Jump to content

Laptop/appliance anti-shock: simple project to avoid getting zapped


lsemprini

Recommended Posts

Here is a public service announcement for folks whose laptops, refrigerators, toasters or other appliances buzz or electroshock their hands on a daily basis. This happens with many many devices when used in Thailand, in particular devices whose power plugs have 3 prongs instead of 2.  To completely and permanently fix this problem, you need to make sure the 3rd prong of your device (the ground/earth prong) is actually connected to the literal ground below your house.  If your device's plug only has 2 prongs but your device has a metal case, see below.
 
Thailand makes it challenging to properly ground any device because a) 99% of power strips that have 3 holes DO NOT CONNECT THE GROUND PRONG TO ANYTHING, in order to save money, EVEN if the male end has 3 prongs, and b) 99% of 3-prong plugs in walls DO NOT CONNECT THE GROUND PRONG TO ANYTHING, in order to save money.
 
To fix this problem, you need to do a little project that takes about 15-30 minutes once you have the stuff.
 
COST: The parts cost about 100-500B at a local hardware or electronics store.
 
GROUND LIFT PLUG: You need to get a ground lift plug like the orange one pictured here:
spacer.png
Then you need to buy some electrical wire and run the wire from the little ground terminal on the top of the ground lift plug---buy one small bolt/screw and one nut that can fit through that little hole, to make sure the wire has a solid connection--you don't just want to coil the wire haphazardly around the loop without a fastener because your knot will eventually come undone and slip down and short out the power prongs and burn your house down.  The secure setup should look like this:
 
spacer.png
 
WIRE: For the wire, you should choose a thick wire with a solid core, not stranded. In the event that something actually does go wrong with your laptop/appliance (e.g. water getting inside), the full 240V of electricity may end up on the outer case that you touch and you will need a wire thick enough to carry all that current to ground rather than through your heart in order for you to be safe (acknowledging that yes, before you did this project, 100% of the electricity would have gone through your heart). The wire should be at least 16 gauge (at least 1.3mm in diameter not including the plastic insulation, at least 1.3 square millimeters cross-sectional area, which is probably the number you will see in the Thai shop (get 1.5 or 2 sq mm)). The wire should be thick enough that you can mold it and it will keep its shape like a coat hanger (but not quite as thick as most coat hangers). Green color insulation is good because it will remind people that it is a ground wire, but of course any color is ok.
 
STAKE OUTSIDE: Now run that ground wire out of your house to the wettest patch of dirt around your house (being consistently damp is critical for good ground--under a drain that you use every day is good) to a copper ground stake, also available at the same shops.  The copper ground stakes already have a handy nut to screw the wire on securely and a handy pointy bottom end that you can drive into the ground with a rock or hammer. Get a stake that is long enough that you will definitely hit some damp soil. Don't be shy, buy the meter long one to be sure.
 
spacer.png
 
spacer.png
 
HOW TO USE: Now you can plug your laptop/appliance into the ground lift plug and enjoy non-masochistic operation forever. If you have many 3-prong devices that shock you, you can seek out a real actual 3-hole-3-wire-3-prong power strip (very hard to find in Thailand: just because the male plug has 3 prongs doesn't mean the ground prong is actually connected all the way to the 3rd prong on the female jacks: you REALLY need to test it in the shop with a multimeter to be sure; on Lazada there is really no way to be sure; more expensive makes it more likely but still not guaranteed) and then plug the power strip into the ground lift plug and your devices into the power strip.
 
IF YOU HAVE 2-PRONG DEVICE WITH METAL CASE: You still might be able to solve your problem using a similar trick.  Instead of attaching the wire to a ground lift plug as above, attach the wire directly to some screw on the outer metal casing of your appliance.
 
