Power supply causes "Tingling" sensation

It is quite simple: The PSUs have input filters that put capacitors between the two power lines and also towards ground. In effect that gives you have the input voltage on the ground line as long as the ground line is not connected. You only get a slight tingle, because these filters are designed to be non-lethal when ground is not connected. The capacitors act as "AC-resistors" and let only a small current through.

It is the grounding. The only way to get rid of this problem is proper grounding. You can usually get propper grounding from a water-line. In fact that is how it is obtained in some houses in Germany: Three phases run to the house and the zero/ground is connected to the water line within the house.

If you have such a water line you can do an external grounding to make things safer: Run a wire from the chassis to a water-faucet. The only propper solution however is to get rid of this dangerous installation and do a propper 3-wire installation with good grounding in your house.

Understandable.

Regards, Arno

I agree, I live in a farmhouse in the uk and around 20 or so years ago the lead and copper water pipes were torn out and new "healthier" plastic ones put in, I still remember mums scream as she got a shock from the cold tap (just a small shock but as we work with 240 volts in the uk it could have been lethal), no one had thought about the houses earth was atached to the water pipe system, to fix it a 6 foot copper spike was driven into a wet area in the garden and the earth atached to it (turned out the washine machine was faulty and leaking power through its earth)

Drew

I have used copper pipes for ground but I suppose there is some risk from a replaced pipe as mentioned ...

The metal stake into the ground does work but the ground must be moist. One time when I used to live in the mountains I developed problems with a washing maching which would not run. When the motor would try to start the lights in the cabin would get _brighter_! When I told my landlord (a jack-of-all-trades mountain man) he went outside and started watering the ground next to the cabin! I thought he was crazy ... until I realized the problem was fixed. Turns out there was a ground stake there ...

Wayne

The problem wasn't fixed by the watering. The symptoms were masked.

The neutral, or one of it's connections, between the source and the panel in the cabin, was most likely bad. This resulted in current flow in the ground (bonded to the neutral in the panel), and most likely a semi floating neutral. Bad - bad - bad!

The issue should have been fixed correctly.

Louis

Good advise, since in many parts of the US the water service to your home is provided by non-conductive PVC or HDPE plastic lines. Worse still, if you do ground to your water pipes, the electrolysis created can eat through part of the plumbing in your home and a flooded basement sometimes results (as it did for me).

In my case, we had received "tingles" from touching the cover plates of outlets and switches. I had determined that the ground plates were at an elevated potential from ground, but it wasn't untill my daughter heard the "hiss" of leakign water in the basement that we realized why. It turned out that the final 6' of 3/4" copper pipe leading to the water meter has been eroded by electrolysis until it was paper-thin and starting to develop pinhole leaks. After installing roughly 80' of new 3/4" copper pipe in the house, and adding a 25' driven ground the problem vanished.

Never trust the judgement of the contractor that built your house, or the electrician that wired it. Neither is an EE, and while some of the craftsmen are more informed than others, not all are aware of the problems (safety and others) that blind, rote practices like this can cause.

Still to this day I simply cannot imagine how stupid an electrician is that can use a water system connected by non-conductive plastic pipe as a ground system, still it happened in the home I purchased.

Harry C.

I can't comment on the intelligence level of the electrician who wired your home, but the situation you describe is dictated by the NEC. To paraphrase the NEC; all piping systems in the dwelling shall be bonded to the main electrical panel. The water service, if it is metallic would qualify. The water service, again if it is metallic, and at least ten feet of the pipe is in contact with the exterior soil, qualifies as one of the required grounding electrodes. There are additional specifics as to exactly where, and how, the connection will take place (one reason is to avoid the example of a plastic section being used for repair setting up a deadly shock opportunity). The electrician was following a mandate.

While the surface of a ground rod will qualify as the second electrode, the best option I believe to be a Ufer Ground (a copper conductor in the footing of your structure) as long as possible. This works in new construction, or a buried copper wire ring surrounding the dwelling, and brought up to the service panel.

Louis--

The NEC does address the water service, and requires it to be bonded to the service panel.

The NEC allows an 8' ground rod to be used as the required supplemental grounding electrode. In some soils, and moisture levels, the effectiveness is less than stellar, but it does qualify.

The purpose of radials used by radio stations are related more to acting as an RF Counterpoise. They can however be a superior grounding electrode, as they are required to be bonded to the service panel.

Louis

Most PSUs have 5nF RF filter capacitors between the AC lines and the chassis ground that can cause a tingle, but I think Antec is one of the few exceptions and has RF filter capacitors only across the AC lines. The circuit board of my Antec SmartPower is designed with room for those 5nF capacitors, but it doesn't have any installed. Shouldn't the lack of line-ground capacitor prevent tingling, assuming that nothing else is wrong?

His rationale was that a section of water pipe might be replaced with PVC at some point in time.

If you use a self-grounding receptacle for the new 3-prong receptacle, no grounding jumper is required. If you do use a grounding jumper - that's what I do, whether or not it's a self grounding receptacle - you should use the right size: #14 for 15 amp circuits, and #12 for 20 amp circuits.

In article , Louis Bybee writes

What the NEC is doing here is not addressing the water service it is addressing the plumbing installation. The purpose of the equipotential bonding is to hold all extraneous conductive parts of the plumbing installation to ground potential. The water providers make their own arrangements and have their own codes of practice.

Yep they are for ground planes, nominally multiples of 1/4 wavelength.

Thanks for everyone's help and suggestions! I wound up borrowing around 5M worth of insulated copper cable from a mate, wrapped it around the house's "grounding port"/screw thingy, and then wrapped the other end to the grounding wire coming out of the surge protector. My computer is now zap-free. (Now if I could only solve the problem of the CMOS corrupting every time the system restarts... but that's another thread.)

Thanks again!

As I said, I don't know about Japan. My reference is several older houses in Germany. So it is definitely something to check locally.

It does. Again it does not work well when you really need low-resistance grounding, as for example when your utility delivers 3 phases but no zero (as in the cases of the houses I mentioned). Here I would strongly recomend having a qualified professional install and check the grounding. But to get rid of the voltage on his chassis, the original poster does not need "real" grounding, since it is high-impedance.

Arno

Funny _and_ impressive story. A true enginer... ;-)

Arno

Yes, it will. No tingeling in this case.

Arno