Disclaimer delidding is a very dangerous process! Only delid if you truly believe you are ready after reading and watching and viewing all the necessary materials that apply to delidding and getting all necessary equipment ready as well!
MOD NOTE: Delidding your CPU does void its warranty. If you RMA the chip it is classed as RMA fraud, please do not discuss such actions in this thread or anywhere on OCN.
Second off is this!
There is a new way to delid! so go share some love at these guys's home and come join the Official Club here!
Hey guys! Two days ago I posted saying "YOLO, going in the vice now". Within 5 minutes of posting that I was done.
I have some tips for and info on the vice method for people who have never done it, or are still afraid of this method. For the record, I successfully delided my 3770K with a THICK box cutting razor. There was not a scratch on the PCB, and it took me all of 5 minutes. I'll admit I was careless, but I'm damn good with my hands.
Onto the vice method!
For those that don't want to buy a Vice, or think that the razor method is cheaper and safer, THINK AGAIN! Two days ago I bought a 4" Drill Press Vice from Lowe's for $19. The Part number is BV-DP40. This is probably the best vice for the job. And for all you people who don't want to drop $20 on a vice just to delid, Lowe's has a 90 day return policy! I don't feel guilty about returning the vice because: 1. The box is in perfect condition, 2. The vice jaws were covered in blue painters tape to protect both the jaws, and my IHS, and 3. The vice has not a scratch on it, and was never bolted to a bench/ table. So after you're done, return it! That's a free delid folks.
Here's the vice I used. It's 30 on amazon, but only $19 at Lowe's! And at Lowe's you have 90 days to return it.
Onward. Here's how I positioned the vice. The operation was completed on my kitchen counter. I have a board underneath the vice to protect my counter, and a board held vertically behind the vice that serves two purposes. 1: To protect my kitchen counter molding, and 2: To give the vice a solid backing, since all of the force will be directed that way.
With this setup, the vice will not move, and will be just as secure as if it were bolted to a work bench.
Now for the positioning of the IHS in the vice. I haven't seen this discussed much so I'll add my suggestion for how it should be placed. For Haswell, secure the IHS in the vice so that the VRM side is facing you, meaning you will be hitting the VRM side with your block.
Why? Because this minimizes the risk of both the VRM's and the Die itself from hitting anything after it is free of the IHS. Take a look in my pictures at the layout of the 4770k after I crack it open. You'll see that the side of the chip which has the most clearance between the die and the edge of the PCB or IHS is the side opposite that of the VRM side. You want to hit the VRM side because once your chip is free it will move forward, and when oriented this way your VRM's and die are furthest from the threats they are approaching. For Ivy, just make sure your vice is clamping the flanged sides of the IHS, this will allow the most safety clearance for the die. Phew that was a lengthy explanation for such a simple concept. Here's an example of how a 4770K/4670K should be clamped.
Now for what material your hitting block should be made of. PINE. That's it, no other woods, don't even think about it. Pine is soft enough to not damage the PCB, yet strong enough to break the seal easily. Use a section of 2x4 (Two by Four) with nice square edges. I'd recommend a length no shorter than 8", and no longer than 14". The shorter the block, the less force it can absorb meaning the PCB will take more brute force. Too long, and it may become difficult to hold and control.
Now hold the block, with its clean flat and square side, directly parallel to the PCB. Apply pressure towards the PCB with your hand, this eliminates the chance of the block slipping off, or a slapping effect on the PCB which we DO NOT WANT.
Hit the thing, hard. However technique is also involved in this stage. When you're hitting a nail, you swing hard and follow through; the point being to drive the nail with the follow through of your hammer's blow. This isn't the type of swing you want. We want concise yet powerful taps, no follow through. If you hit the PCB with a full blown nail driving follow through, you'll send it flying. Bad. So to reiterate: Strong, yet controlled and concise hits. Your PCB should be free in a few hits, and it shouldn't even go flying if you follow this technique. When I did mine, the PCB broke free and rested right on top on the IHS, no drama whatsoever.
After the PCB was free I inspected the edge that was hit. It was immaculate, not even a micro-scratch or dent to be seen. My block of wood (2x4 cut in half laying in my garage, PINE) took all the force. You can see here how the block fared afterwards. This proves that the PCB is much stronger than pine.
As for cleaning off the glue, I've heard some people complain saying it's hard to get off. >_> Get real. Use the corner of a credit card and scrape it off. Scrape as hard as you want, I promise you the plastic your card is made of WILL NOT scratch the PCB, even if you tried.
In conclusion, I've done both methods. Use a vice. It's safer, its faster, and it can even be cheaper if you have no qualms about returning your $20 vice.
