Sign in with
Sign up | Sign in
Your question

Fans spin, then everything dies

Last response: in Systems
Share
June 1, 2006 6:12:20 PM

So, I'm building this computer I've been scrounging together components for, and I think I've got everything done, right? Well, comes time to turn it on for the first time, and then... the fans spin up for about a second, and then it shuts itself off. Every time. No video display, no beeping, nothing. Just, the fans spin up, then they stop.

I've got a PSU that's plenty powerful right now, so I know that's not the problem unless it was actually defective on shipping. Any suggestions on what could be the problem?

More about : fans spin dies

June 1, 2006 6:20:07 PM

Oh, and on checking the FAQ in the MoBo section of Hardware, the first question addresses this problem, but all of the mounting posts are properly aligned, so I wouldn't expect it to be a problem with being grounded out by that route, at any rate.
June 1, 2006 6:36:04 PM

Quote:
So, I'm building this computer I've been scrounging together components for, and I think I've got everything done, right? Well, comes time to turn it on for the first time, and then... the fans spin up for about a second, and then it shuts itself off. Every time. No video display, no beeping, nothing. Just, the fans spin up, then they stop.

I've got a PSU that's plenty powerful right now, so I know that's not the problem unless it was actually defective on shipping. Any suggestions on what could be the problem?


Can you post your system specs, please?
Related resources
June 1, 2006 6:47:55 PM

Sorry, should've done that before.

3.2Ghz Prescot engineering sample LGA775 processor (the one really off the wall component)
Intel D925XCV MoBo
Scythe Katana heatsink
300Gb Seagate 7200.9 SATA II HDD
BenQ DW150 DVD burner
Thermaltake TR2-430W PSU
CoolerMaster Centurion 5 case
and I'm currently using an old PCI based Radeon graphics card I pulled out of another computer until I get the card I actually intend on using. Forget what model it is, but I do know for certain that it works perfectly.
June 1, 2006 7:08:41 PM

Did I miss it or have you not posted your ram specs? I've come across this problem many times, and it usually realated to one of three things ( for me anyway)

Bad PSU, they do ship bad sometimes.
Back stick of ram that stops post.
Or on occasion a faulty power on button.


I would suggest that you rebuild the current system, idealy outside the case. This way your are certin that no componants are being grounded out by any portion of the case's from factor. Single stick of 'tested' ram, tested video and no HDD. Try to get yourslelf to post from there. If that does not do the trick, your looking at a bad mobo or cpu.
June 1, 2006 7:27:51 PM

Quote:
Sorry, should've done that before.

3.2Ghz Prescot engineering sample LGA775 processor (the one really off the wall component)
Intel D925XCV MoBo
Scythe Katana heatsink
300Gb Seagate 7200.9 SATA II HDD
BenQ DW150 DVD burner
Thermaltake TR2-430W PSU
CoolerMaster Centurion 5 case
and I'm currently using an old PCI based Radeon graphics card I pulled out of another computer until I get the card I actually intend on using. Forget what model it is, but I do know for certain that it works perfectly.


Ok, first potential problem I see it that engineering sample Prescot processor. That could have all kinds of problems inside it, and cause all kinds of problems elsewhere. Prescots were known for problems in the best of times (overheating).

Next is the BIOS, was it set it up correctly? A bad BIOS set up will stop the computer in its tracks. AS PCcashcow said, there could be a faulty PSU or ram stick. Test the ram sticks one at a time. I doubt the faulty "on button" because your computer attempts to start, unless its sticking and you aren't really getting it all the way into an "on" position. Then the contacts might close for a moment, but as soon as your finger moves away, it springs back and the contacts are open again.

A last problem I can think of is a bad motherboard, same as suggested before. Yes, you may have been very careful putting it together, but some boards are DOA, and there's nothing to do but RMA it and start over.
June 1, 2006 8:09:15 PM

Oops, 1Gb Corsair value select DDR2 PC 4200 RAM.

The board has been tested to work, as has the processor (I got them from someone I know, so yeah, they're used, and are going to die sooner, but they're a temporary set up, and they've been shown to both work). Though it's possible something happened, I suppose.

A bad PSU would give power for just a second, and then stop? The indicator light on the MoBo shows that it's at least getting some power from the supply, but I suppose I can test it on another computer.

