DC Cable buried <50mm in wall

[SolarCity]

Another idea would be to distance the cables say 200mm apart in separate single core swa or steel conduit (earthed) this would reduce the risk of a single incident damaging +ve and - ve at the same time
(this is a method I've read is sometimes used in Germany and other countries ,the Americans use aluminium conduits or single core metalclad cable ) be interesting to hear from Bruce to see his take about this idea.

I suppose a problem with this method would be an increased risk of damage from direct lightning strikes on the array.

 

[BruceB]

I agree with the previous post that the reason 7671 deprecates runnng cables of the same circuit separate routes is the potential for induced voltages in for example the case of lightning. It applies not only when lightning strikes directly, but also when there is a lightning strike in proximity to the building as large field strengths are generated.

Picking up Gavin's last post, I still think on balance earthing the armour of a dc swa cable is safer than not doing so. With a galvanically isolated inverter, connecting one of the dc cables to the armour (eg by a nail going through the armour) does not change the potential of the armour, it just brings the potential of that array cable down to earth when it would otherwise be floating. With a transformerless inverter the same will happen. I think both sorts of inverter of the reputable makes regularly measure the isolation of the array cables to earth, either continuously or regularly, so will indicate a fault which is useful. I believe the transformerless ones are effectively monitoring the isolation from earth continuously (but have not gone back to the tech specs to check).

I will acknowledge though whenever you are dealing with earthing or bonding there are many different failure/fault modes, some contradictory. When you protect against one you can make yourself more vulnerable to others which makes the judgement of the best regime in a particular circumstance a tricky one.

Regards
Bruce

 

[dougalthedog]

You put dc through SWA and the SWA is earthed, nail touching one of the conductors nothing is going to happen! No trip - if you hit the + conductor then that would be put the SWA at the same potential as the + conductor and the earth for the whole house would be at the + voltage.

If you hit both the + and - then you would get Isc following but still nothing would trip!

In both cases the fault is not 'removed' by a protective device.

Surely this has "hit the nail on the head" - someone comes along and fits something to the wall and screws into the SWA, shorting one of the conductors to earth. This goes unnoticed (assume the inverter doesn't detect). Some time later an electrician comes along and touches the other DC conductor whilst in contact with earth (assuming it to be floating) and boom!

OK so it's probably not best practice to be touching live DC conductors, but it does happen.

 

[Gavin A]

Surely this has "hit the nail on the head" - someone comes along and fits something to the wall and screws into the SWA, shorting one of the conductors to earth. This goes unnoticed (assume the inverter doesn't detect). Some time later an electrician comes along and touches the other DC conductor whilst in contact with earth (assuming it to be floating) and boom!

OK so it's probably not best practice to be touching live DC conductors, but it does happen.

my point in a nutshell.

I can't see any justification at all for doing this.

 

[Earthstore]

My two pennith worth.
I am not a sparky, but I have a basic understanding of it, and there appears to be different opinions on this, a little strange as you all have the same guidelines.
I am aware that lots of regs can and often do apply, however, DC has no real protection, no trips, no RCD, no MCB, but in the case of PV it is high voltage, so why even route DC where there is potential risk?
I was always lead to believe that electrical safety is about potential, and as electricians, you need to reduce or minimize that "potential".
I can not see how burying High voltage DC in a wall or running it through the house is going to minimize "potential", whether in SWA or otherwise.

This is not a dig at electricians, and I hope my comments are constructive.

 

[Gavin A]

I've been digging through the 17th edition to try to get a definitive answer on this, and this is where I'm at with it FWIW.

On the DC side of the PV circuit basic and fault protection is supplied by 412 double or reinforced insulation and a second method of fault protection is provided by the protective measure of Electrical Separation (413).

413.1.1 Electrical separation is a protective measure in which:

(ii) fault protection is provided by simple separation of the separated circuit from the other circuits and from earth.

Anyone connecting the armor to earth will remove that separation from earth in fault conditions, which is just when it's most needed, and neither will it be protected in any way by the connection to earth.

Granted, 413 is directed at current using equipment, but it works equally for current generating equipment IMO.

Further point to note though is

413.3.4 Flexible cables shall be visible throughout any part of their length liable to mechanical damage.


Neither of these measures is allowed under either 522.6.100 or 522.6.101 for cables concealed <50mm in a wall or floor.

An earthed metal covering does appear to be allowed under the earth metal covering regulation (522.6.100(i)), but as the earthing of the conduit has zero positive impact on the safety of this system and negatively impacts on it by removing the protection from Electrical Separation I'd think it would fall foul of 133.1.3

"where equipment to be used is used outside the scope of its standard, the designer shall confirm that the equipment provides at least the same degree of safety as that afforded by compliance with the regulations."

or some similar regulation along those lines which essentially means you shouldn't rely simply on the letter of the regulations when working with a situation that is outside of the scope for which those regulations were really intended, and where the level of protection afforded by that measure in normal circumstances isn't present that you shouldn't simply obey the letter of the guidance and ignore the purpose of them.

