Why is a hot site a problem?

Hugh takes us on a deeper dive into hot sites and answers the question ‘Why is a hot site a problem?’ You can find the webinar replay here.

Hugh: Hello, everyone, thanks for joining our webinar this morning. This is a continuation from our previous session where Ian introduced the concept of a hot site and talked about how it came about. You can find a replay of the first webinar ‘Why am I getting a hot site?’ here.

And what we are going to do in today’s session is we are going to develop on from that and talk about what the problems associated with hot sites are. Talk a bit about some of the things you can do to fix hot sites when they come about as a problem.

Firstly, we are going to remind ourselves what a hot site is. We’re going to talk about the mitigations that you need to put in place on hot sites. We’re going to talk a bit about working on hot sites and the issues around that. And then we’re going to have a Q&A session. So, if you have any questions that come up at any point in the webinar, if you just put them in the chat box as a chat to everyone. It means that we can all see the questions and we will answer them at the end.

So, with no further ado, let’s just talk about what a hot site is.

What is a hot site?

A hot site is a very UK centric concept. The definition comes from an organization called the Energy Networks Association, the ENA. They define a hot site as being a site where the rise of earth potential exceeds 430 volts under an earth volt condition. Although it can be extended to 650 volts for sub substations with only high reliability circuits. So, ROEP, Rise of earth potential, earth potential rise is fundamentally the product of the earth fault current and the earth resistance of our substation. Depending on the site, we can use some split factor or some network contribution to reduce that. But it’s fundamentally simple Ohms law contribution.

So, what we’re saying is that for most substations, if the ROEP the EPR exceeds 430 volts, they are then classified as hot sites. But if we have a substation with what we call higher reliability circuits, which is generally circuits above 33 kV, with a fault clearance time less than 0.2 seconds.

Where did these limits come from? Actually, they date back to some work done by British Telecom. This fed into a couple of standards generated by an organization called the international telecoms union. Which are part of the UN. Their goal with these thresholds is to protect people working on telecoms equipment from harm, and particularly touch voltages. While it’s quite easy for us to model touch voltages, within our substation that we’re designing. It’s a lot more difficult to apply that modeling to a third-party site of unknown design. So, applying these blanket thresholds, gives us a risk management tool. Which is much easier to apply on a blanket basis.

What are the hazards?

What is the hazard we’re concerned with? Let’s dig into that a little bit more detail. These are a couple of scenarios that are from the ITU standard. And it just gives us these quite useful little graphics to get our head around what a hot site does and what this transferred potential is.

You can see here we’ve got our substation that we’re looking at substation A. Which has an earth fault that occurs and because there’s a power cable that goes from substation A, over to substation B. Some proportion of the EPR at substation A is transferring along this cable sheath to the earthing system at substation B. Generally, we would say, the EPR associated with that fault at substation A is going to be lower than the EPR associated with a fault at substation B. But without doing the calculation, we won’t know for sure. It’s always worth looking into that in a bit more detail.

You can see here we’ve got another example. We have a cable with a multiple earthed neutral, and you can see that the transferred potential is lower at our substation C, and so that can potentially be a solution you can use. Having multiple earthed neutrals often causes other issues so as not a common solution for these transferred voltages.

Transferred voltage

Another really common example of where these transferred voltages can become a problem is where you have a transferred voltage to your secondary distribution substation. And then that secondary distribution substation then has a combined MV LV earthing system. What this potentially means is that some proportion of your HV EPR, is transferring all the way down this MV cable screen into the secondary substations earthing system. And then on into some member of the public’s house into their low voltage installation. That’s something that we’re always going to try hard to avoid because we don’t know the quality of the electrical installation in that third party person’s home. We don’t know if the appropriate measures are in place to keep them safe.

Now, here’s an example of a telecoms installation. And this is where this standard particularly comes into play. You can see where the EPR occurs on the HV system, it can not only be transferred through the power cable. But if you’ve got a telecom system with a shared earth, it can also be transferred over the telecoms cables. So again, you can see there’s really high EPR transferred through the power system, and also through the phone system. And this can be really problematic if you have more significant currents flowing in the telecoms cable. Those telecoms cables often aren’t designed to deal with higher levels of current.

Here’s another example where you have a shared overhead line where the telephone cable and an MV or LV, overhead lines share the same structure. And so that they can be bonded together at each of the neutral locations. And in that situation, there’s an even larger potential for hazard because the telecoms cables are connected together more often.

How to we mitigate the hazards?

