USB static discharge causes instant reboot on GA-MA785G-UD3H

[scottlindner]

I have already posted this in this thread (http://forum.giga-byte.co.uk/index.php/topic,735.0.html), but to keep the threads simple and clean I'm starting another one specific to this problem.

When I connect a USB device my system instantly crashes and reboots.  After testing this for a while I have determined it is caused by static discharge.  Even if I turn the USB connector 90 so the shroud on the case prevents even the outer shielding of the connectors from touching, the static causes the system to reboot.  This could be caused by my motherboard (GA-MA785G-UD3H), my case (Antec Nine Hundred Two), my power supply (CORSAIR CMPSU-550VX), or some combination of them together.  I have read in other forums about this problem being caused by power grounding of the motherboard to the case.  In those cases it was due to the installed using the red insulating washers on the screws that hold the motherboard in place.  I did not use those.  I have good electrical contact with my case.  However, I cannot assume this means that everything is grounded properly.  Maybe my case's front USB panel isn't grounded as part of the case itself?  Maybe the mobo is overly sensitive and I need some form of USB buffer in between?

Does anyone have any good suggestions for this problem?  I suspect at the end of the day this will be a case design problem, but I'm posting here since the mobo is involved.

Cheers,
Scott

[Pottypete]

Hi,
Looks like your going to have to check every item component on your pc for correct earthing continuity,
Starting from your power supply socket,
And have you tried a static band on your wrist?,
Don't look like any other way to do it easy,
Definitely an earthing fault somewhere,
Let us know how you get on,
Years ago I would have used a MEGA,
Regards

[scottlindner]

I have found several posts about this issue regarding the case that I have (Antec 902) so it is most likely the case.  However, I'm confused why that would matter.  I did some static testing (I know, bad idea) on the USB ports.  I zapped the USB ports on the back of the motherboard and the system continued running just fine.  I zapped the front ports on the case after connecting the outer shielding to another ground and it rebooted instantly.  These connect directly to the motherboard.  Why would the motherboard behave differently depending on which USB port is zapped?

Scott

[Pottypete]

Hi,
Glad your getting somewhere
Don't know how your front ports on that case are set up , it might be worth taking the front little panel assembly out of the case and having a look for shorts or duff/dry solder connections,
You could try also removing reset switch panel connection from the mobo, and the power one in turn to see if your still have the same problem,
before removing front panel assembly,
Have the other peps with the same problem resolved it,?
Has Antec support been contacted about this problem, maybe they have a solution if it is a faulty assembly that they are aware of,
Questions Questions I know , but it is very strange fault,
Post back how you get on,
Regards

[scottlindner]

You're making good points.  As far as I can tell there is no know solution.  I'm confused why it would be the case and not the motherboard and that's why I'm posting here.. just in case.  But it looks like the problems are with the case just by looking at the numbers reporting the problem.

I couldn't figure out how to remove the assembly.  I tried.  I think it's rivited in place and there's no way to get at it.  Maybe I'm wrong on that.

I'm going to call Antec on Monday to see what they have to say about it.

I just thought of something I didn't try to remove the assembly.  I'll try that tomorrow and see how that goes.

Scott

[Pottypete]

Hi, Scott
Did you try disconnecting the power and reset cables where they are plugged into the mobo ? and testing?
Have just seen peps reporting on ebuyer.co.uk with that case having bad connections to power connections wires from switches and STATIC buildup from the front headphone sockets, check it out,
That switch/ socket must be accessible somehow
Cheers
Ps
As you say other peps are having simular problems with Static on that case, IE,.

For static issues, I think you can get one of those circular washers that block out radio waves if you loop your audio ports around them.
--------------
My cases for some reason do not have that problem (Intel Gigabyte motherboards, perhaps). Those "circular washers" are called ferrite toroids (powdered iron doughnut).
------------------------------
.I googled for it and it appears to be caused by putting a common earth wire to all the front panel ports. In other words, Antec's fault. If anyone have a solution for this problem I'll be grateful.
----------------------------
As you say give them a buzz,
Regards
Pete

[scottlindner]

I'm still not convinced it's the case, but odds are it is.

I checked my theory out, there's no way into the front panel.  It's all rivited together.