So, to answer the original question... No, IMO you can't use SWA (earthed or otherwise) as protection for cable buried <50mm in a wall. As someone stated earlier, the only way of doing this safely in line with the BS7671 would be 522.6.101 (iv)

'mechanically protected against damage sufficient to prevent penetration of the cable by nails, screws and the like'

or

(Vi) form part of a SELV or PELV circuit...

I'd be interested if anyone knew if this had been defined anywhere into the thickness of metal capping that would be needed to comply with this.

 

[Gavin A]

I just realised that on the first job I did in this house I've buried the DC cable in the wall, protected by earthed metal capping...... I'm pretty sure though that it actually just classes as SELV / PELV as it's only 5 small panels usually operating at around 110V.

 

[BruceB]

I agree that the protective measure double or reinforced insulation is the norm.

Electrical separation is not normally by design being used as a protective measure with solar pv. Whilst transformered inverters are separated dc from ac, the requirements of that protective measure when you have multiple loads as opposed to a single load (ie the whole house not a shaver point in the bathroom) are not normally put in place (418.3) and arguably are difficult to do successfully. And with a transformerless inverter there is no electrical separation between ac and dc (the exemption letters issued by manufacturers under the final part of 712.411.3.2.1.2 are not about electrical separation, just not passing dc fault currents onto the ac side). So I do not think the electrical separation arguments help much.

In Gavin's scenario of a fault between an earthed metallic covering and a dc cable, it takes two separate faults before there is a danger to personnel. My argument is that if the metallic covering is earthed then although the first fault is not cleared it will be detected so action can be taken.

ELV goes up to 120V dc and 50V ac so a small number of panels could well fall into that category.

I'll ask a question of the IET site to see what they come up with.

Regards
Bruce

 

[yellowvanman]

I agree that the protective measure double or reinforced insulation is the norm.

In which case take a look at 412.2.4.1 (ii) - requiring non metallic mechanical protection of the basic insulation

My argument is that if the metallic covering is earthed then although the first fault is not cleared it will be detected so action can be taken.

How will it be detected?

Not getting at you

 

[moggy1968]

The regs only require protection for single fault conditions. Even though you masy be able to predicsat a chain of multiple faults the regs don't require you to allow for that.
metal capping very definately does not fulfil the requirement for mechanical protection of cables.

 

[BruceB]

It is alright I do not feel got at! I like debating the points to get at the truth.

Double insulated solar cable meets 412.2.4.1(ii) by virtue of its outer sheath. So does swa by virtue of the internal sheath outside the insulated inner cores but inside the armour.

The first fault will be detected because the inverters (at least SMA & Power One do) test insulation resistance between the dc side and earth before they start up and will not start up if the value is too low. Also the transformerless inverters generally monitor the isolation between the dc side and earth continuously so will stop and alarm if the resistance goes too low. I have not checked transformered behaviour.

I did ask a question on the IET site, but nothing startlingly new came up:
IET Forums - Running solar pv dc cables inside a house

I also asked NAPIT technical and they said earth the armour. Report of discussion in link above.

Regards
Bruce

 

[Earthstore]

May I
Even if the inverter detects a fault and shuts down, would this only protect the AC from the DC.
Surly the panels would still be producing so the fault would still be there?

 

[BruceB]

Correct, but at least you know there is a fault so can do something about it. The electric shock danger does not arise until there is a second fault.

 

[yellowvanman]

Double insulated solar cable meets 412.2.4.1(ii) by virtue of its outer sheath. So does swa by virtue of the internal sheath outside the insulated inner cores but inside the armour.

Yes the SWA has 2 layers but my interpretation was that there shouldn't be any metallic protection, because in (b) of the above reg it specifically says non-metallic.

Read the summary of your discussion with Napit guy, sounds like he was thinking on his feet and may have gone with your well reasoned argument because he hadn't come across this sort of question before!

I don't think we will get a definitive answer to this in the short term!

 

[Earthstore]

Bruce, I take your point if a sparky is looking at the system, but if you had a fault on AC, the trip would (well should) do it's job and everything downstream would be dead.

However, in the case of the DC, just because the inverter stops working, Joe public may start to fiddle, and hey presto, live DC.

Of course this is doubtful, but it is a potential.

My thought is, as before, despite which regulation may apply, avoid running DC where there could be a risk.
I like the debate though.