Happy surviving!

ground_lift_plug.png

groundlift.png

stake.png

stake_ground.jpg

  • Thanks 2
Link to comment
Share on other sites

3 hours ago, lsemprini said:

Thailand makes it challenging to properly ground any device because a) 99% of power strips that have 3 holes DO NOT CONNECT THE GROUND PRONG TO ANYTHING, in order to save money, EVEN if the male end has 3 prongs, and b) 99% of 3-prong plugs in walls DO NOT CONNECT THE GROUND PRONG TO ANYTHING, in order to save money.

Once you fix that problem with power strips and externals, realize that 99% of the homes in Thailand don't have electric receptacles that connect to an external ground. 

  • Like 1
Link to comment
Share on other sites

1 hour ago, ArcticFox said:

Once you fix that problem with power strips and externals, realize that 99% of the homes in Thailand don't have electric receptacles that connect to an external ground. 

Most definitely ... That's why the project includes running the ground wire outside since the house wiring almost certainly won't do so even if the receptacles have 3 holes 

Link to comment
Share on other sites

1 hour ago, Led Lolly Yellow Lolly said:

You're actually opening yourself up to a different set of hazards by having multiple paths to earth that bypass your CU. The above can only be considered to be an emergency/temporary kludge.

Well in this case there would be only one path to ground, since the laptop/appliance cannot be grounded through the house wiring, and it's fair to assume nothing else in the house is actually grounded at all (except perhaps one shower water heater, which seems to be the only thing that Thai electricians feel they need to ground).

 

Of course it would definitely be better to redo all the house wiring and choose power strips that are truly 3-conductor but that is not always practical, especially for rental units.

Edited by lsemprini
Link to comment
Share on other sites

Quote

Well in this case there would be only one path to ground, since the laptop/appliance cannot be grounded through the house wiring,

 

No, you're creating multiple paths. Even if your kludge ground is the only ground in the house, you have the utility side grounds on power poles (Thailand generally employs a Multiple Earth-Neutral bonded system).

Even worse, if your kludge ground REALLY IS the only ground, you may also have a situation where the ground bar in your CU is bonded to neutral, but no ground rod connection exists. . . It's very hazardous for a number of reasons.

 

All of the above is particularly dangerous in electrical storms where ground differentials can be massive. . . So, you get a couple of million volts spreading across the ground under your house, finding a less resistant path to the utility side ground via your notebook, and your shower... Now imagine that in action!

 

 

 

 

Edited by Led Lolly Yellow Lolly
Link to comment
Share on other sites

14 minutes ago, Led Lolly Yellow Lolly said:

 

No, you're creating multiple paths. Even if your kludge ground is the only ground in the house, you have the utility side grounds on power poles (Thailand generally employs a Multiple Earth-Neutral bonded system). You may also have a situation where the ground bar in your CU is bonded to neutral, but no ground rod connection exists. It's very hazardous for a number of reasons, particularly in electrical storms where ground differentials can be massive.

I don't understand what scenario you are describing here.  I would guess that the ground bar in the CU is not connected to anything at all in most Thai houses (I've seen at least some that are that way myself), but either way,

  1. what exactly is the path of the potential dangerous current flow that you are warning about (please describe the journey of current from point to point)?
     
  2. does the current path you are describing only exist in lightning situations?
     
  3. are you describing a hazard that would not exist if no devices in the house were grounded at all (i.e. the typical situation where nothing is grounded and people receive constant shocks from their laptops)?
     
  4. are you describing a hazard that would not exist if the ground bar in the CU is completely disconnected?

It sounds like you are describing a situation where the dirt containing the kludge ground stake is at a very different potential (i.e. at least 240V and possibly a lot more) than some other surface in the house that a person might be in contact with (in addition to being in contact with the case of the laptop/device that is grounded with the kludge ground stake), but it's not clear exactly what surface and in what situation.

Link to comment
Share on other sites

I don't have time to write out a science lesson for you but the soil has resistance. Copper has resistance but much less than soil, so in a nearby lightning situation, the surge of energy propagates outwards across the ground (the order of thousands, even tens of thousands of volts per metre). It get's to your house, finds a less resistant path in your internal wiring, takes that path, and fries everything en route.

