Intel Core i7-5960X De-Lidded Haswell-E Uses Soldered TIM

[Det0x]

http://www.guru3d.com/news-story/int...dered-tim.html

For its most powerful Haswell-E based processors Intel will use Soldered TIM based on a strong epoxy. OCDrift discovered that when they removed (read forced off) the heatspreader. Beneath the adhesive layer that secure the IHS to the package, Intel soldered the chip?s die to the IHS with a strong epoxy.

As OCDrift reports; this is a good news for overclockers, simply because soldering the CPU die to the IHS allows for much better heat conductivity than if the gap was to be filled with thermal interface material (TIM) like we?ve seen on the Core i7-3770K, 4770K, and 4790K.

Haswell-E is expected to be released in September 2014, and the rumoured specifications suggested that the top model namely the Core i7-5960X will have 8 cores, 16 threads and clocked at 3.0GHz with a 3.3GHz Turbo peak.

Now the question begs, who the heck did they get that processor from ?

Good news for us waiting to updated our systems

[Kersh]

Can't wait to get one of that sexy 8 cores paired with a 20nm maxwell.

[NKrader]

why is this keep getting posted everywere like a suprise? the last 2 gen of -e cpus have been soliered. his one is EPOXIED, same differance, but the title of this thread is incorrect.

[lowfat]

why is this keep getting posted everywere like a suprise? the last 2 gen of -e cpus have been soliered. his one is EPOXIED, same differance, but the title of this thread is incorrect.

Exactly. I didn't expect Haswell-E to be any different.

[NKrader]

Exactly. I didn't expect Haswell-E to be any different.

glued to IHS is only thing that matters call it what you want.

[STEvil]

"Beneath the adhesive layer that secure the IHS to the package, Intel soldered the chip?s die to the IHS with a strong epoxy."


So its glue, not solder. Heat activated epoxy is not solder.

[rge]

all you have is a picture of a forced delidded chip showing what looks like solder at edge, and die cleaved down the middle. And then the person that delidded clearly doesnt know the difference between epoxy and solder via the comment "Beneath the adhesive layer that secure the IHS to the package, Intel soldered the chip?s die to the IHS with a strong epoxy."

So that person thinks solder and epoxy is same thing, hence other than picture, any comments are worthless.

At tdp of 130W, im going with solder at 87 w/mk. There are epoxy's that are typically 1-5W/mk ie no better than paste. And you wont convince me that the thick bright silver is anything but solder.

[G.Foyle]

What kind of epoxy (or whatever heat-activated polymer/metal composite stuff) holds stronger than die underfill and bond pads? This looks to me like metalised wafer soldered to IHS.

[STEvil]

I've torn gpu die's off using silicone and 24 hour epoxies.. yeah.

[LordEC911]

The top of the die, still stuck to the IHS, appears to be sitting on a cushion of whatever they are using...
I hope that isn't solder.

[rge]

it is a blob of excess solder squished out the side after ihs put onto die.

[zalbard]

So its glue, not solder. Heat activated epoxy is not solder.

It should still be pretty good, though.

E.g., some silver epoxy results are HERE.

Thermal conductivity of solder: 41 W/m*K.
Thermal conductivity of ag-epoxy: 29 W/m*K.

The paper is from 2001, so perhaps there is something better by now.

Edit: HERE is a paper from 2007.
They measure absolute temperatures of an ASIC die using epoxy (EDA) vs solder (SDA). They show that both solutions demonstrate very close results.

[STEvil]

It should still be pretty good, though.

E.g., some silver epoxy results are HERE.

Thermal conductivity of solder: 41 W/m*K.
Thermal conductivity of ag-epoxy: 29 W/m*K.

The paper is from 2001, so perhaps there is something better by now.

Edit: HERE is a paper from 2007.
They measure absolute temperatures of an ASIC die using epoxy (EDA) vs solder (SDA). They show that both solutions demonstrate very close results.

Didnt say it couldnt be good, just that its not solder

[zalbard]

Didnt say it couldnt be good, just that its not solder

I am curious why Intel no longer uses solder, though. Is it just cheaper, or can't chips handle soldering temps anymore?
If it's much cheaper, it would be nice to see this epoxy used in more affordable chips as well.

[NKrader]

I am curious why Intel no longer uses solder, though. Is it just cheaper, or can't chips handle soldering temps anymore?
If it's much cheaper, it would be nice to see this epoxy used in more affordable chips as well.

they do use it, 3930k and 4930k off the top of my head used it. prob expensive. so they are trying to cut costs to pad their profit margins.

[To(V)bo Co(V)bo]

I think the whole solder vs tim issue comes down to product differentiation. Too many people are investing in the mainstream oc parts instead of going with the hedp. Intel is just creating a reason to spend more on higher end parts, and I am sure the cost savings is just a bonus to them. This upcoming release has a lot riding on it for enthusiasts. This is about the only generation I am excited about since nahalem. My i7 980 is still working great for me, but having a superior chipset with more cores sounds tempting. The only negative is being one generation behind mainstream parts. Intel should reverse that trend, and I bet sales would be significantly higher with better margins too.

Damn technology stagnation! It really hurts the scene for us hobbyist. Seems monitors, ssd, and perpherials are the only thing innovating lately. Even SSD has been super boring the last 2 years.

[rge]

Intels indium solder is 87 w/mk, it is listed in many different white papers, and at indium site. In the recent video which most that have DC has seen, the intel product manager that works with intel engineers states they did not use solder on DC because of cost constraints where use of solder requires a difference in manufacturing process (alluding to need to metalize the surface of silicone die so solder will attach without voids as stated in white paper) since solder would not be used in this chip across multiple platforms (only in DC ie less than 1% chips), but that they do still use solder on the enthusiast platform as all uses require same manufacturing across platforms (ie all need solder, and all will hence need metalize die). Though it looks like some will just have to wait until someone applies heat to it, where solder will be unmistakeable.

The primary role of the IHS is to spread the heat out
evenly from the die and to provide a better bondline
control of the interface material. This can be achieved by
increasing the area of the IHS and by using a high thermal
conductivity thermal interface material with low
interfacial resistances. In order to meet thermal dissipation
targets, Intel introduced polymer thermal interface
materials (PTIM) initially with 3-4 W/moK bulk thermal
conductivity and then successfully transitioned to Pb-free
solder-based thermal interface material to meet the ever
increasing demand for thermal cooling capability as
shown in Figure 16.....
thermal conductivity and the mechanical compliance requirements resulted in
the development and qualification of low melting
temperatures (157oC Tm), low mechanical yield strength
(4-6 MPa), and relatively high thermal conductivity (~87
W/moK) pure Indium (In) metal for STIM applications.

http://www.lsi.usp.br/~acseabra/grad...0Packaging.pdf

[PatRaceTin]

Good or bad ?... I am waiting for the review result soon.

[rge]

Not that this should be news to anyone..but just in case:

Editor's Note: We just got confirmation from Intel that Haswell-E uses a soldered TIM, which is even better than the NGPTMI found on Devil's Canyon. This is great news.

http://www.hardwarecanucks.com/forum...review-14.html

[Metroid]

It's good overall, 130 watts to be dissipated, that kind tdp on a fragile aircooling is hard,