Overclock.net banner
1 - 20 of 123 Posts
Discussion starter · #1 ·
So what I decided to do it to replace pads on my Asus Ares III to get better temp results.
Just to make sure everyone knows what Ares III is, it's 2x290X on one PCB, similar to 295X2.

Firstly I used Phobya XT which is 7W/mK. This is about the best you can rly buy in EU(thermal grizzly pads are 8W/mk).
However after installing those I was rather disappointed as VRMs on GPU2 were hitting 100C under load, with +200mV on the core and +100mV AUX.

http://postimg.org/image/igdbwho0z/full/


Temps with no OC and no volts were rather "OK". Max ~65C.
(Ares adds +25mV by default)

http://postimg.org/image/d68d573s3/full/


As you can imagine I wasn't satifsied with the results. Who would want 100C VRMs on Ares III?!
So I decided to get Fujipoly. You can't really get fujipoly in the UK nor EU(I believe) except ebay, but even then, you can't get the best version only the second best.

The best I could get is Fujipoly Extreme Plus which are 14W/mK.
I installed those on all VRMs and I also used Thermal Grizzly Kryonaut on VRMs and on the waterblock to even improve temps more!
(Kryonaut=the best non-conductive thermal paste)
(Fujipoly=the second best version, the best one is 17W/mK but you cant get it in the UK).


The difference was insane.
OC Results:
(ΔT, Phobya XT-7W/mK vs Fujipoly Extreme Plus-14W/mK+Thermal Grizzly Kryonaut on pads as well)
GPU1_vrms
VRM1=Δ21C
VRM2=Δ7C

GPU2_vrms
VRM1=Δ34C
VRM2=Δ17C

Phobya XT(+200mV Cores, +100mV AUX) ~100C max
http://postimg.org/image/igdbwho0z/full/


Fujipoly Extreme plus(+200mV Cores, +100mV AUX) ~66C max
http://postimg.org/image/gaefo8v6r/full


Non-OC results:
(ΔT, Phobya XT-7W/mK vs Fujipoly Extreme Plus-14W/mK+Thermal Grizzly Kryonaut on pads as well)
GPU1_vrms
VRM1=Δ11C
VRM2=Δ5C

GPU2_vrms
VRM1=Δ19C
VRM2=Δ10C

Phobya XT(+25mV Cores, +0mV AUX) ~65C max
http://postimg.org/image/d68d573s3/full/


Fujipoly Extreme plus(+25mV Cores, +0mV AUX) ~46C max
http://postimg.org/image/sg88cdzab/full/


I recommend it ! Go for it ! Its worth every penny
thumb.gif
 
Discussion starter · #3 ·
Quote:
Originally Posted by FlanK3r View Post

wow, soemone with ARES III here! Can you show some photos?
smile.gif
Yeah no problem
tongue.gif
















(asi otvorim thread aj na pct.cz
wink.gif
)
 
  • Rep+
Reactions: FlanK3r
Fujipoly / mod/smart Ultra Extreme XR-m Thermal Pad - 60 x 50 x 1.0 - Thermal Conductivity 17.0 W/mK
Gelid Solutions GC-Extreme Thermal Compound TC-GC-03-A - Thermal conductivity: 8.5 W/mK

How should understand these numbers? is this pad even better than thermal compound?
 
Quote:
Originally Posted by fewness View Post

How should understand these numbers? is this pad even better than thermal compound?
For a given thickness, thermal resistance is lower than most non-metallic thermal compounds.

However, for surfaces flat enough to be used with paste, you still want paste, because the bondline of paste is far thinner than the thickness of any thermal pad. Less material, less total resistance.
Quote:
Originally Posted by pdasterly View Post

ultra extreme fujipoly is the best, frozen ships to uk
FrozenCPU is still defunct, is it not?
 
Quote:
Originally Posted by Blameless View Post

For a given thickness, thermal resistance is lower than most non-metallic thermal compounds.

