Removing the A64 IHS

[IvanAndreevich]

mad mikee
Taking the IHS off on 90 nm only yielded you 1C?!

[CrazyXP1700]

mad mikee... you got any pics of it?

[funkflix]

Hi

I can't go any higher before, then 2800MHz doesnt matter how much Volts!
Now i'm benchstable on 2930 MHz @ 1,66V

Gives me 6 - 8 °C under load!

[burnhead]

hey...how fragile are the cores? how far can i go to screw my watercooler on it? until the screws are tighter than tight or whatever??

what about the idea of taking this little piece of metal from 3,5" disk instead of the razor blade?

[krampak]

Does anybody removed the IHS of a FX-55?

[=[PULSAR]=]

GravediggA what kind of cooling are you running?

[funkflix]

Just normal Watercooling!

[burnhead]

GravediggA, how tight have you screwed your watercooler to the cpu?? btw. translate your sig into englsih, my friend... i guess not everybody understands german...

[funkflix]

Screwed it so hard as it goes.

[german]

Lol! Da sind gerade mal 2 deutsche Wörter drin.. Punkte u. einstellbar,
das werden die Leutz hier gerade noch überleben.

[/german]

[saaya]

Does anybody removed the IHS of a FX-55?

its just a normal clawhammer core, ´there are many posts of clawhammers without ihs in this very thread

and about the nekked winchester, WE WANT PICS!

so 6-8°C lower load temps with the winnchester without ihs?

please keep it in english :P

[IvanAndreevich]

>>so 6-8°C lower load temps with the winnchester without ihs?
That is what I would expect - MORE temp drops as opposed to the 0.13 cores.

[funkflix]

Yes! My Loadtemp was ~ 50°C before and now it is 42-43°C!

[p4z1f1st]

btw:

[funkflix]

U should search for an new Webspace m8! Most of ur links in the last 2 weeks
won't work!

Edit: Yeah, after 2 min watching this thread i see 1 cm of ur pic!

[burnhead]

Screwed it so hard as it goes.

[german]

Lol! Da sind gerade mal 2 deutsche Wörter drin.. Punkte u. einstellbar,
das werden die Leutz hier gerade noch überleben.

[/german]

do you want to p´lay a trick on me? then the core will break!? or is it that what you did?

[german] scheiß drauf, hast recht, die wern das schon verstehen... sag mal, willste mich etz echt verarschen? die schrauben so fest anziehen wie möglich?!?! dann bricht doch der core, der soll ja bei A64ern besonders empfindlich sein!! [/german]

@ all others: is it true? screw the cooler on as tight as possible? and that WITHOUT IHS??!

[funkflix]

No joke m8! Damn u really think i want that u break u core?

As hard as it goes means, that i screwed it so far that the resistance, is
so hard, that u can't even do an half turn more on the screw without destroy
ur core! U must test that by urself, it's just a thing of feeling!

[burnhead]

...just a thing of feeling! ouhouhouh... i haven´t got a "good feeling"...

[Huhn]

No joke m8! Damn u really think i want that u break u core?

As hard as it goes means, that i screwed it so far that the resistance, is
so hard, that u can't even do an half turn more on the screw without destroy
ur core! U must test that by urself, it's just a thing of feeling!

just tighten the scrwes so the cooler has good contact.
[/ german]
ich glaube nciht dass viel druck viel hilft sprich dreh die schrauben so fest an bis der kühler fest sitzt und die ganze überschüssige wärmeleitpaste wegdrücken kann. wenn du dir nciht sicher bist ob du genügend druck ausübst: bau den kühler auf mit moderatem druck und schau auf die temps. nun dreh den kühler fester und achte wieder auf die temps. irgendwann ändern sich die temps auch nciht mehr
[\german]

[funkflix]

Got that experience from my XP-M and the 1A HV2! Temps rise if i lower
the pressure of the HV2!

[reject]

IvanAndreevich
remount your wb i get that on quiet air (turned up loud) and 1.78v pretty stable

[p4z1f1st]

U should search for an new Webspace m8! Most of ur links in the last 2 weeks
won't work!