 

There is still a ground differential hazard even when there is no storm, but for different reasons and scenarios.

 

Also, if you don't know what E-N bonding is, why it exists, and why the integrity of such a system is important (broken neutral = death), you're really playing with stuff that can kill you or your loved ones. The best solution is to just wire the house up properly. If you have multiple ground rods (which is acceptable and even desirable) they MUST be properly bonded together so that they are all held at the same potential. 'Properly bonded' means big heavy cables and exothermic welds. I use copper clad steel wire to bond multiple ground rods. A slither of THW cable ain't gonna do it.

 

Don't misunderstand me, I applaud you for taking this important safety issue on yourself, finding a competent electrician in Thailand is pretty much impossible. . . but you have a steep learning curve ahead.

 

 

 

 

Link to comment
Share on other sites

 

Quote

The best solution is to just wire the house up properly....

 

Don't misunderstand me, I applaud you for taking this important safety issue on yourself, finding a competent electrician in Thailand is pretty much impossible. . . but you have a steep learning curve ahead.

I'm totally game for that learning curve.  I want to learn more about the possible dangers so I can make a realistic assessment of the relative risk of doing the kludge above or not doing the kludge above in a typical Thai household, and also see whether there are refinements to the kludge that can strike a better balance between protecting against as many hazards as possible and also being realistic and practical to achieve in Thailand when most of the people doing it are tourists/expats in rental units (not homeowners with access to high funds and experienced electricians) who are just trying to stop themselves being shocked continuously by their laptops and appliances. 

 

It sounds like you're an experienced engineer and of course it makes sense that you would always come back to fully following well-established Western-standard codes as the best solution (I do the same when presented with a Thai-style solution to programming problems and I want to do it the right way that I learned), but this is often simply not realistic for reasons both of us have mentioned above. 

 

Put simply, in 95% of cases, we cannot choose between leaving the house as-is or doing a perfect Western-standard job.  Instead, out here in District 12, we can choose between leaving the house as-is, doing the kludge in the OP above, or doing a different mitigation of the laptop-shock problem that is within practical possibility.  The best we can do is to try to understand the specific hazards of each choice and their reasons/scenarios and find the optimal solution that meets the unfortunate practical limits.  That's what I'll try to do below.

 

First, for this discussion, in order to keep the permutations manageable, let's assume that the Thai household has zero ground wires whatsoever (the ground prong of all receptacles is unconnected, and all ground bars in the CU and elsewhere are empty), meaning all laptop/appliance device cases are normally floating (when no fault occurs), and that there is no ground stake anywhere on the premises (only ground stakes in the power poles and other parts of the distribution network as part of the Multiple Earth-Neutral bonded system).   I think this represents by far the common case.

 

Second, let's define some terms we can use to be clear about which hazards we are discussing since we kind of mixed them up above:

  • We can use the word lightning hazard to talk about the massive static charge from lightning that desperately wants to flow to ground, and we may need to be specific about sub-cases where the strike is on the lines vs. nearby ground.
     
  • We can use the word fault to talk about cases where conductors in the house (line, neutral, ground) unintentionally become disconnected or become connected to each other in various combinations that need to also be specified.  This category of hazards has it in common that the current wants to flow back towards the generating station (which it may do through wires or the soil in various scenarios)

OK, now for folks other than Led Lolly Yellow Lolly who are following along, here are some basics to study and a glossary because sadly many key terms here are completely different in US vs. UK vs. AUS/NZ:

When Led Lolly Yellow Lolly refers to "L-N bonding" I'm assuming you are referring to how a Multiple Earth-Neutral bonded system calls for bonding neutral and ground lines in (only) the main panel, and/or the potential danger of that system when neutral is disconnected (as discussed under "Open-circut PEN conductor" in the medium article above), is that right?  Or are you referring to something else?

 

Ok, with those assumptions and basics now covered, we can get back to the kludge in the OP.

