Is this installation safe or not?

[JamieK]

Hi Guys

I am hoping to get some outside opinion on an installation that I don't think is safe but others do, I've drawn a shocking pic which I have attached to help describe the issue but I will try and detail it aswell.

This was a Solar PV installation in which the AC side was fed into the board on the left. The board is then fed underground using 16mm 4 core and earth SWA back to the main board in the customer building which is about 20m away under a bit of grass.

The SWA has a proper gland fitted where it enters the board on the left and has a 16mm supplementary bond connected to the MET inside the board. However, the left hand boards only connection to earth is through the SWA earth cable going underground to the main board MET 20m away.

The SWA feeding the main board is not glanded due to it coming through a ceiling in steel box trunking and the customer being adamant they did not want any damage in the newly decorated ceiling so the electrician just cut the SWA and terminated into the board without earthing the SWA with a gland - make sense!!??

The spark said that it is safe because the SWA is glanded at the left hand board side and a connection to earth is made through the internal 16mm earth in the SWA to the MET of the board on the right hand side.

My comment is that if (in the unlikely event) some form of excavation takes place and penetrates the soil above the SWA and the sand and the tape that was installed and causes damage to the SWA earth cable this will cause serious risk???

The spark says that is that happens the MCB will trip posing no risk to the person excavating but again my response is that this might not happen soon enough not to pose a danger.

Hope I have explained this as simply as possible I know its an odd one but any input very welcome indeed even if it is just to say I am wrong and it is safe.

 

[Pete999]

Couple of questions
1 how deep is the SWA buried?
2 See no reason why the SWA can't be glanded
3 Is the supply at the left or right hand board?

 

[jackhammerJIM]

drawing shows 5 core swa , text states 4 core .

 

[telectrix]

both ends of the SWA need glanding. generally, if the armour is not uses as cpc, it only needs earthing at the supply end.

 

[Lucien Nunes]

In effect what you are asking is: "Is it OK to earth SWA armour only at the destination, provided a copper core in the cable connects the destination to the source?"


I assume when you say MET in the left hand board, you mean the earth bar: The MET is at the origin. You have to start asking specific questions: What is the Zs and what protection is offered to the cable? What is the method of earthing and is the submain arrangement satisfactory as far as extending or exporting the equipotential zone is concerned? I don't think it is a plain yes or no kind of question.

 

[JamieK]

Couple of questions
1 how deep is the SWA buried?
2 See no reason why the SWA can't be glanded
3 Is the supply at the left or right hand board?

500mm
Spark said he couldn't fit the gland inside the containment from ceiling into DB
Supply on right hand board

- - - Updated - - -

drawing shows 5 core swa , text states 4 core .

It is 4 cores plus earth so in effect 5 core

 

[Silly Sausage]

both ends of the SWA need glanding.

I missed that.
Alarm bells...

 

[Rockingit]

It sounds like it could use some finesse, but if your concerns are that it won't meet disconnection times under a fault load then the only way of knowing is to do a Zs on the circuit and see if it complies with the upstream CPD as dictated by the BGB.

 

[GLENNSPARK]

In effect what you are asking is: "Is it OK to earth SWA armour only at the destination, provided a copper core in the cable connects the destination to the source?"


I assume when you say MET in the left hand board, you mean the earth bar: The MET is at the origin. You have to start asking specific questions: What is the Zs and what protection is offered to the cable? What is the method of earthing and is the submain arrangement satisfactory as far as extending or exporting the equipotential zone is concerned? I don't think it is a plain yes or no kind of question.

this is how i read it Lucien....

as you say there is now not only the length of the 16mm core to consider....but the length of the armourings as well when it comes to meeting a disconnection time...

better to earth the armourings at source..

 

[telectrix]

if the guy has got the SWA into the right hand CU through a hole, then there's no excuse for not fitting a gland.

 

[JamieK]

In effect what you are asking is: "Is it OK to earth SWA armour only at the destination, provided a copper core in the cable connects the destination to the source?"


I assume when you say MET in the left hand board, you mean the earth bar: The MET is at the origin. You have to start asking specific questions: What is the Zs and what protection is offered to the cable? What is the method of earthing and is the submain arrangement satisfactory as far as extending or exporting the equipotential zone is concerned? I don't think it is a plain yes or no kind of question.

Thanks for the reply you are right that I should have stated the earth bar on the left hand side, I don't know what the zs is but the earthing arrangement was TNS. Looking it this I don't see how a sage disconnection would occur if the cable got damaged especially damage to the internal earth cable in the SWA. If you inspected this install you would have concerns wouldn't you?

 

[GLENNSPARK]

if the guy has got the SWA into the right hand CU through a hole, then there's no excuse for not fitting a gland.

perhaps he wasn`t capable....

 

[GLENNSPARK]

Thanks for the reply you are right that I should have stated the earth bar on the left hand side, I don't know what the zs is but the earthing arrangement was TNS. Looking it this I don't see how a sage disconnection would occur if the cable got damaged especially damage to the internal earth cable in the SWA. If you inspected this install you would have concerns wouldn't you?

get a Zs on the furthest point (this will be the armourings top end) to confirm....

 

[Silly Sausage]

perhaps he wasn`t capable....

See Amendment 3 for the new version of competent.

 

[JamieK]

if the guy has got the SWA into the right hand CU through a hole, then there's no excuse for not fitting a gland.

There was just an open hole inside the steel box trunking and so allegedly nowhere to connect the gland and thus earth it, the option was to make a hole in the ceiling and gland onto the trunking that way but client forbid it , not right I know which is why I think it should be redone

 

[GLENNSPARK]

There was just an open hole inside the steel box trunking and so allegedly nowhere to connect the gland and thus earth it, the option was to make a hole in the ceiling and gland onto the trunking that way but client forbid it , not right I know which is why I think it should be redone

well for a start...

regardless of what a client thinks...this should always come secondry to safety, conformance and common sense...

