Once I had a situation where the length of the ground rod mattered.
We were building a fusion reactor prototype, and as far as I am aware the NEC code doesn't really have a section for that. I'm sure that much of what we did was probably against best practices.
We had a number of high voltage power supplies running in one room of our lab, and on days when the humidity in the air was low, the ionization in the air would charge up all the metal surfaces in the room. So, even if something was not connected to a high voltage, you would get a nasty shock when you touched it. All these devices were connected to the normal, "third prong" wall socket ground, but apparently the route through the building to ground was too resistive to handle the large amount of ionization in the air.
When working in a room with 180,000 volt power supplies, random shocks rapidly become more than annoying.
Spence, our electrical engineer, had the idea that the quickest way to fix the problem was to run a grounding rod into the soil right outside the window. We pounded a rod into the ground with a sledgehammer as far as we could, which was about 5 feet, I guess. Then we ran a 1/4" copper cable from the rod, through the window to connect to the main reactor chamber.
This worked great! At least for a couple of months. Our lab was located in a swamp, and the soil was basically a few meters of dried leaves. You could actually feel the soil move up and down when you walked on it. When August came around, it had not rained for quite a while, and the water level in the soil had dropped below the level of the grounding rod. The dried leaves in the soil insulated it quite well. We started getting shocks again, and it took us many hours of troubleshooting before we realized that the ground was not actually ground anymore.
We went out and pounded the rod a few feet deeper, which was easier now that the soil was dry. This reached the water table, and fixed the problem.
Also helps if you dump a large quantity of salty water. The salt eventually leaves too, but helps retain some of the water and makes it more conductive.
Exactly. We had the first generation color copiers in the California high desert and whenever paper jams became frequent we knew it was time to pour salt water on the building grounding rod!
I was building a workshop last year and as I have few old 3-phase metalworking machines and the power comes through a 50m (150ft) buried cable I had to install proper ground for the building and another two independent ground connections for the lightning protection. I have no bedrock, but my soil is pretty much 100% clay plus a football sized stones here and there around the building. The code around here is that the ground connection should be no more than 4 ohms resistance. I was lucky with two lightning protection grounds. For one a single 2.4m rod was enough and for the other two rods (I had the type that screws into each other so you can make the rod longer and longer if necessary) were required.
However, I wanted to locate the electrical ground few meters away to avoid the possibility of a lightning strike raising my ground potential. I ended up having to install 6 rods to get down to 4 ohms. The first two rods were 7 ohms, the next two 2 meters away were 15 and the final two were 25 if I remember correctly.
The take away from that story is: always measure your ground resistance. Never assume 8ft is enough if you have a certain type of soil.
How exactly is that measured? If you’re trying to measure the resistance between the ground rod and actual ground, how do you get the second reference point?
There are two standard ways to measure. The meters that do the measurement usually support both. First method involves two mobile shallow ground electrodes you place 20m away from your ground rod (so electrodes are 40m apart and your ground rod in the middle). I'm not sure how exactly it is possible to measure this when the quality of ground those two extra electrodes is unknown, but it works. The meters cost few thousands $ so I assume it is not trivial.
The second method involves an already known good ground - like for example one supplied by the electricity company. Here they usually bury galvanised steel thick tape along their ground cables. That tape is hundreds of meters long and underground so it is assumed to be a really good ground. It is only used to ground outside metal bits of the power delivery equipment that are likely to be touched by people and it is separate from the ground connection supplied in the cable itself although one is supposed to connect it to the delivered ground when connecting the house cable to the electricity delivery point. Then every house is supposed to have at least two lightning protection grounds and an extra "electricity supply" ground that is again connected with the ground in the supply cable. Please bear in mind I'm no electrician and all of the above is based on me observing electricians at work when they were connecting my house, my workshop and working on lightning protection for both. I may be mistaken on some details.
L.A. County made us drive FOUR 8 foot ground rods into the ground, no less than 8 feet apart, for our ground-mounted 13 kW solar array.
This is ridiculous beyond description and it is NOT in the N.E.C. at all.
The problem you run into with places like L.A. County is that incompetent plan checkers just want to play "cover my ass" and make up rules. There is no practical way to challenge them. They are the absolute authority (unless you have millions of dollars and YEARS to fight them).
I checked with contractors and the recommendation was "just do it and don't complain or they'll make your project a living nightmare".
So, we did.
They still made the project a nightmare to complete. They added a massive chunk to the overall cost and it took six months longer than it should have taken to complete it. Their interference made it such that this solar system will likely never recover the construction costs. Well, maybe in 30 years. It's a shame. Had I known the nightmare it is to deal with this bureaucracy I would have invested my money elsewhere.
