De-Lidding and Overclocking Core i7-7700K with Water and LN2

[Guest]

We're testing Kaby Lake's maximum frequency at various core voltages, the influence of de-lidding, and even applying a bit of liquid nitrogen. Get ready to learn more about overclocking Intel's latest architecture with a bunch of Core i7-7700K CPUs.

De-Lidding and Overclocking Core i7-7700K with Water and LN2 : Read more

 

[SinxarKnights]

Interesting find with the PLL SFR setting. What is that exactly?

 

[rantoc]

Intel should be ashamed of themselves for applying such useless TIM to save a few cents in a "high end" product as they like to call it!

 

[AgentLozen]

Very fun article to read.
You guys were using water cooling and liquid nitrogen for this experiment. I wonder how well aggressive air cooling would have held up.

 

[InvalidError]

Intel should be ashamed of themselves for applying such useless TIM to save a few cents in a "high end" product as they like to call it!

The quality of Intel's TIM is perfectly fine. The problem is the thickness as the gap between the die and IHS is many times wider than necessary. Most of the cooling performance gain from de-lidding comes from reducing or eliminating the adhesive thickness.

To make a fair comparison against Intel's stock TIM, you'd need to shim the IHS to achieve the same total package thickness. Any TIM will perform much worse when it is 5X thicker than the minimum necessary.

 

[Xajel]

I miss the days of Phase Change cooling, while LN is more extreme, you just can't use it in real world usage, just to see how far you can go with it without going for more extreme cooling ( eg. LHe ), With Phase Change, you can even go 24/7 If you have the build correctly ( isolation from humidity )

 

[Math Geek]

i'd love to know how much power it took to get those results

folks ask all the time how much extra power they need for overclocking and this looks like a good opportunity to get some solid numbers if they'd like to share them. what kind of wattage does this use at 6.6 ghz???

 

[damric]

1.9v DRAM on KL is what made me happy. Suck it noobs!

 

[rantoc]

Intel should be ashamed of themselves for applying such useless TIM to save a few cents in a "high end" product as they like to call it!

The quality of Intel's TIM is perfectly fine. The problem is the thickness as the gap between the die and IHS is many times wider than necessary. Most of the cooling performance gain from de-lidding comes from reducing or eliminating the adhesive thickness.

To make a fair comparison against Intel's stock TIM, you'd need to shim the IHS to achieve the same total package thickness. Any TIM will perform much worse when it is 5X thicker than the minimum necessary.

Sure the distance the TIM have to transfer the heat is one of the variables for thermal conductivity but there are tons of documented straight out switches of only the TIM after deliding (without any changes in the IHS height above the CPU) that have resulted in similar improvements like this articles when overclocking the CPU. Sure the tim is sufficient for non overclocking use but intels stock tim is clearly holding the CPU back when pushing that overclock.

 

[InvalidError]

what kind of wattage does this use at 6.6 ghz???

If you include the nitrogen separator and helium chillers needed to generate a sufficient stream of LN2 to keep the overclock running, probably in the neighborhood of 10kW

 

[InvalidError]

Sure the distance the TIM have to transfer the heat is one of the variables for thermal conductivity but there are tons of documented straight out switches of only the TIM after deliding (without any changes in the IHS height above the CPU) that have resulted in similar improvements like this articles when overclocking the CPU.

I have only seen one site report on attempting to maintain the same thickness with Ivy Bridge and that site's results showed worse performance with most aftermarket pastes. All other deliddings I have seen removed the glue and did not put anything back in its place to duplicate its thickness or when they did (ex.: deBauer), they minimized that thickness by pressing the IHS against the substrate before curing.

 

[techy1966]

Nice Article gave a lot of good information about overclocking the 7700K and what delidding and better heat transfer can give you. Actually it gives you information for doing this to more than just the Kaby lake CPU's this would also work on the sky lakes as well heck even the Haswell's or Ivy's would get a boost from this if you still have those CPU's and wanted to get that extra bit of performance to hold out for the next greatest thing from Intel or AMD. Sandy would not get much of any boost I don't think since their IHS was installed a lot better than it's bigger brothers.

Thank you for a great article gives a lot to think about. For me I am still holding on to my i7 2700K @ 5.2Ghz 1.40v it will do more but whats the point the temps are great with water cooling and it performs good in my games. I do plan on getting new CPU but I do not want to have to delid it just to get good temps and the highest OC. Intel starting making your CPU's right again please it seems Sandy and Devils Canyon were the last of the properly IHS installed CPU's not sure about the socket 2011-v3 CPU's and if they are done right for the money I would hope Intel would spend the 4 cents and do it right on those 2011-v3 CPU's.

 

[Max_x2]

I've read somewhere that aftermarket TIMs need to be re-applied every 6-12 months too, because they tend to stay wet and eventually become inefficient. Don't take my word for it as being the absolute truth, but that would explain why Intel puts a TIM that tends to become more dry, in order for it to be useful for the whole life of the chip.

