Delidding gone wrong..

[dighost 25]

My one day old 3770k is dead. I purchased the cpu, got it home and proceeded to "delid" it as per the many instructions here on the website and on Youtube. I used to be an A&P tech at an airline and have always been careful and deliberate with repairs-- from fixing the lawn mower to my computers. So I didn't just tear at the cpu like a hack..

 

The process was as follows. First I used a grounding strap on myself, then marked ultra thin razor blades so they would only go in just enough to penetrate the adhesive. I slowly started on each corner, then the sides. With the lid easily removed I noticed a nick near one of the top corners... and by nick it was the size of a pin head and no deeper than a small surface scratch. Nothing bad enough to make me worry, so I just sealed it to make sure it wouldn't ground on the lid by chance. Got the computer build done and went to boot...and nothing-- just the sound of the system fans. I suddenly got this bad feeling in my stomach and hoped that maybe it was a memory chip or even a bent pin on the motherboard. Still no go. I popped in my "old" i7 2600 and it booted right away. That confirmed it all.

 

So I won't advise anyone not to delid because there are benefits, but make sure you know that even when you are careful and follow directions, at the end of the day it's still a gamble.

 

post. I examined the razor blade and although you can barely see it, there was a super small "crook" on the edge that likely caused the nick.

[Chris198 1]

Fortunately I didn´t use a soo "ultra thin" razor blade....

[Great Ozzie 101]

My one day old 3770k is dead.

 

Man, very sorry to hear that. I'm an A&P too (worked in GA)... and I was ultra-careful during the de-lid. Thing was, I think I got a bit careless during the scraping off of the gasket material. My cpu still worked but only was able to use single channel mode for the memory.

 

Told the wife my sob story (sad face and all)... she was like, "you can order another one but DON'T do anything too it" :Big Grin:

 

(Bimmer feel free to give me grief now :LMAO:)

[ryanbatc 12,797]

I admire the people who do this delidding stuff but I don't have the nuts to do it... especially when I can only get a new PC every 3-4 years....

[RWFeldman 159]

Always figure Ivy Bridges wasn't worth the risk delidding them. Sandy Bridge I7 2700K.

[ryanbatc 12,797]

Granted I'm an idiot for plugging in the USB front panel connectors in the wrong polarity to my mobo and shorting them out when using an external HDD for the first time with my old build.... :ph34r:

 

Turned that external hdd into an internal hdd real quick LOL! And learned to buy a freeking mobo that has bigger than those stupid 1mm connectors for front panel, speaker, hdd lights etc...

[Noel 4,486]

I have a hard time understanding why Intel engineers did what they did here. Is this because total heat content w/ IB is actually matched to cooling needs, i.e., the high 'temps' you see do not translate to 'heat content'. That is one theoretical explanation. Another is that they needed to artificially restrict its performance so they could role out more performance w/ less development effort for future releases like Haswell. Perhaps Haswell will really be IB w/ some new microarchitecture PLUS superior thermal interface so that delidding no longer is necessary. IOW, I can't see how this was not by design. They are engineers extraordinaire!

[TechguyMaxC 516]

Sorry to hear about your bad experience with de-lidding. I'm curious as to the location and size of the scratch, also what you chose to seal it with. My first de-lid job (my own 3770k) has some scratches on it which are quite noticeable but I left them alone and have no problems with my chip. I'm inclined to think your sealant may be the culprit here, remember, everything under the package's top layer conducts electricity so if you interrupt that flow in the slightest you could be shorting out your chip.

 

All that being said, in my experience systems with dead CPUs won't even power on so I would be double and triple checking my configuration were I you. I understand your 2600k works fine, maybe it's not a "dead CPU" but rather one that is simply inhibited in some way. Did you check the pin pads on the bottom of the CPU for any foreign substance such as thermal interface material or your sealant? Worth doing IMHO, I've seen it happen many times.

 

Good luck, I hope you haven't lost a chip!

[Chris198 1]

Man, very sorry to hear that. I'm an A&P too (worked in GA)... and I was ultra-care during the de-lid. Thing was, I think I got a bit careless during the scraping off of the gasket material. My cpu still worked but only was able to use single channel mode for the memory.

 

Told the wife my sob story (sad face and all)... she was like, "you can order another one but DON'T do anything too it" :Big Grin:

 

(Bimmer feel free to give me grief now :LMAO:)

 

And you bought another Ivy just to run it in dual-channel mode ?

[Nuno Pinto 1,394]

They are engineers extraordinaire!

 

That's why they fail on simple things

[Noel 4,486]

That's why they fail on simple things

 

Quaint thought, but I don't think so ;o) This image serendipitously illustrates the argument for heat content versus temp. I say serendipitously because the two colored objects aren't a bad representation of SB v IB! With all of the clamor over IBs prodigious peak operating temperatures I wonder if there has been much evidence that these temps constrain performance as much as people might assume is the case, ie their heat content may be quite comparable in the end.

 

 

[PatrickM 2]

What about the Mobo? Is it Ivy compatible? Which chip do you use?

[dighost 25]

Thanks for the sympathy guys! I went and picked up another chip this morning and it is working just fine. And yes, it is lidded. Lesson learned!!

 

TechguyMaxC-- the nick is super small about the size of a pin ###### and is located near the outer top corner of the chip.

[TechguyMaxC 516]

Quaint thought, but I don't think so ;o) This image serendipitously illustrates the argument for heat content versus temp. I say serendipitously because the two colored objects aren't a bad representation of SB v IB! With all of the clamor over IBs prodigious peak operating temperatures I wonder if there has been much evidence that these temps constrain performance as much as people might assume is the case, ie their heat content may be quite comparable in the end.

 

 

 

Ivy Bridge's thermal density "problem" is a myth which has already been busted, by way of the very procedure outlined in this thread. The problem lies solely in Intel's choice of thermal transfer material between the CPU core and the IHS. Temperatures on de-lidded Ivy Bridges are comparable to equivalently configured Sandy Bridges.

 

Thanks for the sympathy guys! I went and picked up another chip this morning and it is working just fine. And yes, it is lidded. Lesson learned!!

 

TechguyMaxC-- the nick is super small about the size of a pin ###### and is located near the outer top corner of the chip.

 

Mind taking a pic? What did you use to seal it?

[Great Ozzie 101]

And you bought another Ivy just to run it in dual-channel mode ?

 

Yep... my shorts got all bunched up and nothing else to do...

 

Besides... I have everything at this point in reserve parts save a motherboard for another ivy build.

 

My thinking a little warped too after having purchased (for past builds) two AMD 285s ($1100 ea.) and two intel 9775s ($1600 ea.).

 

 

I went and picked up another chip this morning and it is working just fine. And yes, it is lidded. Lesson learned!!

 

Yeah... "fun" to do one... but too scared to join some of those nuts that have ruined two... (let me find that link...)

 

My first de-lid job (my own 3770k) has some scratches

 

You know... now that I think about it... would have been a decent idea to get a cheap(er) ivy ($50) and practice on that one first.