[Reborn] OC3D Delidded Ivy Bridge Club

FTLN

Active member
OC3D Delidded Ivy Bridge Club

Warning - Delidding your CPU can cause serious damage or failure if not done correctly. Your Warranty is now Void.



Please use this thread to share you delidding experience, frequencies and temps with other forum members


**Big Thanks to Genesius for his great photos**









Delidded Club Members
  • Genesius 3570k @ 4.7ghz / 1.225v Max Temps 62c with custom 6 litre loop
  • FTLN 3770K @ 4.7ghz / 1.39v Max Temps 64c with Corsair H100i (ambient 20c)
  • JamesRiley94 3770k @ 4.5ghz / 1.205v Max Temps 63c with Alphacool UT60 360mm,Laing D5, XSPC Raystorm CPU block, XSPC Raystorm 680 GPU Bloc
  • PapaSmurf6768 3570k @ 4.8ghz / 1.38v
  • Master&Puppet 3570k @ 4.6ghz / 1.355v Max Temps 63c Ambien Temp 17c
  • XANADV 3770k @ Awaiting Results



Tools needed
  • Single Edge Razor Blade
  • Credit Card
  • 2 Steady hands
  • Decent Thermal Paste
Optional Tools
  • Anti-Static Bracelet
  • Rubber Gloves
Reccomended Thermal Paste
Delidding Videos



http://www.youtube.com/watch?v=Ppql7sTmQIc http://www.youtube.com/watch?v=XXs0I5kuoX4
 
Last edited:
Hi all, I delidded my 3570k and along with FTLN thought I would share it here.
This is just the method I used, feel free to use your own or others, if I were to do this again I would use a single edged razor and not the double edged razor pictured.
The double edged razors not only increase the chance of slicing your hands off but also bend. They blade is so thin that if it catches the IHS it will bend and scratch the PCB as it is removed.

OK so now for the pictures
DSC_0121_zpsefbd4b7f.jpg

Assemble all of your tools. Take your time and be careful. I had an image on a screen in front of me the whole time to show me how far I could push the blade in without hitting the die – I suggest you do the same (I can’t post the picture I used here because I don’t own it, but do a google image search)
DSC_0124_zps49e80aaa.jpg

Start at the corners and then work the blade round. Push slowly, don’t saw! Try and keep the blade upwards if possible. The blade will get covered in glue so change blades often.
DSC_0125_zpsc62bae63.jpg

Once apart it will look like this.
DSC_0129_zpsd7e1e0ba.jpg

Clean that old TIM off using some decent TIM cleaner.
DSC_0130_zps2c2d6024.jpg

I removed the old glue from mine as well. Some people leave it as it stops the IHS moving around so much when you reseat it in the socket. The glue can be removed using a credit card or like me a Dremel. If using a Dremel, remember to use a polishing bit, get a slow setting, move the tip so that you don’t overheat one area, and remember its not a race, just go slow, hit that die and there's a very high chance that you’ve just created a very expensive keyring!
DSC_0139_zpsaa329b67.jpg

If you have nicked the board a bit like I did. Don’t panic – yet. Get some clear nail polish and dot it on the bare copper to insulate it. Get some TIM on the die, lid back on and put it in your PC to test it. If all is well, then pull it back out and carry on. If all is not well open many many beers!
DSC_0138_zpsaf1be488.jpg

Assemble your next bits. I used Coollaboratorys Liquid Pro as I have heard good things about it uses on the die.
DSC_0140_zpsc7db8ab4.jpg

Apply the Liquid Pro.
Don’t let the needle touch the die.
You actually need less than you can get to drip out of the needle. I very carefully squeezed a tiny bit out and had to carefully jolt the syringe so that it dripped off onto the die.
Be very careful with this stuff, its awesome but its very conductive and a pain to clean up so do it right first time, watch their application video and read the manual. (Im just going to say carefully one more time because I don’t think I said it enough ;-])
Once the drip is on move it around with the supplied cotton bud until it eventually starts spreading.
DSC_0141_zps4c13788f.jpg

TA –DA!!!!!!! Looks like Terminator!
DSC_0142_zpsb6c4d1ce.jpg

Place it in the socket
DSC_0144_zps41334eca.jpg


Hold the IHS in place on the die whilst you close the leaver and lock it into place.
Then add your TIM, im using MX-4
That’s it!


****************EDIT**********************************EDIT**********************************EDIT******************

So now to my temps. unfortunately this isnt going to be a fair representation as i have just upgraded for a H100 to a custom 6Ltr crazy cooling loop but here we go:

So, i just checked an old post to get these temps:

H100 stock fans on max @4.7Ghz 1.225v
IBT Very high 40 passes
77 87 82 81

Custom Loop, delidded @4.7Ghz 1.225v
IBT Very high 40 passes
55 60 59 62

Thats a difference of 25 degrees C!

