[Alienware 17R4 / 15R3] - Disassembly + Repaste Guide + Results

You mean switching to a different brand?
Jokes aside, I'll try to get ICD tomorrow anyway, the only question I have is about its application. On their site it says I shouldn't be spreading it and instead just put a drop in the middle and press the HS down, but will it work fine that way with my uneven HS mount?
edit: by fine I mean will that application method work as intended in my case?

 

Spread it. As the TRIPOD heatsink is too weak to spread the thermal paste itself. Heat up the paste as I said in previous posts although you have to spread it yourself. Too much ain't good, same with too little. Apply paste, screw down then open again and see if you have applied correct amount of paste. Clean up then applying new paste again. Aka what you think is enough after measured first time.

 

So just to get a baseline, should it be roughly the same amount as I did with Kryonaut? Also if I put a bit too much won't the excess just squeeze out kinda?

 

Don't use more than needed. Yeah, it will squeeze out if you have too much. Do as I said.... A test first. Then you will see how it is and then put correct amount paste in the re-apply.

 

Alrite looks like I have some fun times ahead of me. Thanks a lot for all the help! hopefully it all works out and I won't have to be bugging you any more

 

I don't know about spreading ICD man. I think the weak tripod will still spread the paste fine.

 

Read... This article is for DESKTOP heatsink and pressure...We talk about TRIPOD and weak HS pressure.

Spoiler: Thermal paste Viscosity(see ICD)

 

I tested the line method on the tripod XPS15 and it spread fine. I did make sure to heat the paste tube for 5+ minutes on hot water though.

 

See your own post #2405 "I don't know about spreading ICD man". I think also @iunlock who work with this mess would spread the ICD paste by yourself on die and not used the TRIPOD HS for the job. ICD is damn thick aka very high viscosity.

Edit... FYI

 

Yea this is why I am saying spreading IC Diamond is crazy. Super thick that will probably not spread properly if you try to do it manually. I just put line of ICD on cpu (bga) and use the X method on GPU.

 

See THIS from the TRIPOD master fixer @iunlock

 

Ehh the foam I was going to use didn't work out. Its not the same as the ones you guys used and I was aware of that but I thought if I made it extremely thin it would work out. It did not and the hs didn't make contact properly.

On the bright side my LM was reusable and no where near dried out and I had no leakage whatsoever. Surprising because I store it in a hanging back pack over night and it shakes around while on my commute. I originally thought I used way to much LM but being generous turned out fine.

 

A few people in that thread are saying ICD dried out for them extremely fast, literally in the matter of weeks. Did they do something wrong? How can that even be possible?

 

Very bad fit between die and the heatsink + high heat will destroy all sorts of thermal paste. Brand name doesn't matter. But ICD is probably the best paste you can use if the heatsink fit isn't very good.

 

If it's the best option but according to some may dry out in like two weeks I'm really lost, fingers crossed this won't happen in my case

 

On a slightly related topic, through a happy accident I managed to drop my cpu and gpu temps by several degrees with the use of isolation pads.

I purchased some isolation gel pads for my DAC/Amp, and had some leftover for my AW17 R4 1080GTX. My initial thinking was that I could further reduce vibrations so as to minimize any transference of electrical noise and jitter through the USB cable and thus improving audio quality. What I did not expect was that my laptop idle and load temps would get much cooler with the isolation pads installed.

As can be seen in the pictures, my CPU temps dropped from 76-77C to 70-71C max and my GPU temps dropped from 66-67C to 61-62C. My PCH temps also dropped from 75.5C to 68-69C. The isolation pads pics show a slightly lower CPU package power but the temps stay true with CPU package power of >55W. Both these results were obtained by playing The Witcher 3 for about 45mins. Idle temps also dropped by 2-4C. Sorry no pics of the PCH temps.




I'm not quite sure why the isolation pads help so much so because before I used small copper heatsinks to slightly elevate the laptop above the laptop cooler, though granted the heatsinks are not quite as tall as the isolation pads so the drop in temps could be attributed to the slightly elevated position.
Another reason could be that the pads help isolate the cpu/gpu chips from added vibrations from the internal ;aptop and cooling pad fans. Vibration = added energy = rise in temps? I am not quite sure.

I used Atacama Isolation Gel Pads. Comes in clear color or black. Highly recommended for the restless overclocker.

 

@iunlock @Mobius 1 @Papusan Which guide to follow on changing thermal pads if I have a 15 r3 i7 7700hq gtx 1060? Also anyone got any idea where to buy a pressure paper here in the UK. Thanks..

 

I have no clue what's going on with my 13R3. The GTX1060 is getting to 86C even after a repaste. Maybe not enough pressure or air trapped underneath? Either way, going to repaste again and hope for the best.

 

What's your roo temps?

 

Use the guide in op. Just be sure you have thermal pads in several thickness lying when you dismantle the laptop(if 1060 TRIPOD don't use same thickness of pads). You find Fujifilm Prescale film on Ebay. But Damn expensive. Just use a thin thermal paste or my solution Basiron cream/thin low viscosity toothpaste for measuring the pressure as the pressure film is expensive or sometimes can be difficult to obtain. Don't make it so damn complicated.

 

Room temp is around 22C, although it was a bit higher these days. Still, my Skylake chip was doing well in 25C+ weather. Just finished the second repaste and it seems to have brought down temps by 4~6C. Going to test more but looking at the previous past there was definitely enough coverage. Maybe some air trapped?