 

[JamieK]

well for a start...

regardless of what a client thinks...this should always come secondry to safety, conformance and common sense...

Could not agree more

 

[jackhammerJIM]

Looking it this I don't see how a sage disconnection would occur if the cable got damaged especially damage to the internal earth cable in the SWA.

think about it jamie

if a spade was driven through the cable shorting a line conductor to the armour or cpc ,

the OCD would operate under fault conditions if the zs was met .

If you inspected this install you would have concerns wouldn't you?

i would be wondering why the installer had entered from above and not glanded into the bottom of board . seems crackers to me .

 

[Silly Sausage]

There must an RCD somewhere in this place, so it's perfectly safe!

 

[Lucien Nunes]

I guess Jamie's point is that if the cable were partially cut through, you could leave the line and part of the armour intact but sever the CPC, hence losing the earth connection to the armour and the sub DB without the circuit protection operating. Normally if the CPC is earthed at the source, you can't cut through all of the armour without hitting the line, so the protection will operate before the last of the CPC is broken. It's a far fetched scenario but it's needless because it's a bodge.

 

[Paul.M]

Just gland both ends and make the installation safe. Its not that difficult. Do the job better than the regs advise, that way I can sleep at night.

 

[hasel5]

I guess Jamie's point is that if the cable were partially cut through, you could leave the line and part of the armour intact but sever the CPC, hence losing the earth connection to the armour and the sub DB without the circuit protection operating. Normally if the CPC is earthed at the source, you can't cut through all of the armour without hitting the line, so the protection will operate before the last of the CPC is broken. It's a far fetched scenario but it's needless because it's a bodge.

And theres always the risk also of a meteorite strike as well

 

[Rockingit]

Heck, if you genuinely can't get an armoured gland on to anything, you could at least bodge it with an earth clamp (NOT suggesting this as good practice). Let's face it - you can't make it any worse.

 

[shanky887614]

Why didnt they gland into the trunking? Then have just the cores going into the board. Even terminating swa into a 1 gang mc box above ceiling then into trunking would be better

 

[Gavin A]

It shouldn't have been done like that, but there's no way you're going to get a spade through a 5 core 16mm2 SWA cable to the point where the earth core is cut, even if you've eaten your spinach.

Now, if you've got a mechanical digger or something then maybe, but it's pretty safe from spade strikes IMO.

Besides, the mcb would trip from the short circuit.

 

[Knobhead]

Other than the poor glanding this is being over complicated. How in the name of god are you going to damage a cable to the extent the CPC and the armourings are severed but the phases remain live?

Fit a feeder directional relay, should only cost a couple of grand + the pilot cable.

Crazy thread.

 

[uksparks]

if the guy has got the SWA into the right hand CU through a hole, then there's no excuse for not fitting a gland.

Totally agree, even if the gland was hanging loose inside the containment area with a piranha nut or something and 16mm from there, just no excuse however you look at it. The only thing he seems to have got right is his choice of transport saving on fuel bills, he probably rides a horse.

 

[micknew]

perhaps he wasn`t capable....

Welcome back Glen.
Your honest/true posts have been missed by most I'm sure.

 

[micknew]

That post will probably earn me thousands of thanks and likes.
See you in the bar.

 

[Gavin A]

Other than the poor glanding this is being over complicated. How in the name of god are you going to damage a cable to the extent the CPC and the armourings are severed but the phases remain live?

Fit a feeder directional relay, should only cost a couple of grand + the pilot cable.

Crazy thread.

with a spade apparently.

Is the OP available for trenching work next time we've got a long cable run to dig in? Could do with someone with that sort of force behind the spade on the job.

 

[Darkwood]

My take on it---

The Swa should be earthed correctly at the supply end it does not need to be earthed at the load end as you have a separate earth core.

If the cable suffers damage to the outer sheath then corrosion to the SWA has less likely chance of isolating the steel from earth also if the is damaged in your spade scenario nearer the supply side as oppose to the load side and it was glanded only at the load then zs values may not meet as the fault has to travel nearly full length to the load end gland then back down the earth core effectively doubling the distance... you can take a zs at the unglanded end down to the armour to see if it complies or not.

All in all their really isn't an excuse for not earthing both ends of the swa as this would be good practice although earthing at one end is acceptable and pref' the supply end which will ensure zs is met. Their is no reason why a section of the ceiling board couldn't be removed and made good after IMHO this is just lazy attitude and loose excuses.

 

[Engineer54]

I think Darwood has summed it all up nicely!!

I too also support the glanding/earthing at both ends of a SWA cable, there's rarely any excuse not to in my opinion.

 

[GLENNSPARK]

My take on it---

The Swa should be earthed correctly at the supply end it does not need to be earthed at the load end as you have a separate earth core.

If the cable suffers damage to the outer sheath then corrosion to the SWA has less likely chance of isolating the steel from earth also if the is damaged in your spade scenario nearer the supply side as oppose to the load side and it was glanded only at the load then zs values may not meet as the fault has to travel nearly full length to the load end gland then back down the earth core effectively doubling the distance... you can take a zs at the unglanded end down to the armour to see if it complies or not.

All in all their really isn't an excuse for not earthing both ends of the swa as this would be good practice although earthing at one end is acceptable and pref' the supply end which will ensure zs is met. Their is no reason why a section of the ceiling board couldn't be removed and made good after IMHO this is just lazy attitude and loose excuses.

indeed dark...its what i was getting across at #9