California is sucking more and more with time. L.A. County is even worse.
It depends on the impedance of the soil. If impedance is too high, you need additional and/or longer rods, spaced accordingly.
The code says MINIMUM of 8ft in contact with soil. MAXIMUM earth ground impedance of 5 Ohm. So you need to keep driving in rods (or use an alternate grounding system) until you are below 5 Ohm - hence your multiple rods.
I don’t recall the NEC saying that. My recollection is that code is two ground rods, spaced 6+ feet apart, is sufficient for code without measuring resistance at all.
Alternatively, if you can prove that a single ground rod has less than 25 ohms to earth, then a single rod suffices.
If I have time later, I’ll try to look up the actual reference, but I remember reading and thinking “that makes no engineering sense!” so it stuck in my head.
Its actually IEEE and NFPA that recommend a maximum of 5 Ohms for the grounding system.
NEC state less than 5 Ohms for sensitive equipment, 25 Ohms otherwise. (NEC 250.56) - but as you mentioned, the measurement is only taken from first rod and a second can be inserted without testing (so it could be over 25 Ohms still, but based on the code it doesnt even matter....)
As an aside, its 200 Ohm in the UK (but it HAS to be measured - cant just blindly drive 2 electrodes and call it a day).
This discussion triggered in me the question, "Obviously, I can't just stick a meter probe in the ground [as I'd be measuring the dominant resistance of that probe rather than the grounding rod in series]. So, how would I reasonably measure the resistance of an earth connection?" I came up with the (mildly hazardous) solution of putting a 120VAC line from the power company supply down the ground rod and measuring the current flow back to the PoCo (grounded) neutral conductor and then went googling.
> It depends on the impedance of the soil. If impedance is too high, you need additional and/or longer rods, spaced accordingly.
I did the calculations. One rod would have done it. Two if you wanted redundancy. The plan checker wanted four. Want to know why? He wanted one next to each post. As simple as that.
My county also has a bureaucratic nightmare for a building permits department. I avoid interaction with them however I can. Sheds less than 120sqft do not require solar permits, so I am trying to cram 1.7kw of panels on to my garden shed. Also looking at getting another few kw mounted on a wheeled trailer which the building dpt does not have dominion over.
They want $1000 for a permit which would increase my payback time by at least 3 years.
I closely follow the relevant codes that improve my children's safety, but about half of the codes in my county are more about bureaucracy then safety.
That's what happens when bureaucracies need to justify their existence. Sad.
Not saying safety isn't important. Quite to the contrary. However, there's safety and then there's making life impossible with things that don't really add to or enhance safety.
Yeah - 5 Ohm is the bit that matters. I’m in the process of installing an off-grid solar energy system at my home, which the previous owner had wired up, ungrounded, with a generator.
I am not exactly following code, as my earth comprises a 5 meter RSJ left over from construction buried 30cm down under a dirt road - we have very little soil over bedrock, but the soil is full of iron. It’s only half buried but I’m at 2.2 Ohm, so it should do the job - have soldered a nice fat cable to it to be my common ground.
A 5/8" x 8' copper grounding rod is $14 at Home Depot; 3\4 x 10 is $33. I'm very confused on how the extra $100 would meaningfully affect the project in either time or BOM.
> I'm very confused on how the extra $100 would meaningfully affect the project in either time or BOM.
reply
That one change was the tip of a very large iceberg. They easily added somewhere in the order of $50K+ to the total cost of the system. I could have installed triple the power generation capacity for that money.
BTW, I did the bulk of the work on this installation myself (with the help of my kids here and there). No general contractor involved. Had there been a contractor on the job the cost would have been far greater.
Before someone jumps in and says "You probably screwed-up the design" (after all, this is HN). I designed the entire system, hired a civil engineering firm to review the design and calculations as well as provide stamped blueprints and calculations (you can't get a permit unless you have this).
I also consulted with the same plan checkers throughout the design process (before hiring the civil engineering firm) as well as with my suppliers and buddies who had built similar systems.
I only hired the civil engineering firm one the plan checkers told me verbally that all looked good and I should move forward, hire the engineering firm and submit the drawings and calculations for official approval and issuance of permit. I got verbal "All is good" from the structural, electrical, planing/zoning and fire plan checkers.
Three days after I submitted everything for approval the very same plan checker who told me all looked good rejected the very same plans he and I had discussed during weekly meetings for about two months. Things got surreal from there.
I spoke to multiple contractors from different trades during my many visits to the plan check office. I did not find one who did not have horror stories --multiple horror stories-- about dealing with these people.
