Early last month, I decided to take the chance and repaste my laptop using Conductonaut. I don't live in the States so it's hard for me to get thermal pads in my area, so I didn't do any repadding.
Whilst idle temps are good (between 20 -40), I've recently run problems with temps under load. Core 0 almost always hits the high 80s / small 90s, with Core 2 not far behind. Cores 1 and 3 almost never hit the mid 70s. I'm wondering why I'm getting such a high differential a month after repasting?
I heard this might have to do with the heatsink not being properly aligned. I happen to have a CCI heatsink and I was wondering if I should go ahead and get a Sunon one, since they appear to make better contact with the die.
The PCH mod I installed didn't help much either. I used the largest aluminum heatsink from the Raspberry Pi heatsink set, and dabbed a bit of thermal paste under it, but it seems to raise the temperature of the PCH even more than when it wasn't covered.
Would anyone be willing to help me out on Discord to try to troubleshoot this problem? I don't want to keep tinkering around the laptop and break more components (I already broke the Wifi pins and apparently I short-circuited the right side lamps). I would really appreciate the help!
-
I dont use discord.
Anyways, sounds like a classic case of the heatsink not applying uniform pressure. This is caused generally by a contact plate that might not be flat as it should be (convex/concave). The other issue can be not enough pressure. From what I understand R3/R4 both have issues with this out of the box due to the heatsink design.
There are strategies to combat this, search the subforum for stickies as I imagine one of them have already displayed how to optimize the system.omy808 likes this. -
I did search around the forum, but unfortunately I couldn't find a consensus regarding the solution to this problem. Certain people suggest repadding, others suggest bending one of the heatsink's tripod arms, and others say to change the heatsinks entirely. As a novice who is uncomfortable with this process, I find myself stuck and not knowing what to do.
I also can't be sure that I even tightened the heatsink on correctly in the first place. I am using an automatic screwdriver, so I might have overtightened in one of the screws at the expense of another area left too loose.Arrrrbol likes this. -
Yeah it's a problem for sure and no one seems to want to test the aftermarket solution either.
-
-
Falkentyne Notebook Prophet
@omy808 please check your PM and contact me.
-
Falkentyne Notebook Prophet
@Papusan please PM omy808. Keep discussion in PM you know what it's about.
-
-
Build date? How do I check for that?
This is a second-hand AWR4 but I believe it's one of the earlier revisions. CCI heatsink + 6820HK. -
omy808 likes this.
-
Vasudev, Vistar Shook and omy808 like this.
-
@ Papusan: I reseated the heatsink earlier this afternoon while keeping uniform pressure on the heatisnk, and screwing in each one bit by bit. Unfortunately I didn't take any pics. The temps haven't improved. Think it's an issue with fit. I'll snap a few pics tomorrow after work when I redo the reseating.
Vasudev, Papusan and Vistar Shook like this. -
-
omy808 likes this.
-
Vasudev, KY_BULLET, Vistar Shook and 2 others like this.
-
Hi all, I think I'm done trying to reseat the current heatsink because I believe that the pads on the chokes above the CPU mount are causing uneven contact. I want to buy a new heatsink but I can't find a legit source for it. Does anyone have any pointers?
If I can't find a new heatsink, I guess I'll have to find 0.1mm pads. Anyone know where I can get those? -
Last edited: Apr 5, 2018omy808 likes this.
-
@Rei Fukai I heard of this! But isn't it risky because the constant heating and cooling of it would lead to seepage over time, which would cause most of it to spill out after a few months?
-
-
Last edited: Apr 5, 2018omy808 likes this.
-
gallium not only corrodes, it forms an oxide and an alloy. after a few months the alloy is the material that can be scratched of the heatsink ass corrosion. Hence why people use an IHS to inhibit that behaviour. -
-
-
-
UPDATE: So I went ahead and reseated the heatsink a few more times, and accidentally rubbed off some of the liquid metal from the CPU. I decided to do a repaste on the CPU using MX-2 while keeping the GPU covered with Conductonaut. I did use quite a bit of thermal paste as I had a lot of issues with my core differential, but now all cores are within 6-7 degrees which I'm more than happy with.
However, I think by overtightening the CPU end, or by applying too much thermal paste there, I'm getting higher GPU temps. It used to be that it would never go above 70C, but now temps shoot up to 81C after a few minutes of Overwatch. Should I be worried? -
Falkentyne Notebook Prophet
I would fix it. I don't like when I get worse temps after doing work. Need to make it right and do a good job.
But MX-2 is quite bad paste.
You don't have anything better?
If you don't want to use LM you should try Coolermaster Gel Maker Nano.
Probably the best paste for bad heatsinks. -
For now, I'm limited to what local stores here sell, unfortunately. I wish my GPU temps were better, but if this doesn't throttle, I'll hold off on repasting until I get proper pastes. -
As I said in the repaste thread my two biggest advices are to lap the heat sink, and be really sure all your pads are the right thickness. -
I have read a lot about K5 PRO indicating that you have to re-do it every 3 months and that it runs off (loses contact) after its composition deteriorates.
I used IC7 Diamond instead of processor pads as a replacement with good results. During disassembly of heat-sink, after 2 months of IC7 being on, the area was definitely holding its form AND good contact, because I had to use a bit of force to 'tear IC7 gum in half' when taking heat-sink off motherboard.
See pictures 1 and 2:
http://forum.notebookreview.com/threads/official-alienware-13-r3-owners-lounge.797884/page-709 post #7088 -
c69k, Vasudev, Papusan and 1 other person like this.
-
Good they fixed the formula. So, after tearing heat-sink off motherboard, does it turn into 'chewing gum' or stays paste-like? It has higher Wm/K than IC7 so maybe it does its business good. Do you have any pictures of K5 PRO how it looks after some use?
Rei Fukai likes this. -
but that's my experience, people could have different experiences with it. -
Good post, thank you. Can you please confirm that it stays as sticky paste after longer use (toothpaste-like) or it becomes solid (1 hour old used chewing gum on the table = IC7 lol). Thanks, good info. +REP
Last edited: Apr 10, 2018Rei Fukai likes this. -
AWR417 - Repasted with LM, but high core differential?
Discussion in '2015+ Alienware 13 / 15 / 17' started by omy808, Apr 1, 2018.