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

Ive ruled out the psu....

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

knoober

Member
Joined
Mar 18, 2015
Hi guys, Im back again with another problem. Im currently trouble shooting it but hoping to save some time by throwing in a post before I throw in the towel. Here is the system and the problem:
System:
Asus M4N78 pro
Athlon II x2 270
350W Seasonic PSU (SS-350ET)
I don't remember exactly what the cooling is called but its basically a little bigger than a stock cooler with a 92mm fan
Ive got 4 various sized of HDD in the box (3 sata, 1 IDE) and no ssd
2x2 corsair dominator ddr2 and 2x2gb Kingston valueram

No OC, nothing special. Just a bunch of HDD and the bare minimum otherwise.

Problem:
This box started powering down after a few minutes running (say about 10min - I haven't timed it yet)

Troubleshooting down so far:

Changed the psu: I immediately blamed the psu because Ive had this box running for a while now with no troubles. It seemed to me that if anything was going to take a sudden dive it would be the psu. New psu but the problem perists. I should mention that even though it is a quality brand of psu it is off the used shelf at the local pc repair shop. That's just to say it isn't "known working"

reseated the cpu: why? because there is a problem that's why :) Ive got the box open so I might as well make sure everything is copacetic right? problem still exists though

started moving the ram around: I haven't run any memtest yet, but its on the way. Ive only got 4 sticks so I am trying to see if I can run 1 of them without having the error. Maybe memtest would do this quicker.... but what if I get another shutdown? :)

And that's where I am folks. This machine was running decently until I shut it down last night. Ive been trouble shooting most of the morning and would appreciate any helpful advice. Thanks in advance

Edit: Hope I caught that before anyone read it but it is definitely an m4N78 pro, I had typo'd an m4A78 .
 
I would check for overheating to make sure temps are in good order.

Would also check the motherboard for popped capacitors, they can cause issues and instability.

I would do the memtest as well, I don't like to see two different sets of RAM mixed together, can cause weird issues.
 
Caps *look* okay, but I will have to admit Ive never seen a bad one personally. Ive read a lot of accounts that a bad cap will have a bulged top or be weeping. I am seeing none of that kind of stuff.

I cant get it to run with either set of ram alone, so I feel that the "mixed set" issue can be set aside safely.

As Ive been typing this post the problem has gotten worse. I cant even get a full boot anymore. The fans will spin up and then shutdown. it will try to spin back up after that.... unsuccessfully. Is it reasonable to suspect the board at this point? I realize that further testing will reveal more, but Im kinda leaning towards just swapping the board at this point. New psu, and the ram are tentatively ruled out. what else does that leave? I don't have a cpu to swap in there for testing so... I guess Ill let you know. Please let me know if you think of anything else that I should look at before I smash the board :)
 
That's good about the caps, yeah looking for bulging or weeping stuff out of them is what I've seen in the past.

HXKKl89.jpg

What all is the box used for? I imagine the 350W is enough for your system if it is literally just cpu + fans + hard drives.


I don't follow you with what you said that you can't get it to run with either set alone? you mean by themselves the problem persists [even down to a single 2GB stick]?


Can try resetting the cmos by pulling the battery to see if that helps.
 
That's good about the caps, yeah looking for bulging or weeping stuff out of them is what I've seen in the past.

View attachment 171330
thanks for that pic. Mine are in good order then.
What all is the box used for? I imagine the 350W is enough for your system if it is literally just cpu + fans + hard drives.

NAS. CPU + Fans + drives is all it needs. Ive been putzing with this one for a couple weeks now. I finally got the software/OS situated the way I like it an now this. There should be an emoticon of a guy headed to the liquor store to get a six pack, because this is why to drink :)

I don't follow you with what you said that you can't get it to run with either set alone? you mean by themselves the problem persists [even down to a single 2GB stick]?
Correct. Either set by themselves (2x2gb corsair OR 2x2gb Kingston) are producing the error still. I have run all 4 sticks singly as well (1 x 2gb stick for each stick) still get the error. I will have to think of a way to verify these sticks after this because they don't fall into the "known working" working category anymore.

Can try resetting the cmos by pulling the battery to see if that helps.

I kind of sidestepped this one because I hadn't changed anything in the bios but it is worth a show. I am going to change the board anyway but I hadn't thought of this since I started my troubleshooting. Thanks
 
I had a friend who once had his PC randomly powering down for no apparent reason. Brought it over and I tried everything I could imagine. Stress testing the system for up to 12h, running memtest, checking hdd, etc... everything was running just fine. Couldn't replicate the problem.

Brought the computer back. Guess what? Lost power in first three hours.

Eventually we found the culprit. It was a faulty wall socket. A screw had became loose and would throw an odd spark every now and then. That spark was enough for the PSU protection to kick in and power off the system.
While this might be a long shot, try plugging in the system in another socket or if you're using an extension lead, try changing that. Never know, might be something before the PSU that's causing the problems.
 