Now i did that 40 pass run at the weekend, but just to get a screen shot and probably seeing as most people will do a 10 pass on very high i did that too and here it is....

47GhzScreenshot_zps2333fd11.jpg



Oh and just to mess about, one a 5Ghz as well (Personal note - Yeah yeah yeah 5GHz, wahoo! Time for many beers lol)
5GHzScreenshot_zps07aa8a78.jpg
 
Last edited:
man you guys are brave i wont be doing this anytime in the near future maybe in a few years when my intel warrenty runs out :D
 
Test set up:
Parts: Switch 810, Asus P8Z77-v Deluxe, GTX670, AX850, 8gb Mushkin Blackline 1600mhz
Cooling: Alphacool UT60 360mm,Laing D5, XSPC Raystorm CPU block, XSPC Raystorm 680 GPU Block
Fans on the radiator are Silverstone AP121 in push/pull.

I did a number of tests, at different overclocks, different fan speeds, and also disabled the GPU on the majority of tests so I could reach higher overclocks, to see what different the de-lid makes at the higher levels.
I ran Prime95 for half an hour (certain tests recorded at different stages between that too)


So to start of, at stock clocks (3.5ghz (Turbo 3.9ghz)) with core voltage at 1.025v - fans at 12v, pump at speed 5 - with GPU disabled:

TRqKJxZ.png


Then, at 4.5ghz, core voltage at 1.205v - fans at 5v, pump at speed 1 - GPU enabled:

qKNv7ml.png


Next, 4.6ghz, 1.325v, fans at 12v, with the pump at max speed - GPU disabled

6KrRzG8.png


Lastly, 4.8ghz, 1.4v, fans and pump at max speeds - GPU disabled:

xvi4wuG.png


Temperatures were measured with HWmonitor, and represent each individual core of the 3770k.


Ambient temperatures were roughly the same - don't have equipment to measure properly, however.

So, to conclude, the de-lid has definitely made a difference to temperatures. Whether it's worth the risk is another question though.
 
Last edited:
Delidded 3770K @ 4.5ghz / 1.28v Max Temps 78c with Corsair H60 (ambient 22c)

Using Crappy thermal paste from coolermaster, will post again when I receive my Liquid Ultra later this week.


hmveCDU.png


5jaLq4q.jpg
 
i always kinda figured the guy on youtube who did this was a troll and just trying to get folks to break their intels but seeing an oc3d crew guy doing this i know it's legit :p
Dont think i have a reason to do it until maybe custom loop time though

With a 'measley' h70 core on xigmatech fan push generic pull.
DtWydJp.jpg
 
i always kinda figured the guy on youtube who did this was a troll and just trying to get folks to break their intels but seeing an oc3d crew guy doing this i know it's legit :p
Dont think i have a reason to do it until maybe custom loop time though

With a 'measley' h70 core on xigmatech fan push generic pull.

You got a nice chip there mate, looks like you dont need to delidd whats your voltage ?
 
clickbios2 cpu vcore = +0.04000 i dunno, 2nd or 3rd option down.

During p95 test CPUZ is saying 1.176 volts under full load.
When i stop the test, the volts actually go up. 1.232 when it's about 1% load idle. .

I figure its a possibility im doing it wrong, or cpuz has an issue reading vcore. The first option seems more likely.
 
clickbios2 cpu vcore = +0.04000 i dunno, 2nd or 3rd option down.

During p95 test CPUZ is saying 1.176 volts under full load.
When i stop the test, the volts actually go up. 1.232 when it's about 1% load idle. .

I figure its a possibility im doing it wrong, or cpuz has an issue reading vcore. The first option seems more likely.


I dont know at board your using but you need to adjuct your vDroop to extreme or high. The reason your voltage drops under load is called vDroop and its meant to happen. Adjusting the vdroop control will cause the board to do this less. That means that you can probably reduce your voltage down to around 1.176 and still have it stable. I have to say that those are pretty good volts to be running 4.5ghz at.

What do you use to stress test?
Have you made sure you dont get WHEA errors (i presume your using windows)

..................................................................................................................................

Intel should have just soldered the dies to the IHS in the first place. I have no idea why they didnt, maybe someone can shed some light.

All i know is ive dropped 25 degrees and i can now hit 5Ghz without even getting to 70 degrees.

I think its really a risk & reward thing. If your always going to wait for a time when you need to do it then its never going to happen. For me it was a case of, i didnt need to overclock, but i did and now i didnt need to delid, but doing so has paid off.

