Neat enough job and probably technically compliant, but is it sensible to put the DC side conductors in metallic trunking and conduit even if they do meet the class 2 or equivalent requirement?
I am aware that there are SWA cables specifically for solar PV, so I guess there is not much difference if the SWA is earthed. However, on a slightly different point, if these SWA DC cables are buried, what is the point of earthing the wire armour?
Well from an EMC perspective, not having wires exposed so they can radiate is a very good thing, and the insulation is less likely to get damaged on the inside of a steel box than running the outside, as I assume the trunking would be needed anyway for the LV stuff. I also assume the batteries are themselves in an earthed metal box, this just extends that protection.
The only thing I'd want to be sure of is how the DC cables leave the steel tube, to be sure that is not going to chafe.
Mike.
Good points, by default, DC conductors in metallic trunking is standard for the majority of commercial installs, mechanical protection of the cable being the primary driving factor with little concern given to the risk level of insulation breakdown. On the point of buried SWA DC cables, most cable runs will have a portion that's touch accessible either at the inverter or panel end, which could be classed as a potentially exposed conductive part, hence the earthing consideration.
Just thinking aloud...
A lot of the early regs on earthing talk about preventing metalwork becoming "charged" - these days with almost everything a.c. we tend to think of that happening either because of insulation breakdown or capacitive coupling as any tiny leakage of a.c. currents will discharge on the next half cycle as easily as it charged on the first half-cycle. With d.c. power though I guess there's a potential for a charge to slowly build up on the unearthed metalwork over a very long time just due to less than perfect insulation (it may be many mega Ohms, but it's still not totally perfect) - if the metalwork is substantial then it could hold a reasonable charge (just like a capacitor) - maybe enough to give a jolt if it discharged to Earth via human contact?
- Andy.
is it sensible to put the DC side conductors in metallic trunking and conduit even if they do meet the class 2 or equivalent requirement?
What's your thinking, why do you think that would be a problem?
When I was researching into a home solar installation, one of the topics that caught my eye was inverter design, that the older/early inverters used transformers and the DC side would, typically, be galvanically isolated from the mains/AC side of the inverter, so the DC side was a relatively simple/pure DC circuit, unearthed or earthed in some way - earthing one pole, resistively earthed with monitoring etc.
The later designs of inverter, virtually all of the ones sold today, are transformer-less with the DC-side connections and all of the panels, cables etc connecting into the inverter's switching circuits which are themselves then connected to the mains/AC side, so the DC-side connections are not galvanically isolated from the incoming mains/AC, and depending on the design of the inverter, the DC-side will be at some fraction of the incoming AC voltage.
So with the DC-side not a simple isolated DC circuit, but actually a DC circuit with an mains AC potential from the inverter switching super-imposed on top, it then raises issues around capacitive coupling and leakage, particularly on the panels themselves and why its advised/required to earth these, otherwise their frames could also have an AC potential appear, with respect to earth, posing a hazard to anyone making contact with the panel metalwork. I understand this is one of the reasons why inverters typically have such high leakage currents.
For metallic containment, I can see the positive of a continuous metal covering, reducing likelihood of contact with a damaged cable, providing a return path to earth to trip the inverter and the non-combustible metallic covering providing some containment for a fire/release of energy. But, depending on workmanship, sharp edges cutting into the DC-side cables, it could increase the likelihood of a DC-side earth fault, or a short-circuit, from pole-to-pole, or pole-earth-pole on the DC-side . You would expect the inverter to detect such a fault and shut-down, but the solar modules would continue operating, feeding current into the short-circuit (unless fused close to the panels).
could hold a reasonable charge (just like a capacitor) - maybe enough to give a jolt if it discharged to Earth via human contact?
That is 'static electricity'.
One can estimate an upper bound for any given situation. The capacitance of something isolated to free space is always less than that of the sphere that surrounds it, and the capacitance to local ground can be estimated as a series of oddly shaped parallel plate capacitors, To give a feel, people are typically 100pF or so, the more ah 'corporeal' (being more like that ideal sphere) perhaps a few hundred, a car may be 1000pF and a lorry a few thousand.
The energy on a capacitor is 1/2 CV^2 so on 1000pf you don't get that much energy per kV
So how much hurts ? a whole joule is a bite like an electric fence, and the twitch/ startle response can be enough to cause injury.
less than 0.1J is still an 'ouch- bleep' but not usually a problem, and around 1mJ or less (0.001 joules) is the sort of crackle from combing hair vigorously or putting your socks on on a dry day and unremarkable.
Having said that, note that under the right conditions, petrol vapour can be ignited by about half a millijoule, and methane (domestic gas) about half that again.
There is a reason for the earthing worries near petrol pumps and anti-static flooring in places handling volatile chemicals.
Generating that level of stranded charge accidentally from a DC supply can be very much a problem with kit that throws a lot of voltage, like particle accelerators X ray machines and so on. Luckily this is only normally found in bigger hospitals, university research sites and a few 'ministry' type places that are quite coy about what they actually do... In such cases a gentle bleed down, with braids and metallic or carbon fibre brushes bridging between places where you don't want a voltage to build is the solution. I have seen sparks drawn from metal steps left on a rubber mat before...
However, at voltages below a thousand, as currently solar systems are, the capacitance needed to pack a dangerous punch requires charging an unreasonable volume of unearthed metalwork. A long length of floating armour at the odd kV could I suppose be bad news. (perhaps 50uJ per metre. I'd be wary of 100m length of SWA being enough to ignite a vapour, but not to deliver a serious shock .)
The other place where problematic charge build up happens, that is probably more familiar, at least to factory sparks, is where different insulators are constantly making contact and separating - rubberised conveyor belts and plastic rollers are a classic, but things like paper mills also have this issue. Here the voltages can be tens of kV and up, and will both damage electronics and certainly make people jump if not carefully managed.
Mike
What's your thinking, why do you think that would be a problem?
I wasn’t thinking safety. I was thinking that using metal trunking/conduit might increase risk of insulation faults which in turn could cause the inverter to drop out or lose grid connection. That might have financial implications, especially if, like this system, no one goes near it for months on end.
I was thinking that using metal trunking/conduit might increase risk of insulation faults which in turn could cause the inverter to drop out or lose grid connection. That might have financial implications, especially if, like this system, no one goes near it for months on end.
Cable faults are just one potential fault cause, there's various other possible faults which could occur. If the system is going to be unattended, then perhaps set-up some form of remote monitoring, either via Wifi/LAN and the inverter manufacturer's cloud platform, or a more traditional approach, see if the inverter has any alarm outputs or volt-free contacts which could be wired up to provide an indication.
if a cable that is supposed to be insulated well enough that it is perfectly safe to touch and generally handle while live, can be made more likely to fail by earthing the metal nearby, by any mechanism other than mechanical damage / abrasion, there is something very much below par with that insulation...
So long as there are grommets/ glands or bushes in all the right places, cables in trunking should be the best solution,
Mike.
but the solar modules would continue operating, feeding current into the short-circuit (unless fused close to the panels).
Fusing PV output for d.c. fault protection can be tricky - as the available current is limited by what the panels can generate. Rate the fuse so it won't blow in normal service and it's very unlikely to blow given a short as the available current, even during bright sunshine, won't be much larger (max 1.2x comes to mind from somewhere, but I may have imagined that).
- Andy.
Cheers Mike - it's good to be able to put some numbers to an idea!
- Andy.
We're about to take you to the IET registration website. Don't worry though, you'll be sent straight back to the community after completing the registration.
Continue to the IET registration site
