• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

70 V AC from Ungrounded Case to Ground!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

shagrat

Registered
Joined
Dec 6, 2003
Location
Norway
I fried my amplifer today when contacting a RCA from my motherboards sound-out to it. As I contacted it I felt the current go through me. After that only one of the channels on the amp was working:( .

Anyways, I started to investigate this strange behavior. When connecting a multimeter from the case when the computer was powered up to the amp, I got 70V AC :eek: :eek:! I coulden't belive what I saw, so I connected another multimeter and a analog voltmeter too. Still 70V AC.. The same voltage was also read from the sound connectors and other connectors on the case, to the amp.

After alot of testing I came to the conclusion that since my computer is hooked up to a non-grounded power outlet, it was grounded through the phono-wire, to the amp, from the amp to my receiver and from my receiver through the coaxial cable thats grounded :rolleyes: . Yeah, I know that this sounds crazy, but take a look at this:

When diconnecting the coaxial-cable the reading on the voltmeter went from 70V to 25V AC! Obviously it still shorted to ground somewhere else, but it was now halfed. I then ran a powercable from my bathroom to the rom with the pc and connected the voltmeter from the case to the ground wire on the powercable. Same results, 70V AC from the computer to ground.

This lead to me experimenting a bit more: First I thought that my PSU was faulty and swapped it for another one. Still the same results. Then I disconencted everything from my PC exept the powercable ofcource. Still 70V AC. Then I swapped out the whole PC. Still all fuc£ed up. I then tested my two linuxboxes to ground from the case and also here the 70V AC was seen.

My conclusions is that this floating AC current is supposed to be there. If you use grounded powercables and outlets you'll have no problems. If you hook it up non-grounded you'll have no problems as long as you have no way for the current to ground.

So for my big questions:

Is this normal (it has to be ;))?
What is this current for/the result of?
Have you experienced anything like this?
 
Same here I don't know why it's 70VAC but I would never run a computer that wasn't grounded because it's supposed to be grounded for a reason.
 
This doesn't sound normal. If the case constantly has 70vac potential on it, you would have gotten nailed the first time you touched it, and were grounded. Here is a little test, meter the frame (unpainted metal) of your fridge, or any other household appliance, to ground. You should have 0 volts. If you measure anything other than 0 volts, you need to have an electrician check your breaker box/entrance service for proper grounding.
I have seen some older mobile homes with similar problems, before the more stringint grounding rules were developed for such structures. Since you are not using a grounded plug on your computer, because your outlet is not a grounded outlet, that tells me your not in a mobile home, but more likely in an older home, built before grounded outlets were common (possibly the older knob style wiring).
If I'm correct, and your home is older, and you do not have grounded outlets (except for the bathroom), I strongly urge you to carefully conduct the tests above.

"After alot of testing I came to the conclusion that since my computer is hooked up to a non-grounded power outlet, it was grounded through the phono-wire, to the amp, from the amp to my receiver and from my receiver through the coaxial cable thats grounded . Yeah, I know that this sounds crazy, but take a look at this:"

You are correct, that sounds like a perfect path to ground. Be careful when metering potentials to ground, as you discovered earlier, the human body is an excellent conductor of electricity and accidentally touching the metal tip of your meter leads or the frame of the appliance your testing could lead to another shock if, in fact, you do have a problem.

One last note, the best ground to use for testing is not the ground from your bathroom outlet. Use a metal cold water pipe even if you have to run a length of wire from it to your test site.

Good luck and for goodness sake BE CAREFUL!
 
Deathknight said:
Wow no way on earth I would run my machine on an outlet that was not grounded.

You always want to ground a circuit properly. It's actually pretty dangerous if a circuit is not grounded properly.
 
Could it be the low-value capacitors connected between each AC line and ground, like C2 and C3 in this ATX PSU schematic (English also available) ? With no connection to earth ground, won't these capacitors cause the case to be at approximately 1/2 the AC line voltage?