 

[WINTERLORD]

what if I don't delid my cpu is thermal grizzly still a good paste to use? or stay with as5

 

[acme64]

i thought delidding was removing the lid entirely and applying the heatsink directly to the chip?

 

[jtd871]

@InvalidError I'm with Acme64 on this one: I think you need to better explain what exactly your delidding process consists of, and why it isn't just a straight 1-for-1 replacement of the TIM. How much is your process reducing the internal gap between IHS and die? How do you accomplish/ensure/measure this? Why isn't the gap discussed in the article, if it is so important to the results you achieve?

It would be interesting to back-calculate the effective thermal conductivity of the stock TIM using the comparative data to back up the assertion you made in the comments:

"The quality of Intel's TIM is perfectly fine. The problem is the thickness as the gap between the die and IHS is many times wider than necessary. Most of the cooling performance gain from de-lidding comes from reducing or eliminating the adhesive thickness."

I nitpick to 1) improve the quality and integrity of the article and 2) improve the understanding of the technically-inclined but otherwise lay reader - especially me.

 

[ATI9800Pro]

I'm wondering what is the effective area through which heat is produced on the die, and the TIM's thickness between the die and the IHS.

Because if S1, the effective surface area of the TIM through which heat flows from the die to the TIM, and S2, the surface area of the TIM in contact with the IHS, are quite different values, this would require the use of some formulas to estimate the thermal spreading resistance of the TIM : https://www.electronics-cooling.com/2004/05/simple-formulas-for-estimating-thermal-spreading-resistance/

And this would explain why the thermal resistance of this TIM is much greater than one would expect when deriving it from Fourier's law : https://en.wikipedia.org/wiki/Thermal_resistance#Derived_from_Fourier.27s_Law_for_heat_conduction

Note I'm facing the same problem with my i7-4790K, which I delidded, slightly lapped the IHS's rim to make it sit closer to the die, and applied some Thermal Grizzly Kryonaut for both the TIM between the die and IHS and the TIM between the IHS and my waterblock, and the results have been something like a 2°C improvement over the pre-delidding temperatures.

 

[InvalidError]

@InvalidError I'm with Acme64 on this one: I think you need to better explain what exactly your delidding process consists of, and why it isn't just a straight 1-for-1 replacement of the TIM.

This may be the story I remembered that from:
https://forums.anandtech.com/threads/delidded-my-i7-3770k-loaded-temperatures-drop-by-20%C2%B0c-at-4-7ghz.2261855/page-23#post-34053183

Noctua NT-H1 performed 2-4°C worse than the stock TIM when duplicating the OEM gap using shims.

 

[mapesdhs]

"... the process became more popular when it became clear that Intel's Haswell-based CPUs ran hot ..."

No, it become more popular after IvyBridge was shown to be such an awful heat monster, long before Haswell.

Ian.

 

[bit_user]

Sure the tim is sufficient for non overclocking use but intels stock tim is clearly holding the CPU back when pushing that overclock.

One thing to keep in mind about Thermal Interface Material is longevity. Intel's will last well beyond the warranty of the CPU, while many aftermarket compounds exhibit significant performance degradation after extended use. Also, keep in mind that direct contact with the die places more stress on the TIM, and many compounds do not withstand it well.

So, it's not only about cost...

 

[bit_user]

i thought delidding was removing the lid entirely and applying the heatsink directly to the chip?

They both involve removing the lid, but what you're talking about is often referred to as "direct-die cooling". It's not very common, because there's much greater risk of cracking the die. The shim used to rest the heatsink (although people doing this almost always go with a waterblock) must be machined with requisite precision - too thick, and you lose the benefits of delidding - too thin, and you'll crack the die.

 

[bit_user]

what if I don't delid my cpu is thermal grizzly still a good paste to use? or stay with as5

Arctic Silver 5 has long been surpassed. If your needs are modest, it's a reliable option. Otherwise, check a recent TIM roundup.

The best I've seen is Indigo XS. But that's not for the faint of heart...

 

[bit_user]

"... the process became more popular when it became clear that Intel's Haswell-based CPUs ran hot ..."

No, it become more popular after IvyBridge was shown to be such an awful heat monster, long before Haswell.

The generation before Ivybridge used solder. So, it was the combination of that and dissipating a similar amount of heat in smaller area (160 mm2 vs. 294 mm2).

And because it wasn't soldered, delidding was actually an option. With Sandybridge, it wasn't practical.

So...

Sandy would not get much of any boost I don't think since their IHS was installed a lot better than it's bigger brothers.

Thank you for a great article gives a lot to think about. For me I am still holding on to my i7 2700K @ 5.2Ghz 1.40v it will do more but whats the point the temps are great with water cooling and it performs good in my games.

Don't do it. You'll break your CPU!

 

[Gunkk0]

SFR:
https://en.wikipedia.org/wiki/Special_function_register

 

[CoachAub]

Jumping on Rantoc's bandwagon with the idea that Intel does not consider the TIM material OR thickness as mentioned by a couple others. It still boils down to the fact that for their top CPU (mainstream), they should put more effort into the TIM and lid. It's not like they don't charge enough for it.