I'm stuck.. Please Help!

fqh · Post by **fqh** » Sun Apr 13, 2008 9:55 am

Dear all,

I wasn't sure where to post this, but here goes:

I've just taken delivery of the following system, in bits:

Antec P182
Q9300 CPU Boxed
Asus P5K-E/WIFI-AP
GeForce 8800GTS 640MB
Western Digital Caviar 500GB SATA2
2x 1GB Corsair DDR2
Antec Phantom500 PSU
Samsung DVDÂ±RW burner SATA

After putting everything together, it didn't start...
On pressing the on/off button theb system turns on for a second (all fans spin) but the goes off.
The green LED on the MOBO is on.

After taking it all out of the case, it seems that the issue is with the CPU or cooler.
When powering up just the motherboard without the processor (with all fans connected) it does start and stay on.
When putting the processor and stock cooler on, it only powers up for a second, as described above. This happens inside and outside of the case, with or without other components connected (ram, vga,e tc).

I've cleaned the processor and headsink surfaces (w/ Artic clean) and applied thermal paste (OCZ freeze), and mounted them on the mobo a number of times, but the result is always the same.
I thought the issue could be related to poor mounting of the cooler, but however awquard I find the mounting hole procedure I can't see what else different I could be doing..

Being a bit of a noob with this sort of stuff I'm kind of stuck and would really appreciate some help!

Cheers

widowmaker · Post by **widowmaker** » Sun Apr 13, 2008 10:50 am

Your processor should have a built in feature to automatically shut itself off in the event that it overheats to extreme levels. I had this problem once with my Athlon 64 as well. It turns out my cpu cooler was not securely attatched. Check the base of the heatsink with a blade or a known flat surface. Make sure it is flat and not concave or convex. If it is, this may be the cause of your problems. Do the same with the processor. The processor also may be defective. If possible, return the processor to the store and just tell them that it's defective and you want an exchange. Chances are they'll allow that. If the problem persists then the problem is probably elsewhere. Test the components one at a time and use process of elimination to determine where the problem lies.

fqh · Post by **fqh** » Sun Apr 13, 2008 11:49 am

Thanks for the reply

As far as I can see, the surfaces of the processor and cooler are flat, using a blade to check.

I am puting a blob (rice grain sized) of thermal paste on the surface of the cooler, as instructed on te OCZ website. Is this the best procedure?

Has anyone had mounting problems with this stock cooler on a similar mobo?

Not sure where to go from here...

Arvo · Post by **Arvo** » Sun Apr 13, 2008 12:21 pm

Have you connected 4pin +12V CPU power cable?

I've heard and seen numerous cases, when this cable is not connected at all or is not connected properly. On one system PSU had 4pin connectors like CPU1 and CPU2 (joinable into 8pin CPU power connector) - I did connect CPU2, exactly same symptoms. CPU1 connector got system to life.

Bob The Pc Builder · Post by **Bob The Pc Builder** » Sun Apr 13, 2008 2:47 pm

Arvo wrote:Have you connected 4pin +12V CPU power cable?

I've heard and seen numerous cases, when this cable is not connected at all or is not connected properly. On one system PSU had 4pin connectors like CPU1 and CPU2 (joinable into 8pin CPU power connector) - I did connect CPU2, exactly same symptoms. CPU1 connector got system to life.

yeah this sort of thing sounds about right, cpus generally start up and let you get into bios at least for 20 seconds before the overheat protection kicks in and it does sound like you are seating your cooler properly.

kittle · Post by **kittle** » Mon Apr 14, 2008 12:06 pm

Also check which fan plugs you are using for your CPU fan.

some motherboards have a auto-shutdown when the cpu fan fails. and if you are plugging the cpu fan into the wrong plug, the motherboard wll think the cpu fan has failed.

mark314 · Post by **mark314** » Mon Apr 14, 2008 1:45 pm

I have been having the exact same problem with a computer I built for my uncle. It worked perfectly the last few months with occasional problems, and those were fixed by clearing the CMOS and starting anew.

Yesterday I went to install a new hard drive, and I did nothing more than plug in the SATA connector and power, and the computer wouldn't start.

The PSU fan would spin up, stop, spin up, stop. Spin up, stop. Over and over. Mobo light was on. In playing with a number of things, I could make it start, occasionally. For instance, I took the side of the case off. System started just fine. CMOS was cleared. Re-set my settings, restarted - system wouldn't start.

I have tried re-seating my cables, re-seating the video card, re-seating the RAM. CPU is not overheating because the system has been running fine for months.

I took the CMOS battery out completely. System started fine. Put the battery back in. System wouldn't start. "Oh, the battery must be bad!". Took the battery back out. System still wouldn't start.

So, to make a long story short, I suspect either the motherboard has a hairline crack in it, or the PSU is faulty, or there is a short somewhere.

Check your motherboard carefully for a crack, blown capacitor, cut wire, or short.

Let me know what you find - I am at a loss as to the cause of my problem.

PS. I am also on a P5K motherboard.

andyb · Post by **andyb** » Mon Apr 14, 2008 3:38 pm

Hi Mark.