This, unfortunately, seems to be the norm here in Austin as well. There's an entire industry of building permit "expediters" you almost have to hire to avoid the nonsense. This 3rd party handles all communication with the relevant permitting authorities, and mostly correct plans get approved almost immediately.
> I checked with contractors and the recommendation was "just do it and don't complain or they'll make your project a living nightmare".
> So, we did.
> They still made the project a nightmare to complete. They added a massive chunk to the overall cost and it took six months longer than it should have taken to complete it.
I read it that the excessive grounding requirements were just the beginning of a series of silly and expensive demands, which in aggregate made the project way more expensive.
It's that or it took six extra months to install four grounding rods.
> It's that or it took six extra months to install four grounding rods.
Given some contractors I've dealt with --- I wouldn't be shocked. (or maybe I would be shocked, ... if they were allowed to do electrical work unsupervised. :) )
I’ve installed a bunch of these for myself. You always think about what you might be driving the stake through or if you’re going to hit rock... but you do have a point: after the cost of the rods, there’s the cost of the sledgehammer and the ground cable to wire back to the box. Kind of hard to see how this one issue caused a great deal of expense...
They're a real bitch to drive into the ground, though, especially if you have clay rocky soils.
However, what I think he's saying is that despite blindly complying with this in hopes of a smooth project... the bureaucracy found other ways to make his project expensive, miserable, and slow anyways.
Switching from a forward slash to a back slash between clauses really threw me off. It took me three attempts to correctly parse the second clause (with "3\4" in it)
With respect, this is like saying “but the hard drive on Fry’s discount shelf only costs $100, why is your NVME-over-Fiber SSD storage system so expensive?”
Hint: what the consumer can buy over the counter, and what gets used in a professional setting, are frequently two radically different things that may only be tangentially related.
With respect, what you're saying is absolute nonsense in the context of ground rods. They're hunks of metal clad in copper and driven into the ground with a sledgehammer. They're so simple that there's literally no part of a modern PC that's analogous. The ones you can buy from Home Depot for $15 are identical to the ones I buy from a professional electrical contracting supplier for $15 - they come off the same factory lines from the same manufacturer.
There are other components used in contracting that have a bit more product segmentation, but the whole "frequently radically different things that may only be tangentially related" bit is wildly exaggerated. We use the appropriate components for the job, and often that means the same stuff you could buy at Home Depot. You're paying for that knowledge of what's appropriate/necessary and the labor to do the work, not for super-special premium contractor ground rods.
PV array ground buses for something small scale will commonly be a single bare 6 gauge solid copper conductor bonded to the rods with 7 dollar hardware. Unless your setup is really huge, no more than 80 dollars of 6 gauge bare copper.
It's not like you need to buy a $800 thermite welding kit.
Nope, California has been the poster-child for Big Government for _decades_. Each year the State regularly spawns off several tens of thousands of folks who have 'had enough', who migrate to Small(er)-Government states, and then promptly start advocating for the exact kinds of policies that made them leave CA in the first place. It's a stereotype at this point, especially in places like Washington, Oregon, Texas, Montana, etc.
I suspect it's a flaw in how humans (of certain political persuasions) rationalize "how we got here". They think "Well, the old place was OK except for this very small bit that pushed me over the edge into 'move elsewhere' territory. Now this new place is OK, but if we just enacted this little bit of legislation to be more like the place I just came from...." That in and of itself would be fine if it was extremely limited. However, when it's tens of thousands of people, and they all have different ideas of what "this little bit" that needs to change is, it's a recipe for the behavior we see now.
Incidentally, this pattern is where a lot of conservative commentators are coming from when they say "liberalism is a disease". Non-liberals don't tend to exhibit this behavior - they move to an area, hunker down, shut up, and try to integrate.
> Non-liberals don't tend to exhibit this behavior - they move to an area, hunker down, shut up, and try to integrate.
I'm not sure what definition of “liberal” you are using, but there are so many obvious counterexamples that don't fit any of the plausoble ones that I can't see any way that that could be even a useful rule of thumb.
Missionaries tend not to be liberals, and indeed tend to be lionized by conservatives for doing exactly the opposite of “move to an area, hunker down, shut up, and try to integrate.” Conservatives don't object to liberals because liberals proselytize more than conservatives, they object to liberals because liberals disagree with conservatives.
The term "liberal" in the US has lost it's meaning. What we often call "liberals" in the US are more like "leftists" than the traditional liberal way of life and thought.
Classical Liberalism if very, very far from what the Democratic party pushes these days. In fact, I would venture a guess that most thinking people are Classical Liberals. I make the distinction about "thinking" people not to be elitists at all, but rather because it is obvious that there are those who blindly follow and then there are those who choose to be more analytical and actually think rather than follow. Problems, as Einstein famously said, cannot be solved from the state of mind that created them in the first place.
