Sign in with
Sign up | Sign in
Your question
Solved

M2N32-SL Computer turns on for a couple seconds then turns off

Last response: in Components
Share
October 17, 2009 12:33:00 AM

My computer is as follows:

Cooler Master Ammo 533 case
Ultra Products 700W power supply
Crucial 4GB Ballistix Tracer Ram
AMD FX-62
Asus M2N32-SLI Deluxe
Sapphiretech X1900XTX 512MB Video Card

The computer has been running great for about 2 years and have done no changes to the components. Just recently it started to turn off by itself while in the OS. Then after about a week it would turn on after a cold start and then turn off after about 2 seconds. Here is what i have done to troubleshoot:
1. I disconnected all components from the motherboard except the PSU, the CPU, the RAM and the video card. Still the same problem.
2. I then took out all the RAM and used a single stick in various slots and did this until I covered all combinations with one stick. Still the same issue.
3. I bought a new PSU and connected it and still the same issue.
4. I took out the BIOS battery, shorted the CMOS and then reinstalled the battery. Still the same problem.

Im at my end and have never experienced this type of issue where I just cant narrow down the fault point. What is recommended at this junction? What else can i try?

~ag~
October 19, 2009 12:00:47 AM

Im going to now take the mobo out of the case and try to test it that way. Other suggestions were to try another cpu fan, or it could be the power switch. Worst case it will be a bad CPU or bad mobo. Not sure how i can test those two though. Would Fry's electronics be able to diagnose mobo/cpu and tell me if they are bad?

~ag~
m
0
l
October 19, 2009 12:57:54 AM

sounds like the cpu might be overheating? or a falt in the cpu fan, i kno that a computer will shut down if the cpu is over heating and if the fan is bad or has a short, if the fan is fine, remove all crap from between the cpu and heatsink and apply new thermal paste, cheap and might fix it, plus it doesnt hurt to do it anyway
m
0
l
Related resources
October 20, 2009 1:20:59 AM

right before it started wigging out i checked cpu temp and fan speed and they were within tolerance. even took out the fan & block and completely cleaned out all the dust. also reapplied thermal paste. but no cigar.


This time i think i will take out the fan & block and then reset the cpu. one post that i read said the user took out the cpu, reset it, and that cleared up the problem. i will probably also try another cpu fan and see if that works.

~ag~
m
0
l
October 20, 2009 3:01:06 AM

ya give that a try, i dont really know anything else to check, the only other thing is maybe your pc is drawing too much power, try un plugging any opticle drives, i had problems in the past just like this with desktops and it either was the ram, cpu, fan, psu or a bad opticle drive
m
0
l
October 21, 2009 2:33:33 AM

Well the point is that this machine was working with all the same components from the same power plug just fine for the last two years. no changes to the hardware and then it just started turning off after a couple seconds. Right now, everything is unplugged except cpu/fan and video card and ram.

~ag~
m
0
l
October 24, 2009 7:43:43 AM

well 2 years is plenty, things do age in time, age faster with constant use, a brand new computer can have problems, thats why they have warranties, but even after 2 years, a compurters value can go from 2000 to 200, might be time for a new pc, i go threw 1 a year cause they get outdated soo fast no a days
m
0
l
November 2, 2009 10:30:42 PM

ok i bought a new motherboard and cpu. put it together with the stock heat sink and fan and slapped in to the case. turned it on with the same effect. turns on for a second then shuts off. took the PCIExpress video card out and put an older PCI video card in. It booted up and I was able to run BartPE and do some diagnostics. Everything looked ok and the machine was up for about 45 minutes. Shut it off and restarted only to have the machine again shut off after a couple seconds. I was sure it was the mobo and/or cpu but now i just have no clue. memory sticks I alternated a couple of them in one spot and no cigar. What could be causing this...this is driving me absolutely crazy.
m
0
l
November 2, 2009 11:48:43 PM

..and I have taken the switch from another computer and connected the POWER and PWRRESET wires and used that as the main buttons. Still the same result. Turns on for couple seconds then turns off. I think I have ruled almost everything out except the case itself. Maybe something ios shorting but cant see anything. tomorrow ill take it out and put it on a cardboard box and see if it works then.

~AGP~
m
0
l
November 3, 2009 10:26:37 AM

Before buying a new CPU and mobo, I would have tried the less expensive ideas (RAM, or GPU). You could have a bad memory stick; although unlikely because you did test each one individually--but still a slight possibilty. You say when you put in an older PCI card rather than your PCI-E it ran much longer, were there any other symptoms? Did you check GPU and CPU temps? What are your system voltages, and are they in sync with manuf. specs?

