Delid 4930K

I am planning on delidding my 4930K. The IHS is soldered on, so I can either desolder it, sand through it, or destroy the CPU with a vice and hammer or razor blades. I have picked to sand through the IHS because I am not an idiot and I have no soldering equipment. Correct me if I am wrong, but it should work to grind the IHS down to the die with low-grit sand paper and then use a razor to pry the remains of the IHS off. My plan for handling static electricity is to tape the contacts on the bottom of the CPU. Will tape be enough to handle the static or will I need to do something else to protect the CPU?

why?

Why are you removing the IHS on a soldered part? It's not some 4670k where there's just some crappy thermal paste... you're not going to be able to do anything to improve thermal conductivity over solder so why are you risking damaging your CPU for this?

This thread got hostile REAL quick :P

I want to run the CPU bare-die. Blame it on my PowerPC use, there is no need to delid a PowerPC CPU because there no lid to delid! I've never liked heat spreaders / heat boosters because of that. Heat boosters, in my opinion, are an incredibly stupid invention; all they do is increase the amount off material that heat must travel through in order to be dissipated.

Thank you for responding like a normal person, unlike the immature bastards posting junk here. I am planning on removing the IHS to cool the die of the processor directly. I am not used to using X86 processors; I have always cooled the die directly. An IHS is just more material that heat must travel through to get to the waterblock; it makes zero sense to have an IHS. Are you telling me that this: Die -> Solder -> IHS -> TIM -> Waterblock -> Water will be less conductive than: Die -> TIM -> Wateblock -> Water? Heck, on my dual PowerPC 7448 rig, I run water directly over and on the die like this: Die -> Water

OP what do you expect? Delid cpu that doesnt need to be and wonder why you get troll and fun meme responses. You'll get even worse responses on oc net, im on that forum as well so it'll be fun to watch.

I'll put this thread down as a brain fart on your behalf and move on.

then again there was this other thread by the OP... perhaps this is a trend.

https://teksyndicate.com/forum/cpu/why-do-all-these-people-mess-cpu-vcore/166414

Tape doesn't sound like a good idea. Either way, that is a ballsy move. I thought you were moving towards watercooling? I would wait and see what temps you get with that first before getting this extreme.

Yea, it's true that you'll get better results directly from die because of less layers to transfer heat through... but also consider 1.) The highly likely event of damaging this CPU and 2.) Heatsinks and their fastening options are designed with the added height of an IHS, so you'll likely get very little pressure on the CPU just depending on how your waterblock fastens to the mobo.

yeah, deliding a soldered IHS is pointless. you'd get like 1/10th of a degree, at the most.

I personally would like to see a blog progress of what you intend to do, regardless of the results it would be better then this...

if he wants to... but it's not really a good idea, and we're not gonna help him. you really would need extremely precise sanding instruments and such.

A 4670k or 4770k would be excellent candidates for this, as the risk is much lower and the reward can be a good deal higher just depending on the quality of the job they did applying the IHS in the first place.

I always questioned why AMD or Intel wouldn't release a IHS-free CPU for people that do custom cooling and more advanced users with a limited warranty or something along those lines. You'd only gain a degree or two, but there are plenty of users like Rabidz that like doing things like this in order to try to squeeze out that extra 1-2mhz which is why BS thermal solutions like Diamond-infused-watery-bullshit-we-call-thermal-paste sell so well.

Why would they release a IHS free cpu? Extreme oc'ers use dice, phase or ln2. Plus it protects the cpu and allows even spread of surface pressure to the heatsink.

Yeah a haswell/ivy chip benefits from delidding cause it isnt soldered, this is nothing new and is completely removed from the topic of what the OP is wanting to do. Of which makes no logical sense whatsoever no matter how you look at it and smells of being another troll thread.

I'm not sure how much you are going to gain and you may end up destroying the CPU... however, do you plan to document it with images? I'd love to see this.

And to the trolls:

WTF is wrong with with some of you? ... go back to 4chan and YouTube. I came here for a discussion about the CPU. I shouldn't have to deal with this BS on my own damned site. Seriously, keep the trolling and the casual stuff in appropriate threads... and please, if the OP says asks you to chill... do so.

Commissar or other mods... Send warnings to trolls and spammers. 

 

The youtube commenting side of the teksyndicate community has come out.

This is indeed quite brutal and immature.

I was actually just talking about that on another thread

https://teksyndicate.com/forum/inboxexe/forum-has-some-issues/170466

 

I mean regardless of this seemingly being a bad idea, if the guy uses PPC and is used to IHS-less cpus then it's his part to do with as he wishes.

Logan. We, The supposed "Trolls" dont do something like this for no reason. We have along with many of the Mods that you and pistol have nominated to look after this forum have been battling miss information for some time. While we do not have issues with people trying new things. The obvious and general consensus was that it would not work and that the risk to the CPU but as its his hardware and he can do whatever he wants. The issue was not this thread but the person himself.

I know you are busy and of tine dont get time to see the politics that go on in the forum but its there and the largest issue that some of us members have is misinformation. Some people on this forum provide false information as factual or even just ideas as fact. Rabidz is someone who consistently states his ideas as factual, provides no evidence and refuses to accept any information used as rebuttal. 

What we did here was too far, Not all of us agree but it was. However the other issue as stated is false information. It has happened in the past and these people have slipped up and have been banned. But sometimes there is not a lot what mods can do and well, while most issues are sorted. Some Toxic people can not be removed because their not breaking the Rules.

This highlights an issue with the community. Mods do not have enough power to deal with such situations. It got to the point where a Mod supported what we did. Another issue is the lack of mods at this time of day. We were trying to contact mods but very little success. 

If you wish to talk about it. I and some of us are in the Mumble. We dont want this forum to be 4Chan but we were sick of the low quality posts and reverted to a crude method to combat a problem. If a selection of mostly long standing members on this forum did this. Then there is more to the story.

1 Like

rabidz7's idea of having no IHS is nothing new. Imagine if CPUs, motherboards, and heatsinks fit perfectly without a processor's heat spreader. Thermal energy would move from a processor's die, through thermal paste, and a heatsink. This makes transferring energy more efficient, causing a lower core temperature. If this were true, then an overclocked Haswell 4770K that reaches a max core temp of 100C at full load would be about 20C cooler. Therefore that 4770K could be more stable and possibly have a higher overclock.

Yes however the idea of doing it with a system that has a soldered IHS is not a good idea. With a 4770k that uses a thermal paste is commonly done but with a 4930k, the shroud is soldered on. There is very little performance increase if you were to do that not to mention maybe a 90% chance of killing your CPU. And even then you could crush the chip so easily aswel.

I have actuly been thinking though. If say intel a corsair? partnered up and fused a H110 or another high end cooler to the chip itself, What sort of cooling could you get? The issue is the getting heat from the chip obviously but if you could get the surface of the chip going to a small heatsink surrounded in coolent. You need some sort of heatsink but it could be interesting.