Printable View
no shims or anything. just straight BT right onto the top of the processor. it'll be fine. if you don't remember, all processors used to come w/o an IHS. i had an old xp 2000 that i gave hell to, and did the same thing w/ an xp 3000. it'll be just fine.Quote:
Originally Posted by kimnkk
I have the same setup, my opteron 148 has no IHS and I've got a BT clamped right on :D works like champQuote:
Originally Posted by buttmud
Question:
With the IHS removed and a tower cooler installed (zalman CNPS9500) any extra care needed when moving the computer around, like in a car? I usually have set my computer upright on the floor of the back seat (works the best for me) but I'm wondering if that would now be a bad idea with the IHS removed and all that weight directly on the core bouncing down the road...
Naw, it should work fine. :) Just don't hit any huge potholes or something.
OK cool. I try to avoid those potholes anyway, trying to make my worn suspension last as long as possible :p:Quote:
Originally Posted by NickS
if you want to be safe, get some black electrical tape and make it 4 ply. put 1 pad on each corner. it'll help you get that warm fuzzy feeling that your core is safe :)
Just popped the IHS on my X2 4200
All I can say is
DDD
(didn't do :banana::banana::banana::banana:)
Temps nearly identical to previous readings using Core Temp. Which means I allready had really good contact between the core and the IHS/Shin Estu, that was allready there.
Oh well, guess my waterblock just sucks ass. Will replace it at some point.
Will post up pics in a bit here.
I did use a somewhat novel method to cut the sealant. Initially tried an x-acto blade but found out it was too thick. so tore a schick disposeable razor apart and then took one of the blades and broke it in half, mounted it in the x-acto knife handle and damn did it ever go through that sealant fast, 2 passes on each side (first pass for depth gauging) and she popped right off lickity split. Hardly even needed to push hard due to the razor being so damned thin and extremely sharp.
Well first off, just use your basic 1-sided razor blade. That's all thats needed, and it works perfect.
Secondly, people have mixed results. Maybe yours isn't tightened down hard enough.
Back to the mixed results, some people claim no drop, while others claim 10*C drops.
On all three of my Venice CPU's that I popped, I got on average 5*C lower temps.
Allright, here's the pics, rather large so didn't want to put them inline so just clicky.
The victim, Athlon X2 4200+ 236x11 for 2600 MHz 1.525 Vcore (allready previously lapped the IHS with 0 noticeable difference there either)
http://snipershide.us/images/sniper/...cpu_before.jpg
The method
http://snipershide.us/images/sniper/...azor_blade.jpg
Popped
http://snipershide.us/images/sniper/...ed/ihs_off.jpg
Remaining sealant removed and the x-acto blade depth
http://snipershide.us/images/sniper/...r_complete.jpg
Before and after temps
http://snipershide.us/images/sniper/..._and_after.jpg
Note that there is a 3C drop in the main cpu core temp, and a 1C gain in the Tcase temp, both of which can be attributed to just minor fluctuations in the case/room temps.
One other note, as far as holding onto the cpu while cutting. Took the foam pad from the plastic clamshell that the CPU came in and double back foam taped it to a small piece of aluminum, result, no bent pins.
I am hopefull that once the AS5 has had a chance to cure for a couple of weeks I will start to see some lower temps but who knows.
I Also am starting to think that the waterblock included with my Vantec kit just isn't up to the task of removing the heat from the core fast enough, so it will likely get replaced with a good swiftec or danger den at some point.
I hate computers ;)
I didn't feel like runnig down to the 24 hour wallgreens just to pick up a 59 cent razor blade at 6:00 AM this morning :P. So, I made do with what I had on hand, it worked out better than I could have ever hoped for, the blade in the disposeable razor is extremely thin and holding it with the x-acto blade holder worked perfectly. (check out the pics I just posted above).Quote:
Originally Posted by NickS
I did compensate for the loss of thickness due to the IHS being removed by adding some double back foam tape between my water block and the clamshell designed mounting bracket that the Vantec Stingray water cooling kit comes with. It's got plently of pressure on the core, probably just a matter of getting a better waterblock that screws down so I could increase the pressure even more. (pic of the wb installed) , another pic showing the mounting bracket
One thing to consider as far as your venice core is concerned, the X2's core is twice the size of a venice core, which I would have thought could actually cause better heat transfer just due to size but I guess the increased voltages and the simple fact of running two cores produces even more heat that needs to be taken away.
Ahh well, it was fun little project to do in the wee hours of the morning here :)
Decided to check out that theroy for certain.Quote:
Originally Posted by NickS
Didn't have any way of machining out a proper metal spacer to increase the core tension more, so I did the next best thing, I dug into my MR Yacknfutz (brain on too much coffee) machineries handbook and found a way to do it.
Sata cables and a waterblock, my new best friend. :D
I cut two lengths off of a Sata cable I had laying around, which is actually about .030" thicker than even the origina heat spreader was, and held them in place while I remounted the clamshell bracket.
Lo and behold, I hath seen the light :clap: 13C drop in heavy load temps :woot: , running 2 instances of Stress CPU and the real time HDR lighting demo running.
Oy, now I get to come up with a more permanant way of increasing the core tension, I don't think the sata cables should stay there ;) Pics coming in next post
Haha, told ya man! Glad to hear of your good results, and glad to be of help. :up:
I had forgotten that I actually turned off my radiator fan and main system fan to get the temps up faster, temp drop is more like 15C and that should increase even further once the AS5 has had a decent chance to settle in.Quote:
Originally Posted by NickS
Christ, I been kinda ticked off at the temps I was seeing with this water cooling, up untill now, the only real advantage I saw was in the quieter system.