I had another theory it was due to poor grounding of the front panel.  There is a grounding wire that comes out and is screwed with a sheet metal screw into the top part of the case.  It occurred to me last night all of the sheet metal has been painted gun metal black.  I bet there is no real good ground on the top part of the case due to the paint acting as an electrical insulator.  So I added another wire to the grounding screw and connected it directly to one of the motherboard screws that hold it to the case.  This should be a perfect ground.  It is still happening with the front USB ports but not the ones in the back on the mobo.

What is really bothering me is that both are on the motherboard.  Even if the case isn't properly grounded, I still don't understand why it is reseting when the USB ports on the motherboard are zapped, but it doesn't when the ones in back are not.

I'm calling Antec tomorrow.  Thanks for the discussion.  I will be certain to post back the results.

Cheers,
Scott

[runn3R]

(...) I had another theory it was due to poor grounding of the front panel. (...)

You are right. I met similar front USB ports reboot problem before.

(...) I still don't understand why it is reseting when the USB ports on the motherboard are zapped, but it doesn't when the ones in back are not.(...)

The back USB ports are properly grounded, while the font ones are dependent on the chassis and internal USB cable design which seems to be faulty.

[scottlindner]

(...) I still don't understand why it is reseting when the USB ports on the motherboard are zapped, but it doesn't when the ones in back are not.(...)

The back USB ports are properly grounded, while the font ones are dependent on the chassis and internal USB cable design which seems to be faulty.

I thought about it more and realize what I was missing when I wasn't ruling out the mobo.  The USB ports on the back of the mobo have shielding on them directly connected to the motherboard's ground, the USB header on the mobo for the front panel has no shielding, although it does provide a ground.  This could be part of the problem, but as you've noted, it's more likely the front panel.  I was missing the difference between the back USB ports and the headers on the mobo.  Now it makes sense to me.

You'd think I'd be quicker on this since I have a BS in Electrical Engineering.  Guess I got ripped off.   

Antec is sending me a new USB module to try out.  I'll post back when it is resolved.  I'm fairly convinced it's a case problem now.

Cheers,
Scott

[Pottypete]

Hi
There are some 3.5 and 5.1/2 drive bay USB  front panels available, thought about one?
It would knock the case front usb connectors out of the loop,
You could also  try connecting a rear USB bracket to the rear slots,
Did you try a ferrite ring fitted to the  front USB panel wiring?,
Any port in a storm,
Keep us informed,
Regards

[scottlindner]

I did consider a 5.25" front mounted media reader that includes USB ports and just skip the case front USB ports.  They are really handy so I'm going to give this a shot, but if this replacement from Antec doesn't work I'm going to get a good media readier with USB ports just as you suggested.  Part of this is solving the problem now, and not just getting a workable solution.  I'm sure you're familiar with that, just gotta figure it out for the sake of figuring it out.

I'll be sure to leave everyone know what's going on.  I also plan on posting how this works out on the NewEgg reviews since about five people have posted the same issue.

Scott

[Pottypete]

Hi,
Just dug up some old info,
Looks like you have to check your pin out connector with the mobo pinout in the manual,
I know there old posts BUT,

The USB port consists of:

1) Connector shield. That is the metal body of the USB connector.
It makes contact first. The shield on the computer side _MUST_
be connected to ground. When an appliance is "hot plugged",
static electricity on the USB cable is transferred to ground
when the shields of the two devices touching first.

2) Take a look at the pins inside the USB connector. The outside
pins are longer than the inside pins. The outside pins are
+5V and GND. They make contact before the two shorter data
pins. By having the power pins connect first, that prevent
currents from flowing through the data pins without a logic
ground in place.

Everyone should check their USB front port wiring. Make sure
that the shield around the four pins, is at ground potential
with respect to the motherboard. This is complicated by the
fact that the motherboard header only has enough ground pins
for the signal pins.

/ USB+5V (VCC) X X (VCC) USB+5V ----> \
To __/ USB_P5- (D-) X X (D-) USB_P6- ----> \__ To
USB \ USB_P5+ (D+) X X (D+) USB_P6+ ----> / USB
pins \ GND (GND) X X (GND) GND ----> / pins
/ X NC \
_____/ \______ GND must also
GND must also go go to connector
to connector shield metal!
shield metal!