Ultimately the choice is all yours but i just think its amazing what a difference it makes, just dont mess it up!
 
Last edited:
clickbios2 cpu vcore = +0.04000 i dunno, 2nd or 3rd option down.

During p95 test CPUZ is saying 1.176 volts under full load.
When i stop the test, the volts actually go up. 1.232 when it's about 1% load idle. .

I figure its a possibility im doing it wrong, or cpuz has an issue reading vcore. The first option seems more likely.


Defo tighten up that vdroop and lower volts to 1.176 or 1.180 as you may still have vdroop even when you got it fully tightened.
 
Some new tests, looks like im at the limit of the H60 until i receive my liquid ultra.. Max Temps 83c over a 45 min occt / linpack run @ 4.6ghz with 1.312v

4_6.png
 
Some cracking result's there for the people who have done this.

I was tempted to buy a 2nd hand Ivy Bridge CPU and try this but I just cannot bring myself to it as I am quite heavy handed so potentially would break the cpu.

Looking forward to seeing some more results though.
 
Some cracking result's there for the people who have done this.

I was tempted to buy a 2nd hand Ivy Bridge CPU and try this but I just cannot bring myself to it as I am quite heavy handed so potentially would break the cpu.

Looking forward to seeing some more results though.


If your not sure on your abilities then i probably wouldnt do it, but if you want to try there are a whole load of chips you could practice on before hand.

let me find the list..................

here you go:

(S-478) Pentium 4 HT (Northwood "A" and "B" Core)
(S-478) Celeron
(S-775) Celeron
Celeron 420
Celeron 430
Celeron 440
AMD Athlon 64 3200+
AMD Athlon 64 3700+
AMD Athlon 64 3800+ (Venice core)

Dual Cores

AMD X2 5000+ BE (Brisbane core)
Celeron Dual Core E1200
Celeron Dual Core E1400
Pentium Dual Core E2140
Pentium Dual Core E2160
Pentium Dual Core E2180
Pentium Dual Core E2200
Pentium Dual Core E2210
Pentium Dual Core E2220
Pentium Dual Core E6300
Core 2 Duo E4300
Core 2 Duo E4400
Core 2 Duo E4500
Core 2 Duo E4600
Core 2 Duo E6300 (L2 stepping)
Core 2 Duo E6400 (L2 stepping)
Core 2 Duo E7200
Core 2 Duo E7300
Core 2 Duo E7400
Core 2 Duo E7500
Core 2 Duo E7600

All these chips dont have soldered dies so should be good for some practice delidding and some a pretty cheap on ebay
 
Liquid Ultra Results

4.5ghz 1.28v Max temps = 59c = 19c cooler delidded with Liquid Ultra

AMAZING STUFF

Ok4vPzn.jpg


UgQkXqM.png
 
Its pretty cool stuff. How did you get on with applying it?

Your temperatures across the cores are way more consistent than mine.

Although ive had a 25 degree drop in temp from the hottest core to hottest core, my core 1 is still 8-10 degrees cooler than the other 3. Maybe I need to reseat my cpu block I apply some more liquid pro? If I can get all my cores to the same temp as core one then id be looking at a drop of over 30 degrees, and that would make me a very happy chappy!
 
I used the half of the syringe on the die and under the ihs / and the other half on top of the ihs and then seated the h60 on top... i spread it out with little paint brushes that they give you with the pack. Before all that cleaned using the tim cleaner that comes with the syringe and gave the ihs a good scrub with the scrubber that is also included..

I have no liquid ultra left now... They give you hardly anything, lol but well impressed with the results..
 
I used the half of the syringe on the die and under the ihs / and the other half on top of the ihs and then seated the h60 on top... i spread it out with little paint brushes that they give you with the pack. Before all that cleaned using the tim cleaner that comes with the syringe and gave the ihs a good scrub with the scrubber that is also included..

I have no liquid ultra left now... They give you hardly anything, lol but well impressed with the results..


with the liquid pro I hardly used any, I think I could do another 20 dies!

those temps are pretty nice at that voltage! I didn't use the pro between the IHS and my cpu block because I heard its a pain to remove and clean up.

Glad you've got some wicked temps though. is that going to be your new 24/7 OC?
 
Yep, gonna be staying at 4.7 for the moment, until I get a better power supply. At the mooment Im running some crappy no name chinese thing 480w
 
When it comes to PSU's im not sure about the differences between the topend ones and the standard ones. For sure, if its 750w and £30 then its not 750w. but I have a CX750 corsair and I got that because of budget, however im yet to see what having a GS, or AX series PSU would do to benefit me. Maybe Tom will be able to enlighten me once he gets that PSU tester up and running.
 
Back
Top