 

Given that we are comparing the assumed Thai house (no grounding/stakes on premises) either with or without the kludge, which of the hazards that you have in mind apply only when we do the kludge?

 

Let's take faults first.  Do any of the fault hazards apply with the kludge that would not also apply without the kludge?  Suppose the laptop/appliance gets water in or has frayed insulation and faults a live conductor to the case.  Without the kludge, the person gets the full live voltage shock that conducts through their body to the floor and soil and back to the nearest power pole.  The kludge introduces a new path from the device case to ground (but NOT one which is bonded with neutral as normally done in the MEN/PME/MGN system) and that ground connection may take some of the dangerous current through the soil back to the nearest power pole, since the kludge wire has lower resistance than the person's body.  In NEITHER case will the breaker break (due to the high resistance of the soil), so in that sense both scenarios are unsafe.   But having the kludge may take away some of the current from the person's body (or at the very least the danger is the same).  Is that right? 

 

Are there other fault scenarios where there is a big difference with or without the kludge?

 

Second, let's take lightning.  Since the kludge involves driving a new ground stake, even worse one without any kind of breaker or other protective device, I could see some possibility for the kludge making things less safe than not doing the kludge.  You wrote:

 

5 hours ago, Led Lolly Yellow Lolly said:

in a nearby lightning situation, the surge of energy propagates outwards across the ground (the order of thousands, even tens of thousands of volts per metre). It get's to your house, finds a less resistant path in your internal wiring, takes that path, and fries everything en route.

I'm a bit confused by this because all the explanations of lightning above describe how the lightning seeks ground as its destination.  Why would the charge come back up into the house if it is already at the ground?  Especially in our specific scenario---why would it come up through a ground wire to the casing of a laptop/appliance that is not connected to anything else (remember our assumption that the house has no other ground connections on premises) and why would that damage the laptop?  Some damage due to capacitance between the case and internal components?

 

So I'm not yet clear on how the kludge could add additional hazard in this case.

 

You mentioned ground differential hazards from various sources.  I'm guessing that you're referring to ground differentials between the kludge's ground stake and the nearest power pole's ground stake, is that right?  Suppose there is a large ground differential (say, 240-1000V, not the kV-MV of lightning where charge issues prevail).  This will create a small amount of current through the soil from kludge stake to power pole.  While it's not great, how does that present a hazard to the person touching the case?  The person's feet are on the floor of the room, which is presumably nearly identical in potential to the kludge stake just outside the house, right? 

 

Or are you actually referring to the difference in potential between the person's feet on the floor and the kludge stake just outside the house?

 

Quote

There is still a ground differential hazard even when there is no storm, but for different reasons and scenarios.

I'm very interested...what are those reasons/scenarios?  I'd like to go through them and see if they apply and how likely they are in the OP situation.

 

----

 

Another separate but important question is: why do our laptops shock us in the first place?  What is the mechanism by which the current reaches the outer case when the laptop's ground pin is not connected to ground?  Perhaps if we understood this better, we could come up with a better solution that does not introduce new safety hazards (but still a practical one--please resist the urge to go back to "just wire up the whole house properly" for reasons explained above).  Any idea where the zap comes from?  Even brand-new laptops do it, so it is unlikely to be caused by faults from worn-out parts inside the laptop that short live lines to the case.  Could it be related to the "Perceived Shock" issue of MEN/PME/MGN that is described in the Medium article ( https://medium.com/@Voltimum/the-principles-of-protective-multiple-earthing-pme-c068f2f433ac ) ?  I don't quite understand the "Perceived Shock" as they describe it because the potential difference between the ground just outside the house and the floor of a room in the house should be negligible. 

 

Or is the laptop zap explained by some kind of "transmission line" effect where components inside the laptop are producing AC EM radiation that somehow induces current in the casing?  Seems far-fetched, but if it's true, is there another workaround that doesn't involve driving a stake outside?

 

Thanks!

