Static electricity causing reboot?

[Radeon962]

My secondary machine is in our den which is carpeted. More than a few times now, someone has been walking in to use the computer, sits down and touches the case and "crack" (or "pop" whatever you want to call the noise when static electricity discharges from the body to the case) and the machine reboots. I checked the plug today to make sure that it was grounded correctly and that checked out OK.

I checked to see if there was any thing inside that may have come loose. Nothing.

Mobo was installed using appropriate standoffs.

Did not find any wires, metal, etc. that was touching areas it/they should not.

The only thing I can think of is that possibly my PSU may have a faulty ground?

The machine runs fine after it reboots, but I know its not good to have a static charge running through the case with all the goodies inside it.

Could I just run a piece of copper wire from a screw on the the back of the machine to the screw in the outlet to try and ground the thing that way?

Any suggestions would be appreciated.

Thanks, Bill
It is a fairly new build:

Windows 7 Home Premium (32-bit)
Antec Solo
Corsair 400CX
Dell 2408WFP
Gigabyte GA-P55M-UD2
Core i5 750
Stock Intel Cooling
4GB (2x2GB) DDR3 1333 F3-10666CL8D-4GBHK
Western Digital Caviar Black SATA 640GB
Seagate ST3320620AS 7200.10 SATA 320GB
Western Digital Caviar GP WD10EACS 1 TB SATA
EVGA GeForce GTS 250 PCI-E x16 512MB
Scythe S-Flex 120mm/Scythe 100mm
Integrated Sound
ViXS Pure TV-U 48B0 PCI-E (NTSC/ATSC Combo)
Samsung SH-S223B 22x DVD +/- SATA2
HDHomerun Single Tuner (ATSC/Clear QAM)

This thread is canceled on account of trolling.

-ViRGE

 

[mfenn]

My secondary machine is in our den which is carpeted. More than a few times now, someone has been walking in to use the computer, sits down and touches the case and "crack" (or "pop" whatever you want to call the noise when static electricity discharges from the body to the case) and the machine reboots. I checked the plug today to make sure that it was grounded correctly and that checked out OK.

I checked to see if there was any thing inside that may have come loose. Nothing.

Mobo was installed using appropriate standoffs.

Did not find any wires, metal, etc. that was touching areas it/they should not.

The only thing I can think of is that possibly my PSU may have a faulty ground?

The machine runs fine after it reboots, but I know its not good to have a static charge running through the case with all the goodies inside it.

Could I just run a piece of copper wire from a screw on the the back of the machine to the screw in the outlet to try and ground the thing that way?

Any suggestions would be appreciated.

Thanks, Bill
It is a fairly new build:

Windows 7 Home Premium (32-bit)
Antec Solo
Corsair 400CX
Dell 2408WFP
Gigabyte GA-P55M-UD2
Core i5 750
Stock Intel Cooling
4GB (2x2GB) DDR3 1333 F3-10666CL8D-4GBHK
Western Digital Caviar Black SATA 640GB
Seagate ST3320620AS 7200.10 SATA 320GB
Western Digital Caviar GP WD10EACS 1 TB SATA
EVGA GeForce GTS 250 PCI-E x16 512MB
Scythe S-Flex 120mm/Scythe 100mm
Integrated Sound
ViXS Pure TV-U 48B0 PCI-E (NTSC/ATSC Combo)
Samsung SH-S223B 22x DVD +/- SATA2
HDHomerun Single Tuner (ATSC/Clear QAM)

Could be a swapped hot/neutral wire. Does the computer exhibit this behavior when plugged into another wall socket (or better yet, a UPS)?

Also, it could be the TV tuner, does the system reboot when the cable is unplugged from the TV tuner?

 

[Radeon962]

I do have it plugged in (along with everything else) to a Belkin surge protector, so not UPS but computer and other devices are plugged into the surge protector rather than directly in the wall.

No issue with either TV tuner.

Bill

 

[frostedflakes]

What part of the case are they touching? My Antec NSK-3300 has a grounding issue with the front USB ports that causes a reboot sometimes. Newer Antec cases actually addressed this problem by grounding the front ports to the chassis. Grounding myself to part of the chassis before plugging anything into the front ports seems to avoid the problem (also, it's only an issue during the winter when humidity is low, not during the summer).