This is where, I think, Classical Liberalism fits. I happen to think it is an ideology that would do the most good and deliver the greatest benefits across the board.
In case someone reading this isn't clear on the distinction, here's a good explanation:
> What we often call "liberals" in the US are more like "leftists" than the traditional liberal way of life and thought.
Originally, the political Left was a name for classical liberalism based on where the liberals sat, as opposed to the conservative monarchists, in the French National Assembly. So if we’re fetishizing the original sense of terms, the distinction is nonsense.
Most of the political spectrum in the modern West (outside of the
far right, which is classically conservative) is rooted in classical liberalism, which was nearly completely victorious over classical conservatism. They differ in whether they see the progress made by the 18th century liberals as the end goal or the starting point (“Left” evolved somewhat faster than “liberal” because that division became evident during the revolutionary period in France.)
> I make the distinction about "thinking" people not to be elitists at all
Yeah, you just think disagreeing with you ideologically can only be possible if people don't think.
> but rather because it is obvious that there are those who blindly follow and then there are those who choose to be more analytical and actually think rather than follow.
Sure, that much is true. The part that's less obvious is where you say most everyone that falls in the latter camp agrees with you ideologically while most everyone who disagrees with you is in the former camp.
> Incidentally, this pattern is where a lot of conservative commentators are coming from when they say "liberalism is a disease". Non-liberals don't tend to exhibit this behavior - they move to an area, hunker down, shut up, and try to integrate.
This is an interesting comment. Or rather, thinking about how this could be --and, it is-- is what I think is interesting.
I've given this much thought because, as someone who tries hard to navigate an ideological center --borrowing what's good from each side and discarding the rest-- I have watched in horror as some of my friends and even a few family members have effectively become radicalized (the only word that fits) towards each of the ideological extremes.
My conclusion, or rather one of them, is that there must be a genetic element at play. Nobody is born with these ideologies. Just like nobody is born religious. And yet, over time, mental changes occur.
How can I explain my atheism and extreme skepticism? All I can grab for is that my brain is wired just different enough that it is able to go against the grain. Social pressures can be intense. I had absolutely no negative experiences with the church growing up. And yet, somewhere near adulthood I started to question it all and ultimately rejected it.
The same mechanism applies to political leanings. I've gone back and forth thinking that one school of thought was better than the other, only to realize +/- 25% from center is where the good stuff lies and the rest is just grotesquely wrong on both sides.
A long way to say that it is my theory that people who's brains are genetically wired in a certain way self-select towards the left or the right. This isn't an intellectual process at all. The failing being that very few of them ever take the time to, in fact, at some point, look at their positions through the lens of honest intellectual inquiry.
And so, the "Democrats" who leave CA and behave the same way they did in CA do so because, well, that's who they are. Just like "Republicans" are who they are. Genetically.
A lateral example could be had in Olympic class swimmers vs. wrestlers. Their body types are very different. Yet, swimming doesn't stretch someone out, make them taller and more slender any more than wrestling makes them short and muscular. The top achievers in both domains self-selected into those positions because that's the genetics they were handed. They build from there, but you can't take a short person and make turn them into world competitive swimmers unless they were also born with super-human endurance and power delivery capacity...their hull speed will be lower than that of a longer person.
The other interesting data point is found in the Hispanic population in the US. Quite a few of them got here by escaping from regimes that are not that different from the ideology that has currently taken hold in the American Left. And yet, surprisingly, they are voting for and supporting the same kinds of people and policies they left behind. Once again, my thinking is that this is simply who they are genetically rather being "Democrats" or "Republicans". After all, they had no clue about either of these parties before they got here.
It's an interesting problem because this effect has destroyed nearly every single Latin American nation over the last, take a pick, one hundred years. The policies that have permeated places like CA and are being championed as saviors for our country are the very same kinds of policies and ideology that have laid waste to entire nations. This isn't a matter of opinion, it's a matter of Latin American history, something we don't teach here.
I have friends in Argentina --where elections are about to take place before the end of the month-- who are in absolute fear of the Leftists once again gaining power. And when I say "fear" I am not being dramatic. They actually fear they will lose everything they have worked so hard to achieve during the last several years. Venezuela's Maduro regime is sending activists into Argentina to cause mayhem and disruption. In fact, a good deal of Latin America is on fire right now with the political Left trying very hard to either take control or gain greater control. They are attacking journalists and, generally speaking, unraveling the very fabric of civilized society. It's sad to watch this kind of thing form a distance. It's also sad to understand that if we, in the US, don't realize what these people are actually selling we stand a really good chance to end-up on a path that will forever transform this nation and not in a good way.
Anyhow, this stuff is really depressing to me. I can't see why people don't see what's going on. A lack of education in the actual history of other lands might be part of the culprit. I mean, how many Bernie and Warren supporters actually even know about The Gulag Archipelago, much less read any of it.
Not to get political on HN, but, well, California is now in the grips of a truly demented ideology. Due to having overwhelming majority everywhere, they can pretty much do as they wish. It's as close to a corrupt third world government as I hope we are ever going to get.
For example, our roads are in terrible shape. They have squandered or diverted funds that were specifically allocated to road maintenance. By that I mean: We voted to pass measures to collect additional taxes to fix the roads and they stole the money.
Another interesting one: They pushed and passed a law taxing any square foot of land that does not allow a drop of rain water to directly make contact with dirt. The tax is $0.025 per square foot. This means that your home, driveway, patio, patio cover, etc. are counted as part of this. Think of places like your local mall, Home Depot or strip malls, where 95% of the dirt is covered. This is huge.
I could go on. Like the high speed train to nowhere that we approved at a cost of $10 billion that is now near or past $100 billion. It goes nowhere, it isn't finished and nobody is gong to use it. And yet they don't stop. I would not be surprised if this becomes a half billion dollar project. The ROI on these things is so upside-down a fifth-grader could figure out it's a bad deal.
re the rain tax: if that means the mall owner has to open up some of those acres of asphalt to reduce stormwater runoff or landlords end up jackhammering the cement slabs covering the front and backyards (because spraying herbicide a couple times a year is way cheaper/easier than regular gardening service) then I fail to see the problem.
Your giant slabs of cementitious landscaping are contributing to my heat island effect so the tax is a small reminder to rip the unnecessary bits out and plant a couple trees instead.
The rain tax thing makes sense to me. Storm water has to go somewhere. If it hits the ground then some of the water will soak into it. Covering the ground with a solid surface increases runoff. Which in turn increases the amount of infer structure needed needed to deal with the water and prevent flooding.
Covering the ground creates an externalised cost, basically someone else is being made to pay for the economic cost of that persons decision. Without a proper price signal externalised costs result in a net loss of wealth.
This is not a new thing. Laws dealing with the fact that rain water flows from property to property are in some of the oldest recorded regulations.
Which is why it's so important for the, er, loyal opposition party to clean its own house.
If the party in question poisons its brand to an extent that keeps them out of power for generations -- which is what they've already done in California and are busy trying to do everywhere else -- we will all end up worse off for it. The only thing worse than a two-party system is a de facto one-party system.
You should have bribed the inspectors. After multiple FBI investigations, LA County building inspectors have figured out that the fines are smaller than the profit. It’s surreal how openly corrupt the planning and building departments in LA are.
What do you do for grounding if your house sits on bedrock?
We barely had enough for the "buried in trench, 30" deep" option at our house before hitting the solid lava everything here is built on. (And it's buried in mostly gravel, there is practically no dirt.)
A ground driven into rock or buried in dry gravel isn't going to be of much use. Typically electrodes are buried in a mixture of bentonite and gypsum, which is pH neutral and provides low resistance. There are also chemical (salt) earth fills which work well but require thicker copper electrodes to deal with corrosion. Various conductor arrangements are used, but in shallow trenches there is always the problem of the wires being cut. A radial or cartesian mesh burial over a larger area would be typical. You can buy electrical mesh and ground enhancement fill for this purpose. Note that you still need it to be below the frost layer in your region -- which varies from 15cm (California) to 300cm (Canada) depending.
An earth connected via a distant cable isn't going to be effective against lightning (transients are high frequency and hence conductivity is constrained by the skin effect, and long wires have high inductance even if they have low DC impedance), and additionally is poor practice as it is more lightly to be cut without someone knowing in the future. I doubt any EE would approve it when other methods are available.
Run a large gauge wire to another location where you can put a better earth.
In reality, you are likely on a TN earthing system, so you already have a low impedance earth wire supplied from your local transformer and your local earthing is just a "backup".
I'm kind of surprised to read that entire article and not encounter the phrase "ohm meter" anywhere in a search. Rarely does anything serious have only one ground rod. Three legged telecom towers will have three rods, one attached to each leg, linked together by a ring of 6 gauge copper buried outside the perimeter of the concrete foundation.
Multiple 5 foot ground rods in the correct configuration bonded together could offer much more conductivity to the earth than one 8 foot. The article seems to be obsessing about the length of the rod and not the scientifically measurable ohm resistance.
Why do the US regulations specify the method and not the result? That seems bizarre.