Link to comment
Share on other sites

I live in an area where we have lots of thunderstorms and ground lightning strikes often close to the house, one very close strike either on the ground or the HT lines outside the house (cant remember which) a few years back blew out the circuit board on a brand new Mitsubishi air con, which being a new addition they had not connected to the house earth system,  thought I would be smart and  run an earth cable from the air con to its own earth rod.

 

Sure enough more ground strikes and one of them came back up my new earth rod and cable and completely fried the new air con.

 

I seem to remember reading that some countries do not allow multiple earth rods in houses for this reason.

Link to comment
Share on other sites

7 hours ago, Rimmer said:

Sure enough more ground strikes and one of them came back up my new earth rod and cable and completely fried the new air con.

Interesting.  How exactly would the "correct" house wiring have prevented that?  Did your aircon fry because of a ground potential difference between the two stakes (and if so, why? what was the exact electrical path through stakes and L, N, and G wires that caused the problem?) or was it because in the main house wiring from the main ground stake there is a spark arrestor or some other kind of protective gear that would have protected the aircon?  If the latter, that implies that perhaps installing such gear might help the OP kludge be safer.
 

Link to comment
Share on other sites

More than one ground rod will never be safe unless they're bonded together with thick heavy cable. The thick heavy cable locks them at the same potential. Lightning SHOULD travel through the thick heavy cable rather than the internal wiring (if it's straight and direct, lightning has an extremely high frequency/rise profile and can arc across bends in cables). If the lightning gets into the house somehow (e.g. un-bonded rods make that happen), it's trivial for it to jump between ground and neutral connections internally (think about it, LNG connections on circuit boards are just a millimetre or two apart. Even if there's no ground, the lightning will find it's way to ground via the neutral bond further upstream). . .

 

People tend to think about ground connections being for human safety, which is true, but most circuit boards on most equipment have MOVs on them as a last line of defense against surges and they're often arranged in a manner designed to drain surges to the ground connection, but obviously this all falls apart if the ground connection is either not there, or not optimal.

 

Quote

It sounds like you're an experienced engineer

Experienced, yes. . . formally qualified, no, so bear that in mind when reading any of my posts.

 

 

 

 

Edited by Led Lolly Yellow Lolly
Link to comment
Share on other sites

8 minutes ago, Led Lolly Yellow Lolly said:

More than one ground rod will never be safe unless they're bonded together with thick heavy cable.

Interesting info for the case of multiple ground rods, but see my longer post above...in my scenario there is only one ground rod on premises, the one I added via the kludge (nothing is grounded anywhere on the entire property, not even at the CU, which I gather is the most common case for Thai houses).  So the only other ground rod is at the utility pole (I hope).  See the long post for more info and questions.

Link to comment
Share on other sites

I'll read it when I have time. . . but I figured all this out because years ago we were losing some really costly equipment every year during the rainy season (IT stuff, Cisco, Juniper, $$$$$$$$). Eventually I realised it was lightning travelling between buildings via ground rods, making it's way into ethernet cabling between the buildings. The cost of replacing this stuff every year was crucifying us. . . To fix it I had the choice of bonding the grounds (lots of digging between buildings, lots of money) or replacing everything with fibre optics. I took the latter route.

 

In any case, it doesn't matter, lightning hitting the ground nearby causes massive step potential. Google around and you can find images of cows killed in fields. They were electrocuted by the voltage difference between their feet. . . So you're kludge ground is never safe, for this reason, and other reasons of touch voltage safety... and as I point out in my previous post, your neutral will be grounded somewhere upstream.

 

There are also codes on bonding metallic pipes to ground, but I never had to look into that and I don't know much about it. Plastic is king here.

 

 

 

Edited by Led Lolly Yellow Lolly
Link to comment
Share on other sites

39 minutes ago, Led Lolly Yellow Lolly said:

Google around and you can find images of cows killed in fields. They were electrocuted by the voltage difference between their feet

Only too true. Grounding of farm buildings in the correct way is also vital as cows have been killed by bad grounding of the regular mains supply. 

