[JCOC]

Hello fellow OCers!!!

I was wondering if any of us on here run with no IHS. I have two friends that have been running with no IHS on both their i7s. One of them has an i7-3770K. The other has an i7-4970K . They get pretty sweet temps with a Corsair H100i AIO under load with no OC. Like 55ºC in a 67ºF~70ºF ambient temperature. Now both of their rigs sit directly under a vent from the ceiling. They get blasted by those vents. Also no sunlight can touch the rig as the rooms have no windows. So alot of me feels like its the environment their rigs are in that gives those temps. Or is the die being naked and directly touching the water block helping that much? I warned them of corrosion but they said they check every 3 months and apply new kryonaut every time.

Basically I'm asking for everyone opinion on the matter really. I and thinking about doing the same thing on my i7-7700K but use TG Conductonaut instead. I would be running like that for about 3-4 months depending on the results I get.

I would like to thank everyone in advance for their opinions and advice!

 

[frostbite]

You can try it, With an h80i v2 I was getting 23°C idle and 50-55°C load no oc.

But as with the days of Athlon XP, the die will be exposed and you can chip the corners

You shouldn't encounter corrosion, the die and the coldplate aren't affected by the liquid metal

 

[JCOC]

Quote:

Originally Posted by frostbite

You can try it, With an h80i v2 I was getting 23°C idle and 50-55°C load no oc.

But as with the days of Athlon XP, the die will be exposed and you can chip the corners

You shouldn't encounter corrosion, the die and the coldplate aren't affected by the liquid metal

Chipping the die is a concern of mine as well. So I have been thinking of ways to use washers and a small plastic string level to make sure the block will sit flush. Would that be an option that you would consider?

 

[frostbite]

In the old days they had these foam discs, to "spread the load" it didn't work

If you can measure the thickness of the die it may work.

Or you could put the liquid metal on the die and put the IHS back on then more liquid metal then AIO, will save you any damage and straight temps

 

[JCOC]

Quote:

Originally Posted by frostbite

In the old days they had these foam discs, to "spread the load" it didn't work

If you can measure the thickness of the die it may work.

Or you could put the liquid metal on the die and put the IHS back on then more liquid metal then AIO, will save you any damage and straight temps

Thanks for the advice!!!! It will give a whirl and see what I can get.

 

[scracy]

Quote:

Originally Posted by JCOC

Thanks for the advice!!!! It will give a whirl and see what I can get.

Not worth the risk you might see about a 3 degree C drop in temperatures vs a reseal with liquid metal

https://www.hardocp.com/article/2017/02/07/7600k_naked_die_cooling_temperature_followup/

 

[JCOC]

Quote:

Originally Posted by scracy

Not worth the risk you might see about a 3 degree C drop in temperatures vs a reseal with liquid metal

https://www.hardocp.com/article/2017/02/07/7600k_naked_die_cooling_temperature_followup/

I have delidded and relided all of my Intel chips. I used TG Conductonaut on all of them. I think my friends temps were so low because of how the cool air from the A/C vents hits the intake fans on the AIOs they have. Plus no sunlight to give a greenhouse effect. If I come up with anything I will post it!

 

[Bing]

Quote:

Originally Posted by JCOC

Hello fellow OCers!!!

I was wondering if any of us on here run with no IHS. I have two friends that have been running with no IHS on both their i7s. One of them has an i7-3770K. The other has an i7-4970K . They get pretty sweet temps with a Corsair H100i AIO under load with no OC. Like 55ºC in a 67ºF~70ºF ambient temperature. Now both of their rigs sit directly under a vent from the ceiling. They get blasted by those vents. Also no sunlight can touch the rig as the rooms have no windows. So alot of me feels like its the environment their rigs are in that gives those temps. Or is the die being naked and directly touching the water block helping that much? I warned them of corrosion but they said they check every 3 months and apply new kryonaut every time.

Basically I'm asking for everyone opinion on the matter really. I and thinking about doing the same thing on my i7-7700K but use TG Conductonaut instead. I would be running like that for about 3-4 months depending on the results I get.

I would like to thank everyone in advance for their opinions and advice!

Still, protecting the cpu die from the crushing force above, especially at those corners is better.

Just google for "cracked core" or "chipped core" , you will find old posts about it.

It was done decades ago, here is another alternative method to secure the cpu die , more details here -> Protecting an exposed CPU die

One most important tip : Test the epoxy 1st at other surface, or get used how it works at other victims cpu if any

, DO NOT try the 1st time directly at your precious cpu.

