Printable View
Succesfully removed the IHS from my Q6600, and removed all the Intel TIM from the dies. However, my temps aren't any cooler, and in fact are hotter. My original thread is here. Anyone have any ideas on how to get my temps cooler? Thinking I'm gonna have to lap both dies.
I just re-read something yesterday on a easy way to kill a core 2 duo by doing this so I'd say be careful. I'm curious though what cooler do you have/ is it lapped or how smooth? Before you had surface area spreading out some heat now there is just two pieces. With my experiencing lapping smooth as possible is the way to go that and a good mount. But I'm curious what cooler your using since you head the temps have gone up? I'll be lapping a new cooler of mine and a e4300 soon but lol I wont be taking off the TIM I dont have the extra cash to shrug it off haha
You have some serious cojones to do that to bare die.
I bet some people will destroy their procs attempting that after seeing you do it!
Lol, well I read the success rate is only 10%, but I'm pretty sure I have a guaranteed way to get a successful run everytime. And I'm using an Apogee GT waterblock. I'm gonna try some 3000 grit sandpaper on the dies to see if I can get them both level with each other.Quote:
Originally Posted by AngryArtichoke
you sir...have big balls of steel!!
^ seconded that!
Lapping the die is a real easy way to break something, but if you do it right and don't chip any corners or dislodge any of the bumps on the top of the die from the package you are removing the silicon which is probably the material with the lowest thermal conductivity in your system.
Hard to tell from the pics, but is the left over rim from the IHS interfering with your water block making good contact?
Already got it a lot lower since those pictures were taken. It's as far down as I can go without damaging the PCB. I just did a test, and the dies have plenty of clearance over it :up:Quote:
Originally Posted by The_Barnacle
OK I figured as much, but just wanted to make sure it wasn't something silly like that. On the same note have you checked if the MB socket retention thing is above the die height? I've never messed with a naked Core2 so I'm not sure how tricky the mechanics of it all might be.
If you go as far as lapping th actual dies, definitely post pics, cause that's some crazy crap...
I removed the lock down part of the CPU socket, as with that still in, the dies sit below the top and the waterblock doesn't make contact at all. So I have to install my CPU with my mobo lying down, otherwise the chip falls out lol.
You are sanding the CPU dies! :O
I wonder if the waterblock isn't sitting evenly on the dies. Do you have one of those adhesive foam stabilizers around the dies so the block puts even pressure on exposed dies?
http://www.frozencpu.com/products/52...rs_5-pack.html
These will prevent the block from sitting at an angle on the dies and damaging something. They're for northbridges with exposed dies technically, but I think they'd work for CPUs too.
Is the waterblock making good contact with the dies? Are you sure the waterblock isn't resting on the retension bracket for the socket?
this guy needs a Xtreme title
and a pychologist :rofl: only kidding "let nothing get in the way of your die and the heatsink":eek:Quote:
Originally Posted by alexlyc
Quote:
Originally Posted by Glow9
I have a Q6600 that was lapped by someone who really knows how to get it smooth. Some kind of granite stone or soemthing, all i know is my temps went down and it was noticible and I was glad to have had it done
I also had a cpu with the IHS removed and i would have had to take off lock down plate of the 775 socket to make contact. plus you have to worry about the corners of the die, just like on some northbridges, so i jsut added tim to the die's and put hte IHS back on.
This is what I was thinking. Are you absolutely positive that there is good contact between the waterblock and the bare dies? I'm thinking that in removing the IHS, there might be a bit of clearance between the dies and the waterblock, because even a slight clearance can kill temperatures.Quote:
Originally Posted by Polizei
Yes, I removed the locking mechanism so that way there's no part of the socket higher then the 2 CPU dies. I'm not entirely sure how that styrofoam setup is supposed to do?
Actually fellas, from my experience removing the IHS is the "scariest" part of the whole effort. The lapping a die(s) is the easier part if you have the proper equipment...and time . I almost think it is mandatory to lap the die on dual dies, as they just aren't exactly the same mounting on the pcb and you don't get good heatsink contact without the lap at all. Plus I really don't find it is very beneficial on recent Intel CPU's and the "solder" TIM. The older AthlonX2's made sense as the TIM was a paste and it's characteristic could change over the life of the chip.
Granite is good, but a piece of glass is pretty good substitiute. I'd ditch the sandpaper and use compounds it worked better for me. Mad Mikee here has a pretty decent tutorial for using calcium alum and zirc ox if you search the forum. Far superior results. It takes hours to do it right. I just listen to music and do an hour here, and and hour there. It's not very dangerous if you go slow. A few swirls, rotate the chip 45 degrees, repeat, and change out compound once it gets contaminated.
I'd think you would be good with the GT mounting, but mounting is where alot of people run into mounting problems because the clearances have changed, or water blocks that are bowed at the center, etc, etc. That could be your problem but looking at the "patterns" you have left on the chip you are not getting good contact as those dies look very different in height. So I'm leading towards a good lapping helping(caveat is that if you do a poor lapping job you are just going to make it worse.
Yea, I'm working on that :shakes:Quote:
Originally Posted by aicjofs
I have to say the lapping of my q6600 made me a true believer in the fact that the increased contact DOES cut some degrees off the heat load. I would think that would
matter the most under load, when you need it the most.
So now I'm a proud lapper.
just ask my wife :yepp:
Regards
SystemViper
Speaking of lapping, here's what it looked like before I exposed it :D
Very nice looking.Quote:
Speaking of lapping, here's what it looked like before I exposed it
I hope you didn't think I was knocking the look of your die, I know you are working on it!! :p: You look like you did a fine job on the IHS so I'm sure the naked cores will turn out fine...slow and patient and I bet you'll see some better temps.
haha, funny guyQuote:
So now I'm a proud lapper.
just ask my wife
No offense taken, I came here asking for advice so any given is appreciated.
Lapping your dies... Whoa :shocked:Quote:
Originally Posted by ghost_recon88
You do know why it's called a heatspreader right? Heat-spreader... Had the same back in the Northwood days. Lapped IHS = small decrease in temps
No IHS = higher temps. Maybe a silver HS with AS would give a gain...
But you have got biiiig cohones my friend :D
Someone on another forum modded his Apogee GT to provide more mounting pressure, showed a significant decrease in temps because his block was making better contact with the CPU.
http://www.xcpus.com/forums/case-mod...-pictures.html
I'll ask him exactly what he did and post back later.
Looks like instead of using capped bolts, he just used open bolts so he could keep tightening them down :up:Quote:
Originally Posted by YMAA
If I can't get this quad cooled tonight, I'm just gonna sell it on eBay or something :(
Just thought I would give an update on this. Finally got both dies flat. I'll test it shortly, but I just got in several low VID Q660s I have to test (one is @ 4GHz with 1.400 volts currently :rocker:)
Awesome pictures, and if that doesn't drop your temperatures, I don't know what will.Quote:
Originally Posted by ghost_recon88
You know you want to send that Q6600 my way. :up:Quote:
Originally Posted by ghost_recon88
2nd pic = awesome. Great job lapping :up: and let us know temps :cool: