Computer randomly turning off Seasonic X750 Gold

T

TempestBlayze

n00b
Joined
Dec 11, 2007
Messages
35
For the past couple months my computer has been randomly shutting down when I simply surf the web, play games, when I am doing nothing at all, in safe mode and even in BIOS. The computer does not blue screen, it just turns off. Sometimes it restarts itself after a hard shutdown but sometimes I am unable to turn it back on for 5 minutes.

I have run many tests to make sure that is was the PSU.

-My 3750k CPU temperatures are normal at no more than 53c at load
-I tried taking out the video card.
-I am also not overclocking my CPU.
-I have tested my RAM sticks by removing them and my computer still turns itself off.
-I reset my motherboard CMOS to make sure nothing weird was conflicting with my computer
-I even reformatted my computer to make sure it wasn't a driver problem.
-I also made sure all the connections to the motherboard is secure.

So after all of these tests I believe that my problem is the PSU. I bought a Seasonic x750 Gold PSU 5 months ago because I wanted to avoid something like this since I know Seasonic is one of the best out there.

What does everyone think? Should I send the PSU back?

I have:
AS Rock Extreme 4
3570k CPU
Coolermaster Hyper 212 EVO Cooler
Seasonic x750 Gold PSU
 
Removing RAM sticks is only a small part of testing RAM. Run Memtest86+ for at least three passes. You can find it here or on the very useful PartedMagic. The behavior you describe sounds like bad RAM or your motherboard not being compatible with your RAM. Exactly what RAM do you have, and have you updated the motherboard's BIOS and chipset drivers to the latest versions?

Failing that, consider whether the power coming from your wall might be the problem. It's not impossible that the PSU is bad, but being a Seasonic X, it's pretty much the last thing on the list to consider.
 
I highly doubt it's the PSU. Your RAM and motherboard are more suspect, and quite frankly, your tests doesn't rule either out.
 
While it's quite possibly your RAM, I wouldn't be so quick to dismiss the SeaSonic power supply. They're generally very good but no company has perfect production. It could also be a poor connection on the 8 pin or 24 pin main connector. Did you inspect them to make sure that they have good contact and no obvious signs of burning?
 
Also if it crashes and then you sometimes have to wait five minutes this strongly suggests a heat issue. A poor electrical connection could still be a culprit, but possibly a hot controller on the motherboard somewhere? Check the NB temperature with your hand (don't just touch it, it could be very hot.. approach slowly!) next time it happens, along with the CPU VRMs, and any other heatsink you see.

Might want to try putting your X750 into regular mode instead of Hybrid mode and feeling the exhaust temp of the PSU as well, although that's not a definitive test of a failing component in the power supply it could tell you if the temp sensor was bad and the fan wasn't kicking on to where it needed to be. The five minute thing actually makes me suspect a power supply problem more than anything, do you have a friend that might let you borrow one to help diagnose your issue? A bit of a pain, but parts swapping IS the best way to verify a problem.
 
Seems like Memtest86 is going to take a long time since I have 16gb of RAM. I ran is for 2 hours and no errors but only one pass. I will run it overnight to see.

My RAM is Gskill Ripjaws 2x 8gb 1600

However, by default my motherboard reads it at 1333.
 
FrankD400 said:
Also if it crashes and then you sometimes have to wait five minutes this strongly suggests a heat issue. A poor electrical connection could still be a culprit, but possibly a hot controller on the motherboard somewhere? Check the NB temperature with your hand (don't just touch it, it could be very hot.. approach slowly!) next time it happens, along with the CPU VRMs, and any other heatsink you see.

Might want to try putting your X750 into regular mode instead of Hybrid mode and feeling the exhaust temp of the PSU as well, although that's not a definitive test of a failing component in the power supply it could tell you if the temp sensor was bad and the fan wasn't kicking on to where it needed to be. The five minute thing actually makes me suspect a power supply problem more than anything, do you have a friend that might let you borrow one to help diagnose your issue? A bit of a pain, but parts swapping IS the best way to verify a problem.
Click to expand...

Any idea how to get it out of hybrid mode?
 
There's a switch near all the modular connectors, Hybrid fan mode runs the fan slower. Off/O is Hybrid, On/I is "normal" mode.
 
Unfortunately it seems I have a old model. I see the switch on Newegg but mine does not have that :(

The fan does not even turn on when I run Intel burn test and Fur Mark at the same time. I did notice a very low electrical interference like sound coming from the PSU when it was running the burn test but not sure if that's normal or not.
 
Last edited:
Are you running off of the iGPU?

