RCDs and isolation transformers

[Paul C]

What are the various failure modes of a marina-side RCD which a boat owner hooked up to it should be worried about?

 

From a bit of looking around, it seems that galvanic isolators on average cost about £80-100, where an isolation transformer costs around £700 (and weighs much more, and is bigger in size). Anyone know where you can buy an AFFORDABLE isolation transformer?

[GUMPY]

From a bit of looking around, it seems that galvanic isolators on average cost about £80-100, where an isolation transformer costs around £700 (and weighs much more, and is bigger in size). Anyone know where you can buy an AFFORDABLE isolation transformer?

http://www.airlinktransformers.com/cat1/35/boating_transformers.html

[Grace and Favour]

The Airlink Isolation Transformer is, I believe, regarded as being amongst the very best available

[Paul C]

http://www.airlinktransformers.com/cat1/35/boating_transformers.html

 

Ouch! Still £250 vs £80 and a whole lot more bulkiness.

[Grace and Favour]

Ouch! Still £250 vs £80 and a whole lot more bulkiness.

 

Plus VAT & delivery!

 

 

But it does a superlative job!

[smileypete]

What are the various failure modes of a marina-side RCD which a boat owner hooked up to it should be worried about?

Good question - none, really.

 

Why not try the RCD test button before connecting up and at intervals afterward?

 

Seems most boats have an on board RCD too.

 

cheers, Pete.

~smpt~

[GUMPY]

Good question - none, really.

 

Why not try the RCD test button before connecting up and at intervals afterward?

 

Seems most boats have an on board RCD too.

 

 

The RCD test button only confirms that the mechanism of the RCD is working

It does NOT tell you if the RCD is working and tripping at the correct level for that you need an RCD tester.

I have seen RCDs where the test button does not work but they still trip at the correct level and also RCDs that trip fine on the test button but need 3-4 times the "safe" current to make them trip.

 

Now on to the subject of GI

Diodes (which are the main part of a GI) can and do fail either open circuit or short circuit one way will leave your boat in a dangerous state and the other with no protection.

I am still not convinced of their legality and would refuse to fit one if asked as technically you are not supposed to fit any kind of switch or device in the earth path and GI are both.

Edited June 4, 2012 by idleness

[Paul C]

The RCD test button only confirms that the mechanism of the RCD is working

It does NOT tell you if the RCD is working and tripping at the correct level for that you need an RCD tester.

I have seen RCDs where the test button does not work but they still trip at the correct level and also RCDs that trip fine on the test button but need 3-4 times the "safe" current to make them trip.

 

Now on to the subject of GI

Diodes (which are the main part of a GI) can and do fail either open circuit or short circuit one way will leave your boat in a dangerous state and the other with no protection.

I am still not convinced of their legality and would refuse to fit one if asked as technically you are not supposed to fit any kind of switch or device in the earth path and GI are both.

 

 

Hiya, interesting comments on the effectiveness of a galvanic isolator. Are you still selling isolation transformers?

 

Blatent plug time

Declares finacial interest

 

There will be an insulated one on the market soon for about £275 cant say more than that.

 

beware the victron one as its wired incorrectly and could cause the boat to go live....

 

see here for details

 

http://www.smartgauge.co.uk/iso_wire.html

 

Julian

 

...from http://www.canalworld.net/forums/index.php?showtopic=6101&st=0&p=84609&hl=+galvanic%20+isolator&fromsearch=1entry84609

[GUMPY]

Hiya, interesting comments on the effectiveness of a galvanic isolator. Are you still selling isolation transformers?

Nope it never happened the manufacturer I was trying to get to go into partnership didnt come up with the goods.

Airlink built one which filled the gap and I have no connection with them.

Edited June 4, 2012 by idleness

[Robbo]

Nope it never happened the manufacturer I was trying to get to go into partnership didnt come up with the goods.

Airlink built one which filled the gap and I have no connection with them.

 

How come the Victron one is wired incorrectly?

 

My Airlink went dead with about 3-4 months of use. But as I had it in a box for 2 years it's out of warrenty.

[GUMPY]

How come the Victron one is wired incorrectly?

 

My Airlink went dead with about 3-4 months of use. But as I had it in a box for 2 years it's out of warrenty.