 

Probably because the TRIPOD wasn't screwed down properly (uneven attached) when you mounted the HS.

 

I'm not talking about the CPU temps here, but about the GPU temps. CPU temps haven't differed at all. I used the 1-7 sequence and went back twice to re-tighten. Did the same both times I re-pasted except this time I also put a bit of pressure on the HS and moved it around a bit before screwing it down.

 

So those temps are after raising the back of the PC too?

 

TRIPOD is the whole HS!! See... when you screw down the HS, it can differ. You haven't much headroom before the pressure being uneven and you get worse temp. Either for Cpu or gpu. Or both.

 

Yep, the back of the laptop is always raised when I'm gaming. I put it on a cooling pad with mesh underside so airflow shouldn't be a problem. And keep in mind it was reaching 86C on the GPU with an undervolt on the CPU, without that it would probably reach 90C on the GPU.

 

Ah, I get what you mean now . That's a really poor design if all it takes is a little bit of uneven pressure to offset so much.

 

Is speed shift enabled through TS? Another issue is shared heatsink issue, so nothing can mitigate that.

 

Speed shift seems to be disabled. Doing a re-seat of the heatsink has definitely brought down the temps. CPU is now Idling at 41C vs 52C before and in game the GPU is 4~6C cooler, hitting a max of 82C. Not too happy with 82C though since my Skylake would hover around 78C but I guess you can't have it all.

 

Enable SST in TS.

 

Damn what games are bringing your 1060 to 82C?

 

GTA:V and Battlefield 4. Before it would take it to 86C and without repaste it would hit 88C. Still way too high for my liking but probably not enough to warrant a replacement HS. However, when playing America's Army the GPU stays at around 67C. Maybe it's the Nvidia drivers, not sure what's really going on here.

Ah, I'm not using Throttlestop, I'm using XTU. I'll download it and see if it makes any difference.

 

UK, try to find a distributor and ask for a sample. Say that you're a computer repair person looking to test new tools.

Not entirely sure what company does distribution in the UK though.

 

I don't think the liquid metal would just drip out on it's own. First when the laptop is off the cool LM is not as runny and second the surface tension should keep it in place. The biggest danger with LM is during application and then when you clamp it down if some is squeezed out due to over application it can start to roll around the inside of the laptop.

 

I don't know if you have seen this but here is a test of the various thermal compounds to see how they perform in real life.

 

Results from custom test bench - the better desktop cooling will never give same results with the lot weaker cooling on laptops. We have several threads about Thermal paste in the forum. One to watch.
You find a lot info in the forum if you search

 

Given enough time any LM that squeezes out after the initial application should dry out and no longer be a problem, correct?

 

Don't have too big hopes for that

 

Wait you're telling me you had this piece of metal out and about exposed.... and its still good? Wow

 

Yeah, 3.5 months since I painted the hammer head for testing. Still liquid. And Liquid metal will be even more liquid in a hot laptop

 

Tried foam from my comforter and it didn't work. My PCH mod was a bust, I believe I didn't put the paste on properly but **** it I threw it away.

Now... I ****ing broke the black pin on my wireless card. Not sure why but the wireless still works fine. Bluetooth seems ****ed and randomly drops now.

 

Indeed...always have lots of thermal pad sizes around for sure...especially when dealing with a tri pod mess of a design.

They all vary so much that it definitely is not a one size fits all so to compensate, some areas may need up to 0.5mm more ...crazy I know... The only time this isn't the case is if the heat sink is properly reformed, which is very tedious and the average person wouldn't even bother to do this because it can take several hours...

Oh nono...LM stays LM.

Those things are a pain....

Before:


After:

 

It gives me a excuse to get the intel card anyway haha

 

Not sure why, perhaps it's enabling speedshift in ThrottleStop, but my GPU temps have dropped down and stabilized around 78~79C. This compared to 86C it was hitting before, pretty decent change.

 

What is Enhanced Intel Speedstep® Technology
Enhanced Intel SpeedStep® Technology allows the system to dynamically adjust processor voltage and core frequency, decreasing average power consumption and heat production. Means the TRIPOD will work better for the Gpu

 

Flood dams have been erected and Wireless card black pin has been fixed! I have also tried a PCH mod yet again but temps are 5c higher with my testing and PCH temps have a higher average as well. I believe the aluminum heat sink I applied is actually hindering the PCH from cooling down. Going to wait and see how temps settle over time. Currently hitting 71c, I believe I should be worrying once it gets close to 80c.

At least I no longer have to worry about LM leakage.

 

Alright guys - could do with some advice i have carried ou the PCH mod and basically all mods for the 17 R4, my PCH temps under gaming are getting upto 80 degrees, is there anybody that has a better method to cool it ? and what does everybody else get...


It currently has a HS sitting on the PCH

 

I don't believe the PCH HS is getting enough airflow on either the 15 or 17 but I haven't gone over 71c so I think mines is doing well it just stays hot longer. When did you apply the PCH mod? What heatsink and paste did you use.

 

I just used a self adhesive heatsink is that my issue do you think

 

Ehhh you need to use a quality thermal paste for best heat transfer. That'll definitely knock the temps down a lot.

If the self adhesive is similar to the stuff that came with my little heatsinks then its not a good idea to just stick it on your PCH. Clean off your PCH and use kryonaut or really any thermal paste you can get your hands on.