PC reboots if someone walk by

Lava King · 01 Mar 2010

If I may jump in here...

I've read this entire thread and I agree that it seems to be a grounding problem, or more specifically, an electrical noise problem.

I see that you have made sure that the motherboard is truly grounded to the case, but have you made sure that the case is grounded when it is plugged in?

To test this, first REMOVE the AC cord from the PSU. Now, set your DVM to Ohms at the lowest setting (probably 200 or 20) or continuity if it has that option. Touch the probes together to make sure it's working, you should see a very low reading (ideally 0 but more likely something less than 1 Ohm). Now touch one probe to the bare metal of the case and the other probe to the "prong" in the CENTER of the AC connector on the PSU. A reading of 1 Ohm or less indicates that the ground connection is good. Additionally, you can check your AC cord (disconnected from the wall socket and the PSU) for continuity while you're at it.

If the case is grounded correctly to the PSU, there is one more thing I can think of to check. Get an outlet tester (about 5 or 6 dollars) from your local discount store or hardware store. In case you haven't seen one, it looks like an AC plug without a cord. There are 3 neon lights in it that can reveal any wiring problems depending on which ones illuminate and which ones don't. If there's a wiring problem, then it's time to call a professional electrician.

If, however, all the wiring is correct, I don't know what the next step would be. Call an exorcist, perhaps. :)

An important safety reminder: NEVER attempt to use an ohmmeter or DVM set to Ohms on a powered circuit!

Keep us informed, I'm sure that everyone reading and responding to this thread is curious as to what the real problem is.

Kent

Dwarf · 01 Mar 2010

Darryl Licht said:

Have you tried the on off switch in place of the reset switch... they are the same except a reset is a momentary switch... while on off is a regular switch. This might help show if it is just a bad/shorted reset sw... You can turn on system by using a spare jumper and shorting the on/off switch pin header on mobo.

I've already suggested swapping the switches over. Incidentally, they are both momentary push-to-make switches. The only difference is that the operating button of the power switch is usually bigger than that of the reset switch.

stormy13 · 01 Mar 2010

Just wondering if you have tried contacting CM and seeing what they have to say,

Cooler Master Forum • Index page

Maybe even try their live chat,

http://www.coolermaster-usa.com/live...department.php

churin · 01 Mar 2010

Lava King;
The resistance between gound pin of the AC connector and one of the nine screws to fasten the mobo to the stand-offs is measured to be less than 0.2 Ohms(Reading resolution of the meter is 0.2 Ohm). I tested the AC cord as suggested and found no problem.

I have the AC wiring tester and it indicates normal as shown below picture.

churin · 01 Mar 2010

stormy13 said:

Just wondering if you have tried contacting CM and seeing what they have to say,

Nobody was on-line so I left a message.

Dwarf · 01 Mar 2010

Ok, here's another suggestion. You will need some wire for this. Strip the insulator from the wire at one end. You need to strip sufficient so that you can wrap the bare wire around the switch and then twist it together to ensure that it is firm. Route this wire away from the main reset cable and connect the other end to a point that you know is connected to ground such as the case of your PSU (or indeed use one of its mounting screws).

Darryl Licht · 01 Mar 2010

I advocate you try sandpaper, star washers, or similar around ALL grounding points... mobo standoffs, psu, switch mounting screws, etc. I bet the paint or powdercoated interior is the issue somewhere, this isnt a very old case and may it be an issue.

Its worth a try... and you've gained skill no doubt at removing and replacing your mobo.

churin · 01 Mar 2010

Dwarf said:

Ok, here's another suggestion. You will need some wire for this. Strip the insulator from the wire at one end. You need to strip sufficient so that you can wrap the bare wire around the switch and then twist it together to ensure that it is firm. Route this wire away from the main reset cable and connect the other end to a point that you know is connected to ground such as the case of your PSU (or indeed use one of its mounting screws).

I am unable to visualize exactly what you suggest that I do but what has been done and was posted before is the following:
The wire strand for the reset switch was wrapped around by aluminum foil for about four inches from the mobo side and the foil was grounded via a copper wire to the nearest mounting screw(one of 9 screws for ATX).
You stated "wrap the bare wire around the switch" but the switch assembly itself is not accessible. My experiment so far done indicates that the rebooting(resetting) does not happen if the discharge current is bypassed the reset cable.

churin · 01 Mar 2010

Darryl Licht said:

I advocate you try sandpaper, star washers, or similar around ALL grounding points... mobo standoffs, psu, switch mounting screws, etc. I bet the paint or powdercoated interior is the issue somewhere, this isnt a very old case and may it be an issue.

I would think the reasoning behind this suggestion must be that the mobo may not be adequately grounded to the case. However, doesn't the the result of the test suggested by Lava King prove otherwise?

Dwarf · 01 Mar 2010

Is it possible to have a shot of the inside of the case showing where the switches are mounted with the switches in situ? I need to be able to see the back of the switches clearly.