Edit: Yeah, after 2 min watching this thread i see 1 cm of ur pic!

give me a suggestion m8

i don't know y, but how u said, the last 2weeks my webspace is crappy slow....

[IvanAndreevich]

reject
My chip is , I think. Might require removing the IHS. But I am not sure I wanna with it, since I have a S939 board coming and going to be picking up a 90 nm chip very soon.

GravediggA
Excellent! Nice decrease in temp.
>>I can't go any higher before, then 2800MHz doesnt matter how much Volts!
>>Now i'm benchstable on 2930 MHz @ 1,66V
Wow that's a hell of an increase. I might have to do it when I get one

p4z1f1st
Imageshack.us for hosting pics

[Liquid3D]

Be every careful tightening down those blocks boys/girls. This is why higher tension Springs are so essential. I've tried mounting blocks without springs and in my experience if your needing to tighten down ANY block using that much pressure, the problem resides in the four mounting bolts themselves, and/or rentention mechanism itself.

Here's what I do. When I mount a block, first I insert the four mounting screws/bolts, through the back of the motherboard flip the board over and screw down the surface tightening nuts, BUT without tightenting them. Leave them about 3-mm above the mobo surface. With the CPU in the socket (no paste) slide the waterblock down onto the four mounting screws/bolts which will then be equaly spread by the waterblock mounting hardware and postion the mounting screws equi-distant from each other. Slide the WB all the way down until it rests evenly on the CPU's surface. Then (holding the block firmly to the CPU surface taping it over the top may help) tighten down the surface fastening nuts which secure the mounting screws onto the board surface. This should require a few turns as you've slide the block over them already. With the block resting flatly on the CPU, tightening down those surface nuts willl ensure the four mounting screws/bolts will remain equi-distant, as guided by the waterblock mount.

It takes some time and nimble fingers to get used to this method, but recently I ran my P4 530 at 89C, panicking once I'd discovred this getting to the temp screen while setting up the BIOS. I shut-down the system to see that I'd tightened down those retention nuts so that they compressed the springs 100% and I'd placed a lot of pressure on the block believing it made solid contact, only to find the block hadn't even made contact with the thermal paste, which was applied in a dollup meaning it had about 1mm of height off the CPU!

The tension I did feel was the waterblock as it dropped down on the four mounting screws which were obviously spread further apart nearest the bottom (surface of the mobo). Even though those screws fit perfectly through the board's mounting holes, they still had enough play to make it impossible for the block to make contact with the CPU as it neared the bottom of the mounting screws/bolts. Feeling tension as the block was sliding down the screws (you'll feel the vibration as the block drags along with the screw threads) I thought I was ok because I'd applied so much pressure. I'd even stuck my head in the case to get as close as a look as I could to verify the Block was making contact. So visually inspecting it isn't always reliable.

This also happened to my friend JNAV89GT recently, who ran his PC for almost 10-minutes without the the waterblock making ANY contact with the CPU surface. As a result I'm writing an article in which I plan on running py P4 530 without an IHS! Thermal throttling and VID kick-in at a specific temp which is why everyone experiences the same 89C temp.

If the waterblock doesn't glide easily from the top of the mounting screw or bolts to the CPU's surface then the screws inserted through the back of the motherboard were secured off-center (inntheir perspective holes) by the mobo surface nuts.

By the way although somewhat off-topic try measuring the distance between Socket-775 mounting holes. Their just about equi-distant, meaning it's very difficult to decide how to rotate the block if you don't align the impingement area over the core under the IHS.

[IvanAndreevich]

Right, so here I am - I picked up a 3000+ 90nm and struggling with it to get it stable @ 300 HTT. The temps I am getting is 50C with water. That's high after the ASUS which was showing 40C under same load on a Newcastle. However, obviously you can't compare temps going from board to board.

What kind of speed increases did winch users get from taking the IHS off again? 50 Mhz?

[IvanAndreevich]

STUCK @ 297x9 = 2670. ANYBODY who has tried taking off an IHS on a Winchester? Do you think it will let me run 305x9 or so stable?