Link to comment
Share on other sites

8 hours ago, sometimewoodworker said:

Only too true. Grounding of farm buildings in the correct way is also vital as cows have been killed by bad grounding of the regular mains supply. 

 

Even tiny amounts of leakage can significantly reduce milk yield, even to the point of stopping some cows altogether.

 

There was a race horse that dropped dead whilst being led, turned out to be a bad underground HV joint. Other horses have been spooked by standing on both rails at a level-crossing. The track-circuit voltage is small but enough.

Link to comment
Share on other sites

Sorry but if I had a house without an earth system (and my Thai house was like that) I wouldn't be installing any earth wires around the house using 50 baht plugs. I'd rewire it with a proper earth to ground no matter what the cost (and I did). I value my life enough to justify the cost.

  • Like 1
Link to comment
Share on other sites

4 hours ago, KhaoYai said:

Sorry but if I had a house without an earth system (and my Thai house was like that) I wouldn't be installing any earth wires around the house using 50 baht plugs. I'd rewire it with a proper earth to ground no matter what the cost (and I did). I value my life enough to justify the cost.

Fair enough, and I'm sure there are thousands of expats who would make the same choice, but for the purposes of this post I'm referring to the much larger group who are renting and/or unable to make large modifications and/or who don't have access to good electricians or sufficient DIY knowledge.  So the relevant questions are (see long post above for important details): For a house with no grounding at all, is the kludge in the OP better than nothing or not, and either way why exactly?  And second question is is there a refinement to the kludge that can solve laptop zap in a safer way that is still be practical for the target audience defined above?

Link to comment
Share on other sites

The kludge ground for your lappie to stop the tingle is the best way to do it if your home has no grounding at all. However I would NOT do that if you already have a ground (for the reasons detailed in other posts).

 

Whether you have a ground or not adding earth-leakage protection (e.g. a Safe-T-Cut) is a valuable addition if you don't already have one. Most landlords would be happy to let you do this, they might even help with the cost.

 

Modern equipment that requires a ground falls in to two categories:-

  1. Those that need a safety ground (e.g. your toaster, kettle etc.) Mostly metal cased white goods.
  2. And those that need a functional ground (mostly PCs).

Those in group 1 MUST be grounded to be safe. No choice.

 

Those in group 2 need a ground to meet their EM emissions spec. most of them also have mains filters in order to meet that spec. It's these filters that provide the leakage which gives the tingle. In itself it's not hazardous but it's annoying and could constitute a hazard by causing you to jump resulting in injury. 

 

If you have a number of devices that need a ground it's best to use a single rod for all of them, reasoning also in earlier posts.

  • Thanks 2
Link to comment
Share on other sites

6 minutes ago, bankruatsteve said:

The concern for multiple ground stakes is not a huge issue, IMO.  But, to avoid that one could install one stake and then make a ground bus around the structure which could then connect to outlets as required.

 

Like many things safety related (not just 'lectrical) it's not an issue, until it is.

 

Many whizz around on motorcycles with no lid for years and years, then one day, they come off...

 

Simlarly, there are loads of Class 1 devices installed in Thailand with no ground. They all work just fine. Until a child dies! (water cooler and ATM accidents spring to mind)

 

Multiple rods will work fine, until the day that there's a close ground-strike leading to a significant potential between the rods. This may (or may not) exceed the limits of the appliance insulation leading to pain.

Link to comment
Share on other sites

7 hours ago, Crossy said:

Multiple rods will work fine, until the day that there's a close ground-strike leading to a significant potential between the rods. This may (or may not) exceed the limits of the appliance insulation leading to pain.

Is this an issue with “properly” (US NEC) laid out systems?  Ground rods collected at a common bus and grounding electrode connections tied to that bus?  For data centers we would generally run a ground ring on the exterior of the building with ties to the column footings.  Individual substations would tie to the ground ring (plus a supplemental local rod and a local water pipe for compliance).  

