3770k IHS Removals - CPU temp dropped from 79C to 71C

[MaynardX]

Hersounds

You're right, but nobody has ever done comparable tests within lets say the same batch CPUs. That is why I wonder.
Maybe not the climate but ambient temp and humidity (humidity I think has less impact).
My ambient temp for the test was 24 Celsius.
If you don't mind I'd like to see your results just for curiosity.

[hersounds]

UPDATE
4.05 Ghz - 0.98vcore - 16 gb full bank 2400 - maximus V Gene Z77



Batch 3226B987

[docwhom]

Today the patient was admitted to the operating room

successful surgery!

Dear Sir,

I'm presuming the Super Thin Gillette's were of great need when staging this operation?

Because I have never used them before and they are double-edged, you hold onto the non-sharp edges with two fingers and saw the blade into the space between the IHS and pcb of the cpu?

I want to do the same thing but because I have never done it before the idea kind of freaks me out. YES I'm a total noob when it comes to this, hence any subsequent tips and/or suggestions would be gratefully appreciated.

The thought of possibly destroying a $330 cpu with one tiny mistake really weighs heavily upon me, but if I could get a similar finish to your excellent work, I would be very happy.

[hersounds]

you just have to go and start at the corners, and see which of all the Gillette glides smoothly.
In some cpu's cost more in another less, in my case was completely sealed, only the last corner let me enter for lucky.

once the gillette is inside the corner. simply slipping slowly,carefully taking control

you enter more than allowed and your chip is died be careful with that,just let the gillette slipping do the work

[SonDa5]

you just have to go and start at the corners, and see which of all the Gillette glides smoothly.
In some cpu's cost more in another less, in my case was completely sealed, only the last corner let me enter for lucky.

once the gillette is inside the corner. simply slipping slowly,carefully taking control

you enter more than allowed and your chip is died be careful with that,just let the gillette slipping do the work

Also you shouldn't use the corner of the blade.

Corner of the blade can damage very easily. http://hardforum.com/showpost.php?p=...3&postcount=86

[hersounds]

Also you shouldn't use the corner of the blade.

Corner of the blade can damage very easily. http://hardforum.com/showpost.php?p=...3&postcount=86

that's right

101_0049.jpg

2.jpg

[owikh84]

well it makes sense what Zeus says if you take IHS out gives 55 then either is not tighten enough without it or overtightened...when you put IHS all things are normal as it should be,thats the reason most of us use it with IHS on but paste changed

I think I should try other waterblock and see how. I believe MVE needs more mounting pressure on the bare die. I don't wanna use back the IHS as it will hit TJmax easily under load. Now using EK Supreme HF.

Edit:
Edit: I've figured out how to fix this issue - removed the tiny washers that came with the waterblock.
Debug code 55 disappears so I can run my chip without IHS

[andressergio]

guys i have a weird situation, my 3770K lowered 12C temps using MX2 for more than a month now it has a core going to 97C with hyperpi summer aproaching more humidity but i guess something's wrong i want to try the liquid pro ultra, as i know its a metal liquid doesnt scratch the die ? what are the recomendations for using this kind of tim ???

thanks !!!
Sergio

[andressergio]

that's right

101_0049.jpg

2.jpg

hersounds do you have any pic of how you aplied that tim ? im not sure how to do it never used liquid tim in ma life

thanks
Sergio

[C-N]

Sergio, the clue was in hersounds picture... the cotton thingys/cue tips

Just whack a tiny spot of lm on the die & rub it in/spread it with the cue tip.... I also done the underside of the IHS

[hersounds]

hersounds do you have any pic of how you aplied that tim ? im not sure how to do it never used liquid tim in ma life

thanks
Sergio

if my friend, as said CN, using cotton swabs and stretching across the surface like you're painting.
careful to use not too much, just a drop

Sergio, the clue was in hersounds picture... the cotton thingys/cue tips

Just whack a tiny spot of lm on the die & rub it in/spread it with the cue tip.... I also done the underside of the IHS

That's right, just a drop on 4 sides, gives the best results

Die --- Ihs --- ihs top --- base cooper --- enjoy!

[MaynardX]

Hersounds

Could you post your temp results with 4.888 vcore please?