I'm not saying that you should get rid of those capacitors, but some PSUs, like Channel Wells and Antecs made by them, don't have those capacitors. Of course, if any other ungrounded equipment connected to the computer has similar capacitors (the monitor may), then they'll put 1/2 AC voltage on the computer case, too.
 
The home is about 12 years old. When it was built it was not common (atleast here in Norway) to use grounded utlets for anything except "wet-rooms". It's still no regualtions forcing to use grounded outlets except for these kind of rooms (but most new buildings use it anyways. I seriously have never tought about this being a potencial risk before this. Aa a mather of fact my dad is an electrican (he works as a teacher in the field now), and he installed all the electrical components in our home. When I told this to him he was shocked to say the least. He knew there would allways be a potential risk with not using ground, but he had never heard of this floating current in computers (I actually had to show it to him on all the four computers and three voltmeters :p). Almost all that I know uses their computer without ground. Haven't got around to measuring from the fridge yet. And also, my father told me that we haven't got a "ground-failure-switch" (don't know the english word for it ;)).
 
GFCI is I think what we call it. GFCI = Ground Fault Circuit Interuption which pretty much senses when someone or something is about to get shocked and shuts off the power before this happens. It does this by sensing minut changes in the current bettween the two leads. These can be installed in ether the main circuit panel itself or the actual plug on the wall. The thing with those are that if you have a UPS you don't want it pluged into a GFCI protected outlet because normal UPS units always have a small leak into the ground circuit that can sometimes trigger the GFCI protection.
 
Last edited:
Grab a drill, some cocking (sp?), a shovel and some wire and make your own ground. Just run the wire from you house, to the outside and then run in around in the dirt (and cover it).
 
9mmCensor said:
Grab a drill, some cocking (sp?), a shovel and some wire and make your own ground. Just run the wire from you house, to the outside and then run in around in the dirt (and cover it).

Much easier to just ground the outlets in my room then :p
 
9mmCensor said:
Grab a drill, some cocking (sp?), a shovel and some wire and make your own ground. Just run the wire from you house, to the outside and then run in around in the dirt (and cover it).
Caulking?

I dont think thats the best way to ground. Especially when you can use less effort and ground to a water pipe like mentioned before. Its copper and travels underground for miles. Sounds like an easy way to ground to me.


I dont think sticking a wire a few feet under the ground will work properly.
 
ways to ground or ways to spell caulking :D

I think the proper way to ground without using water pipes is to drive a pole into the ground. IMO sounds easier than digging.
 
Grounding should be either through the cold water pipes, or through a grounding rod driven into the ground. Ground rods are copper or galvanized steel 1/2 inch diameter round stock which is 6 to 8 feet in length and should be driven into the ground until only projecting out about 3-4 inches. Your ground wire is then strapped (using a bolt fitting) to the rod creating a proper grounding circuit.

The dedicated ground created by the grounding rod is the preferred method in my part of the USA. We have one on every house.

Don't skimp on this part of the project, do it right.
 
eaglescouter said:
Grounding should be either through the cold water pipes, or through a grounding rod driven into the ground. Ground rods are copper or galvanized steel 1/2 inch diameter round stock which is 6 to 8 feet in length and should be driven into the ground until only projecting out about 3-4 inches. Your ground wire is then strapped (using a bolt fitting) to the rod creating a proper grounding circuit.

The dedicated ground created by the grounding rod is the preferred method in my part of the USA. We have one on every house.

Don't skimp on this part of the project, do it right.

We already have a device like this installed. Just need to hook it up to my outlets :eek:
 
stompah said:
Caulking?

I dont think thats the best way to ground. Especially when you can use less effort and ground to a water pipe like mentioned before. Its copper and travels underground for miles. Sounds like an easy way to ground to me.


I dont think sticking a wire a few feet under the ground will work properly.

In the US, as I understand the codes (father is contractor), you need to drive a metal bar (possible copper) 7 ft into the ground. TO be a acceptable ground.
 
Back