Well mine has been running since 2006 and its been abused

as for the Victron

http://www.smartgauge.co.uk/iso_wire.html

its wired with the interwinding screen connected to the boat and hence can be dangerous if there are certain faults

[Robbo]

Well mine has been running since 2006 and its been abused

 

Mine did get abused also, and it died slowly that I didn't notice it was cutting the electric off, just my laptop complaining that I removed a external disk sometimes. (which it does if electric is removed from the external disk) but I never thought it was the IT until then it totally went and was quite hot when it did. I think it just got overheated as there's basically no escape for the heat just a enclosed box.

[GUMPY]

Mine did get abused also, and it died slowly that I didn't notice it was cutting the electric off, just my laptop complaining that I removed a external disk sometimes. (which it does if electric is removed from the external disk) but I never thought it was the IT until then it totally went and was quite hot when it did. I think it just got overheated as there's basically no escape for the heat just a enclosed box.

mine is the "prototype" and is in a metal case not a plastic box and I have seen it sit at 25+ amps for long periods

[Robbo]

mine is the "prototype" and is in a metal case not a plastic box and I have seen it sit at 25+ amps for long periods

 

Mine is in the plastic box. When it did die it was basically too hot too touch. I did pull 16amp (if it was more the shore line RCD tripped) a lot over the winter for good periods of time as I use a fan heater. (fitting out, no stove, etc.).

[Bob-18]

One important thing to remember is that an RCD is not an overload detector, it is an earth current detector.

As such an RCD will "hold in" quite happily for a very long time despite the circuit it is protecting being seriously overloaded.

But the same RCD will "trip out" when the current flowing in the earth circuit is above a certain level, despite the power drawn being at a low level.

 

There are devices which combine the function of an overload trip and an RCD into one unit "RCBO". These are gaining popularity as you only need one "slot" in the fuse box for them, but needless to say they are more expensive.

[Paul C]

Yeah, hence I need to install a box which has an RCD and MCB. Typically, a "garage consumer unit" has 1x RCD and 2x MCBs of 6A and 10A, and isn't too big. In an ideal world I'd like a little box with 1x RCD and 1x MCB (for the inverter) and another with 1x RCD and 3x MCBs (sockets, battery charger, immersion heater element) but since nobody does a 3x MCB unit, I'll have to combine battery charger and immersion heater off of one MCB.

 

I've seen compact RCDs which fit into a normal domestic sized switch box, eg

 

http://www.screwfix.com/p/volex-13a-rcd-fcu/83049#

 

but not an RCBO in a similar size, which would be handy.

 

I guess I could have a custom panel made up, but that's a bit more £

[smileypete]

The RCD test button only confirms that the mechanism of the RCD is working

It does NOT tell you if the RCD is working and tripping at the correct level for that you need an RCD tester.

I have seen RCDs where the test button does not work but they still trip at the correct level and also RCDs that trip fine on the test button but need 3-4 times the "safe" current to make them trip.

Not saying the RCD or it's test button is super perfect, just that for the boater as long as the test button trips the RCD, it gives reasonable enough assurance the RCD will work as intended. There's a comprehensive report on RCD reliability right here:

 

http://www.esc.org.uk/fileadmin/user_upload/documents/industry/rcd_research/Final_Report.pdf

 

'This report contains a review and analysis of the results of tests conducted on RCDs ... 607 electromechanical RCDs were tested in properties in the UK owned by Housing Associations and Local Authorities.

 

Six RCDs were found to have been shorted out to avoid what was perceived to be nuisance tripping. If these six RCDs are left out, the pass rate was 97.2%...

 

It was concluded that the RCD faults could have been detected if the occupier had operated the test button on the RCD.'

 

(By the way the test button is not just mechanical, it creates a measured current difference between live and neutral to simulate an earth leakage.)

 

 

Now on to the subject of GI

Diodes (which are the main part of a GI) can and do fail either open circuit or short circuit one way will leave your boat in a dangerous state and the other with no protection.

I am still not convinced of their legality and would refuse to fit one if asked as technically you are not supposed to fit any kind of switch or device in the earth path and GI are both.

I think you're trying to imply that GI diodes are likely to fail in practice which simply isn't the case.

