XP 1800+ killed stone dead in GA-7VTXH - help!

[Thorin]

My system is as follows

AMD XP 1800+
Gigabyte GA-7VTXH
Enermax 431W PSU
ThermalRight SK6 with a 60mm Delta Fan

I used Arctic Silver II as thermal compound and set the system up, so far as I am aware entirely correctly. I had the motherboard set to 133FSB and 11.5 multiplier. A soon as I turned my system on - and I mean almost *instantly*, there was a "pop" sound and an acrid burning smell, I shut down the power almost immediately, and when I took the heat sink off, I found a very small amount of a sort of browny redy sticky liquid. I'm assuming that the CPU is dead, so haven't tried it again.
Having looked at things, the only thing that I think might have been wrong is that my arctic silver II looked *slightly* too thick, but would this really cause the CPU to blow so quickly while the HSF was properley and securly clamped onto the chip?
If anyone has any ideas as to why this would happen I would greatly appreciate their input. This is my second motherboard and second CPU in my ongoing saga of attempting to set up this computer, and I'm *so* hacked off about it at the moment it's not true! <g>
Please ask if you want any further information about my setup.
Hope someone can help,

Thorin

 

[peteb]

Um - when you bought the SK6 - did it have a square of firm tacky material on the base - or was it polished/flat metal surface?

Is the 'browny redy sticky liquid' still liquid, or has it now taken on the consistency of burnt plastic?

If I were asked to *speculate* on what happenend (more info will help), your heatsink shipped with a thermal interface material/pad on its base that was protected by a polythene strip. The ASII/Polythene/TIM/Heatsink combination was not very effective at cooling your cpu and it blew, melting the polythene in the process.

ASII is to be used on a naked processor die to facilitate heat transfer to a clean, naked metal surface - the heatsink. It should not be used in conjunction with a TIM. Also all polythene protective coverings should be removed prior to being inserted in a 60 watt thermal junction....

Let me know how it went.

-* <font color=red> Under Offer </font color=red> *-
email for application details

 

[Thorin]

I'm afraid I cleaned the liquid off, but I can tell you that the SK6 shipped with *nothing* on the bottom and a little sachet of silicon compound - which never went anywhere near the heat sink. The only thing that I used on it was the ASII.
Thanks for your response.

Thorin

 

[Guest]

I am sorry to say that, most probably your CPU is fried. I once cracked a core during heatsink installment, its pretty annoying. I understand your situation.

Anyhow, installing the HS is a very delicate job. You should pay a lot of attention to it. One little mistake and your AMD goes into flames.

I wish we could have some sort of overheat protection. The motherboards should finally support the thermal diode of XP.

<font color=blue>Get a T-Bird...
Impressive CPU + House heater in one package. What do you need more?</font color=blue>

 

[xxsk8er101xx]

sorry to laugh at your misfortunes but it is pretty funny. It's probably under warranty so get it replaced and you should be fine.

thats the only thing that keeps the Pentium 4 to look most appealing. P4 wouldn't do that.

so when you got the heatsink there was NOTHING on the bottem of the heatsink? no plastic or anything?

what may have happened is that the heatsink may not have been on completely... meaning there was a small gap where there was no heat transfer and it just popped... that liqued is probably the core was it blue or brown or green? the core is blue.

i would guess that you forgot to take the plastic off the heatsink perhaps? ... or you had a wire under the chip when you put on the heatsink causing a gap?

<A HREF="http://www.anandtech.com/mysystemrig.html?id=9933" target="_new"> My Rig </A>

 

[Thorin]

Laugh away - it is pretty funny, since it was my second chip that's gone up (the other was in different circumstances), if I didn't have a sense of humour about this I'd have taken a pick axe to something by now! (second motherboard too actually).
There was absolutely nothing whatsoever at all on my heat sink - I promise. There appears to have been nothing at all between my CPU and my heatsink, apart from my ASII. The surface of my heatsink is made up of very very small ridges. As per the ASII instructions on their webpage, I put some ASII on the heatsink, rubbed it in well with a plastic bag over my finger, wiped off what was left on the surface with a lint free cloth, and then spread a minute amount over the die on the CPU.
I know it sounds like I'm trying to make out I didn't do anything wrong. Presumably I *did* do something wrong, but I really want to know exactly what it was so that I don't do it next time!
The only thing I can think at the moment is that the ASII was a bit too thick - but it was pretty darn thin, and would it really burn out *so* quickly (instantly really) if I had an extra fraction of a mm there?
The liquid looked browny red - to me it looked very much like I'd expect the brown covering on the XP chips to look like if it were to melt - so at a guess that's what it was. I wiped off the liquid and underneath it was a teeny bump on the brown chip covering, which hadn't been there before - it looked like something had exploded outwards basically.
Thanks for the feedback - keep the suggestions coming... please?

Thorin

 

[Thorin]

I am sorry to say that, most probably your CPU is fried. I once cracked a core during heatsink installment, its pretty annoying. I understand your situation.

I thought it would be fried, but thanks. For next time what is it I need to do to not crack the core (if that's infact what I did)? I placed the heatsink on the CPU very gently. Would purchasing one of the copper shims designed for the XP chip prevent me from doing this again?

Anyhow, installing the HS is a very delicate job. You should pay a lot of attention to it. One little mistake and your AMD goes into flames.

Any tips for what specifically I should be careful not to do next time?

I wish we could have some sort of overheat protection. The motherboards should finally support the thermal diode of XP.

Amen to that!
Cheers for your reply,

Thorin

 

[AmdMELTDOWN]

LOL! AMD cpu's are a POS, they offer no thermal protection; none whatsoever! and to add to that, their PR system is a sham.

well, don't feel bad most of the AMD ppl here have gone through several cpu's, mainboards and mem.