The RAM is going to be harder to test. I don't have another computer that accepts DDR2 ram, so I'm guessing I'll have to find someone else willing to let me try it in one of theirs.

Thanks, guys.
June 1, 2006 8:10:15 PM

I had symptoms like these after I built my machine. The psu turned out to be faulty.
June 1, 2006 8:12:56 PM

By the way, is it possible that I just misconnected something? Or would that not cause this sort of problem?
June 1, 2006 8:51:37 PM

Quote:
By the way, is it possible that I just misconnected something? Or would that not cause this sort of problem?

could be. That 4 pin 12v extra plug.
you can try clearing the bios.
try to use just one stick of ram. test one at a time
June 1, 2006 9:43:25 PM

if your fans turn on then it is powering on, the problem wont be a pwer button.

when this happens in pc's, especially ones that are powering on for the first time it means that one of your major components isnt being read correctly.

1st,
take everything out of the pc that isnt needed for it to boot. i.e. 1 stick of ram, take out the vid card (if it has onboard video) use a single hdd, which i think you only had one.

2nd,
clear the bios

if it still acts the same way, then start swapping things out.

the thing that would make the most sense would be your processor, if it is indeed faulty in some way, but id guess that its not mounted correctly. the prescotts are bastards when it comes to mounting them correctly due to those stupid lil plastic pins (yay intel cheapies). if one of those lil pins isnt in all the way the processor wont get the "needed" pressure to contact all the pins on the mobo and will not post.

i experienced this exact thing when i switched to water cooling, only the fans would spin for a few secs then it would power down until i got the correct pressure on the cpu and water block.

good luck man, this situation deffinately sucks.
June 1, 2006 9:45:17 PM

Quote:
By the way, is it possible that I just misconnected something? Or would that not cause this sort of problem?

could be. That 4 pin 12v extra plug.
you can try clearing the bios.
try to use just one stick of ram. test one at a time That's my guess..the atx 12 connector. Also, don't forget to set PCI for graphics
card, in case you haven't. GL :) 
June 1, 2006 9:47:10 PM

if it wont post and go into the bios then you cant set which video mode to start in.

and yes if you dont plug in the 12v atx connector it would cause this, and possibly other problems.

double check all your wires and connections.
June 1, 2006 9:51:12 PM

Quote:
Oops, 1Gb Corsair value select DDR2 PC 4200 RAM.

The board has been tested to work, as has the processor (I got them from someone I know, so yeah, they're used, and are going to die sooner, but they're a temporary set up, and they've been shown to both work). Though it's possible something happened, I suppose.

A cpu can go bad bad in a second while importing into a new enviorment. Remember, this day in age we are still ruled by the laws of static transfer....and there is always Moores Laws.
Quote:


A bad PSU would give power for just a second, and then stop? .


Yes! The system ready light only tells you that there are no voltage and interrupts passing through various sensors. Power-on tells a diffrent story.


But again, lets focus on the ram, since we are phishing data from you, once again is that a single stick or pair, and what is the test that you know it works.

Is there a way to put those componants on a diffrent mobo? You need to check with the intel site to identify if that ram is a tested and working DDR2 chipset.


I recommended a rebuild with the 'min' amount of 'parts' once youve done this you should atleast be getting to post at some point, even with no video or cpu no ram youll get system beeps. So play around with that.
June 1, 2006 9:54:07 PM

Quote:
if it wont post and go into the bios then you cant set which video mode to start in.

and yes if you dont plug in the 12v atx connector it would cause this, and possibly other problems.

double check all your wires and connections.

Yes, i realize this. I didn't say that i thought the wrong video card setting was the problem, so i just thought i'd throw that in, in the event he gets it
working, but has forgotten to set this.. It's an easy enough thing to forget.
June 1, 2006 9:59:07 PM

Quote:
....and there is always Moores Laws.

That would be Murphy's Law. Moore's Law is that # of transistors in a CPU will double every 18months. :wink:
June 1, 2006 9:59:41 PM

A bad PSU would give power for just a second, and then stop? The indicator light on the MoBo shows that it's at least getting some power from the supply, but I suppose I can test it on another computer.

The RAM is going to be harder to test. I don't have another computer that accepts DDR2 ram, so I'm guessing I'll have to find someone else willing to let me try it in one of theirs.