 

[Blameless]

There is also the tricep workout way, where you don't cut off the IHS, but spend three hours and twenty big sheets of automotive sandpaper to lap all the way through it.

You can also just glue washers to the corners of the substrate and lap the washers down.

 

[ozlay]

And lap the die to get even better results.

 

[Bing]

Quote:

Originally Posted by Blameless

You can also just glue washers to the corners of the substrate and lap the washers down.

It was heavily discussed and tested in the past (man, I feel old), the main question still ... how do you know the lapped washer has "good enough" same height as the cpu die ?

 

[Blameless]

Quote:

Originally Posted by Bing

It was heavily discussed and tested in the past (man, I feel old), the main question still ... how do you know the lapped washer has "good enough" same height as the cpu die ?

You lap the whole thing after you attach the washers.

 

[frostbite]

You can do, the fine grit shouldnt do much.

Amd's had chipped corners all the time

 

[DStanding]

MSI provided a Delid Die Guard with some of their high-end 1150 motherboards which replaced the standard socket cover for a delidded CPU (bare die). I wonder if there's a way to get one.

 

[JCOC]

MSI provided a Delid Die Guard with some of their high-end 1150 motherboards which replaced the standard socket cover for a delidded CPU (bare die). I wonder if there's a way to get one.

I have a MSI die guard that I got from an OCN HWbot teammate. They do not sell the MSI die guards anymore; however, they do pop up ebay from time to time. You're still in luck if you do want a die guard.

Here is a verbatim copy of the MSI Direct Die Guard.

https://www.ebay.com/itm/Alloy-CNC-...hash=item1a3fa04554:m:mttwzbiu5803L6yMttpdZ8g

I will warn you though. With the MSI and the generic version, you may have to use a piece or tape on top of the CPU's PCB. This helps the die guards push the CPU onto the pins evenly. Tape doesn't always have to be used but it sure does help.

Also, in that link it says it is compatible Intel 4, 6, 7, 8.

I know for sure that you have to shave off some plastic on the Z270 series sockets to let the die guards press the CPU onto the pins correctly. I am almost positive that the Z370 sockets are the same way, but I don't own a Z370 motherboard. So don't quote me on that.

 

[fx3861]

I have a MSI die guard that I got from an OCN HWbot teammate. They do not sell the MSI die guards anymore; however, they do pop up ebay from time to time. You're still in luck if you do want a die guard.

Here is a verbatim copy of the MSI Direct Die Guard.

https://www.ebay.com/itm/Alloy-CNC-...hash=item1a3fa04554:m:mttwzbiu5803L6yMttpdZ8g

I will warn you though. With the MSI and the generic version, you may have to use a piece or tape on top of the CPU's PCB. This helps the die guards push the CPU onto the pins evenly. Tape doesn't always have to be used but it sure does help.

Also, in that link it says it is compatible Intel 4, 6, 7, 8.

I know for sure that you have to shave off some plastic on the Z270 series sockets to let the die guards press the CPU onto the pins correctly. I am almost positive that the Z370 sockets are the same way, but I don't own a Z370 motherboard. So don't quote me on that.

What do you mean by tape on top of the cpu pcb? Any pic to show as im kinda of interested to purchase

 

[JCOC]

What do you mean by tape on top of the cpu pcb? Any pic to show as im kinda of interested to purchase

I'm sorry that my camera has dirt on the inside of its lens. Lol

But in the picture below, is what I mean by putting tape on the PCB of the CPU.

You don't want the tape to overlap at any spot. However, you do want to cover an area on all four sides.

Sometimes you won't have to use any tape though.

 

[fx3861]

i see.. thanks guess any medical tape will be suited for this

 

[JCOC]

I would use blue painters tape. It is made to peel off easily. It also is very, very thin.

Duct tape is too thick. Way to thick. Stick with a tape that is very thing.

I would also make sure that you have to use tape in the first place.

I know it may seem frustrating, but it is worth it to try with out tape.

If you don't need to use tape and you do use tape, then it may interfere with the contact of the pins in the socket and the CPU.

Also

DO NOT

Tighten any form of cooler down very tight. Be very careful.

You want the exact amount of rotation to be applied to each thump screw or reqular screw.

In order to see the benefits of bare die cooling, there has to be an even amount of pressure applied to the die.

This can easily be screwed up if you are paying attention to how much you are turning each thump screw or screw on the cooler.

 

[AlphaC]

Shim

http://shop.aquacomputer.de/product_info.php?products_id=3378

Spoiler

DELID DIE GUARD