If static discharge on the chassis itself is causing reboots, though, I don't know what to suggest. I've never run into that before.

 

[Radeon962]

Just the outside of the case. It's in a desk with the front part overhanging by about 3", so when they sit down to plug in an iPod or whatever it happens. Granted, it does not happen all the time and it seems that it has been more frequently recently.

I'll check the inside this weekend, when I have a chance to take things apart.

Thanks, Bill

 

[yottabit]

My Antec Sonata case does this on the front USB/mic jack ports... I know it is a reported problem on the older cases, but I can't help but think maybe it's more widespread than a lot of people realize

 

[Radeon962]

My Antec Sonata case does this on the front USB/mic jack ports... I know it is a reported problem on the older cases, but I can't help but think maybe it's more widespread than a lot of people realize

Were you able to figure out a fix for that? Once I have it open this weekend, I'll have to check it out.

Bill

 

[corkyg]

I have had it happen when working on the system and closing the side door. That caused a reboot. My fix was easy - I don't do that anymore.

 

[ecom]

Could be a swapped hot/neutral wire. Does the computer exhibit this behavior when plugged into another wall socket (or better yet, a UPS)?

I was thinking the same thing but I am suspecting open ground. If you have a multimeter or voltmeter you can do some simple tests to check to see if your outlet is wired right. Easier to use an outlet tester with 3 lights, but either tool will work.

Smaller slot is hot and bigger one is neutral. You should have 120V between hot and neutral and 120V between hot and ground. Neutral to ground should be 0.

 

[Radeon962]

I was thinking the same thing but I am suspecting open ground. If you have a multimeter or voltmeter you can do some simple tests to check to see if your outlet is wired right. Easier to use an outlet tester with 3 lights, but either tool will work.

Smaller slot is hot and bigger one is neutral. You should have 120V between hot and neutral and 120V between hot and ground. Neutral to ground should be 0.

I picked up an outlet tester on the way home from the office last night and the outlet checked out fine, so it is not that.


Bill

 

[yottabit]

My solution was unplugging the front panel ports. On my case, that was the only issue (case grounding itself seems to be fine)

 

[oynaz]

Do your machine have a reset switch? If yes, the discharge might short the switch, thus triggering the reboot.

 

[westom]

I picked up an outlet tester on the way home from the office last night and the outlet checked out fine, so it is not that.

A two part answer. The first will describe why so much posted is wrong or irrelevant - including that tester. Second will describe how to solve that problem.

What is the static electric discharge circuit? Outgoing electricity from hand. Through something conductive - ie computer, wooden table, power cord. Through floor to charges located between shoes. Long before discussing solutions, first the current path must be defined. Notice what is irrelevant. Earth ground. Receptacle safety ground.

Fundamental to eliminating static electricity are simple solutions such as humidity. Humidity is necessary for human health as well as making static electric charges difficult. No charge storage means no electric current from hand to ground beneath the shoes.

Meanwhile, static electricity is also a perfect test tool for defective wiring. For example, build up charges with leather slippers. Which wall switches and outlets are not properly grounded - do not make more conductive connections to the floor? If you touch something and it does not create much pain, then the conductive parts may not be sufficiently safety grounded - therefore also not make a good connection to the floor.

Why the #1 defect causes computer crashes. Motherboard ground and chassis ground should be separate except where both meet at a single point. You can prove this yourself with those leather slippers AND if the computer is mounted on a glass tabletop (because other materials such as wood are too electrically conductive). Simply static shock the chassis at various corners. The connection to floor (and charges beneath the feet) is probably through the power cable. So static discharges that cross the chassis will also pass through the motherboard.

Yes, the motherboard is one big copper ground sheet. Therefore the static shock creates a voltage difference across that motherboard. Current passing into the motherboard at one end and out at the other are causing a massive voltage difference across the board. That is causing your crashes.

Now, moving on to the second part – a solution: your working experiment (leather slippers and the glass tabletop) can reproduce that crash. Now change the circuit.

