Applying Shin-Etsu

Hi guys I have set this thread up as some users are wanting to change their thermal paste on the CPU and GPU. My personal experiences have been with the Shin-Etsu X23, so I will be talking about that particular paste.

Before I go into detail I would like to say that the tempratures I talk about are estimates and from memory and I forgot to get some pictures of the before and after temps before I shipped the notebook back. When I recieve it back I will get propper pictures of temps up but for now we are generally interested in the application.

Tempratures at Idle with stock paste:

CPU: ~55*C
GPU: ~50*C

Tempratures at Idle with X23

CPU: ~45*C
GPU: ~40*C

I played Batman AA for around 2 hours and monitored temps, both the CPU and GUP never wrose above 60*C which I think is impressive.

Application of the paste:

I tired both dot and spread methods and found the dot method and allowing the pressure to spread the paste to work the best. Another reason for this is Shin-Etsu drys very quickly and becomes hard to spread.

Tips for Application:

1.Before application place the tube of thermal paste in the microwave and heat on incriments until warm, this makes the paste easier to apply.
2.The heatsink on the GPU needs some force to remove dont be worried about breaking it, it is stuck down and wont come off with ease.
3.If you want to perform the spread method then I found the best way to spread it was with a Joker from a pack of cards.
4.Apply a line instead of a dot to the CPU as it is rectangular and not square.
5. To remove stock thermal paste I used microfibre free cloth and Isopropyl.
6. A phillips screw driver is required for removal of all screws.


If anyone has any question please dont hesitate to aks. I will do my very best to find an answer.

 

Just a tip, but if you apply a line on one of these i7 quads you are going to use way too much.

If you are spreading it anyway it doesn't matter if you use the rice grain or pea, but just make sure not to use too much.

I dabbed 3 small dots/peas (no spreading) onto my 720 with MX-3 and my temps are fantastic. I don't spread anymore as I have found it can trap air when the heatsink is applied.

 

I dont do a line all the way down I should have explained. its basically a dot but elongated.

 

Thanks for the guide fluff, I will post back how my results are on Wednesday.

Just to clarify, you are saying place a small dot or "line" of paste and then use the heatsink to basically spread it? Wouldn't this provide an uneven apply? Typically in the past I have always spread using a razor, credit card, or my finger inside a plastic bag. Would this not be optimal?

 

The spread method can introduce air pockets when the HS is applied, which isn't good.

I'd recommend a THIN line down the middle or 2-3 small dabs (very small if doing 3).

Then just press the HS down firmly.

 

width='480' height="385"><param name="movie" value="http://www.youtube.com/v/ffK7L0Qj13Q&hl=en_US&fs=1&"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/v/ffK7L0Qj13Q&hl=en_US&fs=1&" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width='480' height="385"></embed></object>

 

I tired both and found the pressure method you said above to be the best. it wont be that uneven if you put the dot in the centre. However its all about how you apply it, if you are more confortable with the spread method you can do that, but spreading with Shin Etsu can be quite tricky.

If you really want you could try the spread method and if you get the temps you want keep it, if not try my dot and pressure method.

The video above shows a square chip and doesnt really lend its self to the line method.

 

The video was just to give you an idea of what happens with each method.

That's all

 

I agree thanks for the contribution it really helps! I just didnt want any confusion lol

 

Great video sleey, I will prob go with the 2 small dabs and away she goes

 

Viscosity plays a lot in determining whether the thermal paste should be spread or left up for the heatsink. In this case for X23, because it is a very viscuous paste, spreading it would be the better method than just applying it in the center. Low viscosity pastes such as MX-2 in the video above can be applied by just applying paste onto the component as MX-2 can "flatten out" much easier than MX-2. Keep in mind that the user in the video above has to use a lot of force, as displayed by pushing down the glass pane with both his hands using a lot of upper body strength, to flatten out the paste. That amount of force is not going to be seen by the retention screws in notebooks. The springs in the CPU heatsink do not have anywhere near that kind of retention pressure and the screws on the GPU heatsink are just there to hold down the paper thin metal arms that have almost no retention whatsoever. The screws in notebook heatsinks are there just to make sure the heatsink doesn't fall off. http://forum.notebookreview.com/showpost.php?p=5411393&postcount=159

I've went through more syringes of various thermal pastes than you can imagine. I know the differences. Because of the lack of retention, there will be air bubbles to a certain extent no matter what method or paste is used. It's just that some thermal pastes work better with the spread method and some thermal pastes work better with a direct application.