 

To get an earth fault current that exceeds the surge rating of the GI diodes before the RCD/MCB trips would require a very low 'earth fault loop impedance'.

 

Even the distribution board at a fairly large install is not that likely to see such a low earth fault loop impedance, let alone a boat on the end of a shoreline in a boatyard.

 

I'm not saying that it could never ever happen, just that it's not worth losing any sleep over.

 

cheers,

Pete.

[Paul C]

Cheers for the replies so far, keep them coming. So far I think this is a reasonable conclusion:

 

Isolation Transformer

 

1 Cost a lot of money

2 Are bulky and heavy

3 For best protection, need to be off the boat (theft risk? see 1)

4 They can fail, and when they do, get hot (fire risk if on boat?)

 

Galvanic Isolators

 

1 Break 'the rules' on electrical wiring on a technicality

2 Can be rendered ineffective by devices which are switched mode, produce noisy RFI AND have an earth connection

3 The above can be indicated by specifying a GI with indicating diodes (only slightly more expensive)

4 They can fail, either shorted out (leaves the boat vulnerable to galvanic corrosion) or open circuit (leaves the boat without grounded connection)

5 Can be tested, although the test is onerous and likely to not be done (frequently) (involves disconnecting shoreline, then earth connections, then wiring a battery, etc etc?)

6 'Safer' GIs exist, eg USA-spec ones which have tighter rules on them, but are as expensive or more expensive than isolating transformers

7 If they do fail, the unsafe condition only occurs if another electrical device also fails and sends the boat hull live. Ie a simultaneous double failure would need to occur, although if the GI had failed and been undetected for some time, this is plausible (although rare)

 

Other options are:

 

1 Make the boat entirely 2 wire (double insulated) in its AC wiring (illegal due to no ground connection but theoretically safe?)

2 Not use 230V AC at all, ie use engine power, gas/diesel/solid fuel space/water heating, solar, wind etc and all equipment is 12V DC

[GUMPY]

Just don't moor anywhere near me if you go for your other option 1 alternative its both dangerous and stupid

Edited June 5, 2012 by idleness

[Paul C]

Just don't moor anywhere near me if you go for your other option 1 alternative its both dangerous and stupid

 

I'd not go for it, but I've heard it as a suggestion so thought it was worth mentioning to stimulate debate on it.

[jonathanA]

Cheers for the replies so far, keep them coming. So far I think this is a reasonable conclusion:

 

Isolation Transformer

 

1 Cost a lot of money

2 Are bulky and heavy

3 For best protection, need to be off the boat (theft risk? see 1)

4 They can fail, and when they do, get hot (fire risk if on boat?)

 

Galvanic Isolators

 

1 Break 'the rules' on electrical wiring on a technicality

2 Can be rendered ineffective by devices which are switched mode, produce noisy RFI AND have an earth connection

3 The above can be indicated by specifying a GI with indicating diodes (only slightly more expensive)

4 They can fail, either shorted out (leaves the boat vulnerable to galvanic corrosion) or open circuit (leaves the boat without grounded connection)

5 Can be tested, although the test is onerous and likely to not be done (frequently) (involves disconnecting shoreline, then earth connections, then wiring a battery, etc etc?)

6 'Safer' GIs exist, eg USA-spec ones which have tighter rules on them, but are as expensive or more expensive than isolating transformers

7 If they do fail, the unsafe condition only occurs if another electrical device also fails and sends the boat hull live. Ie a simultaneous double failure would need to occur, although if the GI had failed and been undetected for some time, this is plausible (although rare)

 

Other options are:

 

1 Make the boat entirely 2 wire (double insulated) in its AC wiring (illegal due to no ground connection but theoretically safe?)

2 Not use 230V AC at all, ie use engine power, gas/diesel/solid fuel space/water heating, solar, wind etc and all equipment is 12V DC

 

I think you've succumbed to the 'transformer' propaganda...

 

there is NO evidence that GI are 'technically' or acutally illegal - so thats just rubbish

 

There is a risk that some Gi can be affected by switched mode devices or devices with a high earth leakage. The better ones have a higher trigger voltage (usually just means a couple of diodes in series) and as you say the ones with some form of meter or led indicators would alert you to this possibility. (so thats 2,3,4 and 5 taken care of these are all the same really anyway)

 

safer spec GI's lot of maufactuers with lots of specs... there are 'safer spec' transformers available that are rated for larger currents and therefore less likely to burst into flames...

