Plug in Galvanic Isolator.. which one

[Sofia]

Hi everyone,

 

I am looking to purchase a glavanic isolator plug in one for the boat as it is currently on shoreline.

 

I am currently looking at them but there is different amps etc. Which one should I get it?

 

I have seen some from 500amp or 1000amp etc.  an example https://www.ebay.co.uk/i/291931699806?chn=ps

 

Do you have any recommendations ?

[Flyboy]

36 minutes ago, Sofia said:

Hi everyone,

 

I am looking to purchase a glavanic isolator plug in one for the boat as it is currently on shoreline.

 

I am currently looking at them but there is different amps etc. Which one should I get it?

 

I have seen some from 500amp or 1000amp etc.  an example https://www.ebay.co.uk/i/291931699806?chn=ps

 

Do you have any recommendations ?

500 amp is more than adequate, 100 amp would be fine though.  I suggest you get one with a status monitor.

[cuthound]

The fault capacity of the galvanic isolator depends on the maximum prospective fault current of the electrical supply.

 

For a domestic single phase 230 volt supply this is a theoretical maximum of 4.6kA (4600 amps) but in practice usually lower.

 

So if you are often connected to a mains landline, then go for a larger capacity one.

 

Personally I would go for the largest rating you can, because circumstances can and do change.

 

My galvanic isolator (Victron) is rated at 1600 amps, but I believe they do even greater capacity ones than that.

[Sofia]

Thank you

[David Mack]

24 minutes ago, cuthound said:

The fault capacity of the galvanic isolator depends on the maximum prospective fault current of the electrical supply.

 

For a domestic single phase 230 volt supply this is a theoretical maximum of 4.6kA (4600 amps) but in practice usually lower.

 

Can you please explain, for the benefit of the ignorant, how you can get a fault current of 4600A from a typical marina 16A or 32A bollard.

[cuthound]

31 minutes ago, David Mack said:

 

Can you please explain, for the benefit of the ignorant, how you can get a fault current of 4600A from a typical marina 16A or 32A bollard.

 

The maximum prospective fault current is the maximum current that can flow until such time as the fuse blows or the circuit breaker trips.

 

It is calculated by measuring the resistance between line and neutral and line and earth, and using the lower of the figures obtained.

 

 For example if the line to neutral resistance is the lower of the two resistance measurements and is measured at 0.05 Ω and the supply voltage measured at 230 V:

 

Maximum prospective short-circuit current (line-to-neutral) = 230/0.05 = 4600 A (or 4.6 kA)

 

This current will flow until the protection operates, typically in 10 to 20 milliseconds, and will damage any components not designed to withstand it.

 

These days it is measured by a tester, which displays the maximum prospective fault current on a digital display. At the dawn of my career you had to measure cable lengths and look up the resistance values for the various cable sizes in tables and manually calculate the maximum prospective fault current.

Edited January 16, 2019 by cuthound
To add the last paragraph

[Chewbacka]

52 minutes ago, David Mack said:

 

Can you please explain, for the benefit of the ignorant, how you can get a fault current of 4600A from a typical marina 16A or 32A bollard.

Assuming the bollard can supply infinite amps (which obviously it can’t) then if the bollard is connected to the boat with 2.5mm2 cable that is 15m long and there is a short between live and earth on the boat, then the fault current would be about 1000A.  As the bollard supply will have a resistance as will electrical cable connections, fuses, switches etc this will further limit the current, so a GI rated at 1000A is, I think, just fine. 

[ditchcrawler]

59 minutes ago, cuthound said:

 

The maximum prospective fault current is the maximum current that can flow until such time as the fuse blows or the circuit breaker trips.

 

It is calculated by measuring the resistance between line and neutral and line and earth, and using the lower of the figures obtained.

 

 For example if the line to neutral resistance is the lower of the two resistance measurements and is measured at 0.05 Ω and the supply voltage measured at 230 V:

 

Maximum prospective short-circuit current (line-to-neutral) = 230/0.05 = 4600 A (or 4.6 kA)

 

This current will flow until the protection operates, typically in 10 to 20 milliseconds, and will damage any components not designed to withstand it.

 

These days it is measured by a tester, which displays the maximum prospective fault current on a digital display. At the dawn of my career you had to measure cable lengths and look up the resistance values for the various cable sizes in tables and manually calculate the maximum prospective fault current.

Fault currents and protection are a world of there own

[WotEver]

1 hour ago, David Mack said:

Can you please explain, for the benefit of the ignorant, how you can get a fault current of 4600A from a typical marina 16A or 32A bollard.