It is called a single point ground. Motherboard ground must connect to chassis ground at only one point - where the power cable and IO cards are also located. Only that one standoff should be conductive. All others nylon.

The tabletop experiment should no longer cause computer crashes. Static electricity is electricity. Without both an incoming and outgoing path, no static discharge passes through the motherboard - no high voltage differences on the motherboard ground plane.

Single point ground is routine in so many solutions for the same reason. And why, for example, so many interconnected grounds are also electrically different.

Your problem exists because a static electric discharge was passing through a motherboard. Install that single point ground solution so that static currents no longer exist on a motherboard. Do the experiment to both learn AND to confirm your solution has been properly implemented.

What could compromise that solution? How is a power switch or LED mounted? If not properly isolated from the chassis, then the experiment will identify defective connections that compromise single point ground. Just one example of why your solution is best confirmed using the above 'leather slipper and glass tabletop' experiment.

Discharge static electricity to any electronic device with great pain. That electronics must work 100% without crashing. If not, a human has made a wiring / design mistake. Static electricity is also a powerful diagnostic tool.

 

[Radeon962]

Do your machine have a reset switch? If yes, the discharge might short the switch, thus triggering the reboot.

Yes, the Solo has a reset button on the front below the main power button.

Bill

 

[Radeon962]

westom, thanks for the lengthy post... short and sweet are you saying to remove all of the brass mobo standoffs except for the one by where power cord/IO cards are located and replace them with plastic/nylon standoffs?

Worth a shot, but I've never heard of that before.

Bill

 

[westom]

Worth a shot, but I've never heard of that before.

Few first learn simple electrical concepts - such as why single point ground is a common solution. One can become an A+ Certified computer tech without any electrical knowledge. Which is why so many computer techs recommend all those conductive standoffs. Meanwhile, doing the experiment is important and educational. Fixing something without first making the problem predictable and repeatable can create confusion.

Then move on to the second solution. Room relative humidity should be 20% or higher.

 

[Googer]

My 1st guess would be to swap the power supply for a bigger and higher quality unit.

 

[Radeon962]

My 1st guess would be to swap the power supply for a bigger and higher quality unit.

That's what I was leaning towards. I picked up a Corsair 650HX as I wanted to replace the 400cx with a modular as I may just swap the whole thing out and put it in a Lian Li PC-V351B once they go on sale down the road.

Bill

 

[westom]

That's what I was leaning towards.

What is a larger power supply going to do? Vacuum up stray electricity? Nothing says and nothing is posted that says a larger power supply solves anything. That larger power supply is a classic example of junk science reasoning.

Static electric discharge is a current. Where does that current go to cause computer interruption? Provided was that answer (the reason why) AND a solution to eliminate that current AND how to determine if the problem is or is not solved.

Where is the reason that a larger supply does anything? None. Absolutely not even one reason provided. Not even one reason is the first indication that a larger power supply does nothing.

Scary - and demonstrated in this thread - is why the Silicon Valley must now hire mostly from India and China. Too many Americans know only because they 'feel'. So many today never learn how to think; that an answer without reasons why is popular junk science that even proved Saddam's WMDs.

Only one answer was provided with the always necessary reasons why, how to test that solution, and how to gain knowledge as a result.

 

[piasabird]

Maybe it is a bad power supply. A faulty power supply can cause a static problem. Might be a bad ground. Using one of those mats that go under a desk might help. Having some metalic surface to touch might help also.

Make sure the wall socket is really grounded. Wouldnt hurt to test it. However, My bet is a faulty ground in the power supply.

 

[Radeon962]

What is a larger power supply going to do? Vacuum up stray electricity? Nothing says and nothing is posted that says a larger power supply solves anything. That larger power supply is a classic example of junk science reasoning.

Static electric discharge is a current. Where does that current go to cause computer interruption? Provided was that answer (the reason why) AND a solution to eliminate that current AND how to determine if the problem is or is not solved.

Where is the reason that a larger supply does anything? None. Absolutely not even one reason provided. Not even one reason is the first indication that a larger power supply does nothing.