It does make me wonder if increasing the core tension with the heat spreader on there would have made a difference or not, my guess would probably be not.
At any rate, just got through takin some pics, will post em in a few min.
Sounds good, cant wait!
Quote:
Originally Posted by NickS
Ok hwg, not going to put them inline cause I'll probably get :banana::banana::banana::banana::banana:ed at so just clicky ;)
Can see the two green strips of SATA cable i put under each side of the waterblocks hold down mechinisim. (note that there is a third strip along side the block cause the block scoots around on the core a lot more without the surface tenstion area of the heat spreader)
http://snipershide.us/images/sniper/...terblock_1.jpg
Here's the results.
http://snipershide.us/images/sniper/..._tension_2.jpg
As much as 15C drop, oy, maybe I can get this damn thing to 2800 MHz like I allways intended to now ;)
Nice :) Good temps for just a Vantec Stingray kit.
@ 2.81GHz, 1.63v I get about 54*C/56*C in Core Temp, Socket AM2 w/IHS Lapped (Soldered onto Die)
They started soldering the IHS down? I know Intel was doing that, didn't know the new AM2's were.
Yeah some of teh new AM2's are.
Heres my temps after about 30 mins of WCG.
http://upload.nickfire.com//files/1/temps%281%29.jpg
Not bad for IHS still in tact, with two vModded 7600GT's in the loop.
http://www.nickfire.com/public/ihs-lap/ihslap%20022.jpg <thats my IHS
Those temps look damn near identical to mine (crosses fingers for another 200 MHz).
Went the extra mile with that IHS I see lol.
Did quite a bit of reading before I lapped mine and found out that basicly anything over a 600 grit finish was likely just overkill so thats where I stopped at. Edit: heres what mine looked like when it was done. http://snipershide.us/images/sniper/...e/cpu_done.jpg
Oh and as far as your earlier coment about the stingray kit, I researched that damned thing from one end of the internet to the other, and basicly it allways came within 1-2C of the Astek kit that I had been looking at for quite some time and the Astek kit was $100 more.
As I commented earlier in the thread I thought that maybe the waterblock was just junk so I went out and found even more reviews, one inparticular caught my attention. It compared several of the leading waterblocks including the one from the Vantec kit and it came in second place for temps, the only one that beat it was the Swiftec Storm at $85 bucks just for the friggin block!!! ;)
Overall the stingray kit is a decent starter kit, which is kind of what I wanted, it does have some fit and finish problems that were not noted in any of the reviews I read, like the screws that hold the chipset spring arms on the nb, sb and gpu waterblocks all stick up above the surface of the contact area, thus you gotta file off the tops of the screws or run the countersinks deeper.
At any rate, I'm just glad I finally got the damnable temps to where I thought they should be, thanks in part to you ;).
Yeah the Vantec kit doesn't seem to be bad at all. Only thing bad about it I can tell is that the block is clip-down, not bolt-down.
As for the lap job, I also read that over 600 grit doesn't really help, but I wanted a mirror finish bad. BTW, what voltage are you running your CPU at? I wouldn't expect your temps to be too pretty after going as high as ~1.63v like me, sadly. Especially with a single 120MM rad handling CPU, GPU, and Chipset. (May be wrong on that one, having trouble recalling your exact setup.)
Anyway, glad to help! :)
Nick
1.50 Vcore 236x11 for 2600 MHz, ram is 4x512 sticks of corsair 1:1 @ 2.9 Vdimm, Northbridge is at 1.8 volts. 7900 GS 575/730. and a buch of other stuff.
Kicked it up a notch.
1.625 250x11 for 2750 1:1 with ram, backed off the ram timings to 10,4,4,3 for now and will dial it back up later, temps very very similar to yours.
New temps, http://snipershide.us/images/sniper/...bbadabbado.jpg :bday2:
One thing I did just notice when I brought up CPUz, friggin cool and quiet must be enabled cause it keeps dropping back to 1.350 volts, but so far appears to be stable, at least stable enough to run stress cpu for 10 min.
Dunno why its enabled, it shouldnt be, even bios says its disabled, hrrumph, oh well, if its stable, who cares!!! ;)
Edit: ACKKKKK that didnt werk lol, backed off the ram to 4/5 works fine now :) I really need to get pair of 1 gig sticks vs 4x512.
Wow, great improvement there Sniper_Merc :D . Makes me want to pop the IHS off my x2 3800+, but I'm afraid I'll crush the core when mounting it. Any tips to mount it as securely as possible?
Will I be able to mount my Storm water block without any modifications to the mobo? The reason I want to remove the IHS, is that one of the cores peeks 60c during load, while the other one keeps a lot cooler. Not sure if it's due to the IHS though.
I have a razor blade next to me, so it's hard to control :D
Yeah, the storm will mount fine. :)
GO FOR IT!
I just started on it, but gave up halfway. I accidentally cut of a tiny tiny piece of the edge of the green pcb. I just tried it, and it worked just fine though :) . Not sure if I'm going to finish it or not. Maybe I'm scraping the pcb all the way? I won't find out until I've chopped it off :p . I used a extremely thin and flexible two edged razor blade. It seemed to cut through the black rubber thingy pretty easily.Quote:
Originally Posted by NickS
But the worst thing was that my hands were shaking like a leaf... Didn't make it any easier lol :D