Like I've said many times before, in the end it really doesn't matter what paste is used, rather it comes down to if the paste has been used correctly. All comparison reviews out there will show you the deviations in performance between pastes dependent on the reviewer's handling, but ultimately all of the comparison reviews just prove that all of the pastes perform well and are extremely close to each other in regards to temperatures.

 

The springs might not have the force but I do

I always apply solid pressure before screwing down the HS.

I don't just apply paste and screw down - I make sure to put some good elbow grease into first.

 

It's not always a flat and even push. Even with the screws lightly mounted to center the heatsink and physical force applied by the user to flatten the paste upon tightening of the screws, it's the retention in the long run that matters because the screws cannot hold the same level of pressure that the user initially used.

In my case, it's harder for me as my heatsinks have to carry much more weight as I have a lot more mass mounted on them.

 

what a timely thread! just about to apply some X23 that i have on my hand. and according to soviet sunrise, i'm gonna spread a little bit haha

 

Spread a little bit? Spread it entirely.

 

I don t disagree with what soviet is saying regarding other pastes, however this thread was made soley for the Shin Etsu, This is a very viscouse paste and although you state the pressure method as not being the best for the application of pastes with high viscosity, I would reccommend it as Shin Etsu drys very quickly and will become hardto spread within about 10 seconds. For someone who isnt use to applying thermal paste I think, applying a small dot and a decent amount of initial pressure will work fine.

 

Yes. Either way isn't wrong. It's not like one method is going to be disastrously poor in performance compared to the other. Both methods will yield great results as long as they aren't done extremely improperly. For lazy f*cks people that want to save time, applying an amount of thermal paste on the die and allowing the heatsink to do the rest of the work is just fine. For people that want to go the extra mile, spreading the paste will probably deliver another degree or two in load performance. Though universally, there is no completely succinct, scientifically qualitative way to yield precise results for comparison for the end-user as we cannot afford the time, effort, materials, and equipment to thoroughly put this difference between two "schools" of thought to rest. In the end, user variability is the far most influencing factor in this.

It seems that X23 is starting to snowball a bad rep for being absurdly difficult to apply. This needs to stop. Over time, the user will learn how to handle X23 and apply it more easily each time. Of course, at first it will be a pain, but that's with all pastes. A user applying thermal paste for the first time will have a hard time applying even MX-2. I can say this right now, X23 for me is no longer difficult to apply once you have accrued a preferred way of spreading it on the die.

 

True , but I think even you will agree with me in saying that the X23 would be a difficult paste to learn on. I have applied many pastes also and although I cant say im a veteran I have had a couple and the X23 was the hardest to apply. But like you said practice, practice and more practice.

 

It takes about 3 applications to learn how to do it. For me now it's not a problem anymore and I can do it very quick as well. First time took me like 20 minutes ).

But I prefer to use it only in my machine because it requires so much attention. In all other machines I put MX-2.

 

Took me a while too lol, with a fair amount of cursing.

 

I've been ranting that X23 is a difficult "first time" paste before you even joined this forum.

 

So thats a yes lol

 

I remember the days of people warning to never push on the cores when putting on paste or a heatsink because you could crush them Hell, I remember taking a small nick out of my my older Athlon64 back in the day.

Since I've never handled anything like X23 before (I'll admit, always an AS3 or 5 guy) I've decided to practice on a spare proc I have laying around (i got 2 1g tubes) before I try on my laptop.

 

That's a good strategy.

 

Exposed dies back then were tiny and the base of the heatsink were huge in proportion compared to the dies on notebook CPU's nowadays and the much smaller base area. Back then, desktop CPU's had to be babied because retention clips are such a pain to put on carefully. The sheer mass of desktop heatsinks also contributed to the potential chipping of the CPU die because of the leverage. Now, all current generation desktop CPU's have nickel plated copper IHS's to protect the die.

We can push on notebook chips all we want as long as the bolts are already fitted on somewhat as we screw them in equally.

 

Back to original post, instead of microwave, I just put my tube in a plastic baggie and ran it under hot water for a couple minutes. I got some unknown variety of Shin-Etsu (someone sent it to me free of charge except shipping), and it was much thicker than the AS5 I've dealt with for years. TBH my temps were a bit better with the AS5 and I tried the dot method and spread method with both the Shin-Etsu and AS5.

If you spread it though, be sure it's relatively thin and even. I don't know how you did yours so well Soviet Sunrise, that's quite impressive. What did you use to spread it? I usually use a flat plastic toothpick. But in my laptop it was hard to get a good horizontal access to the CPU and GPU.

 

I was reading this morning throwing it into a cup of boiling hot water for 10 minutes is very helpful.

 

Try taking the CPU out of the system to apply the paste lol

Then you might be able to achieve a better application layer.

 

I would run it under hot water if your going through that route, I was worried about applying the tube directly to water, and when I placed it in a zip lock bag and placed it in boiling water the bad melted, so be careful.

Microwave doesi t quick and you can keep an eye on it, how water will also work but I wouldnt boil it.

 

and does anyone know the inbetween time for every apply with this Shinetsu x23?

 

Shin-etsu should outlast your system.

10 years maybe.

 

Shin-Etsu applied back in the Red days still works well.

 

hahahhaha in soviet russia, thermal paste apply you!

 

OP, why have you chosen X23 to begin with?

 

CPU wasn't so bad, but GPU was a pain because I couldn't get a good angle on it and isn't really so easy to remove.

Either way, next time I get a chance I'll try some X23 or some ICD7.

 

Next time try MX-3.

Just as good and way easier to apply.

 

I haven't heard of Shin-Etsu before this thread really....how does it compare with AS5 and IC-diamond? xoticpc says ICD is the best thermal paste...but then again it doesnt list Shin-Etsu.

 

As many have said, Shin is not forgiving as most thermal pastes. I tried 4x on my GPU and switched over to OCZ Freeze and my temps went like this:

1.) 90c on GPU under load
2.) 105c on GPU under load
3.) 98c on GPU under load
4.) 89c on GPU under load.

I just swapped to OCZ and already at idle my temps are 10c lower so I imagine the load scores will be much better. It is odd because I actually don't have a hard time spreading the X23, but I think I am either using too much (doubtful) or not enough (plausable)

I will say on my CPU I seemed to of got the perfect application because it idles from 39c-43c on average.

 

I question whether or not its possible to not apply enough paste. Theoretically, if the layer is even and visible (ie you can't see the metal underneath), that should be sufficient. It's intended to be a microgap filler, not a spacer. My guess is that if that isn't sufficient, the heatsink isn't seating as well as one would hope.

 

stop spreading the paste yourself, you'll get air bubbles! let the pressure do the job for you http://www.youtube.com/watch?v=ffK7L0Qj13Q&annotation_id=annotation_461720&feature=iv

 

I agree, but this isnt a desktop and the fixing on the notebook dont allow that type of pressure. Just allowing the pressure of the heat sink to do the job apparently wont work. You need to apply pressure yourself.

 

Wow, obvious TL;DR. Did you not read the initial pages of this thread? The same video you linked as well as the counter argument to it are there.

 

Yeah, it has to be under pressure:

width='480' height="385"><param name="movie" value="http://www.youtube.com/v/d-xVb1qsPCw&hl=en_US&fs=1&"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/v/d-xVb1qsPCw&hl=en_US&fs=1&" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width='480' height="385"></embed></object>

 

LOL liked the video, I actually stopped to watch it!

 

ok, then its not like some other paste that's gonna require an apply every month or so...which is great, once and for all, isnt it?

 

great vid btw

 

No paste requires a reapplication every month.

 

Yeah, not monthly. lol

I have heard reports of IC-7 drying up and needing redone about once per year.

That is why I don't use it.

 

Drying up? Did users report flaking or something?

 

From what I have heard, yes, IC-7 can dry up and cause increases in temperature.