If you have another PCI-E card, maybe give it a shot...

Last ideas..maybe you somehow had two bad PSU's in a row?! Tough one man..keep us posted on this please!
m
0
l
November 3, 2009 3:01:59 PM

I hear what you are saying. I did try the RAM swapping with four sticks in all slots to no avail. What are the chances that 4 sticks of RAM have suddenly gone bad? I did check system temps and they were in the operating range. Voltages I did not look at but I was running an app that is supposed to alert you if temps or voltages are out of wack. Nothing stood out. The PCI video card i tried is an old, old Rage128 card that doesn't even have a fan just a cooling block.

Strangely I did all this before I bought the new mobo/cpu and every time i swapped something out it did not resolve the issue. Im just at a loss as to what else to check. Could be the memory or PSU since those i am reusing. The PSU I already bought another one with no change. The memory is good brand Crucial Ballistix. I unplugged everything from the motherboard except the video card, power, 1 memory stick, cpu fan, and the exteranl power switch. Still no resolution. I even disconnected mouse, keyboard at least i should get a post. Arghhhhhh, im about to throw it out the window.

~AGP~
m
0
l
November 3, 2009 4:09:35 PM

Geez, that is a definite window-thrower lol. I'm racking my brain trying to think of what could possible be doing it--you've already cleared CMOS as well as replaced the mobo so I think it is safe to rule that out. The RAM could have been overvolted by your old PSU, causing it to still be bad--however unlikely.

You said you tried taking the whole rig out of the case (to be sure it isn't somehow grounding out or shorting due to a bad contact on the mobo/case) correct??

There is still a bad component somewhere in your loop, sadly I think you may need to try new memory and another psu... :( 
m
0
l
November 3, 2009 4:30:49 PM

sindizzy said:
The PCI video card i tried is an old, old Rage128 card that doesn't even have a fan just a cooling block.



It's funny you mention that, I literally JUST took apart an old Dell Dimension (Pentium II swineflu edition) and inside it is a nice dusty Rage 128 AGP lol. I guess it isn't that funny, but extremely coincidental, I had to post back just to tell someone hahahaha. :sol: 

EDIT: mine is an AGP, "pro" edition. Good stuff lol. Why we even still had this amongst our hundreds of PC's is beyond me..
m
0
l
November 4, 2009 11:22:10 PM

best bet to avoid more problems, just part out that pc and start fresh with a new case and build up

1 thing i found interesting that you did not test is the hard drive just cause you have to use it to try and boot every time, i remember way back having a problem like that with an ide 100gb western digital, run for a bit then would over heat and shut down

replacing the hard drive is the last thing on the list, bit of a hassel having to reinstall windows but may solve the problem, if that doesnt work, then it wont go to waste, an extra drive is always good
m
0
l
November 6, 2009 12:13:10 AM

the reason i want to stick with it is that its a very well executed modded case. but if it doesn't pan out i may just have to put it in the closet and do what you suggested.

the hdd's i believe are ok. those i disconnected at the very first sign of failure. as soon as i get the machine working they will be reconnected. again I am not even able to post so booting to an OS is not my issue at the moment.

~AGP~
m
0
l
November 7, 2009 5:18:15 AM

did you try to flash and reinstall the bios?
m
0
l

Best solution

a c 144 ) Power supply
November 7, 2009 5:24:39 AM

Try breadboarding. It will look something like this:
http://www.tomshardware.co.uk/forum/page-262730_13_0.ht...
You can turn on the PC by shorting the two pins that the case power switch goes on.

That's part of our checklist and troubleshooting thread.
Boot problems:
http://www.tomshardware.com/forum/261145-31-read-postin...
Share
November 9, 2009 5:52:09 AM

I will breadboard it the next chance that I get. I will connect PSU, CPU, HSF, video card and thats it. I'll try it with and without RAM and see if what happens.

~AGP~
m
0
l
November 9, 2009 10:23:27 AM

Another thing I thought of, kind of random (and not likely)...you may have a short in a system connection or a wire in your case (such as the power switch) causing instability.