As the previous three posts explain, it’s the fault current that can flow prior to the MCB tripping that is relevant here. It only takes milliseconds to ‘blow’ a couple of diodes if they are over-stressed. 

[cuthound]

27 minutes ago, ditchcrawler said:

Fault currents and protection are a world of there own

 

Tell me about it. The first major job I designed and project managed, way back in 1985, involved 2 x 1600 amp 415 volt 3 phase supllies, 2 x 1000kvVA diesel generators and a switchboard to control, parallel and ditribute the load from, via 32 x 400 amp breakers.

 

I was so worried that I hadn't got the discrimination right that I deliberately shorted the 3 phases and neutral together on one of the 400 amp output breakers, to prove that only that breaker would trip. I repeated the test  about 10 times (until the breaker welded itself together) and videoed the results.

 

Fast forward 20 years and a young graduate consultant was employed to check the electrical distribution for the building, by running breaker settings, cable sizes etc through some design software.

 

He contacted me to say the breakers couldn't achieve discrimination. When I met him to discuss this, he was adamant until I showed him the video footage of it discriminating on a massive fault.

 

He couldn't accept that his software was wrong, presumably because at each level it errs on the safe side to minimise the chance of failures and hence risk of litigation.

Edited January 16, 2019 by cuthound
Clarification

[ditchcrawler]

26 minutes ago, cuthound said:

 

Tell me about it. The first major job I designed and project managed, way back in 1985, involved 2 x 1600 amp 415 volt 3 phase supllies, 2 x 1000kvVA diesel generators and a switchboard to control, parallel and ditribute the load from, via 32 x 400 amp breakers.

 

I was so worried that I hadn't got the discrimination right that I deliberately shorted the 3 phases and neutral together on one of the 400 amp output breakers, to prove that only that breaker would trip. I repeated the test  about 10 times (until the breaker welded itself together) and videoed the results.

 

Fast forward 20 years and a young graduate consultant was employed to check the electrical distribution for the building, by running breaker settings, cable sizes etc through some design software.

 

He contacted me to say the breakers couldn't achieve discrimination. When I met him to discuss this, he was adamant until I showed him the video footage of it discriminating on a massive fault.

 

He couldn't accept that his software was wrong, presumably because at each level it errs on the safe side to minimise the chance of failures and hence risk of litigation.

Someone will ask what discrimination is and why you use it in a minute. If ever you post the videos let me know so I can have a look

[Big Bob W]

What's discrimination and why do you use it?

[Flyboy]

2 hours ago, cuthound said:

The fault capacity of the galvanic isolator depends on the maximum prospective fault current of the electrical supply.

 

For a domestic single phase 230 volt supply this is a theoretical maximum of 4.6kA (4600 amps) but in practice usually lower.

 

So if you are often connected to a mains landline, then go for a larger capacity one.

 

Personally I would go for the largest rating you can, because circumstances can and do change.

 

My galvanic isolator (Victron) is rated at 1600 amps, but I believe they do even greater capacity ones than that.

So none of these 70 to 100amp GI 's are any good then, please explain why?     https://safeshoremarine.com/products/

[WotEver]

13 minutes ago, Flyboy said:

So none of these 70 to 100amp GI 's are any good then, please explain why?     https://safeshoremarine.com/products/

The explanation has already been given and yes you are correct, none of those are any good. 

[cuthound]

56 minutes ago, ditchcrawler said:

Someone will ask what discrimination is and why you use it in a minute. If ever you post the videos let me know so I can have a look

 

Unfortunately they were the intellectual property of one of the companies I used to work for, so remained with them when I left.

 

Pretty interesting though, watching huge copper bus bars rated at 4000 amps move under the electromagnetic forces as the breaker closed (I used a high speed camera to film the bus bars, and a normal video camera filming each of the upstream and downstream breakers, plus Dranetz Power Quality meters to see the effects of any resulting transients downstream.

Edited January 16, 2019 by cuthound
Spillung

[ditchcrawler]

23 minutes ago, WotEver said:

The explanation has already been given and yes you are correct, none of those are any good. 

I wonder how they worked out their rating

1 minute ago, cuthound said:

 

Unfortunately they were the intellectual property of one of the companies I used to work for, so remained with them when I left.

