[Diver9]

Hi,

I have a very strange situation that I have been asked to look at by a friend:

It is and eMachines E4252 AMD 54 x 2 Processo 4000+, nvidia Geforce 6100 onboard Graphics, 1 GB DDR2 Ram (2 x 512), 150GB Sata HDD with Windows Vista Home Premium Installed.

Basically, as I understand it, the computer started shutting down as it leaves the "safe/with networking/last known good configuration" screen, all power stops all fans stop. Sometimes, after repeated attempts, it would start up and work normally.

I took it home and first changed PSU, same issue; I then removed/swapped RAM for some new ones I have here....same result. I tried CMOS battery and removing IDE/power to cd drive

It will sit at thesafe mode screen for hours with no problems and also in the bios with no problems, but as soon as it progresses past the safe mode choice screen it shuts down and on the odd occassion it does "go all the way" it will shut down at any point in time.

I have checked the CPU temp and it is running at about 35c so not overheating.

My only thoughts left are mobo??? Anyone here have any further suggestions as I am going bald scratching my head.

Thanks in advance

Diver9

[Advertisements]

Take the side of the case off and look at the power supply.

If it mentions the word "Bestec" on it, which it probably will do as they tend to be found in eMachines, and if the machine is about three years old, the problem is almost certainly caused by that power supply. In fact I'd be tempted to replace it anyway for the simple fact the power supplies in stock eMachines are rubbish and I wouldn't even prop my internal doors open with them.

This is a common problem on this particular model, but the Bestec thing is common to all eMachines.

[Neil Jones]

Take the side of the case off and look at the power supply.

If it mentions the word "Bestec" on it, which it probably will do as they tend to be found in eMachines, and if the machine is about three years old, the problem is almost certainly caused by that power supply. In fact I'd be tempted to replace it anyway for the simple fact the power supplies in stock eMachines are rubbish and I wouldn't even prop my internal doors open with them.

This is a common problem on this particular model, but the Bestec thing is common to all eMachines.

[Diver9]

Thanks for the heads up, in fact it was the first thing I tried, but still had the same problem even with a brand new one. Thanks again.

[D-Berd]

Can you boot off of a cd or floppy without the pc shutting down? Can you put a pci or pcie or agp video card in and test it? Can you update the bios? Or reset the bios? Does the Hd make any funny sounds or can you swap out the hd for a different one for testing?

[Diver9]

Thanks for the fast response - I tried booting from the factory disc and it went through the language set up and then it shut down.....the graphics card was the next option...not sure if I have one here so will need to rport back on this. Hdd makes no funny noises - due to the owners trying to do an install to fix it a few days ago, I thought I would try by mounting it as another drive on my system and back up all the user data...will swap out for another sata drive to see if there is an issue with it. I have changed the CMOS battery, but have to admit I did not reset the jumers at the time....that's 2 things to look at.

Thanks again,

diver9

Edited by Diver9, 17 November 2010 - 02:26 PM.

[D-Berd]

Yeah, if a different video card or HD don't work then it could be the motherboard. You could check and make sure nothing is grounding the mobo out to the case, like a screw or something. Disconnect any case fans, modems, network cards, you know. You could also try booting a disk from another cd rom drive if you have one. not much else you can do besides change out the cpu. probably not worth the effort thou.

[phillpower2]

I have checked the CPU temp and it is running at about 35c so not overheating.

Without the pc operating under normal circumstances the temperature you are seeing is no proof really, as the temps can go from 35c to 100c+ in the blink of an eye, if the MB has a thermal sensor and this happens it will shut the system down to protect the CPU.
With the power cord removed from the wall check that the CPU fan is connected securely to the header on the MB and that the HS and fan are mounted correctly on top of the CPU, re-connect the power and whilst keeping an eye on the CPU fan attempt to boot up and see if the fan is working and that it continues to work, if it does but the system shuts down I would suggest removing and re-seating the CPU, to do this you will need some TIM (thermal interface material) such as Arctic Silver 5 and the correct cleaning and preparing fluid, clean off the old stuff from the CPU and the HS face, remove and re-seat the CPU and then re-apply the correct amount of TIM to the top of the CPU, too much is as bad as too little so just spread a nice thin even covering over the top of the CPU and reassemble the HS and fan etc, hopefully this should do the trick, just an added note let the computer run a while before you remove the HS from the CPU as it will help to separate them due to them being warm, it is not unknown for the HS to pull the CPU from the MB socket when cold, make sure you earth yourself before reaching inside (touch a bare metal part of the case) and that the power cord is removed from the wall.

[Diver9]

Thanks for the reply.

The fan cpu fan is connected properly and the heatsink appears solid - upon boot up the fan spins up as normal and stops the instant the computer shuts down.

I have now tried another new hdd and still the shut down occurs. I have also tried a new graphics card and this does not cure the problem. I think it is and indicator that the motherboard may be toast?

Thanks

diver9

[phillpower2]

Possibly but without checking out the re-seating the CPU option it is inconclusive.

[Diver9]

Thanks

I will order a mobo just in case (thankfully not too expensive) and also some thermal paste and thermal paste remover - will then try reseating the cpu prior to replacing the mobo.

Again thanks for the fast replies.....it helps reinforce some of my own ideas and point me to a few un thought of ones as well.

Cheers

Diver9

[D-Berd]

whoops

Edited by D-Berd, 18 November 2010 - 10:05 AM.

[D-Berd]

You can pull the heatsink and cpu, then reset the cpu and heatsink without replacing the TP. I wouldn't run a system like that as a final build, but for testing it will be fine. That way, when you get the new motherboard you wont have to go through the process of applying TP and then removing it only to have to do it again to install on to the new motherboard.

[Diver9]

Cheers for the reply, I really appreciate it - I am going to work on my other system tonight and look at my friends tomorrow when I can give it my undevided attention.

Many thanks

Diver9

[Diver9]

Thanks to all for your help - new mobo arrived and low and behold....Problem Solved!!!

Thanks All

Diver9

[phillpower2]

Congratulations and thank you for posting your solution