I am still rather confused how this can be happening to ya, after switching so many components! I admire your patience to continue working on this one, and I hope you pinpoint the issue soon!
m
0
l
November 25, 2009 3:18:04 PM

Well I got some time and I breadboarded the system. Its on a wood desk and connected are the PSU, CPU with fan and block, 1 stick of RAM, video card, mouse and keyboard. Turn it on and still the same thing. Stays on for a second then turns off. I just dpnt see how this can be happening with two different mobos. Could two mobos have the same defect? Im grasping at straws. Could the PSU be bad even though I already tested with another new one? What are the chances that both are bad? The RAM I took out and the same thing happened. I even used the power swicth from another machine. What could it be? Im tempted to test one more PSU and completely rule that out.

Any other suggestions let me know.

~AGP~
m
0
l
November 25, 2009 4:11:13 PM

You're 100% sure the PSU's you have used are both correct power and amperage (per rail) for the system?? If yes, test another motherboard. If no, test another power supply..

The CPU itself is properly seated and placed in correctly as well, I assume--but may be worth a check (maybe find a bent/burnt/damaged spot).


You have more patience than I do, Sindizzy--I woulda tossed that whole system and started over by now! Good luck man.
m
0
l
November 25, 2009 5:26:12 PM

The PSU that's in there and have been using for 2 years is a 700W Ultra V Series. The replacement I bought is also a 700W supply. What specs should I be looking at to buy a comparable supply?

I also bought another mobo/cpu combo with similar specs. the mobo is the same as my old one except with built in wireless. the cpu I believe is the same. I did look at the seating and found no anomalies.

Strange that it does the same thing on both mobo/cpu combos. That to me tells me they both suffer from similar defects OR a common component is faulty. The common components are the PSU, RAM, and video card.

AGP
m
0
l
November 26, 2009 12:27:34 AM

Hard to believe a video card would cause all of this, but if you've got a spare try it out ;) 

I cannot say as I have ever used a PSU like that, you may benefit from a more well known brand. If you look at the total power consumption rating for your CPU, Video card, and whatever other devices you have I usually like to have at least 100W more than that available.

Some people prefer 4 rails @ 12v each, some prefer 1 rail. To me, honestly, it makes little difference. I have seen "700W" power supplies perform worse than 550W supplies because they are not properly rated, and are just not made with quality components. May I suggest a different brand PSU?

Personally, if I have a component that I suspect is defected (in your case you replaced the cpu/mobo/psu) I would not buy the exact same model from the same manufacturer...maybe that is the issue...you are the one unlikely poor s.o.b. that got two of the same bad mobo's and psu's?!

Possible, but not probable. Anything is worth a try at this point, am I right?
m
0
l
November 26, 2009 1:34:15 AM

I have tried two different video cards. One PCI-E and one older PCI. Same results. What PSU brand do you suggest I use? I actually like the mobo quite a bit before it started freaking out. If this doesnt pan out I will probably go the distance and just buy a new mobo/cpu.

You are right though...the chances of me getting bad PSUs, mobos/cpus, is slim but maybe I am that poor sob. At this point its partly more curiosity thats driving me to figure out what the problem was. I'll try the PSU and maybe try another video card.

~AGP~
m
0
l
November 27, 2009 11:08:51 AM

Corsair makes a quality PSU, as does Enermax, and even OCZ seems to have come a long way in a short time. BFG tech seems to have decent products as well.
m
0
l
December 18, 2009 6:55:42 PM

ok im back and done some more tests. i bought another PSU and hooked it up and got the same result. turns on for a second then turns off. so at this point i have to say that the possibility of three faulty PSUs is remote.

So now what Ive done is switched the cpu from the mobo/cpu combo i bought on ebay. ive also switched my testing stick of ram. at this point i turned on the board and it beeped once stayed on until the first post and then turned off. i then made sure ram and cpu were all seated properly. the cpu block/fan right now is not anchored down but it is sitting on the cpu. i understand its not advisable to do this but its just for troubleshooting. first time it booted up and i had a message "BIOS checksum error - Defaults loaded". i was then able to post all the way through and look at the hardware monitor. at present it lists
Vcore voltage 1.4
3.3V voltage 3.26
5V voltage 4.94
12V voltage 12.16

CPU temp 56 to 60 deg C
Mobo temp 34 to 42 deg C
CPU fan 3358 rpm

Ive been able to reboot three or four times with a single beep at the beginning. I've breadboarded the whole thing and its on a wood desk with PSU, PS2 keyboard, PCI video card, one stick of RAM in B1 slot, cpu block/fan and a salvaged power button assembly from an old case.

Im not sure what it could be but could it be that all 4 sticks of RAM i have have gone bad? Ive also tested with no RAM before and i get the same result. right now it seems to be able to boot but im a little hesitant to call it good as once before it seemed to work only to start freaking out later. would an attempt at a BIOS update help the situation? Seems like its random so it could be almost anything.