Hope this post wan't too long, and that it helps some of you out. Happy deliding. I'll post my results and validation here in a few minutes.
YOUR HASWELL CAN BE DELIDDED AS WELL!!
So the main reason for this is that there is a certain way to delid with a blade on this one.. with the caps under the IHS now... Make sure to cover the caps with either LET (Liquid Eletrical Tape) or clear nail polish
As you can see the caps are on the same side as the line of gold pads... so you NEED to start the cut on the opposite corner to keep those caps safe for the initial cut.
Also read this as well To get a good idea of how good of a cpu you have try a low over clock like 4.5GHZand see how low of voltage you can get it stable at. 1.2v and below without delidding for 4.5GHZ
is usually a decent chip and when you delid the performance increases.4.5ghz @ 1.2V vcore or lower is prolly a good chip,but the oc needs to be stable, not a quick run, like for a validation, then it could do 1.2V vcore or less mostly
There will be a few things first off that I need to get out of the way for this. I want this club to not only show what we have done through our time and effort but to be a helpful club to others and direct them to the right people on how to properly delid and get the information out their now so no one makes any mistakes! Delidding information and guides!(Click to show)
First off here are a few guides that we have been looking at on how to delid our Ivy Bridge processor, these will hopefully help you in this decision!: Warning: Spoiler!(Click to show)
Very interesting results from a very good delid test of deliding and TIMs on an anandtech forum thread where the OP Idontcare found that the IB temp problem was not from the intell TIM at all, but from the gap between the die and IHS.
- if interested, see post #570 on link below: http://forums.anandtech.com/showthread.php?t=2261855&page=23
And the results?
And if we remove the paper shim and drop that IHS down onto the CPU (not perfectly zero of course, there is still some NT-H1 CPU TIM there after all) reducing the gap to as close to zero as possible then we get the "c" cases...and the temperatures show the expected fantastic drops we have all come to expect from delidding our Ivy Bridge chips.
Conclusion: The Intel stock CPU TIM is not the reason Ivy Bridge's run hot, and replacing the Intel stock CPU TIM is not the reason a delidded Ivy Bridge runs so much cooler - the benefits of delidding are entirely due to the resultant reduction in gap height between the CPU silicon die and the underside of the IHS
>>>>> Surprising also was his finding that direct die to HFS did not help to reduce temps as much as he had thought they would.
see #583: http://forums.anandtech.com/showthread.php?t=2261855&page=24
The take home message there is replacing the CPU TIM doesn't really provide any benefit. Reducing the gap between the CPU and the IHS does. And removing the IHS entirely doesn't really provide much benefit either. Unless you have a custom loop and everything is set up just right then your results can really shine. but it takes time and practice for more information ask SonDa5.
And that stands to reason. The heat is going to flow through the copper of the IHS the same as it does through the copper of the H100 waterblock.
But if there is a thick pad of thermal paste in the way, as was the case with my 3770k at time of purchase, then it doesn't really matter how good the TIM itself is (unless it too is made of metal) because the mere presence of that thick pad of thermal paste becomes the weakest link in the thermal conductivity equation.
Once you eliminate the bottleneck that is the gap between the IHS and the CPU, or if you happen to end up with an Ivy Bridge CPU which doesn't have much of a gap to begin with (Yuriman ), then you have pretty much optimized the system at that point regardless which CPU TIM of choice you employ and regardless whether or not you leave the IHS in place.
Now the choice of CPU TIM still plays a role in terms of the robustness in maintaining those nice low temperatures. If the so-called "pump out" effect is real then we can expect it to bite us unless we choose a substitute CPU TIM that is designed to avoid such thermo-mechanical effects.
I haven't really got into testing that part out yet, but I expect IC Diamond and the metal TIMs like Liquid Ultra to be key there.
So I'll explain what this means quick as well. what that W/mk means is this. It's the absorbance and moving of heat that the TIM or Thermal Interface Material can move from place to place. As you see Liquid Pro is at 82 and AS5 is at basically 9. So this means that you get 9 times the heat movement by using Liquid Pro over AS5.
Example pictures of Liquid Pro/Ultra applications:
If you need a better explanation of this just PM myself.
Now for the requirements to join the club are as follows. Post a picture of you delidded chip and write your OCN name on a piece or something of the like. For those already running delidded chips and don't feel like taking them out and re-installing I understand so take a picture that you most likely took when you delidded it and use what every program you'd like to put your name somewhere on the picture.
For the layout of the chart and what I want for submissions to be accepted.
OC after delid:
CPU-Z validation of max OC:
For the submitting process of your temps use the following rules.