It needs to be said that capacitors can fail without showing obvious signs of bulging or leaking. Certainly, on today's modern boards with solid caps you won't see that. Besides, there are other components that can fail on a motherboard.
 
Caps *look* okay, but I will have to admit Ive never seen a bad one personally. Ive read a lot of accounts that a bad cap will have a bulged top or be weeping. I am seeing none of that kind of stuff.

I cant get it to run with either set of ram alone, so I feel that the "mixed set" issue can be set aside safely.

As Ive been typing this post the problem has gotten worse. I cant even get a full boot anymore. The fans will spin up and then shutdown. it will try to spin back up after that.... unsuccessfully. Is it reasonable to suspect the board at this point? I realize that further testing will reveal more, but Im kinda leaning towards just swapping the board at this point. New psu, and the ram are tentatively ruled out. what else does that leave? I don't have a cpu to swap in there for testing so... I guess Ill let you know. Please let me know if you think of anything else that I should look at before I smash the board :)

I literally saw bad caps with my own eyes all the way back in 2008 and at least one cap on my Asus A7N8X-X suddenly bulged and leaked when sitting! (Even though the cap is from Chemi-Con)
 
Ive switched the board and ram, and psu.... now its time to redo all the work. The big plus to the board I was using was that it had IDE support so I could use my 320gb WD blue. I ended up replacing a 320gb with an 80gb so the loss in storage space is a real :nuts: but atleast this project will be coming to a close. I tried cloning the drive by sticking the 320gb into an external enclosure, but it wouldn't boot correctly.... so like I said, I guess I get to start from scratch. Good thing I kept decent notes (I hope). Most of the more difficult stuff I actually did take notes on (like installing a dummy driver for the gpu) but who knows what I forgot to write down eh?

Thanks to everyone who chimed in for the help! :)
 
Im resurrecting this thread because this issue has reappeared. The specs in my sig are fairly up to date (NAS/Etc), with the exception of the CPU being an Athlon x4 630 now (running at bone stock). At this point Im leaning towards the PSU not being enough for the system (but the online PSU calc says Im well within range for the system Im running)? It was a bargain price on a psu that had a reasonable expectation of being good enough (Seasonic Psu with nothing bad to be heard about it in a basic search), but id hate to get one from the tried and true list and not have solved the issue. Thanks in advance once again

Edit: When I was thinking about this again I came to a startling realization that I thought I had better put right out in the open for this issue: Motherboard changed: check. PSU changed: check/ CPU changed: check. RAM changed: check. HDD changed: bzzzt. Nope same drives. Im down to the OS and the Drives as the potential causes of the issue here.... with a possibility of the PSU as well. Let me know if you do that math any differently please. Thanks again :D
 
Last edited:
Have you checked the SMART data on the drives? If one is failing, it can surely cause this issue, even if it isn't the system drive.

Beyond that, you could try swapping out the PSU for a known good and see if it's that.
 
Have you checked the SMART data on the drives? If one is failing, it can surely cause this issue, even if it isn't the system drive.
SMART checks out healthy with smartmontools in Linux.... not 100% that Im reading it right, but the output is right inline with CrystalDisk on Windows. It just doesnt have a little picture on the GUI that says healthy or otherwise, so Im not absolutely sure on the drives, but it is reasonable to rule them out as well.

Beyond that, you could try swapping out the PSU for a known good and see if it's that.
Well I only have one other known good PSU and its in my daily driver. Id take it out and swap for a bit to check but the issue is so random there is no way to tell how long Id need to have the swap last. For instance, its happened 3x in a day and then been a 1.5 weeks between issues. Looks like im headed to the PSU list :shrug:
 
Have you bench tested it? Taken the board out and tried running it on a piece of cardboard or something?

I had a somewhat similar sounding issue. There was a standoff in the case i had forgot to remove and it was causing the board to short out or something. Had many successful boots prior to that, but moving RAM around or a slight jolt made the issue come up.

Maybe also a power supply tester can help you out since you don't have another one you can swap in

http://www.newegg.com/Product/Product.aspx?Item=N82E16899705002&ignorebbr=1
 
Have you bench tested it? Taken the board out and tried running it on a piece of cardboard or something?



Maybe also a power supply tester can help you out since you don't have another one you can swap in

http://www.newegg.com/Product/Product.aspx?Item=N82E16899705002&ignorebbr=1

i havent put it on the bench yet, but I suppose its worth a shot. The case is in a stable spot (doesnt get bumped or anything) and so I think I wouldnt see boot at all if it was shorting though.

I went and got a cheap multimeter today as well, so we will see if the PSU is putting out the rated voltage too. I will report back when I check the PSU. It will be a few days before I can toss it on the bench :)
 
Back