Here in the UK, most earth connections are provided by the power company, but in rural areas you may be too far from the substation, so a ground rod is permitted - as long as it provides an earth loop impedance (Z) of less than 200 Ohm.
So if you want to provide your own earth, you have to keep driving ground rods deeper, or adding new ones until the "Zeds" are good enough.
It looks like the US regulations permit you to bang an "8 foot" rod into dry sand, and walk away as though that's job done??
I think the greatest problem that troubles me personally is that the single-point grounding required by lightning protection is almost never implemented properly in the vast majority of installations.
I realized the existence of this problem when I wanted to install a mesh network in a school campus, so I started reading documents about lightning protections, amateur radio publications have good resources on it, and all of them indicate a correct installation should eventually bound all grounds, including mains ground, telecommunication ground and lightning protection ground at a single-point via low-impedance connections (even if they have their own ground rods), also, all metallic objects and structures, such as gas pipes, fire escapes, etc, should be bounded to ground as well. This is to prevent a voltage gradient developing between different grounding points in case of a lightning strike, and it's usually required or implied in electrical or building code.
However, who ever obey this rule in practice? I've seen various installations of outdoor/rooftop Wi-Fi networks, and all the lightning protection they have is a gas discharge tube and a ground connection of unknown quality, I don't think the installation is legal. Also, many buildings were in compliance of the code when they were built, but even adding an extra metal frame or a cable-TV box will invalidate the entire lightning protection, at least from a technical or legal perspective.
I guess the reason behind the non-compliance is that the likelihood of a lightning strike is already low, even if it does strike, it's possible that even a non-compliant lightning protection system can work to an extent. And if it doesn't work, most people would just accept the damage. As a result, it's a concern for radio hobbyist only, whose antenna increases the likelihood significantly.
But I was still shocked when I first understood the fact that most (low-rise) buildings don't have a technically sound lightning protection system and bet the safety of the building on luck. And its practical consequence is: if you want to put a rooftop antenna for amateur radio, with good lightning safety and in compliance with the code, in many cases, the answer is you can't - the building often doesn't even have a technically sound lightning protection system to begin with. I read many hardcore radio amateurs redo the entire electrical wiring at their homes to the proper standards.
High-rise buildings are much better, lightning protection is serious business, but you are not usually allowed to put a mesh network antenna on top of them. But if you are allowed, I've heard that some people even got a chance to have a friendly conversation with the civil engineer who's in charge of the building's lightning protection system and work out a solution together.
This is called equipotential bonding and a failure to implement or maintain it leads to occasional but tragic or fatal outcomes, we had one recently locally where a 12 year old girl was seriously brain damaged and needing life long care because she touched an outdoor tap.
This possibility is obvious from the technical documents I read, and it was what I worried (and I'm still worrying!) about. Thanks for the comment, I now know it's a real danger, not just a theoretical one.
Because of this, I eventually abandoned the mesh network experiment. I don't want to blow the school up and pay for it, and potentially have someone killed, just for a faster Intranet.
And I decided that if I'm going to do the experiment in the future, I'll just setup a gateway in an open space, somewhere middle but away from the buildings (so the gateway can pick the signals from those buildings), and run the system on solar power. In this case, lightning protection becomes easier - just ground the system and it should be fine.
That's probably why the two faucets in the UK are required to be electrically bound with a counductor, under the wash basin or sink. If the mains is in contact with a pipe and not the other, touching both faucets would result in electrocution. Also an extra argument for a single combined faucet for both cold and hot water.
No, the real reason had nothing to do with electricity. In fact dual taps were entrenched before electricity was common place.
Mythbusters have done several demonstrations over the years of what happens when you take a hot water heater, disable both of the over pressure valves, and break the thermostat so that it doesn't turn off. They explode spectacularly.
Early water heaters had zero overpressure valves. And thermostats are not particularly reliable even today. So exploding a home wasn't particularly uncommon.
In the UK, they made a low tech solution to this problem. Look in your toilet, in the back. There is a level sensor. If there is insufficient water, more is added. Imagine taking a large one of these and putting it in your attic. When you drain out of it, gravity provides the pressure. Now put a heater under it attached to a thermostat.
If the thermostat fails, either you have cold water (failed closed) or boiling water in your attic and there's fucking steam everywhere (failed open) but you notice and shut it off. If the level sensor fails, you either get water pouring into your house and you turn it off, or the hot water turns off and you investigate. Under no circumstances can pressure build up and explode the house.
Of course in like the 1880s better safety mechanisms were developed so it's fucking ludicrous it's still like this but whatever I'm not the king of england.
edit: I forgot the punch line. These are open top devices so all manner of shit got into your hot water. They kept separate taps to ensure that when you were pouring cold water for food, you knew it was clean and not contaminated with your hot water. Why they're ok with showering in shit water is something you'd have to ask a Brit.