Scary - and demonstrated in this thread - is why the Silicon Valley must now hire mostly from India and China. Too many Americans know only because they 'feel'. So many today never learn how to think; that an answer without reasons why is popular junk science that even proved Saddam's WMDs.

Only one answer was provided with the always necessary reasons why, how to test that solution, and how to gain knowledge as a result.

Easy there, you sound like you might explode yourself. Not all of us are as smart as you and make posts to see if we can learn from others, not be talked down to.

Thanks for your input and opinions in any case.

My reasons for going with a "larger PSU":

The larger power supply is because I wanted a modular one and Corsair does not produce the 520HX or 620HX anymore both of which I have used in the past and still have 520HX in my other machine. They have a 650HX as their modular supply of choice in the states now. The 450HX is available outside of the US and would have been more than enough to run my or most rigs. The 400CX runs it fine, but I have read other posts on the web where other users have had issues with the grounding of the 400CX. So I started thinking about swapping out the PSU and that with wanting a modular one for a future move to a SFF case led me to my current reasoning.

I have been thinking of moving my system to a Lian Li PC-V351B and would definitely need a modular PSU to make the most of that box, so I figured why not pick it up now and swap out the 400CX to see. Nothing to do with size of PSU rather my upgrade path. Whether it fixes the problem, it's one less variable to deal since I wanted to change the PSU.

I do intend to go through your suggestions and already have dealt with turning up the humidifier. Newer house, so we have been toying with the settings to avoid moisture on windows when too high and static electricity when too low.

Thanks, Bill

 

[arnoldap]

I would like to suggest the following few points to short out the computer restart issues.

1- Proper grounding and earthing is necessary.
2- Touch the Plug with empty hand before inserting into your AC power source.
3- Check all the wires in your PC cabinet to be certain that no wires are touching the metal case of the Cabinet.

 

[Radeon962]

I would like to suggest the following few points to short out the computer restart issues.

1- Proper grounding and earthing is necessary.
2- Touch the Plug with empty hand before inserting into your AC power source.
3- Check all the wires in your PC cabinet to be certain that no wires are touching the metal case of the Cabinet.

Item #3 - How would you recommend making sure none of the wires are touching the metal case? I have the majority routed between the side of the case and under the mobo, so I know there would be wires touching there, but there are no bare wires touching anything as ALL wires are coated.

Are you saying that ALL wires need to be away from the case sides/walls even if they are normal coated wires?

Thanks, Bill

 

[Zargon]

this happens during winter around the office all the time.

 

[westom]

Are you saying that ALL wires need to be away from the case sides/walls even if they are normal coated wires?

Nothing visual can tell you anything useful. Wires can contact that case and cause no problems. Or wires can contact that case and cause problems. Parameters such as insulation's breakdown voltage, purpose of that wire, etc must be known. But then the 'glass top and slipper' test will make that obvious as posted earlier:

What could compromise that solution? How is a power switch or LED mounted? If not properly isolated from the chassis, then the experiment will identify defective connections that compromise single point ground.

Add a 'wire to chassis due to defective wire' as simply one of maybe 35+ other problems that were not listed because the list is too long and because that test would identify any or all.

BTW, reporting back the result of that testing is beneficial to all. Every so often, someone discovers a 'new' leakage path.

I will never understand so many who assume a larger supply means implies higher quality. The reverse is too often true.

A typical computer consumes about 200 watts. That means a 350 watt supply is more than sufficient. And since so many third party supplies measure that numbers differently (to hype a larger number), then some 500 watts supplies are the same 350 watt supply. But to increase profits, some manufacturer sell 600 and 750 watts supplies, forget to include essential functions, sell it at a lower price, THEN reap significantly higher profits.

No relationship exists between more power and higher quality. Neither is relevant to this topic. Your have defined a problem traceable to where a static electric current flows - and must not be flowing. A problem complicates and confused by other obviously irrelevant assumptions. Confusion so often traceable to A+ Certified computer techs that a national consumer magazine could not get even simplest computer problems properly fixed. Too many computer techs immediately wanted to replace the power supply because swapping parts is only what they understood. Power supply discussions are irrelevant to this topic.