How do we mitigate this hazard? Well, there’s a couple of things we can do that essentially fall into two categories. So, the first one is the reducing the EPR. So, either we reduce our Earth resistance, or we reduce our Earth fault current. Or in a number of cases, we can reduce our fault clearance time. Push our substations that high reliability category, and then potentially not have a hot site to deal with. And for those of you who’ve been on health and safety courses, you will probably have heard about the hierarchy of controls. Where we start by reducing the hazard or where we can eliminate the hazard entirely. Then we reduce it severity and only once we’ve ruled out measures that do that, do we consider engineering controls and PPE because they’re much less effective in managing that hazard.

That’s this second category of controls or we talk about separation of Earths, isolation of services and counterpoise earthing. Where we talk about separation of Earths, that’s where we have a segregated, high voltage and low voltage Earth, where we may be separate them by a few meters. Isolation of services, there’s a couple of approaches we can use, which we’ll dig into in a minute. And counterpoise Earthing is where we reconfigure the earth electrode to take the EPR contours away from an area of concern.

Modelling

What I would say is that the common thread among all of these is modeling. There are all sorts of rules of thumb out there that have been used back when modeling was a whole lot less accessible and less effective. But in 2022, when modeling is pretty easy and pretty accessible, I would suggest that if you’re trying to mitigate a hot site. Modeling should always be your first port of call to work out the best way to control that hazard.

Reducing EPR

Let’s dig into reducing EPR. Fundamentally, what we can do is we add more earthing. We add more earth rods, maybe we add a horizontal earth electrode. In certain rocky geologies there are commercial products you can use like concrete, rods, which can be effective in certain specific scenarios. But unfortunately, in earthing there is no substitute for just adding copper. While these commercial products are more effective, in some situations, there’s no substitute for just adding more earthing.

In some scenarios, we can use any neutral earth resistors. And we can reduce our fault currents. It might be possible to review our protection settings and reduce that fault clearance time to improve the threshold at which hot site occurs. But that’s the reducing section. Often in hot sites by the time you get to doing the earthing design, a lot of these decisions have been taken. And the cost of reducing the EPR further can be prohibitive. This is when we look at putting into play these engineering controls.

Separation of Earths

The first one everyone talks about is separation of Earths. Commonly in distribution substations, you’ll find that they’re configured like this. Where your transformer star point and your MV earthing system are shared, they’re combined.

And this tends to be a preferable solution because it’s more cost effective and it’s more reliable in terms of the longer-term operation of the substation. Well, it’s very easy to build a substation with segregated Earth’s. If you then reconfigure the site a few years down the line, the records that indicate how that separation was put in place may be lost, and that separate segregation may be compromised. I always recommend where possible to stick with combine earth, because it’s a much more reliable long-term solution.

MV & LV Earths

But where that’s not possible most MV distribution switchgear has a fairly simple capability to segregate the MV and LV earths so that they’re not connected together and the MV faults potential is not transferred onto the LV system.

Let’s have a look at that in practice. This is a little snippet from one of UK PN standards. It shows the MV earthing system with the bare horizontal electrode around the perimeter of the substation, and a handful of earth rods, and then you’ve got a transformer here and that feeds the LV panel.

You can see we’ve got our insulated earthing cable that goes all the way out to this standard separation distance. That’s typically quoted among the UK DNO is or somewhere around eight meters. Then you start with your first rod. Then from that first rod, you go into bare conductors, because obviously, that’s a bit more effective, it’s a bit more earthing. I always prefer to model that segregation and make sure that our LV Earth is genuinely beyond that hot zone and is not going to be affected by this high potential. But I know for a lot of people that’s a step too far, they’re quite happy just to stick with this, eight meters is enough. And it can often be that insights which are dense doing the modeling may allow you to achieve a smaller segregation and still achieve safety.

Isolation of services

Another measure you might need to put into place is isolation of services. If you have a BT phone line coming into your site, and you’ve declared to them that it’s a hot site, you will have to ask them to come out and fit one of these types of devices, which is called a line isolation unit.

This is quite an old model. I think they’re on Model 12. Now, but they all look fairly similar. Essentially, you have your phone line comes in at one end, there’s some electronics that isolates the power supply, and also isolates the phone line completely say that there’s no electrical connection between your site electrical system, and BT’s electrical system. This ensures that there’s no risk of BT’s staff being hurt when they’re working on phone lines downstream of you, for example, in the telephone exchange.