Mobo or PSU are most likely, but without spare parts or a PSU tester both can be impossible to diagnose.

Try this.

Take the mobo out of the case and put it onto a non metallic flat surface, and attach only the following.

CPU, heatsink/fan (plug the fan into the CPU fan socket), 1 stick of RAM (consult the manual as where to put it), Graphics card (including power cable), monitor and keyboard (wired only, ideally PS/2 (purple port) not USB), pull the battery (or nuke the CMOS using the jumper (listed in the manual).

Then hook up the PSU (24pin + 4 pin), and whilst watching the 3 LED's on the keyboard turn it on (you can use a scredriver by touching the 2 pins that the power switch attaches too if it doesnt reach).

If all 3 lights on the K/B flash at the same time, then you should get a screen display, if the 3 LED's flash once, but you dont get a display your graphics card is probably duff, if not it could easily be your mobo or PSU.

To test your PSU, buy one of these (easily the best tool I ever bought, although I use it more than most people use screwdrivers

).

http://www.ebuyer.com/product/99889

Its easier to rule out everything except your mobo as they are often only identifiable as faulty by testing everything else, your CPU and RAM are a thousand times less likely (unless you have bent a CPU pin on the..... motherboard

) You can also swap the RAM around between slots and try other sticks (just dont do it while its on because you can fry stuff).

Andy

BLG · Post by **BLG** » Mon Apr 14, 2008 5:13 pm

My vote goes to the PSU . . . or the motherboard. But I've have heard very similar circumstances, though like everyone has said its almost impossible to easily pin down the problem without spare parts and a PSU tester.

mark314 · Post by **mark314** » Tue Apr 15, 2008 8:39 am

Excellent suggestions, all of them. I'm picking up the computer tomorrow to disassemble it. I have two spare PSU, spare ram, and maybe a spare video card. So I can test it.

I have tested it inside the case in a number of configurations - no video card, no ram, 1 stick of ram, 2 sticks of ram, no hard drives, nothing plugged in, no CMOS battery, etc etc.

Twice I was able to fix the problem by clearing the CMOS. The last time I did that was last night. It fixed the problem right away. Unfortunately, of the two hard drives, it didn't know which to boot from. When I unplugged the power from the other hard drive to prompt the system to re-recognize my boot drive, the HW configuration change prompted it to stop working again.

So to be honest, I am thinking power supply. I am using the default Antec Solo PSU - Antec TruePower P3-430. I have no experience with this one so I don't know if it's prone to failure.

Another distinct possibility that I noticed yesterday is that the RAM I'm using in that machine is OCZ Platinum DDR2-800 - that's 2.1V ram. The ram is supposed to run fine at default settings of 533MHz, 1.8V, but I have had problems in the past. Since I can't get in to the BIOS to change the voltage, the undervolted ram may be destablizing the system. I have 1.8V ram to try.

Your suggestion to test the video card is a good one, though the system should boot okay with no graphics card installed - it would then give the 'no display' POST error.

I will definately disassemble if needed and examine for shorts, hairline cracks, blown caps etc.

mark314 · Post by **mark314** » Tue Apr 15, 2008 8:41 am

PS. I don't have a PSU tested but I do have a multimeter.

But since issues only arise when hardware-level changes are made to the system, and otherwise the computer is flawless (no crashes, no BSOD, nothing) it would seem to me that a configuration issue is to blame (ie CMOS, or RAM voltage) and not the PSU/MB themselves.

CMOS battery has been replaced as well.

xan_user · Post by **xan_user** » Tue Apr 15, 2008 8:56 am

Have you ruled out a faulty power button by starting the mobo with a screw driver? IE short the mobo jumper for the on of switch with the tip of a screw driver momentarily.
If the power button is getting stuck in,or has a short, it could cause the problems you describe.

mark314 · Post by **mark314** » Tue Apr 15, 2008 9:19 am

No, I haven't ruled that out. It seems unlikely to be the case, though, for a low-dust, <1 year old high quality case (Antec Solo)

But it is a possibility, and so trying it outside the case with a screwdriver will rule that out.

Only thing is, whenever the computer is running normally, it shuts off/restarts/turns on without difficuly.

Barkas · Post by **Barkas** » Wed Apr 16, 2008 1:45 pm

Check out have you put your "2x 1GB Corsair DDR2" properly in Asus motherboard.

For example on My Abit IP35 PRO there are 2 difference color. And you must put them in the same color slot.

mark314 · Post by **mark314** » Wed Apr 16, 2008 7:56 pm

So.

Tonight I replaced the RAM with standard 1.8V ram and cleared the CMOS. That fixed the problem.

The system now boots fine. However I only put 1 of 2 Gb of ram in. I am going to have to put the second stick in soon; only then will I know if the system is actually fixed.

I'll keep you posted but for now the message is - check your RAM!

mark314 · Post by **mark314** » Thu Apr 17, 2008 9:22 pm

Aha! And I put the second stick of ram in, and it still works.

So the RAM was the problem...