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

SOLVED Possible Gigabyte GA-990FX-UD3 R5 issue.

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

BoundByBlood

Maybe Something Cool?
Joined
Sep 27, 2009
Location
MS Gulf Coast
I have been having the most frustrating issue with unexpected shut downs and I need more opinions. I had a liquid cooler AIO break recently - small leakage - and when it started to malfunction a whole new problem arose. The cooler probably damaged some parts, but I need to figure out which.

What is happening is I can power my system up and usually I can let it run for about 20-30 minutes before it powers itself off and when it does only the power button LED remains lit. When the system shuts off if I press and hold the power button - like a hard shutdown - the LED won't go dark. If I go and flip the switch on the back of the PSU to reset the breaker the LED will go dark and the system is ready to power up again.

The issue is inconsistent - sometimes the machine will run for 20 min before it powers off and other times it will do it before windows even finishes loading.

Now normally the first reaction would be to look at the PSU since I have to reset the breaker, but I have come across some things which indicate more may be involved such as fluid getting onto my board.

In this thread here the guy is describing a very similar issue, the only real differences are his system can run up to 45 min. before shutting itself down and he will have a residual power letting one fan temporarily spin after the power off. Mine doesn't do that, when it happens it is a straight up power off leaving only the power button LED lit. His issue turned out to be his motherboard either due to faulty VRMs or power control circuitry.

Then I saw this youtube video where a guy is talking about the same issue I'm having - no mention of the power LED staying on - and all of the usual hardware suspects checked out. Turns out the mobo was warped and the cpu socket was separating from the board and he demonstrates this with a straight edge razor. What caught my attention was when he stated the motherboard was getting to a certain point with temps and then just shuts itself off. This kind of sounds like what is happening to me.

I haven't used the PC at all since this started happening to prevent further damage to components, but the last time I did turn it on I was looking at temps within HWInfo. Cpu [with stock air cooler] was fine between 18-23 C, general motherboard temp was 30 C, northbridge was between 40-42 C, and HWInfo under the motherboard section has individual listings for "temperature 4, 5, 6" which were around 45 C. I launched Rise of the Tomb Raider, played for about five minutes, and about three minutes after exiting the game is when the shutdown happened. I didn't get to see mobo temps before it happened.

If it really was just the PSU after I flip the switch I would assume the breaker would need time to cool down and reset before being ready to power on again. When I flip the switch it's ready to power up immediately.

Parts are going to have to be replaced and I need to get the best idea I can what is at fault since I would hate to go and buy another PSU when it may not be the problem. I have limited resources so I can't test the PSU or mobo with a multimeter, but I figure one of them is at fault. The question is which one is it?
 
Last edited:
Doubt it's the psu unless some water dripped inside... not to mention, doesn't seem to do it under load (does it?). Try swapping with a know good psu...

I'd think board first too.
 
PSU would not leave a live rail (verified by LED still on) when it shut down. Therefore, it's not the PSU shutting down. Physically disconnecting the PSU by switch or power cord is resetting whatever is 'sticking' when the board cuts off. You have a bad component/components on the board somewhere.
 
Thank you EarthDog and Scott, you all are top notch in my book :thup:

I had a feeling suspecting the PSU was too easy of an answer and suspected the board, but couldn't place my finger on why.

Yes, it is new enough to RMA so it will be done.

You all want some money or something for your expertise?? ;)
 
I agree, RMA the MB, any time water gets on live electronics, you can expect to see more than steam, all the magic blue smoke can get out as well :rofl:
Now you can get a MB wet as long as you remove the battery and no power is on it, then dry it out with no harm.
 
On this one I'm leaning toward the PSU, I'm thinking if there was a heavy load with the motherboard the PSU would have a hard time stating up, startup is the heaviest load for the motherboard.

Maybe there is a short on the motherboard time will tell.
 
On this one I'm leaning toward the PSU, I'm thinking if there was a heavy load with the motherboard the PSU would have a hard time stating up, startup is the heaviest load for the motherboard.
You make some good points and you aren't the only one to suspect the PSU, Shrimpmeister is with you in that minority :p.

I just completed the first series of testing with a multimeter on the 24 pin ATX connector. All pin locations with a charge were delivering the right amount of volts for their respective location.

Just to be safe I checked it twice retesting each pin and it was all looking good. So series 1 & 2 of PSU tests is a Pass. If anyone is genuinely curious I can post the results.

Next up is to test the molex connectors, but so far Scott is looking good in his infinite wisdom suspecting the board.
 
If you checked everything else, what would checking the molex do? Molex runs off the same rails you checked..as well as not supplying the motherboard with any power, but peripherals only.

As far as 'looking good' I assume you mean they were all in spec (5%)?

So you don't have another known good psu anywhere to test? That will tell you immediately.
 
If you checked everything else, what would checking the molex do?
Probably unnecessary, but ensuring there isn't a short in the line and that the connector is getting the power as the rail.

As far as 'looking good' I assume you mean they were all in spec (5%)?
Yes, the 3V locations were 3.4V, the 5V were 5.11, the 12V were 12.24 and the -12V pin was spot on so everything was well within the acceptable range.

So you don't have another known good psu anywhere to test?
Nope, but this one is as I said looking good :p
 
Well, you really don't know until you put a load on it anyway... it could sag out of spec under load.

While boot isn't the 'most stressful thing' on the motherboard (hook up a killawatt and compare to gaming or anything else), there is a load on it of course.
 
Well, you really don't know until you put a load on it anyway... it could sag out of spec under load.
Of course and if I really wanted to be sure I could test it again behind the wires when it is plugged into the mobo and everything is hooked up. Then I could see if it falls out of the 5% range.

For these tests though I did it with the six chassis fans running to simulate some kind of load. Not much, but it was something.
 
That's how it should have been tested in the first place. ;)

Anyway, are you going to RMA the board at this point?

(How is this thread solved?)
 
The board is already on RMA.

When I started this thread I was still on the fence as to which component was causing the problem so I pretty much came to the conclusion this thread as being solved when we all came to the conclusion the board was at fault and not the PSU.

I have just been checking the PSU for safe measure while I wait on the board.
 
At this point it's starting to look like the PSU is the faulty unit responsible for everything.

Shrimpmeister might be onto something...
 
So...... if I am reading between the lines right, you got the board back from RMA and it still doesn't work?
 
I logged into the Gigabyte website to check on my RMA and for status it stated: "No trouble found after tests with multiple configurations".

They haven't sent me an updated email yet, but it looks like they are about to clear the board.

The GPU and motherboard has been cleared, AMD wants to replace the CPU so I'm in the process of getting an RMA authorization from the PSU manufacturer.
 
Last edited:
"AMD wants to replace the CPU so I am in the process of getting an RMA from the PSU Mfg.."

My basketball is broken, but I am in the process of RMAing my sneakers...

Help me make sense of that while I finish my morning tea and wake up please....:)
 
Back