Link to comment
Share on other sites

8 hours ago, tjo o tjim said:

Is this an issue with “properly” (US NEC) laid out systems?  Ground rods collected at a common bus and grounding electrode connections tied to that bus?  For data centers we would generally run a ground ring on the exterior of the building with ties to the column footings.  Individual substations would tie to the ground ring (plus a supplemental local rod and a local water pipe for compliance).  

 

No issue when done like this with everything tied together with nice fat conductors, the whole lot goes up and down in potential together. The problems occur when you have separate rods that are not properly interconnected, basically two isolated earth systems which could get (briefly) many kV apart during a strike.

 

Some time ago I was working in the PUTRA LRT Operations Control Centre in KL (it's a full-auto railway). We had literally just got everything up and running for testing to start. It was "raining" (a typical tropical downpour) so I was waiting under the porch for it to ease off before heading to the car. There was a f.....g great flash and instant bang when there was a ground strike only about 50m away. Almost before I'd blinked my phone rang "coms is down", back into the systems room and every light I see is red! That building had "state of the art" lightning protection, fancy air-terminals, huge grounding mat etc. etc. It took various contractors two days to get everything back up. The lighting protection contractor was also seen installing more and fatter copper and various holes were drilled in the structure (I suspect for structure bonds).

Link to comment
Share on other sites

Shouldn't we be addressing why the appliances are faulty in the first place ???

 

A "how to" shunt the stray voltage away, is missing the point. And totally irrelevant for any appliance with a plug pack (there is no earth from the appliance to the wall). There are no AC mains voltages inside a laptop, not sure why anyone is earthing the wall socket for a device with no AC voltage and no earth pin.

 

I have 2 appliances in my house that have earth wire/pins, coffee machine and refrigerator. If they happen to go faulty the earth will protect me. I now have a faulty appliance. Otherwise, the other 32 devices I have plugged in have no earth connection.

 

People appear to have an incorrect assumption that an earth magically provide surge protection, it doesnt. In most houses, there is no circuit that trips and sends an over voltage surge (lightning etc) to earth. The earth, in isolation, doesn't provide any surge protection.

 

Yes, install an earth (for earthed appliances) and make sure it works etc, but in the first instance , throw the faulty appliance away.

 

Edited by Peterw42
Link to comment
Share on other sites

1 hour ago, Peterw42 said:

Shouldn't we be addressing why the appliances are faulty in the first place ???

 

A "how to" shunt the stray voltage away, is missing the point. And totally irrelevant for any appliance with a plug pack (there is no earth from the appliance to the wall). There are no AC mains voltages inside a laptop, not sure why anyone is earthing the wall socket for a device with no AC voltage and no earth pin.

 

I have 2 appliances in my house that have earth wire/pins, coffee machine and refrigerator. If they happen to go faulty the earth will protect me. I now have a faulty appliance. Otherwise, the other 32 devices I have plugged in have no earth connection.

 

People appear to have an incorrect assumption that an earth magically provide surge protection, it doesnt. In most houses, there is no circuit that trips and sends an over voltage surge (lightning etc) to earth. The earth, in isolation, doesn't provide any surge protection.

 

Yes, install an earth (for earthed appliances) and make sure it works etc, but in the first instance , throw the faulty appliance away.

 

 

My Dell and Acer notebook have 3 pin power connection with ground.

 

I look inside old faulty notebook power adapter and find the ground is used at the circuit.

 

If ground is not connected to these adapters there is some volts at the ground connection.

 

Many appliance used these ground connection in filter circuit.

 

Some circuit have surge device connected to ground.

 

Please observe this video with diagram of popular power supply brand.

 

 

 

  • Like 2
Link to comment
Share on other sites

15 hours ago, Crossy said:

That building had "state of the art" lightning protection, fancy air-terminals, huge grounding mat etc. etc.