 

in the event of '7' - either your on board RCD would trip making everything safe and alerting you to a problem or if the problem was on the shore side of the GI the shore RCd would trip just as with a transformer...

 

I think gibbo's treatise on the smartgauge pages is the fairest impartial assessment of the pro's and cons of GI vs IT. For me A Gi with status indicator made sense, but if you absolutely want guaranteed no risk then your option 2 is the way to go.

 

I'd say your option 1 is effectively what you get with an isolation transformer anyway...

[Tony Brooks]

Cheers for the replies so far, keep them coming. So far I think this is a reasonable conclusion:

 

 

big snip

 

 

 

1 Make the boat entirely 2 wire (double insulated) in its AC wiring (illegal due to no ground connection but theoretically safe?)

2 Not use 230V AC at all, ie use engine power, gas/diesel/solid fuel space/water heating, solar, wind etc and all equipment is 12V DC

 

 

Others have been eloquent and concur with my views of option 1.

 

Using on board generated AC does not require a GI or IT to prevent hull corrosion even though your on board mains should still be bonded to the hull because there is no connection to other boats or the land so no circuit can be formed. So an on board inverter or generator (if suitably constructed to allow earth bonding) will give you a mains supply, its shorelines that cause the problems (and extremely rarely if your boat is between one with a fault and another that is earthed without a GI or IT).

[GUMPY]

 

there is NO evidence that GI are 'technically' or acutally illegal - so thats just rubbish

 

 

I know that BS7671 is not law, however it is what all electrical regs are based on.

And in this case a diode counts as a switching device.

 

 

BS7671:2008 ammended 2011

 

section 543.3.3

A switching device shall not be insetrted in a protective conductor except for the following

(i) as permitted in 537.1.5

(ii) multipole linked switching or plug-in devices in which the protective conductor shall not be interupted before the live conducters and shall be re-established not later than when the live conductors are re-connected.

 

537.1.5 deals with multiple source switching and is not relevant here

[jonathanA]

I know that BS7671 is not law, however it is what all electrical regs are based on.

And in this case a diode counts as a switching device.

 

 

BS7671:2008 ammended 2011

 

section 543.3.3

A switching device shall not be insetrted in a protective conductor except for the following

(i) as permitted in 537.1.5

(ii) multipole linked switching or plug-in devices in which the protective conductor shall not be interupted before the live conducters and shall be re-established not later than when the live conductors are re-connected.

 

537.1.5 deals with multiple source switching and is not relevant here

 

The BSI publish numerous titles concerning acceptable standards of design/safety/quality etc. in various fields. BS 7671 : 2001 (AMD No 2 : 2004) concerns the safety of electrical wiring in buildings (dwellings, commercial, industrial or otherwise)

 

note the underlined text

the point i made and which is perfectly valid is that both GI'd and It's are accpetable practice for marine installations as far as I know, your quoting the 17th edition wiring regs doesn't alter that i'm afraid.

 

can you actually produce some evidence that GI's might be "technically illegal" or is it simply your opinion to which you are of course entitled ?

 

I can't see thousands of people and the several very reputable manufacturers stopping using them or making them....

[GUMPY]

BS7671:2008 ammended 2011 covers marinas and caravans as well

If you would like to stand up and present a case as to why they are safe feel free.

Personally I cant and wont ever fit one as the repercussions if it did fail and someone was killed or injured would finacially break me.

You are of course correct that BS7671 is aimed ad buildings but it now does cover Caravans and in the next revision will cover boats as well.

If it came to it in a court of law in the absence of any BS that specifically applies to boats then BS7671 would be taken as the backstop and if you hadnt complied then you would be found guilty.

 

The "evidence" is in my post above just because you dont like it doesnt mean to say its not correct

 

Final edit BS7671:2001 is now two versions out of date it was fully revised in 2008 and ammended in 2011

Oh and yes I do hold the latest BS7671 qualification.........

Edited June 6, 2012 by idleness