this has resolved the problem."

 

I assume he knows what he's talking about.

We know chargers nowadays tend to be posh switched mode power supplies and we also know the interference suppression on such devices can create higher than normal voltages on the boat side of a GI.

The Multiplus is a charger inverter which I think should already be adequately "earthed" to the hull. I agree with WotEver more info needed.

I note than nowadays GIs often use two pairs of three diodes instead of two pairs of two diodes so if you "lights" are indicating a voltage on the earth line  the GI may still be doing its job, the question is what voltage that is and how much headroom you have before it is forced into conduction.

 

[smileypete]

38 minutes ago, steve7a3 said:

I was told by Ed Watt from The Galvanic Isolator Company after I checked with him that I'd installed my new GI Status Monitor correctly as it was showing two lights.

His comment was:

"Many people have problems with earth leakage when using specifically Victron chargers. It's usually due to the charger not being adequately bonded to the boat's hull. You need to "Earth" the case of the charger to the boat's hull/ground using a short, thick cable, min 2.5mm cross sectional area or preferably greater.

 

In all the cases I have come across so far, this has resolved the problem."

 

I assume he knows what he's talking about.

He could well be right, reading between the lines it sounds like the boat is lacking a hull to earth bond.

Since it's a £££££ combi why not give Victron a call and see what they say? Personally I'd prefer not to duplicate their tech support, or offer inconsistent advice, (but would be interested to know what they say).

On a bog standard boat without a combi, a hull/earth bond would typically involve a thick green/yellow earth wire between consumer unit common earth block, and a crimped ring terminal bolted to hull, but Victron may have other ideas, they may well require doing it from the combi chassis instead.

Edited May 29, 2017 by smileypete

[steve7a3]

 

35 minutes ago, smileypete said:

He could well be right, reading between the lines it sounds like the boat is lacking a hull to earth bond.

Since it's a £££££ combi why not give Victron a call and see what they say? Personally I'd prefer not to duplicate their tech support, or offer inconsistent advice, but would be interested to know what they say.

On a bog standard boat without a combi, a hull/earth bond would typically involve a thick green/yellow earth wire between consumer unit common earth block, and a crimped ring terminal bolted to hull, but Victron may have other ideas...

The Victron manual states, in the Installation section:

"- In case of a mobile installation (connection to input AC with a shore power cord), the ground connection is lost when the shore power cord is  unplugged. In this case the chassis of the product or the on - board section of the input ground wire must be connected to the frame (of the vehicle) or the ground plate or hull (of a boat)."

On mine, the chassis of the Victron is attached to a timber bulkhead, not to the hull. The earth connection on the chassis is not connected to anything. It is therefore likely that the chassis of the Victron is not connected to the hull at any point. In this case I assume there is no hull/earth bond. This is what I'm wanting to rectify so I can check  whether the indicated AC leakage disappears as suggested. If it doesn't then I'll start looking at other more complicated problems/solutions. 

My original question was, in short, "if the battery negative cable is connected ultimately to the hull, can I simply connect the chassis of the Victron to the same point or the same cable?"