 

Pretty interesting though, watching huge copper bus bars rated at 4000 amps move under the electromagnetic forces as the breaker closed (I used a high speed camera to film the bus bars, and a normal video camera filming each of the upstream and downstream breakers, plus Dranetz Power Quality meters to see the effects of any resulting trznsients downstream.

Yes I have watched films of busbars flexing

[cuthound]

5 minutes ago, ditchcrawler said:

I wonder how they worked out their rating

 

The designer will have made an assumption of the likely maximum fault current the device can handle.

 

He cannot know the configuration of the installation it is connected to, which includes the mains supply transformer rating, it's proximity to the end circuit, size and length of any cables in between and quakity of the connection to earth.

 

As marinas get bigger, the supply transformer will get bigger, which in itself will generate a larger fault current.

Edited January 16, 2019 by cuthound
To add the last sentance

[ditchcrawler]

I don't think there is a UK standard for them but there is an American one. Note that is only a though

[WotEver]

The ABYC is the generally accepted standard across the world in the absence of a country-specific guide. They have a pretty tough spec for GIs. If you’re suffering from insomnia you might like to read this: http://www.plaisance-pratique.com/IMG/pdf/A-28_Galvanic_Isolators_Standard.pdf

 

Here’s a nice GI that complies: https://www.newmarpower.com/wp-content/uploads/2016/02/Manual-GI-3050.pdf

Edited January 16, 2019 by WotEver
Spotted a typo

[Flyboy]

36 minutes ago, WotEver said:

The ABYC is the generally accepted standard across the world in the absence of a country-specific guide. They have a pretty tough spec for GIs. If you’re suffering from insomnia you light like to read this: http://www.plaisance-pratique.com/IMG/pdf/A-28_Galvanic_Isolators_Standard.pdf

 

Here’s a nice GI that complies: https://www.newmarpower.com/wp-content/uploads/2016/02/Manual-GI-3050.pdf

Yes but these Newmar ones are rated at 30-50amps so how come the Safeshore ones rated at 70-100amps are no good.

[TheBiscuits]

12 minutes ago, Flyboy said:

Yes but these Newmar ones are rated at 30-50amps so how come the Safeshore ones rated at 70-100amps are no good.

CURRENT RATINGS (NOMINAL): GI-30: 30 AMPS, GI-50: 50 AMPS

MAXIMUM SHORT CIRCUIT (BOTH MODELS): 5,000 AMPS

 

for the Newmar ones, against

 

● Suitable for use with 110 / 250 volt power supplies and 3 to 63 amps shore power pedestals.

● 70 amp fault current capacity essential for total reliability

 

For the Safeshore 70A

Edited January 16, 2019 by TheBiscuits

[Chewbacka]

12 minutes ago, Flyboy said:

Yes but these Newmar ones are rated at 30-50amps so how come the Safeshore ones rated at 70-100amps are no good.

If you read the data sheet for the Newmar products it states the 30/50A currents are nominal with a fault current of 5000A, so they are just fine.

 

if you want to know why  they are nominally rated,  it’s because the internal conductors are rated for 30A or 50A  so the smaller unit is not suitable for a 32A supply (fine for 16A) for that you need the 50A unit.

 

 

added - I need to type quicker...........

Edited January 16, 2019 by Chewbacka

[Sea Dog]

5 hours ago, Sofia said:

Do you have any recommendations ?

So, anyway, after all that lot ^^^^, have you considered a built in one? There's a greater choice, it's not going to get damaged or dropped in the cut and it's always in circuit so you avoid the chance of forgetting to connect it.  Its a very easy thing to fit taking only minutes, so if you are not able to fit it yourself, it shouldn't cost much - likely enough still within the cost of the in line versions you're considering.

[Flyboy]

36 minutes ago, TheBiscuits said:

CURRENT RATINGS (NOMINAL): GI-30: 30 AMPS, GI-50: 50 AMPS

MAXIMUM SHORT CIRCUIT (BOTH MODELS): 5,000 AMPS

 

for the Newmar ones, against

 

● Suitable for use with 110 / 250 volt power supplies and 3 to 63 amps shore power pedestals.

● 70 amp fault current capacity essential for total reliability

 

For the Safeshore 70A

So the Safeshore one at 70 amp nominal is OK and Wotever is wrong to say that it's not.

[WotEver]

7 minutes ago, Flyboy said:

So the Safeshore one at 70 amp nominal is OK and Wotever is wrong to say that it's not.

No!

 

Safeshore fault current: 70A

Newmar fault current: 5000A

 

Can you see the difference?

Edited January 16, 2019 by WotEver