Your edit is very interesting and explains why my mother (who lived in the UK from 11-25) ingrained in me as a child to let the water run cold for a while after switching from hot before drinking. She said it was due to increased rust in the longer hot water path, but I suspect that it is mostly culturally ingrained due to your explanation and the rust explanation is just working backwards
When hot water was retrofitted in many homes, it was stored in an open top basin which would collect rats and whatnot. Thus the cold water was kept separate in order to be more clean.
it was inexplicable for me as a native too - although I recently discovered that it was specified by older regulations as they were worried about contamination between the hot and cold-water supplies. The hot water supply was traditionally fed from a header tank in the loft (attic) and so wasn't safe to drink.
A British coworker said he thinks it's because authorities were worried about the water supply being intermittent. A rooftop tank solves the contamination back flow issue via an air gap. And provides water when the mains aren't working.
I also suspect the British would have found a old school US style water heater to be somewhat frightening.
In older houses there is only one tap it is ‘safe’ to drink from. Usually in the kitchen, all others, hot and cold, are fed from the tank in the roof. For some reason it’s OK to brush your teeth with water from the tank but not to actually drink it.
> The hot water supply was traditionally fed from a header tank in the loft (attic) and so wasn't safe to drink.
Now that I think of it, I have the same setup in my flat. The tank is supposedly sealed, but I'm not sure how reliable the seal is (it is not a pressurized tank).
I assumed it was a money-saving measure. If you have separate faucets you can't get warm water, only cold and/or hot. So to get warm water you're forced to fill in the sink and make use of that, which saves both water and heating.
Is it really never, or is that in relation to complex low voltage setups? I replaced my main panel earlier this year. They very intentionally grounded to hot & cold water and iron gas pipe in multiple places plus at the copper water supply entrance. Externally there are two buried rods. From what I recall the NEC requires all of that for every installation.
> all of them indicate a correct installation should eventually bound all grounds, [...] at a single-point [...] to prevent a voltage gradient developing between different grounding points
In the UK, when you get an electric car charger installed (or power to a mobile home or similar), the installers will often install an extra earth rod, right next to where the car is parked, and ground the installation to that - a so-called 'TT' earth.
After all, given that it's possible to get a voltage gradient between different grounding points, you want any grounding point you could touch with your hand to be at the same potential as the one you're touching with your feet!
^ THIS - Go watch at least first video "Grounding - Safety Fundamentals" This was a very eye-opening video. I totally misunderstood grounding before watching this.
Are all domestic buildings grounded this way in the US? Here (Sweden) I believe houses that aren’t far from transformers are only grounded at the nearest transformer (where power is transformed from 10kV 3-phase into 230/400V 3-phase+neutral for distribution to each house.
Yes, all domestic buildings in the US have their own earth ground, and this is and has been required by the NEC (National Electrical Code).
Often this is a thick, 8-foot long copper rod pounded into the ground not far from the electrical panel, connected to the panel by a ~1cm diameter copper cable. (There are other allowed methods, including using a similar copper electrode encased in concrete or in the building foundation.)
It looks like this (which shows a galvanized-steel rod rather than the more canonical copper, and shows a further connection to a second ground rod, which is not required everywhere):
Also, certain sub panels (e.g., in garages) will need their own separate earth ground, made the same way.
This "grounding electrode" is in addition to a connection ("bond") to the domestic cold-water pipe, usually done where the pipe enters the building. (That pipe itself may be a hunk of copper buried in the ground.) The electrical bond to the water service is not a substitute for the dedicated grounding electrode mentioned above, it's to ensure that the plumbing does not become electrically charged.
All except homes built many years ago where someone felt it was satisfactory to clamp a single wire to the water line. I’ve seen this numerous times.
Not to mention the numerous instances of 20A breakers on 14awg. It’s simply amazing what one will find in some homes... like the person who clipped both ends of an extension cord to feed an outlet out of a junction box rather than spend $20 on home wiring. Or the person who literally covered a chiseled-out trench in plaster with painters tape to conceal wire. And on and on...
2 is what I’ve installed in the houses I’ve renovated. It’s just not worth the risk to your electronics and home itself to cheap out on proper electrical in your house. Particularly if you’re one of those people with a 12 core 20tb dell r720 in the basement.
Yellow-jacketed CSST (corrugated stainless steel tubing used for natural gas) must also be bonded. This was added to the code in the mid-2000s so homes built from the mid-nineties when CSST became available till around 2005 when bonding requirements for it were added may not have it. I had to retrofit it to my 2004 home.