Similarly, if you’ve got gas pipes, perhaps you can use measures like these insulating flanges if they’re above ground. But to my mind it’s always preferable to have a larger insulating section in a pipe. Here’s an example where we’ve got a plastic pipe and that completely mitigates the hazard. There’s no risk of transfer of potentials at all. But that’s not always possible. You might have a hot substance that can’t be transported in a plastic pipe, for example. So perhaps there might need to be other measures you’ve put in place.

How do we manage a hot site?

What we’ll do next is we’ll look at a quick example project of how we managed a hot site for a utility customer of ours. This is a project where there was a solar farm constructed in north Wales has water pipes running through and nearby the site. There’s a combination of asbestos cement and cast-iron pipes.

You can see here we built this quite large model. We’ve got the solar arrays, the solar array earthing conductors. These will connect back to the DNO substation here, which was the location that was a hot site. And then we built a model that included these water pipes that ran through the site in a couple of locations. And this third one that that passed nearby.

Using this model, we were able to understand the potential on the earthing systems when they were interconnected and consequently to see the transfer of potential from the earthing system of the substation on to these third-party pipes.

Potential on Conductors

The model allows us to see quite visually how big a problem this is going to be. You can see from this colour plot, what we’re plotting here is the potential on the earthing system.

The red colour is a higher potential. You can see that the earth potential rise is about 2.4 kV, on the earthing system. And in a minute, we’ll show you what the potential impressed on to the pipes was.

EPR Colour Plot

We can also use our modeling software to plot the contours that are of interest. So, this is what we call it a colour plot. It’s just a range of colours depending on the potential, we can also plot the specific hot site contours of interest.

Hot Contours

You can see that we plot the 430-volt contour here, which is shown in pink. And then within the site, you can see we’ve got this red line here. And this red line is the 1150-volt contour.

If you remember back to our last session, the 1150-volt contour was the contour which action was necessary. If we go back to that line isolation unit, BT/Openreach say that you need to tell us it’s a hot site above 430 volts. But you don’t need to do anything about that hot site until the EPR reaches 1150 volts. At 1150 volts, you need to fit that line isolation unit.

Let’s have a look at some key data outputs from this modeling. We have our 33 KV fault current of about 3.5 KA and the EPR. The voltage at the point of fault is about 2.3 kV. But the highest voltage on the asbestos cement pipe is about 1400 volts. So, you can see there’s already quite a big drop in potential just because of where the fault location is compared to where the pipe is. And the cast iron pipe, which is the furthest to the right in the model has a much lower potential impressed on it again.

If we go back to this potential and conductors plot, we can actually see that even though within the site, we’ve got this really high potential as you move away from the site and drops off really quickly. So, if we look at where the conductors end, you can see the potential is actually only about 100 volts or so.

What were the conclusions?

What were the conclusions we’re able to draw on this? Well, yes, we’ve got this high potential, there’s a hazard within the solar farm itself. But beyond the site, the risk to people working on the pipes is very low. So, the fault clearance time for that 33 KV fault was about half a second. Our touch voltage limit was 578 volts. Assuming there was no potential transfer between the pipe and the surrounding soil, beyond the sight boundary there’s no chance that someone working on the on the pipe would be exposed to a hazardous touch voltage. But if people are working on the pipe within the array itself, we might need to look at some measures to protect them.

We did recommend providing some isolation to the pipe, the boundary of the hot zone that will just be inserted in the plastic sections within the pipe so that that potential isn’t transferred into any electronics on the pipe or anything that might be a little bit more sensitive. We did show that even though we had this big red ticket problem beyond the boundary of that solar array, there wasn’t a particular hazard to the utilities staff. That’s an example, I suppose a practical example of what a hot site can do.

On that note, I think we can open up to questions.

Q&A

Does a hot site apply to other countries?

Ian: Thanks for that Hugh. There’s one that is coming through from South Africa, which is, “does the hot zone or hot site concept apply to other countries?”

Hugh: It’s a little bit more complicated in other countries because everyone has different figures for these thresholds. There’s a section in EN 50522 which is specifically around when you can combine the HV and LV Earths. Which is around touch voltage thresholds and so that applies to all the countries that use EN 50522. So certainly, all of Europe. I don’t know specifically for South Africa, I think they use the IEEE standard for earthing. I don’t remember exactly what it says. You will find that in most countries, there will be something around transferred potentials but the particular figures I’ve talked about today, are very much UK centric.

Ian: Okay, chaps, are there any other questions?