If so, you're not putting enough of a load on the power supply to cause the fan to turn on. The Seasonic X power supplies can remain in fanless mode up to 500 watt load or higher, depending on ambient temps and case airflow. Your system without a GPU is at best a 250 watt system. If it's a single GPU system, at most it will be 500 watts.

What you're probably hearing is coil whine. It is not an indicator of any kind of failure, and Seasonic PSUs are notorious for coil whine.
 
I am running a 560ti along with the other parts. So far my Memtest is in the middle off pass 3 with no errors. I will give it to pass 4 until I will call it not a memory problem. Good thing is Seasonic does not want the cables so I don't have to rewire everything.
 
With a 560ti, your peak load power consumption should be around 400-450 watts.

Try using hci memtest. Memtest86+ doesn't detect all kinds of errors, and hci can detect certain errors better than memtest86+.

Also, do you have a spare motherboard you can swap out?
 
Nope I don't have a spare motherboard. HCI did not find any error either. Right now the motherboard detected my 1600 ram at 1333 by default. What I'm going to do next it put the ram to 1600 again like I did when I first built it to see if I get any errors or if the computer turns off while running HCI.
 
Put it to 1600, ran HCI and got a BSOD. It wasn't a total shutdown but I will try leaving it at 1333 for a week or two to see if my computer turns itself off.
 
No I'm not. The ram itself is 1600 but I didn't use XMP to overclocking the RAM. Just set that in BIOS. I am now running Memtest86+ to see if I get an error.
 
It's possible that your motherboard was using the 1333 timings for 1600 mhz speeds, which might be different and therefore cause instability.
 
It could be the power supply, but the only way to know for sure is by trying a different one or a different motherboard.

I don't know how defective memory can cause a sudden shutdown, unless it's so damaged that it doesn't work at all.
 
X750s won't stay fanless quite that long.. if your fan isn't running during FurMark + BurnTest you have an issue. Did you try checking the temperature of the NB/SB/VRM heatsinks on your board when it's happened yet?
 
FrankD400 said:
X750s won't stay fanless quite that long.. if your fan isn't running during FurMark + BurnTest you have an issue. Did you try checking the temperature of the NB/SB/VRM heatsinks on your board when it's happened yet?
Click to expand...

Yes, they can, depending on the hardware and case airflow. His 3570k + 560ti system uses far less power than your 8120 + 460 SLI system.
 
FrankD400 said:
Did you try checking the temperature of the NB/SB/VRM heatsinks on your board when it's happened yet?
Click to expand...

I have been looking at the temps but since the moment I started looking the computer has not shut off. They seem normal in HW monitor.

I'd like to think the airflow in my case is really good so I'm not sure if thats a factor. I tried the paperclip jump start found here but the fan never kicked in.

Since I left my RAM @1333 instead of the RAM default @1600 a couple of days ago it hasn't shut off. I am crossing my fingers but there has been times where it wouldn't turn off for a week.

Thanks for the continuous insight! You all have me trying different things before I send it to Seasonic and run the risk of them fining me and telling me it's working.
 
TempestBlayze said:
For the past couple months my computer has been randomly shutting down when I simply surf the web, play games, when I am doing nothing at all, in safe mode and even in BIOS. The computer does not blue screen, it just turns off. Sometimes it restarts itself after a hard shutdown but sometimes I am unable to turn it back on for 5 minutes.

I have run many tests to make sure that is was the PSU.

-My 3750k CPU temperatures are normal at no more than 53c at load
-I tried taking out the video card.
-I am also not overclocking my CPU.
-I have tested my RAM sticks by removing them and my computer still turns itself off.
-I reset my motherboard CMOS to make sure nothing weird was conflicting with my computer
-I even reformatted my computer to make sure it wasn't a driver problem.
-I also made sure all the connections to the motherboard is secure.

So after all of these tests I believe that my problem is the PSU. I bought a Seasonic x750 Gold PSU 5 months ago because I wanted to avoid something like this since I know Seasonic is one of the best out there.

What does everyone think? Should I send the PSU back?

I have:
AS Rock Extreme 4
3570k CPU
Coolermaster Hyper 212 EVO Cooler
Seasonic x750 Gold PSU
Click to expand...


hmm, you have an asrock board, sure its not that? are you overclocking? check your thermal and power settings? both in windows and bios
 
It could be a short/grounding issue if you have this system in a poorly designed metal case.

Can you trigger the failure by fiddling with wires (side panel off), or shaking your case lightly (side panel on) when everything is powered on?
 
Well, it happened multiple times today. I have had the side doors off the case for over a week now thinking it was the same thing and also changed the power cable going into the power supply. It's been working great the past week but as usual my hopes were crushed. Think it's time to RMA.