It is interesting to see what happens when you mix lightning protection systems. I specified the streamer terminal system for Suvarnabhumi early on, but never got to review the shop drawings on it. Wonder how it would hold up in a close or direct strike. 

Link to comment
Share on other sites

On 12/11/2021 at 9:42 PM, tjo o tjim said:

For data centers we would generally run a ground ring on the exterior of the building with ties to the column footings.

This is what I am doing. My own datacentre is in part of a building around 150 x 30 meters (rough guess). The entire building has a buried ring, each corner a triangle of 3 x ground rods. This is also directly connected to the roof structural steel and column rebar. This is also tied directly to the electrical system ground . . .

 

I've been toying with the idea of building a faraday cage around my whole server room, and only allowing fibres through that.  Power would go through isolation transformer and various other line conditioning measures. Just a fantasy at the moment though. . .

 

2 hours ago, tjo o tjim said:

It is interesting to see what happens when you mix lightning protection systems.

I can't imagine any scenario where that would end well. You can even get huge induced load briefly during a discharge through one part of the building (like capacitance) to the other part.

 

 

 

 

Edited by Led Lolly Yellow Lolly
Link to comment
Share on other sites

Thanks all for the responses, keep 'em coming.  I feel we are getting closer to an answer to the key questions for this post given in the OP and especially this post:  Given the assumptions above, such as that the house is not grounded in any way initially (not even at the CU),

  • Is the kludge in the OP safer than nothing or not?
  • Either way, is there a better mitigation?
  • What causes the buzzing in the first place?

Thanks to SomchaiDIY for the pointer to the analysis of where the buzz come from.  The following 2 quora posts add a few more details too: here and here.  Also interesting is that of the Apple-original power supplies that have UK plugs, some of them (those with no additional cable, so-called "duck" or "stubby" connectors) do NOT connect ground (which is INSANE) and the one with its own cable does connect ground, and as a result many Apple users with the former type had to manually connect ground with a hacky wire to kill the buzz.  One of many examples here and here.  And of course Apple attempts to <deleted> and squirm their way out of this on all their official websites and will not replace the defective units for people.

 

As far as better mitigations, for those households that DO have a ground stake at their CU (thus not in the scope of the OP above) it would be good to advise them to run their ground wire to the house's main ground stake rather than create a new stake, correct?

 

For those households that have no ground stake anywhere (the typical case we assume in the OP), can we improve the kludge in the OP?  For example, rather than adding a new stake and connecting to that, is there somewhere else we can connect the power supply ground will remove the buzz?  How about connecting the laptop case to something (perhaps even the floor of the room)?  Would that be any different?

 

I think a few responses above have confused the two different hazards of lightning and faults, as defined above.  In normal situations (houses with proper wiring and no kludge) faults are handled by tying the ground to neutral at the CU (because remember the current in faults wants to go back to the generating station, not necessarily to the ground) and the ground stake exists solely to handle lightning (which wants to go to the ground), NOT because of faults, as explained very well at 9:10 in this video (ignore the US-specific stuff about 2 hot phases in this video):
 

 

However the case of the OP kludge is a little bit special.  In the case of the OP kludge, we are putting in a ground stake (the only one on the whole premises) but it is not bonded to neutral (at least not by us: perhaps there is a hidden path through the laptop power supply?).   Technically the small voltage/current associated with the laptop buzz wants to go back to the generating station (like a fault), not to the ground (like lightning), but apparently the earth is good enough at soaking up voltage/current because the OP kludge does eliminate the buzz reliably.  Given this fact, does that suggest a refinement to the OP kludge that would also work?

 

For the same reason, the OP kludge wouldn't offer any protection against hard faults of 100% mains voltage on the laptop case: the current would flow through the ground wire, through the stake, through the soil (which is very high impedance and thus the current is too small to trip any breakers), to the power pole's ground.  Is there a modification of the OP kludge that would actually offer this protection, like a properly-wired house would have?  Somehow tie it to neutral?  Where?

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.




×
×
  • Create New...