Had the DNO out to a service head issue a few years ago and got to talking about earthing and he mentioned a job he got called out to by United Utilities the customer complained to UU because her cold water was coming out of the tap quite warm, turned out that one of the supply cables close to the water main had a fault on it and it caused the metal water pipe to heat up and warm the water
Consumer units for solar ...
- Thread starter Skipper
- Start date
Help Support Talk Electrician Forum:
Of course, as you leave the bonded area that is when the danger occurs, but, that can happen anyway in a T-N-CS situation ... the well known "cow shed" problem.
It might make sense to bang some earth rods in as recommended by 411.4.2 anyway ... with earth electrodes at the customer premises to support the T-N-CS system rather than converting it to T-T ... whether it will be possible to get an earth impedance value within the limits of table 54 I don't know ... as I said, it is sand around here ... I suspect in summer when it dry as a nun's chuff may prove problematic!
It might make sense to bang some earth rods in as recommended by 411.4.2 anyway ... with earth electrodes at the customer premises to support the T-N-CS system rather than converting it to T-T ... whether it will be possible to get an earth impedance value within the limits of table 54 I don't know ... as I said, it is sand around here ... I suspect in summer when it dry as a nun's chuff may prove problematic!
From past experience of troubleshooting supply issues, one of the common things I come across is the just bang an earth rod in and all will be fine problem
There always seems to be a total lack of understanding and education these days when it comes to installing earth rods / earthing systems and testing them. Even some of the "specialist" lightning conductor companies seem to be lacking having recently looked at the work of one on a local church who couldn't even test the LPS properly and when it was pointed out to them by the church warden they sent out their compliance / training manager who agreed with my findings and mentioned to the church warden that he needed to go and amend the company training manual, a few years ago this company added an extra earth rod to the LPS without any consideration to it's siting in relation to the existing rods and earth mats
The problem with "supporting" the TN-C-S system is it can hide a failure of the PEN and as I and a mate of mine who is also an electrician have found the property can become the earthing point and neutral for a number of properties connected to that DNO supply line which then raises the issue of neutral currents flowing through the property with the earth rod from other co-located installations / properties
If the ground is sand it is unusual to have a TN-C-S supply for the reason you mention, one of the DNO's local to me won't allow TN-C-S supplies in certain parts of their patch because of the ground being too sandy
The installation of earth rods / mats / discs / systems is IMO a subject all on it's own and with rods the assumption that a 4ft rod is all that is needed for an installation to be earthed is in a lot of cases IMO a demonstration of ignorance and the acceptance of some of the impedances that some find acceptable without taking into account the ground conditions at the time can in a lot of cases increase due to the ground drying especially during the summer months
When installing earth rods I generally take more rods than I think I need and always use the extendable type if I can get 8 or 12ft of rod down then I'm likely to get a low impedance if it is likely to need more than one rod I usually stick to 8ft so it is easier to properly co-locate the multiple rods
The problem with BS7671 there is no retest period suggested for earth rods or installations with earth rods and the maximum impedance values make no mention of possible seasonal changes of the measured values whereas BS62305 for LPS suggests an 11 month retest period to take account of the differing ground conditions
Guidance note 8 makes interesting reading
As always, there are compromises. By following the recommendation of 411.4.2 you are not going to make the situation any worse ... all the earth system would be doing is providing some additional protection in the event of a (hopefully rare) PEN failure ... I can see the transformer from here happily balanced on a pole in the corner of the adjacent field, there are limited opportunities for a PEN failure, and an additional earth point is not going to make the situation worse ... and even if in some way it does, by following the recommendations of ammendment 2 you are at least on the right side of the law.
It is worth remembering that 114.1 tells us that the connection between N and Earth (with a capital E) is deemed permanent, so basically it is DNOs problem, all an additional Earth does is provide a bit of non-obligatory protection should the DNO fail to do that.
In practice, all you can hope to achieve with an earth rod or mat is in the event of a PEN failure, to mitigate risks, so if you are in the habit say of standing in a puddle, leaning in through the kitchen window with your hand on the kitchen tap, you might want to put the earth rod near the puddle. If you achieve the values given in table 54, the RCD may even still trip as it should. Thankfully our garden tap is on 20 metres of plastic pipe, before the supply gets to the house. Beyond that, there is limited potential (ho ho) for serious harm in what should be a hopefully very rare event of PEN failure.
It is worth remembering that 114.1 tells us that the connection between N and Earth (with a capital E) is deemed permanent, so basically it is DNOs problem, all an additional Earth does is provide a bit of non-obligatory protection should the DNO fail to do that.
In practice, all you can hope to achieve with an earth rod or mat is in the event of a PEN failure, to mitigate risks, so if you are in the habit say of standing in a puddle, leaning in through the kitchen window with your hand on the kitchen tap, you might want to put the earth rod near the puddle. If you achieve the values given in table 54, the RCD may even still trip as it should. Thankfully our garden tap is on 20 metres of plastic pipe, before the supply gets to the house. Beyond that, there is limited potential (ho ho) for serious harm in what should be a hopefully very rare event of PEN failure.