Yes, a bad PSU can power for a second or more and then die. If a wire or other part inside it has too much resistance, the current can start to flow, and then as the heat begins (almost instantly) the resistance cuts the power flow to zero. As for the memory, if you can test it in someone else's computer, that would be good. At least you wouldn't be guessing about it.
June 1, 2006 10:44:07 PM

Quote:
2nd,
clear the bios

if it still acts the same way, then start swapping things out.


Already done (at least in theory, I really can't make sure yet...).

Quote:
the thing that would make the most sense would be your processor, if it is indeed faulty in some way, but id guess that its not mounted correctly. the prescotts are bastards when it comes to mounting them correctly due to those stupid lil plastic pins (yay intel cheapies). if one of those lil pins isnt in all the way the processor wont get the "needed" pressure to contact all the pins on the mobo and will not post.


LGA775 is "pinless." I suppose it's possible the cpu spontaneously went bad, but I know it's mounted correctly.

It's a single stick of RAM, sorry that I didn't make that clear.

I've got all the power connections properly set up, which seems to leave me at a bad PSU, which will involve partially disassembling two computers, bad RAM, which I'll need someone else's computer to test, or a bad CPU, which I'm not entirely sure how I'll test, but would really bite. Argh, time to rebuild this thing, it would seem.
June 1, 2006 11:19:06 PM

I don't think this "test" would reveal any faults with your psu even if it was faulty. It just simulates being plugged into your mobo, so you can run fans/peripherals off the 12v connectors.

I would not discount the fact your psu may be faulty, even if it can do this. Best way to test the psu is in another PC, if you've got one.
June 1, 2006 11:21:22 PM

Have you tried what was suggested, about removing everything from your case and doing the following:

MB with CPU
only 1 stick of ram
your video card
NO HDD

??
June 1, 2006 11:23:49 PM

Sounds exactly like a bit of fun I had some time ago.

Turned out to be my power button (As PCCashCow suggested) on the case was sticking. I got kinda frustrated and pushed the thing in really, really hard and it powered up fine. Little more tweaking and I realised what was at fault and fixed it.

No more problems, though a somewhat less shiny power button now...

It's worth a try if nothing else.
June 1, 2006 11:27:08 PM

If a PSU would not start up while connected to a MB, then start up with the bypass, the PSU is fine. If you want to take it further, one can actually test the voltages from all pins. If any wire is grounded, a plug is place on wrong, the PSU is designed to sense this and not fire up.
June 1, 2006 11:39:31 PM

The fact that the psu worked properly when the suggested test was performed, does not completely rule out any problems with it. I tried a similar test with my faulty psu, and all the 12v connectors worked, but when it was subjected to the extra load of a motherboard with all it's components, it failed.
June 1, 2006 11:52:54 PM

If your PSU failed because of overloading it, it's not the PSU's fault. Next time try this calculator to determine what PSU Wattage you'll need for your next build.

http://www.extreme.outervision.com/psucalculator.jsp



As far as the Power Switch, the jumper can be removed from the MB and manually short the two pins to rule out the Power Switch.
June 2, 2006 1:22:42 AM

My last post made it sound as if I overloaded the psu. That was not the case, it was shipped faulty. All the fans, hard drives, video, everything, kept cutting out without warning.

The replacement psu (Hiper Type-R 480W) is less powerful, but works fine.
June 2, 2006 3:35:03 PM

Long story short, your looking for a fix here that includes no contrtols for testing on the back end. You need yank another working PSU, or even run to Compusa or BB and test it with that, in less than an hours time youll have your answer. You'll be able to return everything no questions asked as long as its the same day.


I may have miss something you posted, but you also have not tested with your build outside the case as well as with less periphs to get at atleast some beeps codes. A bad mobo would hang on memchks or atleast just hang, not power down. You can also attempt to reinstall the mobo with only two screw posts holes ( and in other configs) to see if there is a ground out issue prompting a hault. Anyway, second hand items come with these ball breaking issues.
June 2, 2006 11:40:45 PM

Update: PSU is perfectly fine (tested in another computer). Rebuilding outside of the case, with only essential parts, I get the same problem. Sometime in the next few days, the RAM will be tested (not all that easy, as like I said, I don't have another computer myself that will take DDR2).
June 6, 2006 3:04:03 PM

RAM was good, MoBo was good. It was that engineering sample chip. Still plan on playing around with it eventually, but apparently the board just doesn't like it. But yeah, got my hands on a different proc, and now everything works just fine.
!