However, for surfaces flat enough to be used with paste, you still want paste, because the bondline of paste is far thinner than the thickness of any thermal pad. Less material, less total resistance.
So only at a given thickness. Still amazing to me, I always assumed pads are inferior to thermal compounds. I wonder if they can slice the pads into really really thin film and steal some business from compounds manufacturers. A film will be much easier to apply and clean; that's the advantage.
 
Discussion starter · #11 ·
Well. can get 1200MHz with about +200mV on the cores. (1.35V load).
However when using >60Hz refresh rate over DP, I can't go higher on GPU1 as "blackscreening" starts.
It's really weird and related to gpu1 core clock + high refresh rate(possibly resolution).
I'm trying to figure that out how to remove this issue so I can clock higher.
redface.gif
 
Quote:
Originally Posted by fewness View Post

So only at a given thickness. Still amazing to me, I always assumed pads are inferior to thermal compounds. I wonder if they can slice the pads into really really thin film and steal some business from compounds manufacturers. A film will be much easier to apply and clean; that's the advantage.
A pad/sold film won't fill gaps as readily, and would fall apart at the thiness of paste bondlines. With two flat surfaces, you are looking at bondlines in the single digit mil range...or on the order of a thirtieth of a millimeter.
 
Wow, I was half considering getting better pads then the stock ones for my 290's block.. You've totally sold me!

Mind you at 130 offset I only get 70c on the VRM's, but lower is better..
smile.gif


+Rep
 
So you used Kryponaut paste between both surfaces of the pad (heatsink and chip)? Or was it placed over the chip only?

Which thickness of Fujipoly 17w/mk is best to use... 0.5mm, 1mm, 1.5mm? Especially if I'm thinking to use a 0.5mm copper shim on top of the pad and then a aluminum heat sink.
 
Discussion starter · #15 ·
Quote:
Originally Posted by dmitry100 View Post

So you used Kryponaut paste between both surfaces of the pad (heatsink and chip)? Or was it placed over the chip only?

Which thickness of Fujipoly 17w/mk is best to use... 0.5mm, 1mm, 1.5mm? Especially if I'm thinking to use a 0.5mm copper shim on top of the pad and then a aluminum heat sink.
yes exactly. heatsink and chip. it fills the pores in the pads/block/mosfets so the heat transfer is even improved. This is also recommended by EK.

Well I think you have to use the exact thickness as the original pads.
The thinner=the better but if it's too thin it won't make the proper contact. Also fujipoly pads are very soft so there's no problem with compressing them.
thumb.gif
 
Hey i got myself some Sarcon XR-e pads but by Alphacool, guessing only difference is the branding.
Could you share your application method of the thermal paste on both sides of the pad. Did you put paste on the pad, or just small amount on each VRM mosfet?

I'm gonna be using it with EK monoblock on MB VRM but shouldnt be any difference.
 
Discussion starter · #17 ·
Quote:
Originally Posted by Kutalion View Post

Hey i got myself some Sarcon XR-e pads but by Alphacool, guessing only difference is the branding.
Could you share your application method of the thermal paste on both sides of the pad. Did you put paste on the pad, or just small amount on each VRM mosfet?

I'm gonna be using it with EK monoblock on MB VRM but shouldnt be any difference.
Well I just put some paste on components not on the pads.
So basically, on vrms and then on the waterblock. Maybe it's not really needed. I dont honestly know if it would make a difference if I didnt put the paste there. However it's always good to have the paste there on vrms, so the pads hold on it then nicely.
wink.gif
 
  • Rep+
Reactions: Kutalion
I want to use some Kryonaut and Fujipoly pads on my MSI 390x but I have to order internationally as none of those are sold locally.
frown.gif


So I'd first have to make sure of what I will need. I'd guess 1mm thick pads would be ok?

Also, @fat4l, have you considered using the Conductonaut instead of the Kryonaut? Or do you think it isn't worth the effort/risk?
 
1 - 20 of 123 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.