~AGP~

m
0
l
December 18, 2009 7:49:44 PM

This link http://products.amd.com/en-us/DesktopCPUDetail.aspx?id=... says that the operating temp should be no more than 55 to 63 deg C. Im wondering if the cooling fan/block is not doing its job as its pretty close right now to max with no load. should i try an aftermarket fan and block?

~AGP~
m
0
l
December 19, 2009 11:09:25 PM

I may have stumbled into something. when i was testing i bumped into the ATX12V connector and the machine turned off. so i did a little experiment and turned it on while putting pressure on that connector. looks like it posts. i then jiggle the connector just a bit and the machine turns off. it looks like its not on there tight enough. the male end looks like its a bit worn out. but the female end on the board looks like its got a little bit of rust. what are my options here to fix this. clean both with alcohol? how would i replace the male end? or at least make sure that its in there tight enough to make a connection?

~AGP~
m
0
l
December 21, 2009 11:47:53 AM

You're referring to the actual connector on the motherboard itself I assume? If so, is it corroded and worn out on both boards?! I would have to say that is by far the most odd manufacturing error I've ever seen! You already have the most frustrating case I have seen, so it would only make sense if you had 2 boards with bad ATX 12v connections ;) 

I am not sure of the best way to clean that out (or how you even got enough moisture in there to allow it to rust!) but I would do my best to clean it dry. I would scrape what I could and use compressed air to clean it out. If the connector itself appears to be the issue, I would attempt to simply RMA the board and make the manufacturer aware of the problem so they can send you one that has a solid ATX12v on it.

I think you may have found the answer here, all along I was certain it was a power issue and once you finally got to using a third PSU I was beginning to think all was lost and that you lived in some magnetic vortex that simply wouldn't allow you to use a PC lol. Appears now that your best options would be to attempt to clean the connections and test them, try to determine if it makes a difference.

As far as heat goes, I always recommend aftermarket cooling--simply because your PC will last longer if nothing else. When your CPU runs cooler, it runs better, but yours staying under 60C shouldn't be an issue with booting up and running.

Possible all four RAM sticks are bad...yes. Especially if your mainboard is having issues with power, it *is possible* that they have shorted out; although, if they had, I would think you would have the same error (and post beeps) with no RAM installed as you would with any one of the sticks installed. Not sure if that came out clearly...basically, your machine would have the same errors with or without the RAM in it, if the RAM was bad. Since your issues are not normal anyhow, it certainly reminds us that anything is possible!

Hope some of this helps, we're still here for ya!

m
0
l
December 21, 2009 4:52:03 PM

on my original board it looks a little oxidized but not too much. i tried to clean it with a bit of alcohol. on the new board it looks clean as a whistle so that may be something else. both boards are out of RMA time frame so no luck there. the connector from the PSU seems like its worn out. i think its the way the cable was attached to the mobo. When it was built (it was a custom built mod I won at a LAN party) it seemed like it was connected to mobo and then tied down pretty tight. maybe that's what caused it to eventually give. anyway I will try to firmly attach it. im gonna try some electrical tape to make sure it doesn't jiggle loose.

im thinking all other components are working and it was this one thing that was taking me on a wild goose chase. i bought a whole new setup and now just have to get a new PSU. this will now be my main machine and the one im trying to repair will be relegated to be the living room machine.
m
0
l
December 21, 2009 5:42:16 PM

Sounds good Sindizzy, glad you are finally on the up side of this battle! Keep me posted man, will be sure to check this out again.
m
0
l
March 27, 2010 5:36:49 PM

So got all my new stuff i7-920 Intel CPU an ASUS P6T Deluxe, 6GB of OCZ memory and an Enermax Infiniti 720Watt PSU. All is working great. took me a while to put together, reformat drive and then setup all my applications.

Now back to my old parts. One day I thought why not try again and I theorized that getting another way to plug in the PSU to the mobo might work. I bought a $7 extension for the PSU to ATX12V connector. The connections on both ends seemed to be a lot more secure. I wrapped electrical tape around the connections several times just to be sure. Ive been testing the rig for a couple days now and all seems well!!!! So I have a new rig and my old parts can still be sued as a pretty good machine as a secondary.

Thanks for all the help.

~AGP~
m
0
l
March 29, 2010 2:22:28 PM

Excellent! Glad to hear the good news :) 
m
0
l
!