All modern concrete foundations are required to have a Ufer ground[1] installed, and this Ufer is required to be connected to the panel. So ground rods are rather obsolete in most circumstances. But as others have noted, it would be better to have an impedance test during the dry season.
In general, in the US, it is required to bond a metallic water pipe system (e.g., copper or galvanized pipe) to the electrical grounding system. This is usually done near the service entrance so that the connection can be verified by direct inspection.
“If metal water piping systems are not bonded to the electrical system in a way that meets the requirements of the Code, they can present a fire or shock hazard within the building or structure because they could become energized. ...”
In some jurisdictions, inspectors may want you to bond around nonconductive plumbing connectors of the kind you mentioned to ensure the whole pipe system is bonded. In particular, around the hot water heater.
Yes, the neutral is tied to the earth at the transformer and also at every house on the transformer. This means that some return current technically runs through the earth back to the transformer. shrug
I recently installed an IKEA kitchen hood, and was surprised that the power connector only had two pins (IEC C17/18). The hood has a aluminium frame (albeit painted) with metal pole switches. Can someone explain why this doesn't need to be grounded?
Perhaps the same reason toasters aren't.
It is very likely that one will have two hands on the toaster when one is shocked. You have two choices: ground it, in which case the current races up one arm, through the torso, and down the other arm (frying your heart in the process). Or, keep it ungrounded and let the current race down your leg to the floor (sparing your torso and hopefully taking advantage of the fact that your shoes will provide a little more insulation that your two bare hands on metal in the first scenario.)
We're currently building a house in Germany and the grounding consists of a couple dozen meters of thick (maybe 10mm) wiring which is laid out in the concrete ground plate of the house.
A couple of dozen meters? Wiring? Encased in concrete? Five to ten meters of zinc coated 30x5mm flat bar steel bound to three to six 3/4" rods is usually enough depending on the soil characteristics. And it's usuall buried directly in the soil next to or around the bulding. I wonder what bulding codes you've got there.
The relevant code is DIN 18014. And it distinguishes between different kinds of ground plate. Some are considered sufficiently "earth-feeling" to be considered ground themselves, others require separate corrosion-resistant grounding to be put in the earth around them. Zinc-coated steel is not considered sufficiently corrosion-resistant.
Allright. This makes sense as it reduces the electro corrosion on the rebar and also prevents other unfortunate accidents such as the hot wire coming in contact an exposed metal part or a metal pipe which is itself in contact with the rebar.
I have seen a scenario where ligtning struck a DHW solar panel installation, skiped from the plumbing to the rebar and also melted hole in the plumbing while doing so, draining all the antifreeze on the walls. In addition it destroyed all the electronic appliances in the house. If the panels along with the rebar would have been properly grounded, such as scenario would have been mitigated.
Note to readers: NEC is National Electric Code, which is the regulation of all electrical installations in the USA.
The NEC here is not Numerical Electromagnetics Code, an electromagnetic simulation program, commonly used for antennas, which often have a ground connection as well. And it's obviously not the NEC Corporation, the Japanese electric company, whose semiconductor department is now a separate company known as Renesas.
When I first saw the title, it took me a few seconds in my mind to solve the hash conflict here. I wonder how many people on HN know all of these entities simultaneously.
We were building a fusion reactor prototype, and as far as I am aware the NEC code doesn't really have a section for that. I'm sure that much of what we did was probably against best practices.
We had a number of high voltage power supplies running in one room of our lab, and on days when the humidity in the air was low, the ionization in the air would charge up all the metal surfaces in the room. So, even if something was not connected to a high voltage, you would get a nasty shock when you touched it. All these devices were connected to the normal, "third prong" wall socket ground, but apparently the route through the building to ground was too resistive to handle the large amount of ionization in the air.
When working in a room with 180,000 volt power supplies, random shocks rapidly become more than annoying.
Spence, our electrical engineer, had the idea that the quickest way to fix the problem was to run a grounding rod into the soil right outside the window. We pounded a rod into the ground with a sledgehammer as far as we could, which was about 5 feet, I guess. Then we ran a 1/4" copper cable from the rod, through the window to connect to the main reactor chamber.
This worked great! At least for a couple of months. Our lab was located in a swamp, and the soil was basically a few meters of dried leaves. You could actually feel the soil move up and down when you walked on it. When August came around, it had not rained for quite a while, and the water level in the soil had dropped below the level of the grounding rod. The dried leaves in the soil insulated it quite well. We started getting shocks again, and it took us many hours of troubleshooting before we realized that the ground was not actually ground anymore.
We went out and pounded the rod a few feet deeper, which was easier now that the soil was dry. This reached the water table, and fixed the problem.