UPDATE: How to tell if mobo is shorting out on I/O panel?

[BZeto]

I emailed Epox tech support about my 8K5A2+ bootup problem (not sure if is the mobo or the PSU).

They said to check an make sure the mobo isn't shorting out somewhere on the case. They said to check the I/O panel (custom metal cover for the ports on the mobo) and make sure its not shorting the mobo.
How do I tell? Just make sure its not actually touching it?
What would cause the short?

SEE LAST POST FOR UPDATE.

 

[RalfHutter]

That's wierd, AFAIK the MoBo is supposed to ground out on the I/O plate, that's why they have those springloaded "fingers" that rest on the mobo ports.

 

[KF]

I don't see how it could short out to the panel around the I/O sockets.

What people sometimes do is short the motherboard by accidentally having an extra standoff besides the ones at the mounting holes. What you can do is put the mobo on a box next to the case and hook up the power and other plugs to see if it boots OK that way. If so, check those mounting blocks.

 

[BZeto]

Hmm.. When putting the mobo in the case I was POSITIVE there were no extra standoffs.

I just used the brass standoff and the screws that came with the case. I didn't use those paper washers but that shouldn't matter anyway, right?

 

[mdcrab]

You should not have to use the paper washers, they are a thing of the past for most mobo's. If you can get it to boot outside of the case, you coud install it in the case with I/O shield removed to eliminate it as a possible source of grounding.

mdcrab

 

[BZeto]

Maybe I'll get around to doing that sometime... Its just such a pain in the ass, I dont feel like messing with it right now.

 

[Buz2b]

Just what the heck do they base their possibility that it could be the I/O panel or case shorting? It sounds to me like they are just "fishing" right now. Can you give some more details as to what the problem is and what you have done to try and solve it? Also would be helpful to provide more details on the system, if you want.
As far as the PS, that would be pretty easy to test if you could borrow or "hijack" a friends and swap it out just to test things and see if the problem still occurs.

 

[mechBgon]

I've seen the springy "finger" things on the I/O panel slip into the USB and network jacks before. Check to ensure they're not getting into the jacks.

 

[BZeto]

Originally posted by: Buz2b
Just what the heck do they base their possibility that it could be the I/O panel or case shorting? It sounds to me like they are just "fishing" right now. Can you give some more details as to what the problem is and what you have done to try and solve it? Also would be helpful to provide more details on the system, if you want.
As far as the PS, that would be pretty easy to test if you could borrow or "hijack" a friends and swap it out just to test things and see if the problem still occurs.

Ok, here's the problem...

When ever I completely remove power from my system (unpluging power cord or flicking the switch on the back of the PSU), the next time I plug it back in and press the power button the machine does not POST. The fans and CD drives spin up but the monitor doesn't recieve any signal. It just hangs there with the HDD activity light lit up (the hard drive is NOT doing anything though). So I manually press the reset switch, this time the monitor gets a signal, but the PC will usually freeze at the BIOS info screen, not even close to loading windows. So on the second hard reset it will usually complete the POST and boot into windows, no problems after that. It has taken up to 3 hard resets to fully boot though.
Now remember, this is only when complete power is removed from the system. If I do a normal shutdown without unplugging the PSU it boots up fine the next time.

There you have it, an extremely annoying problem (for as much as I work on my PC) that I would like to solve.

I'm gonna check to see if somehow the IO panel "fingers" are inside one of the ports...

thanks.

 

[mechBgon]

You might also try a fresh CMOS battery, they're only a couple dollars.

 

[Buz2b]

Originally posted by: mechBgon
You might also try a fresh CMOS battery, they're only a couple dollars.

Agree and if not that, it is sounding like a PS issue. Again, if you can, try to find a friend, family member, etc that you can "borrow" a PS from for a test on your system. Just make sure it is a bit better PS than yours.

 

[BZeto]

Is it true that if I can get the old CMOS battery out and the new one in pretty quickly I wont lose my settings?

 

[BZeto]

Here's a little pic I drew so you can see how the I/O panel is touching the mobo.

Non of the fingers seem to be inside the ports, only between them.

Does that eliminate the I/O panel as the problem?

 

[mdcrab]

Had trouble downloading the drawing, but the mobo should not be touching I/O shield. Is the shield fully seated in the case openning? Sometimes you have to apply quite a bit of pressure to get them to engage the detents along the edge of the shield. You may want to check that it is all the way in (out).

mdcrab

 

[pspada]

You drawing looks like the i/o cover is installed correctly. The little "fingers" are supposed to press on the sides of the parallel/serial ports, and in between the usb ports. It is unlikely to be your problem. But I have seen a similar problem that turned out to be the power switch itself. You can check this easily, just remove the wire to the power jumpers, and short them with a screwdriver.

If this does not help, you are best off removing the mobo from the case and trying it, as suggested above.

 

[BZeto]

Originally posted by: mdcrab
Had trouble downloading the drawing, but the mobo should not be touching I/O shield. Is the shield fully seated in the case openning? Sometimes you have to apply quite a bit of pressure to get them to engage the detents along the edge of the shield. You may want to check that it is all the way in (out).

mdcrab

You mean the mobo itself? Becuase it is impossible for the metal on the ports not to be touching the I/O panel, unless I cut the little fingers off or something. I'm possitive the I/O panel is completely in.

pspada: It pretty safe to remove the mobo from the case and try it, right? Should I just put it on a peice of cardboard but have everything else hooked up the same?