Global Earthing systems

Hugh: Yeah, there’s ones come in about global earthing systems. This is asking for comment on global earthing systems. It’s generally something that’s positive where you can actually have one. So a great example of a global earthing system is a city center. Where you’ve got lots of large buildings, large consumers of energy, all of whom have a little substation in the basement. All of these substations are connected together via the cable screens. You have this large continuous earthing system that covers a huge area. And so, in that situation, you can see that by having everything at potentially bonded, your substation on its own doesn’t have to have a particularly extensive earthing system because the surrounding substations are all going to contribute when there’s an earth fault.

I would always caution against using it as a get out clause for doing calculations. I always want to prove that as a bare minimum, the steps and touch voltages within my site are compliant EN 50522. As a concept, common bonding and global earthing systems are a really good way of achieving safety. Not just for smaller site but for much larger sites like oil refineries. Having a big common bonding network is a really good way of reducing the risk to safety from high voltage.

There’s a good point in about those line isolation units, which is that when you have one fitted, it needs a power supply in order to operate because it’s fiber optic. If you don’t have a local power supply, you won’t have a phone unless you have some sort of battery to protect that supply.

Ground return currents

What have we got next? “I’m after another session on ground return currents.” Yes, that I think you might find there’s one in our previous library on that. But we will certainly add it on our list as one to cover in the future. I know that’s a really interesting topic it definitely affects your hot site classification so it’s quite important.

Ongoing verification of a earthing system

Another question has come in, which is “once earthing system is agreed with a network operator, for example, for a solar farm, does it require ongoing verification?” It’s a little bit of a complicated topic. There’s nothing in any of the standards that specifically says you must test on this frequency. It’s quite common for people to use the five-year figure that people use for fixed wiring inspections as a starting point for your earthing inspections.

What I would say is that it really depends upon the risk profile of your site. And so, if you’ve got a big high voltage substation, city center, you might want to look at testing that more frequently as the risk to the public is quite large. Equally, if you’ve got a large oil refinery, where you’ve got potentially 1000s of people working, having regular assessments of your earthing system is probably a valuable thing to do. In comparison, if you’ve got a solar farm that’s maybe visited once a year, that probably will need less regular testing. The risk of harm to others is a lot lower.

This is what a lot of the electrical testing standards are saying these days is that there is no cut and dried frequency for inspections. Instead, you should be doing a risk assessment as a designer, as an asset owner to say, what do I know about my earthing system? And what is the risk profile of my site? That’s what’s going to drive if and when you do your earthing, testing or verification.

Ian: It’s a really interesting one, because that’s one of the most frequently asked questions over a typical year. Maybe we ought to put something up on an FAQ page for that as it’s a really good question.

How do you know if a global earthing system is achieved?

Hugh: Yeah. Another one that’s come in “how do you tell if a global earthing system is achieved?”

Because the definition in EN 50522 is very broad and to be blunt, it’s very difficult to prove one way or another if what you have is a global earthing system. This is why I’d always encourage modeling to prove your touch and step voltages are compliant. Because what you can then do with your model is you can say, if we go back to this EPR colour plot, here’s a good example of what you might call perhaps a global earthing system, because everything within this boundary fence here is pretty much equipotential. So, you can say that, probably call that a global earthing system and not have to worry too much about step and touch voltages within the solar array. Because as you can see, the potential gradient is quite low within the site.

Particularly if we were in a location with much lower soil resistivity. You’d see that the gradient of potential would be even shallower it is it is definitely a difficult thing to define. It’s why I tend to recommend against just using global earthing system as a get out from doing modeling. The problem an engineer has is that with every design we produce, we have to be able to stand the court of law and say, I did everything that was reasonably practicable to provide a safe site. I’m not sure the going, I thought it was a global earth system would be enough to satisfy a jury. So, for me, I’m always going to do the modeling. But your risk profile as an organization, as a designer, is something that you’re going to have to decide for yourself.

Earth fault currents

The next question that’s come in is around the separation. “How do you avoid circulating earth fault currents of someone accidentally touches both MV and LV equipment?” The simple answer is you separate it. You put your LV in a different run. You ensure that it’s not possible to simultaneously touch an MV and LV earth system simultaneously, you put a brick wall between it, you put a GRP fence between it, you put something nonconductive in the middle. So, there’s no practical way that someone can touch both simultaneously.

Okay, do we have any more questions, Ian?

Ian: No, I think that’s it for this morning. It looks like unless there is anybody else.

Hugh: Well, thanks, everyone for joining us today, thanks everyone for listening.

Ian: Excellent, thank you Hugh and thank you everybody in the team behind the scenes. We will look forward to seeing you next month and another exciting little webinar coming your way. Cheers for now.