"<b>AMD/VIA!</b>...you are <i>still</i> the weakest link, good bye!"

 

[FatBurger]

I've had one set of all of the above, stable and fast. Hmm...

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>

 

[mbetea]

a cracked core i don't think will "pop". the four mounting pads on each corner is enough cushion for the heatsink. are you sure you didn't happen to get a teenie bit of asII somewhere on the bridges?

[insert philosophical statement here]

 

[Matisaro]

As 2 on the bridges would not short it out, as2 is not electrically conductive, also shorting the bridges does not cause a fry. I think this guy is one of those people from the web sent to start flamewars on our site as mentioned in the beware of the trolls thread, and as such I will ignore his post.

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

 

[mbetea]

ah. thanks for correcting me, wasn't sure but i thought i'd throw that out. pop! reminds me of that candy that snaps and pops when you put em in your mouth... you know, the ones that drinking soda with them will cause death.

[insert philosophical statement here]

 

[CoOLMaNX]

rock pops my friend, and it doesn't cause death when drinking soda with them..i tried as for the fried cpu, instantly? i doubt that but from what your saying, it should be under warranty.

 

[Thorin]

I'm not a troll, my problem is genuine (have you never read the hobbit/lord of the rings? Thorin was a dwarf, not a troll! ). You're entitled to your own opinion of course, but I'm not interested in a flame war, I'm an AMD fan, I'm just having trouble with their lack of thermal protection at the moment (or my own incompetence, I'm really not sure yet <grin>).
Cheers,

Thorin

 

[Thorin]

It really did go instantly - which I was a bit surprised about as well. From peoples reactions here, it's starting to sound more and more likely that I did something wrong that knackered it. Fingers crossed the company I bought it from will honour the warranty.

Thorin

 

[Matisaro]

They should honor it.

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

 

[Thorin]

I didn't think I had, and I've just had another look at it, and there's no sign of ASII in the bridges.
How rough would I have had to have been to have cracked the core with the heat sink? I thought I was farely gentle and it went on straight onto the pads - didn't crash down at an angle or anything. Maybe I wasn't gentle enough though!

Thorin

 

[Kelledin]

If you pressed down only via the clip, you should generally be fine. The clip style used on the SK6 is usually designed to put pressure right on the center of the core, whether it's already clipped or in the process of being clipped.

The only ways the HSF could have damaged the core:

1) Something shorted out, like the HSF touching two capacitors on the CPU or two electrolytic capacitors on the motherboard. This is very unlikely, especially with the SK6.

2) The HSF was pushed down via something other than the clip.

3) The HSF was put on backwards (i.e. the clip points were reversed). This would cause the clip to apply pressure off-center to the core.

4) The HSF clip was just too damn tight.

Kelledin
[dave@discovery ~] kill -9 1
init: Just what do you think you're doing, Dave?

 

[Thorin]

2) The HSF was pushed down via something other than the clip.

Nope, I just placed it down and clipped it on.

3) The HSF was put on backwards (i.e. the clip points were reversed). This would cause the clip to apply pressure off-center to the core.

This is a possibility I guess. Any idea where I might find a picture of which way around it should be attached?

4) The HSF clip was just too damn tight.

Again possible - any ideas what I should do though? Do I just find this out when the next one blows as well and I've already discounted everything else?
Thanks for you help,

Thorin

 

[Kelledin]

It's sometimes a little difficult to find out which way the HSF should go. Instructions are, as always, the final authority if you have them.

If the HSF extends over the "hinge housing" of the ZIF socket (the raised portion with no holes in it), it <i>should</i> have one edge notched where this raised area would fit. The SK6 <A HREF="http://www.anandtech.com/showdoc.html?i=1532&p=6" target="_new">appears to have that notch,</A> although I'm not sure why it has it along two edges. =/

Kelledin
[dave@discovery ~] kill -9 1
init: Just what do you think you're doing, Dave?

 

[xxsk8er101xx]

nope ... too much ASII shouldn't do that. What could be is that you didn't have enough of it and left a small gap from the core to the HSF.

Just take a q-tip and some rubbing alchohol. rub the bottem of the HSF with rubbing alcohol and set that aside. Take a different q-tip and put a small amount on th q-tip. then rub on top of the core until the entire surface is covered.

thats how i did it and temps are pretty low for a 1.4ghz machine and considering i have a pretty old basic HSF (global FOP52).

so it has to be that there was no heat transfer. My guess would be you didn't put enough on. Perhaps that HSF you have purposely leaves a small gab from the core to the HSF *shrugs* thats my thoery and i'm sticken to it unless you gots more info for me.

<A HREF="http://www.anandtech.com/mysystemrig.html?id=9933" target="_new"> My Rig </A>

 

[Thorin]

I didn't get any instructions at all with the HSF .
I've had a look at the pictures on anandtech that you linked to. Now I'm pretty sure I had the heat sink itself on the *right* way around, but the clip I think was probably on the wrong way now. One of the pictures on anandtech, clearly shows the notched side of the heat sink, with the clip with the sticky out bit (which I believe is to allow you to attach/remove the clip using a flat headed screw driver) on that side. I had a look at my HSF and the clip is the other way around. I'm thinking this could have meant that the bent bit in the middle of the clip was applying pressure off to the side of the raised die on the CPU, which would have caused only partial contact between the HSF and my CPU as the surface of the heat sink would have been wonky. Now this could have caused my instant burn out I'm guessing!
Cheers for all the feedback, I *think* I've gotten to the bottom of it now - if anyone thinks otherwise please feel free to say so.

Thorin