[hersounds]

Hersounds

Could you post your temp results with 4.888 vcore please?

sure mate

1.488 Vcore - not trying yet Intel burn test

sp 32 5 ghz.jpg

and my new 24/7

Cinebench R11.5 4.7 Ghz - 1.25 Vcore



Intel burn test 4.7 Ghz - 1.25 Vcore bios - LLC Extreme

[andressergio]

hersounds- i would like you shoot a HYPERPI 32M at 1,488 just to compare my actual results as in BS AS and Uruguay almost same temp

[andressergio]

Sergio, the clue was in hersounds picture... the cotton thingys/cue tips

Just whack a tiny spot of lm on the die & rub it in/spread it with the cue tip.... I also done the underside of the IHS

Thanks C-N but my questions if when you put the IHS wich i dont stick with glue again will not scratch the die ??? you got me ?

i just put the IHS over the die not glued and put in socket that moves obvious til it gets in right position this will not scratch the die ?

[st0ned]

My 3770k power , rockstable, just checking max temps.

[hersounds]

hersounds- i would like you shoot a HYPERPI 32M at 1,488 just to compare my actual results as in BS AS and Uruguay almost same temp

not test hyper pi but yes SP32M with more vcore,sure that can help you to have an idea

http://valid.canardpc.com/show_oc.php?id=2548788

sp32m 5.1.jpg

Validate @ 5.2 ( not benched )

http://valid.canardpc.com/show_oc.php?id=2548808

5.2.jpg

[andressergio]

not test hyper pi but yes SP32M with more vcore,sure that can help you to have an idea

sp32m 5.1.jpg

Validate @ 5.2 ( not benched )

http://valid.canardpc.com/show_oc.php?id=2548808

5.2.jpg

yeah but hyperpi will make heat much more than SPI32 than doesnt even load cores...

i would like to see full load on 8 and not linx or IBT

this is mine at 5GHZ HT using OFFSET vcore on first days of september where was cold



today i shoot the SAME but at 4.8 and last core scores 94 to 97C maybe tim moved dunno next step will use your same tim

[hersounds]

try later sergio...what temps have you with 5100 super pi 32?
and how much score in Cpu Queen ?

cpu queen.jpg

[andressergio]

I think I should try other waterblock and see how. I believe MVE needs more mounting pressure on the bare die. I don't wanna use back the IHS as it will hit TJmax easily under load. Now using EK Supreme HF.

Edit:
Edit: I've figured out how to fix this issue - removed the tiny washers that came with the waterblock.
Debug code 55 disappears so I can run my chip without IHS

glad that worked owi !!!

[owikh84]

Thanks C-N but my questions if when you put the IHS wich i dont stick with glue again will not scratch the die ??? you got me ?

i just put the IHS over the die not glued and put in socket that moves obvious til it gets in right position this will not scratch the die ?

I also was wondering the same thing. Will Liquid PRO eat my die. I'm using Liquid ULTRA right now and wanting to try PRO. Just afraid that PRO will stick permanently on the die and irremovable

[hersounds]

I also was wondering the same thing. Will Liquid PRO eat my die. I'm using Liquid ULTRA right now and wanting to try PRO. Just afraid that PRO will stick permanently on the die and irremovable

NO! Liquid pro / ultra & Phoebya are based on GALLIUM...Gallium eat aliminum but

die = sillicon
Ihs = nickel copper
just be careful that the base of your block/ cooler is copper

put a little little drop & enjoy

With less of 28,56 °C Gallium is solid more of 28,56 °C is liquid...is very imposible that stick permanently and irremovable...suffice a session full of prime95 or whatever you use to soften

[st0ned]

I also was wondering the same thing. Will Liquid PRO eat my die. I'm using Liquid ULTRA right now and wanting to try PRO. Just afraid that PRO will stick permanently on the die and irremovable

Liquid Pro as Ultra and other liquid metal based TIMs are mainly made of a gallium alloy. If I would have to guess I would say it's Galistan which can also be found inside modern thermometers. Although it will adhere to some surfaces like glass or even the nickel/copper from your cooler it wont do nothing to your dies sillicon, and will be really easy to remove from there.

With less of 28,56 °C Gallium is solid more of 28,56 °C is liquid...is very imposible that stick permanently and irremovable...suffice a session full of prime95 or whatever you use to soften

As I sad upper on the post it's not pure gallium if it was it would be pretty impossible to apply in most countries because it would have turned solid :P as I sad as well it is morelikely Galistan which has a -19ºC melting point


EDIT:

WARNING !! take my words with a grain of salt, don't go ahead and break your home thermometer, although I'm positive it would work

[hersounds]

WARNING !! take my words with a grain of salt, don't go ahead and break your home thermometer, although I'm positive it would work

Playing with HyperPi

hyper pi.jpg

[Dumo]

Just about to delided this one

Air