F5F Stay Refreshed Hardware Desktop Take your Intel CPU apart for a DIY liquid metal alternative.

Take your Intel CPU apart for a DIY liquid metal alternative.

Take your Intel CPU apart for a DIY liquid metal alternative.

Pages (3): Previous 1 2 3 Next
F
Fretti0_YTB
Member
196
07-12-2016, 09:12 PM
#11
If the LM comes out of the die, you should rip your CPU since the 8700 die is just like any other Intel chip.
F
Fretti0_YTB
07-12-2016, 09:12 PM #11

If the LM comes out of the die, you should rip your CPU since the 8700 die is just like any other Intel chip.

E
Ellie_12
Member
51
07-13-2016, 03:32 AM
#12
People experience inconsistent LM performance when they attempt to use it on smooth or reflective surfaces, which is problematic. Light sanding (buffing) both the CPU and the underside of the IHS using 1500-grit dry sandpaper works best, ideally followed by wiping with 91% alcohol mixed with coffee liners. This creates a slightly rougher texture while keeping the surface flat, which benefits LM. The increased coarseness aids in spreading and prevents the material from pooling or collapsing when it becomes very thin.
E
Ellie_12
07-13-2016, 03:32 AM #12

People experience inconsistent LM performance when they attempt to use it on smooth or reflective surfaces, which is problematic. Light sanding (buffing) both the CPU and the underside of the IHS using 1500-grit dry sandpaper works best, ideally followed by wiping with 91% alcohol mixed with coffee liners. This creates a slightly rougher texture while keeping the surface flat, which benefits LM. The increased coarseness aids in spreading and prevents the material from pooling or collapsing when it becomes very thin.

S
Sven_Weetj
Member
220
07-13-2016, 03:38 AM
#13
Copper IHS has undergone numerous evaluations from reliable references. It slightly raises temperatures when contrasted with the standard IHS version. The performance of your liquid metal application may not have improved, which could explain the lack of progress. P.S. Liquid metal doesn’t need complete replacement every few years. Lightly sanding the IHS ribs—where the original sealant was applied—can help, as rough edges sometimes occur. Sanding the die itself isn’t necessary and may be risky if done improperly. Cleaning off excess grease or paste residue properly ensures smooth spreading. Avoid using too much material; it should form a very thin layer, not drips or pools. Die lapping isn’t suitable for everyone.
S
Sven_Weetj
07-13-2016, 03:38 AM #13

Copper IHS has undergone numerous evaluations from reliable references. It slightly raises temperatures when contrasted with the standard IHS version. The performance of your liquid metal application may not have improved, which could explain the lack of progress. P.S. Liquid metal doesn’t need complete replacement every few years. Lightly sanding the IHS ribs—where the original sealant was applied—can help, as rough edges sometimes occur. Sanding the die itself isn’t necessary and may be risky if done improperly. Cleaning off excess grease or paste residue properly ensures smooth spreading. Avoid using too much material; it should form a very thin layer, not drips or pools. Die lapping isn’t suitable for everyone.

R
ReborntoKill
Posting Freak
821
07-13-2016, 12:35 PM
#14
People still overlook what I mean. I'm not about sanding the die, but about polishing it with 1500 grit sandpaper. This isn't sanding; it's about creating a better surface for adhesion. You won't notice the difference if you try applying LM on a completely polished mirror and then buff it with the same 1500 grit. The results will be clear—see how it spreads and behaves under different conditions. Make sure you try this yourself before judging similar posts. It's the same technique to lift hardened layers from old LM setups on both CPU and IHS.
R
ReborntoKill
07-13-2016, 12:35 PM #14

People still overlook what I mean. I'm not about sanding the die, but about polishing it with 1500 grit sandpaper. This isn't sanding; it's about creating a better surface for adhesion. You won't notice the difference if you try applying LM on a completely polished mirror and then buff it with the same 1500 grit. The results will be clear—see how it spreads and behaves under different conditions. Make sure you try this yourself before judging similar posts. It's the same technique to lift hardened layers from old LM setups on both CPU and IHS.

B
biotit
Member
81
07-13-2016, 05:04 PM
#15
I've heard what was mentioned. I'm not suggesting sanding or lapping here. What I mean is that those who can't even apply liquid metal correctly will cause more damage if they introduce sandpaper near the CPU. As I previously said, good cleaning helps it spread evenly and prevents temperature spikes. I haven't had to reapply it myself, but you might have a valid point.
B
biotit
07-13-2016, 05:04 PM #15

I've heard what was mentioned. I'm not suggesting sanding or lapping here. What I mean is that those who can't even apply liquid metal correctly will cause more damage if they introduce sandpaper near the CPU. As I previously said, good cleaning helps it spread evenly and prevents temperature spikes. I haven't had to reapply it myself, but you might have a valid point.

D
DarkInnocence
Junior Member
4
07-13-2016, 08:07 PM
#16
I’d prefer a straightforward kill with regular tim. If I’m going to die, it should be clean.
D
DarkInnocence
07-13-2016, 08:07 PM #16

I’d prefer a straightforward kill with regular tim. If I’m going to die, it should be clean.

W
wizebana
Member
141
07-14-2016, 01:56 AM
#17
I understand what you're saying. I was trying to clarify with @Plutosaurus about why YouTube results vary for 'perfect' applications. It seems the issue relates to pooling and pressure application. Buffing the surface can help fix this, provided both the IHS and CPU are treated the same. I might test a spare heatsink to compare 'new' versus 'thinned out' LM on polished and buffed areas, but I’m running low on LM and polishing paper.
W
wizebana
07-14-2016, 01:56 AM #17

I understand what you're saying. I was trying to clarify with @Plutosaurus about why YouTube results vary for 'perfect' applications. It seems the issue relates to pooling and pressure application. Buffing the surface can help fix this, provided both the IHS and CPU are treated the same. I might test a spare heatsink to compare 'new' versus 'thinned out' LM on polished and buffed areas, but I’m running low on LM and polishing paper.

Z
Zacherino5900
Member
120
07-14-2016, 08:43 AM
#18
My initial IHS was nearly perfectly flat. A simple test is spinning the chip on a smooth surface and checking for wobble. If it spins, it indicates unevenness and needs correction. I avoided damaging the special edition CPU's IHS, making this a worthwhile $16 purchase.
Z
Zacherino5900
07-14-2016, 08:43 AM #18

My initial IHS was nearly perfectly flat. A simple test is spinning the chip on a smooth surface and checking for wobble. If it spins, it indicates unevenness and needs correction. I avoided damaging the special edition CPU's IHS, making this a worthwhile $16 purchase.

N
NinatoPvP
Posting Freak
899
07-15-2016, 04:10 PM
#19
I apply it to a single unit from rockitcool, though it causes staining; however, it doesn't influence the cooling performance.
N
NinatoPvP
07-15-2016, 04:10 PM #19

I apply it to a single unit from rockitcool, though it causes staining; however, it doesn't influence the cooling performance.

D
dynamitemanYT
Junior Member
6
07-15-2016, 04:33 PM
#20
For a long-term use, this seems like a natural outcome. YMMV
D
dynamitemanYT
07-15-2016, 04:33 PM #20

For a long-term use, this seems like a natural outcome. YMMV

Pages (3): Previous 1 2 3 Next