I am not overclocking.

Here is a SS I took of the tempes right after the turn off/ restart happened.
crash.jpg
 
I had an issue with my system where it started to randomly turn off on me, and wouldn't turn on until I powered off the PSU for a moment.

After trying out all the different HW options I could think of short of the CPU and mobo, it turned out to be a SW problem. AIDA64 + ASUS Fan controller would glitch and make the mobo read no RPMs for the CPU Fan, so it'd shut itself off. The issue had started because I had upgraded to a newer version of the fan controller SW. Going back to the previous version got rid of the problem.
 
Well that's rather unfortunate. At this point, just start RMA'ing different parts and see what happens, especially since you don't have any spare parts to swap out.

The best thing you can do to isolate troublesome components is to swap out with known working good ones. See if you have a friend that can help you there.

As I said before, I think the motherboard is more suspect than the power supply, but you can go ahead and start with the power supply first.
 
Maybe you need to perform kinetic analysis, AKA whack the power supply hard several times while it's running. It may be better to do this with the PSU out of the case (may have to prop it up on a box or some books so the cables can reach) so that way you can slap it on opposite sides simultaneously. Of course vibration can damage the hard disks easily, so first unplug the power cables to them. Run something like a memory diagnostic from a USB stick or CD to load down the PSU and CPU.

You mentioned having a Coolermaster Evo CPU heatsink. Somewhere I read of a person experiencing problems because its metal mounting hardware wasn't adequately electrically insulated from the motherboard and caused a short.
 
I had an accidental discovery yesterday. My computer was turning off and resting like crazy on Sunday so I decided to change the power cable that goes from the wall to the PSU. That stopped the spontaneous shutdowns for a couple of days. Then yesterday, it turned off and restarted about 3 times. So I got another power cable and it stopped shutting off. Don't know how long it will last though.

I am not quite sure what this means though.
 
It means you should replace the PSU and see if the problem still happens, like everyone here has told you to do?

Can't you even buy a cheap seasonic or corsair that's on sale in the hotdeals section, test it and its good, RMA the seasonic and ebay the replacement? I mean, how long are you going to sit there with a broken system, going "I don't know what this means?"

You've been giving good advice by others but you don't seem to want to follow it :(
 
Falkentyne said:
It means you should replace the PSU and see if the problem still happens, like everyone here has told you to do?

Can't you even buy a cheap seasonic or corsair that's on sale in the hotdeals section, test it and its good, RMA the seasonic and ebay the replacement? I mean, how long are you going to sit there with a broken system, going "I don't know what this means?"

You've been giving good advice by others but you don't seem to want to follow it :(
Click to expand...

I actually have been trying all the things that people have been suggesting.

I am takings trip to my brothers today because he generously let me borrow his 550w PSU for a week.
 
So far my brother 500w CPU is working fine. I also got a Kill-A-Watt meter and am very surprised at how little power the 350k and 560ti take. I was running Furmark and Intel Burn test at the same time and my PC never went over 300w of usage.
 
Do you have a UPS you can use? It's possible that the power coming from your wall is unstable.
 
TempestBlayze said:
I decided to change the power cable that goes from the wall to the PSU. That stopped the spontaneous shutdowns for a couple of days. Then yesterday, it turned off and restarted about 3 times. So I got another power cable and it stopped shutting off. Don't know how long it will last though.

I am not quite sure what this means though.
Click to expand...
That the wall outlet socket could be loose? What does it feel like when you insert or remove a 2-prong power cord? Also are you plugging the cord in firmly at the power supply side?

What voltage reading did the Kill-A-Watt show when the computer ran?
 
larrymoencurly said:
That the wall outlet socket could be loose? What does it feel like when you insert or remove a 2-prong power cord? Also are you plugging the cord in firmly at the power supply side?

What voltage reading did the Kill-A-Watt show when the computer ran?
Click to expand...

Nah it's pretty tight. My house is 110 but my wife never has a problem with her computer so I can't imagine it being a 110/220 deal. The Kill-A -Watt reads 122.

Day 2 with my brothers PSU and still hasn't turned off.
 
Just an update for those looking at this in the future.

I was able to use my brothers PSU for a week since he went on vacation and never had a random reboot.

I ended up RMAing the PSU right after that week and Seasonic sent me the new one. I have now had the new PSU in my system for a week and everything is working fine.

Thanks everyone for the troubleshooting tips. Even though they weren't the issues, I learned a lot by testing everything.
 
Glad to hear you got everything sorted out.

I think I said this before, but really the only foolproof way to troubleshoot computers is to have spare known good working parts.
 
You must log in or register to reply here.
Back
Top