1. Get to 4.5 Ghz (if you are having issues let us know on here and we will help you achieve this!)
2. Run Intel Burn Test in standard setup (this will always appear as is when you double click or start the program)
3. Run for 10 times (Also set as the default when program is started)
4. Report your HIGHEST temp from your four cores.
5. You can also show your temp drops but ONLY from the HOTTEST core to the HOTTEST core. (if not available refer to #4)
6. Ohh! If you have Lapped (sanded) your IHS please provide this information also. (Even what grains you used will be very welcomed) What is Lapping? Click the link!(Click to show)
Thanks again! This may be a ridiculous amount of info to be provided to join, but this is for research to see what we can show Intel eventually and how badly they messed up. (Yes I'm serious abort this part)
This is also for people looking into Delidding and seeing just how crazy their temps can decrease by Delidding.
one tip after you done delidding and putting your processor back,
we noticed, that if you put down the bracket again, the IHS will slide a bit forward,
it will do so whatever tim you use,
except if you use a new/other adhesive/glue also
The trick is to start a bit more to the back when placing the IHS back, i think about 0.5- 1mm will do..
lower the load plate until it sits loosely on the top of the CPU package, check if its all good
Now lower it careful until you can snap it under
the stub holder on the side of the socket.
Lowering the lever takes a bit of force because you are compressing the load plate,
which in turn forces the CPU down tightly on the landing pins.
i held it with my finger in the middle if the IHS, but dont forget to clean it again
after youre done(fingerprints)
really, i tried to move it with a screwdriver afterwards, but could not move it,
so you have to get it right, before the lever is under the stub..
I have a vision of a brilliant master mind Ivy Bridge Intel engineer smiling when an IB is delidded in the pursuit of performance. Having access to the bare die of IB is a special gift and if the IB IHS would have used fluxless solder like SB we would not have been given the opportunity to easily access the bare die of IB.
The benefits to have freedom to cool the bare die of IB may be greater than the design of the SB IHS with fluxless solder.
This is real modding and it is very rewarding.
Our Deliddings Crews very own approved [Official] delidding video. All thanks to Totally Dubbed
<div class="post-sig post-sig-limit shazam usersig-click"><div class="reparse-sig-lineheight"><div style="text-align:center;"><span></span><div style="text-align:center;"><a href="https://www.overclock.net/t/1313179/delidded-ivy-bridge-club"><span><span style="font-size: 12px;"><img alt="skull.gif" src="http://files.overclock.net/images/smilies/skull.gif"><b>Captain of the [Official] Delidded Crewmen</b> <img alt="skull.gif" src="http://files.overclock.net/images/smilies/skull.gif"></span></span></a></div><span></span><div style="text-align:center;"><span></span><div style="text-align:center;"><a href="https://www.overclock.net/t/1464058/build-log-sr-2-folding-ultimate-rig-2013"><span><span style="font-size: 12px;"><img alt="post-flame-small.gif" src="http://files.overclock.net/images/smilies/post-flame-small.gif"><b>SR-2 Folding/Ultimate Rig 2013</b> <img alt="post-flame-small.gif" src="http://files.overclock.net/images/smilies/post-flame-small.gif"></span></span></a></div><span></span><a class="spoiler-link H-spoiler-toggle" href="#"><strong>Overclocks etc, etc</strong> <span class="spoiler-help">(Click to show)</span></a><div class="spoiler-hidden"><span></span><div style="text-align:center;"><img alt="post-flame-small.gif" src="http://files.overclock.net/images/smilies/post-flame-small.gif"><b>_.=2 GHz Overclock Club=._</b><img alt="post-flame-small.gif" src="http://files.overclock.net/images/smilies/post-flame-small.gif"></div><span></span><div style="text-align:center;"><img alt="post-flame-small.gif" src="http://files.overclock.net/images/smilies/post-flame-small.gif"><span> 5 GHz Overclock Club </span><b>3770K @ 5.5 Ghz</b><span> </span><img alt="post-flame-small.gif" src="http://files.overclock.net/images/smilies/post-flame-small.gif"></div><span></span></div><span></span><span style="font-size:12px;"><br></span><span></span><div style="text-align:center;"><b><br>We're tuning things, and we'll be right back<br>All the best,<br>Overclock.net</b></div><span></span></div><span></span></div></div></div>
OCN name: Conquistador SW
On-die TIM: Coollaboratory Liquid Pro
IHS TIM: MX4
Mhz gained: +100MHz (for now)
OC after delid: I'm at 4.7GHz http://valid.canardpc.com/show_oc.php?id=2533967 (higher overclock coming when I have time)
Temp drops: On average dropped ~25C with the hottest core dropping 30C