CPU_LED Solid Red/Burnt CPU

Dopeland

Distinguished
Oct 6, 2011
6
0
18,510
So heres the deal pretty much, I have never built a computer before or even so much as installed ram or anything like that.

I attempted my first build today and got it to post and all that and installed drivers and such but noticed that the CPU temp was very high (90 @ idle). I looked into it and saw that the stock heatsink I was using wasn't all the way pushed into the motherboard, and was I tried to do this one of the prongs used to mount it in there broke.

So I went down to tigerdirect and picked up a n520.

I installed it the wrong way (oops lol) and all posted well with idle temps sitting at 70

I installed some drivers and such and everest to check my cpu temp

When everst tried to check my CPU temp I got BSOD

I reset the system and was able to make it to bios, where I changed the ram speed to 1600

I put more ram sticks in (I only had 1 stick in the A1 slot) and tried to boot up but got the good old CPU_LED shining a bright solid red with no beeps.

The fans and everything would power up in the case, then everything would shut down, then everything would power back up but no POST.

So theres no doubt there was some wrong done on my part, here are the issues at hand:

CPU was way too hot and now has a nice fat burn mark on it (I think?)
RAM was not on the motherboard's authorized vendor's list (but I was able to post with this and use it for a good hour or two)

What I want to know here is where I went wrong and whether or not I should be worried about more than just CPU damage.

Here are some pics of my CPU:
i72.jpg

i71.jpg



So pretty much I have a new motherboard and new CPU on the way tomorrow, but I want to know if I need both or only 1, or if perhaps I'd need some more stuff.


Here was my build:
Asus sabertooth p67
intel i7-2600k
visiontek radeon 6950
16 gb corsair vengeance kit
Raidmax 1000w psu

Also wondering if it wouldn't be a good idea to just drop a brand new cpu into this rig as I'm worried I might fry up another.
 

rvilkman

Distinguished
Generally intel CPU's should be able to handle the fact that they overheat using normal voltages and frequencies.
And generally once it does overheat it shuts itself down. Which it probably did with the bluescreen and all.

The ram should be ok as long as you got it to boot. And corsair vengeances are pretty much a good quality ram so i doubt that is the problem.

As long as you didn't kill the motherboard with the HSF installation by breaking it physically, your problem is most likely the CPU.

Still the 70C for the idle temps sounds really really hot. Since I accidently installed a HSF with the rubbery film still connected and was at 45C (i5 2500k & Scythe Mugen 3) or so.
90+ under load but still it wasn't killing things. Went to like 26C idle and 43C under load after i removed the film :)

So most likely the issue that eventually caused the death of the CPU is the memory speed adjustment. It somehow disabled the speed adjustment functionality of the CPU and it ran at the full frequency all the time. ( thus the 70C temps ) then under load you spiked and it fried the CPU.
Well that's one theory anyway.

Basically take a look at the pins on the mobo if they seem ok it most likely is just the CPU.


 

008Rohit

Distinguished
Aug 1, 2011
1,093
0
19,360
Did you place the heatsink properly the second time? also, what exactly did you break? Its highly unlikely for a Intel SB CPU to die like that.

While setting the RAM speed, did you change the RAM voltage to say, over 1.65V?
 

Dopeland

Distinguished
Oct 6, 2011
6
0
18,510
The damage was done before I readjusted the heatsink I believe, because after I put it on properly it still wouldn't post.

Also I didn't change the ram voltage at all
 

t2couger

Distinguished
Mar 6, 2006
135
0
18,680



Don't be an ass the guy is learning and it sounds like the CPU was defective to begin with. Good luck with you build let us know how it goes.