My 4770k delid video (IHS removal)

[Boulard83]

Video @ http://infodupat.com/forum/viewtopic.php?f=37&t=488

@ 4.4ghz 1.3vcore
Load temp with Small FFT was 85+?c
Now i get ~75?c ( with NH1 paste )

** Waiting som liquid pro and ultra TIM for better results **

There is the base line for my futur TIM testing and a possible lapping if the IHS is not 100% flat.

[[XC] Oj101]

This method is far safer:

[Rattle]

I'd rather delid like op, you can see that whole pcb of the cpu bending while he's hitting it with the hammer lol and then it flies across the room...

[CrazyNutz]

This method is far safer:

Wow Maybe if you had someone with a catchers mitt standing by.

[Boulard83]

Totally agree, i'm far from feeling secure with the hammer method ! ! !

[Evantaur]

Wow Maybe if you had someone with a catchers mitt standing by.

or a cat

[cowie]

Much more fun to watch someone Smashing it off with a hammer,but i still think razor is ok too but boring

[[XC] Oj101]

I'm dead serious. Vivi (you probably know him) has done tons using the hammer method (he replaces the thermal paste and sells them under the Ultracooled brand in South Africa). You can't damage a CPU easily by throwing it, and you avoid the risk of damaging SMD components with the blade.

[nickshih]

Nice video ... thanks for sharing

[Boulard83]

Thanks to you.

I'll update with other kind of paste results and including Liquid pro and ultra.

[Boulard83]

There is the base line for my futur TIM testing and a possible lapping if the IHS is not 100% flat.

[The Stilt]

The hammer method is the most stupid thing I've seen for a while.
It is likely to cause some damage to the substrate & die-substrate bonding.

In sub-zero temperatures the substrate / die-substrate bonding can be damaged even by the mounting pressure of a properly installed LN2 pot. And the amount of energy going thru the chip during the "hammering method" is at a whole different magnitude level.
The Dow-Corning RTV is pretty tough stuff afterall.

It is pretty much impossible to damage the chip with the razor blade method if you have any experience at all.

[Boulard83]

+1 The stilt ...

The only thing you really need to care about is one side of the DIE with haswell wich have some capacitor. In fact, you could cut only the 3 side without any parts under the hood, then just pry whit the blad a little bit to pop the IHS of the last glued side. This way, NO risk at all to damage the capacitor.

[madmaxx]

Bien travaille Boulard83!

The hammer method is the most stupid thing I've seen for a while.

+1. Stupid is, as stupid does. Someone once said haha

It is pretty much impossible to damage the chip with the razor blade method if you have any experience at all.

With a vintage style shaving razor the job is near effortless!

[Boulard83]

With a thiner blade, yes it could be even EZer.

[i found nemo]

deliding is so much fun... unless you're the guy who tried it on any soldered cpu lol.

@boulard83
good speed, man. you could be in the finals for fastest delider. when i did it to my venice it took me like 15 min cuz i was so nervous!

[Boulard83]

deliding is so much fun... unless you're the guy who tried it on any soldered cpu lol.

@boulard83
good speed, man. you could be in the finals for fastest delider. when i did it to my venice it took me like 15 min cuz i was so nervous!

loll, thanks

I accelerated few parts of the vid but i dont think it took me more than 4 or 5 min overall.

[tool_462]

With a thiner blade, yes it could be even EZer.

With even a "used" Astra Platinum blade (my preferred CHEAP DE blade for shaving) these chips feel like you are cutting soft butter, not exaggerating! After I did a few Ivy's it takes about 25 seconds.

[Boulard83]

With even a "used" Astra Platinum blade (my preferred CHEAP DE blade for shaving) these chips feel like you are cutting soft butter, not exaggerating! After I did a few Ivy's it takes about 25 seconds.

I noticed that as soon as this thick blade has started to cut, it became EZ. Next time ill use thiner blade for sure !

[Boulard83]

First paste testing.
Noctua NH-1 versus EVGA Frostbite.



Well, the frostbite suck .... loll !
I had good hope on the frostbite paste since it seems to spread well under pressure but it did not deliver good results.


Edit, i also prepared the CPU for the incoming CLP and CLU "TIM".

EDIT 2 : I just redid the Frostbite application but my first ony appeared to be perfect. I had similar result on second application, quickly peaked over 80?c on AIDA. EVGA Frostbite seems is bad at this job...

[runmc]

What is so funny to me is the guy thinks he needs to wear sterile surgical gloves to smash it with a hammer. OMGoodness don't get any germs on it.

[Boulard83]

What is so funny to me is the guy thinks he needs to wear sterile surgical gloves to smash it with a hammer. OMGoodness don't get any germs on it.

LOLL


I just redid the Frostbite application but my first ony appeared to be perfect. I had similar result on second application, quickly peaked over 80?c on AIDA. EVGA Frostbite seems is bad at this job...

[Boulard83]

Well, after a good 20hrs of cure time for the frostbite it's still the same temp under load. As soon as i start AIDA temps are un the mid 70's with peak in the low 80's .... Overall ~10?c worst than Noctua NT-H1.

Will wait my CLU and CLP for more testing. I'll give a shot with CLU on die/IHS but frostbite on IHS/heatsink and see how it goes for the FB...

[Boulard83]

Got something in the mail today

[SonDa5]

First time I saw that vice method video I thought it was a joke. I read there are 1.4 billion transistors on 4770k, no way would I use a vice and hammer to delid.