If your property is the only one off the transformer then the risk should be minimal, adding an additional earth rod connected to the MET will also connect to the neutral on the service head so you will have neutral currents flowing down the additional earth if a PEN break occurs if multiple properties are on the same supply and only one property has an additional earth connection depending on where a PEN break occurs the neutral currents of a number of properties could flow down the additional earth connection
What you actually need to remember is there is no obligation on the DNO to provide an Earth point but if one is provided they must maintain it
What you also need to remember is the pertinent point in 114.1 which is the fact it references to the ESQCR of which section 8 deals with the connection of the neutral conductor to Earth at the voltage source (the transformer secondary centre point), that connection is made to ensure that the neutral conductor reference remains fixed wrt Earth and does not float when the load on each phase of the transformer is not balanced, this is done whether the supply is TNS, TNCS or TT, on overhead TNCS supplies if you look at the poles along the supply from the transformer ISTR every third pole has an earth rod at the base and cable running up to the neutral on underground TNCS supplies each joint has a short length of copper cable connected to the neutral hanging from it for the PME
Not sure of the point you are trying to make
I assume you are actually referring to table 41.5 and the footnote does suggest that the value should be as low as possible and is not considered to be stable if the value exceeds 200Ω. The reality is sub 50Ω is not normally too difficult to achieve
Not sure what you are suggesting here unless you are trying to highlight fault voltage degradation at the earth rod
Last edited:
I've not read the complete thread, but you are wrong on the above statement*. I currently have 11.5 kWp hanging off a 8kW Quattro inverter, well strictly speaking the panels hang off the MPPT solar charge controllers.
It's unlikely due to orientation and pitch that the panels will produce the full power, but even if they could the SCC controllers will only draw what they require (actually what the batteries and inverter require), and no more. If mine do exceed 8kW in the summer then I will balance generation by exporting some and using some to charge the batteries, thus increasing maximum generation.
Due to the way Victron system works, the inverter can have far more solar connected than its rating suggests, and lots of other inverters can be over paneled, you just need to check the specs. Over paneling boosts winter production.
My system gives full house back up, I have a 3.68kW SolaerEdge inverter, 8kW Quattro, 29 kWh of batteries, and 15.5kWp of panels.
*I may have misunderstood, if the 16kW is the load from the showers, then the system will just draw the excess from the grid, unless off grid, then the inverter will shut down due to overload.
Victron a very different to most inverters available. Not many people but them due to the cost. The op was thinking of putting 16 KW of panels on a single 7.5kw inverter.
Undersizing inverers can most certainly be done, there's a general concept around that undersizing inverters by around 20% is acceptable. Trouble is that idea used to be bandied around when inverters still had wire wound transformers, the efficiencies of which could drop off to around 70%, at lower power levels. Inverters these days are transformer less, and the efficiencies don't drop below 92% at low power levels, and run at around 97% the rest of the time. So undersizing tends to achieve very little these days, and I don't see the point of strangling the outputs of the panels, especially as most are +ve tolerance. The real issue will be the string voltages not power as such. Before any string starts up, you tend to get full Voc on the panels because they are in stalled condition. So if you have too many panels the string voltage can be too high for the MPPT, and the inverter will refuse to start up to protect itself from damage.
You can also get some interesting weather influences on solar such as 'cloud edge effect' that can cause a brief spike in power and shut inverters down.
Many in the industry still like undersizing inverters, but that's mostly due to maximizing profits, smaller inverters being cheaper of course. For me, it's an outdated concept for outdated inverters.
Undersizing inverers can most certainly be done, there's a general concept around that undersizing inverters by around 20% is acceptable. Trouble is that idea used to be bandied around when inverters still had wire wound transformers, the efficiencies of which could drop off to around 70%, at lower power levels. Inverters these days are transformer less, and the efficiencies don't drop below 92% at low power levels, and run at around 97% the rest of the time. So undersizing tends to achieve very little these days, and I don't see the point of strangling the outputs of the panels, especially as most are +ve tolerance. The real issue will be the string voltages not power as such. Before any string starts up, you tend to get full Voc on the panels because they are in stalled condition. So if you have too many panels the string voltage can be too high for the MPPT, and the inverter will refuse to start up to protect itself from damage.
You can also get some interesting weather influences on solar such as 'cloud edge effect' that can cause a brief spike in power and shut inverters down.
Many in the industry still like undersizing inverters, but that's mostly due to maximizing profits, smaller inverters being cheaper of course. For me, it's an outdated concept for outdated inverters.
Yes, I had fun finding the right panels for my last upgrade to stay within the voltage limits with the Victron MPPT I'd be using. The biggest problem is voltage peaks at low temperatures, lower the temperature higher the voltage.
Also voltage drop on PV strings is another thing that needs careful consideration, and a good understanding.
Many inverters can now be over paneled more than previously, but it's paramount to read the specs and understand them.
Also voltage drop on PV strings is another thing that needs careful consideration, and a good understanding.
Many inverters can now be over paneled more than previously, but it's paramount to read the specs and understand them.
think they all can be undersized to a point, but I also know from my customers comments how frustrating it is to have a set of panels choked by a smaller inverter - not systems I've installed I hasten to add. One of my customers panels, nominal value 2.7kw will actually output just over 3 Kw on a good day - Japanese panels. TL inverters are so efficient my own runs down to 7w before shutting down, and over that's 10 years old now.
That is exactly correct. You can hang a fair bit more off them, and providing the string remains within the maximum *voltage* rating of the MPPT, you are fine. The MPPT is simply a buck converter, it can adjust the current being drawn from the string to maximie the power, but it can also reduce the current being drawn and let the voltage rise if the output reaches the maximum permissible current for the string.
You only need to peruse the documentation for most inverters to confirm this is true, the Solis 4.6k I am looking at for example has a recommend max PV array of 8kW ... providing you stay within the max PV voltage of 600V you really could put *any* size array on it, it will just adjust the PWM ratio of the buck converter to not draw more than the max rated string current.