Check the design of the front USB adapter plate, and make
sure that the connector shield is grounded. Using an ohmmeter,
you should measure zero ohms between the shield of the USB
connector and the bare metal of the computer case (as the case
is grounded via the PSU case and via the brass standoffs that
contact the ground rings on the bottom of the motherboard).

If the connector shield is well grounded, perhaps future
"hot plugged" USB devices will not destroy any more motherboards!

Or, simply stop using front mounted USB. Only use the USB ports
on the back of the computer, as the ground is established by
the motherboard design. If you don't own a multimeter, or
don't wish to fiddle with wiring, just stop using the front
mounted ports. You can always bring a USB cable from the
back of the computer, up to your desk.

Regards

[scottlindner]

I just received and tried the replacement USB module Antec provided.  I'm having the same problem.  I seriously am wondering if the mobo is somehow involved with this problem considering all of the tests I have tried.

I'm going to try a front mounted media reader with USB ports and see how that goes.  If it works, I'll just disconnect the front ports on the case.

Scott

[troy216]

I am having a similar problem with the same motherboard but a different case.  My case is the APEVIA X-MASTER HTPC.  I get the reboot whenever i zap the front part of the case.  The closer to the on/off button I touch, the less of a zap is needed.  In fact, I can touch the on/off button and not even perceive a zap and it will reboot.  I suspect that the ESD is making its way into the power on/off or reset wires and into the motherboard's header pins.  So it may not be your USB ports, just that they are close to the switches that connect to these headers.

Did you follow the recommendation of the earlier poster to disconnect these headers?  I will try myself to see if it resolves my issue.

[Pottypete]

Hi,
More info from the web

http://tw2005.giga-byte.com/Motherboard/Support/FAQ/FAQ_456.htm

It seems that some (older) cases (and I am not sure it is all Antec here) have an extra ground wire for the 9th pin on the USB header. Now if you read the Asus mobo pinout it says N/C for that pin. So by having this wire hooked up to the 9th pin I am assuming that is is dumping some static electic into the circuit and that is what might be starting off the burning up process. I also assume Static Electric that is setting it off b/c it does not happen all he time. So I call Antect to confirm my theory and I was told that in other words, most mobos/systems would shut down if you had the 9th wire (ground) hooked up. However it causes the S bridge to fail. Antec says this wire is not to be hooked up if the pin out says N/C (no connection) or O/C (open connection). Now some of my new cases that i have recently built with Coolermaster and my own personal Antec P160W does not een have a 9th wire. So it seems that they are dropping this ground out. I wonder why....
Have a look at this:


It seems the ICH5/ICH5R are sensitive to static discharge. If
the static enters a USB port (the USB port is hosted by the
Southbridge ICH5), it causes an electrical phenomenon called
"latchup". Latchup is where a phantom PNPN semiconductor junction
forms - that is the same structure as an SCR. Latchup turns on
the junction, and the junction sits across the power rails of
the chip. A _large_ current flows through the ICH5 and it gets
really hot - so hot in fact, that it can burn. The Southbridge
can be destroyed in a matter of seconds.

One solution,
Just reporting that the solution of grounding the MB by removing the red insulator on one of the screws securing the motherboard to the case has seemed to fix the reboot problem when I touch the USB port.

I am guessing that my MB was not able to dissipate the static discharged because it was not properly grounded. I pretty much use the insulators on all builds, but I think in this case, the head on the screws I used this time were small enough that they never made contact to the MB when I used the insulators.
solution 2,.
Ok, so you don't know how long I have been scouring the internet to try and find a solution to the one single flaw of this case: The fact that the front panel isn't grounded. I would walk around, build up static and then touch the front of my computer and everything would just go blank. It wouldn't restart, wouldn't respond, but it would still be on - like the fans and everything would still be running but just a blank screen. I finally read in these reviews about people having a similar problem and how they manually grounded it. I took a piece of speaker wire and wrapped it around a screw that went into the front panel then grounded it with the psu and ta da! No more random shut downs.

solution 3,. FROM ANTEC CS
UPDATE!!!!

Called Antec customer support. They said they are aware of the problem and have no fix for it. They actually told me that the only thing that seems to work at this point is to tape a wire to the front bezel and connect it to a screw in the back of the case. Also said they are working on a harware fix and will send it once once is found.

The problem seems to effect HPMC cases as well