 

[mdcrab]

The metal on the ports is suposed to touch the fingers on the I/O shield. The edge of the PCB substrate should no touch the I/O shield.

mdcrab

 

[BZeto]

Ok, the PCB isn't touching.

I am going to try and boot it up outside the case (hopefully my power switch wires will reach). Is there any risk to this?

 

[mechBgon]

Originally posted by: BZeto
Ok, the PCB isn't touching.

I am going to try and boot it up outside the case (hopefully my power switch wires will reach). Is there any risk to this?

Not really, except if you accidentally give it a static electrical shock. Just touch a bare metal grounded object frequently while handling it. Good luck!

 

[Buz2b]

Originally posted by: BZeto
Ok, the PCB isn't touching.

I am going to try and boot it up outside the case (hopefully my power switch wires will reach). Is there any risk to this?

If the wires don't reach, that is not a big deal. You can start the system by "connecting" the two pins that the switch plug goes over, wtih a flat screwdriver briefly. Make note of the two pins that the switch plug wires go onto. Then pull them off if they don't reach. When you are ready to test start the system, just take a flat blade screwdriver and briefly connect those two pins by touching the blade to both at the same time. The system will start. I have done this dozens of times, as have others.

 

[pspada]

Originally posted by: BZeto

Originally posted by: mdcrab
Had trouble downloading the drawing, but the mobo should not be touching I/O shield. Is the shield fully seated in the case openning? Sometimes you have to apply quite a bit of pressure to get them to engage the detents along the edge of the shield. You may want to check that it is all the way in (out).

mdcrab

You mean the mobo itself? Becuase it is impossible for the metal on the ports not to be touching the I/O panel, unless I cut the little fingers off or something. I'm possitive the I/O panel is completely in.

pspada: It pretty safe to remove the mobo from the case and try it, right? Should I just put it on a peice of cardboard but have everything else hooked up the same?

Yes, cardboard or foam will work fine. Just make sure that none of the pieces is grounding out on anything. And shorting the power jumpers with a screwdriver is the best way to make sure the power switch itself is not the problem.

 

[BZeto]

I dont know why, but the idea of shorting something doesn't sound too comforting. Anyway, I'm gonna try that today with the mobo still in the case and everything, that way I can be sure if its the switch or not.

 

[Buz2b]

Originally posted by: BZeto
I dont know why, but the idea of shorting something doesn't sound too comforting. Anyway, I'm gonna try that today with the mobo still in the case and everything, that way I can be sure if its the switch or not.

LOL! Really, it is not a problem as long as you short/cross connect the two pins that the switch connects to. Remove the connectors first, since it will still be in the case.

 

[BZeto]

Ok, unplugged the cord from the power supply, unhooked the power switch wires, plugged PSU back in, and shorted the power headers. Same thing, no boot up on the first try. But on the first hard reset it booted up fine! Usually it will take a couple resets.

The odd thing is, at the bios screen it said the cpu was running at 1.8 ghz (oc from 1.4) but in windows cpu id, mbm5 and everything else showed 1.4 ghz, this usually happens after removing all power. So I reboot and go into the bios, it already has my OC settings saved, so I just save and exit and it boots to 1.8ghz! It would probaby boot to 1.8 without going into the bios...
With regular shutdowns (not unplugging psu) it boots to 1.8 everytime I turn it back on.

What could the problem be?
I think its narrowed down to:

CMOS battery
Mobo shorting out
Power supply

 

[Buz2b]

Originally posted by: BZeto
Ok, unplugged the cord from the power supply, unhooked the power switch wires, plugged PSU back in, and shorted the power headers. Same thing, no boot up on the first try. But on the first hard reset it booted up fine! Usually it will take a couple resets.

The odd thing is, at the bios screen it said the cpu was running at 1.8 ghz (oc from 1.4) but in windows cpu id, mbm5 and everything else showed 1.4 ghz, this usually happens after removing all power. So I reboot and go into the bios, it already has my OC settings saved, so I just save and exit and it boots to 1.8ghz! It would probaby boot to 1.8 without going into the bios...
With regular shutdowns (not unplugging psu) it boots to 1.8 everytime I turn it back on.

What could the problem be?
I think its narrowed down to:

CMOS battery
Mobo shorting out
Power supply

Yep. Switch appears not to be the problem as it seems that the behaviour followed. Here's where you have to make a decision. Spend a couple of bucks for a new CMOS battery and try that or take the MB out of the case to try that. Personally, I'd try the battery. It sounds like you got a bit of a "hint" with what you described. I'd also leave the switch off for now while you test the new battery. If that doesn't make any difference, then you would want to pull the MB and try booting outside of the case. If you do this, make sure you keep as much plugged into the MB (IDE and optical drives, fans, etc) as you can. Reason is that if it is a PSU problem and you unplug most things while it is outside of the case, the system might seem like it was fixed, when actually the PSU just had less stress on it and was able to function properly. You don't have to worry about the switch and LED MB connectors, just the drives and fans.