VPC-Z Change Thermal Paste

Fantastic guide jollywombat !

I followed your guide with slight mods:
Copper Shim - 1.5cm X 1.5cm X 0.5mm
Thermal Paste: DeepCool Z9 (24 hrs. curing time)

GPU overclocked: Core: 600 Memory:800 Shader:1000
Power Mode: Performance

Immediately after reassembly, and playing GTA IV, NBA 2K14 while overclocked with the above numbers, GPU temps RARELY go over 83C while CPU temps are around 90-95C, occasionally spiking to 101C.

Initial results seem promising, maybe even better after the curing period.

What I would like to know if anyone has tried using copper shims on the CPU and Integrated GPU? Is it safe and effective?
I would like to see lower CPU temps while maintaining the performance if possible.

 

There is no advantage to be gained by using copper shims with the cpu and intel gpu. The heatsink sits directly on these; thermal paste is enough. The shim is only needed for the nvidia gpu where there is a .5mm gap between the gpu and the heatsink that has to be filled with something conductive.

 

Planing to make a little mod on heat sink

add a new heat pipe ( not a ram sink ), and would like to test the result

 

I was getting temps in the 95C degree range on my i7 cpu when gaming...which is a little higher than I thought they should be considering I had just changed the paste on the cpu's. The next time I had the lappy opened up I pulled the mobo again and checked the heatsink for the cpu's for alignment with the chips. I did this by placing a straight edge (in this case a metal ruler) across the upper part of the GPU heatsink (the area that contacts the video memory chips NOT the area that contacts the GPU itself) and the CPU heatsink. I figured these two areas should be parallel. Low and behold the cpu heatsink was lower on one side than the other....which would leave a small gap between the heatsink and one side of the cpu chip when assembled and installed. A little careful bending of the cpu heatsink to eliminate the angled portion and make it the same height as the raised (video memory chip) section of the GPU heatsink and after reassembly my gaming temps dropped 3-5C! I did the bending with just hand pressure...the copper heat pipes bend easily. This is the second VAIO model where I have found a heatsink out of alignment and dropped the temps by making it parallel to the top of the chip.

CAUTION: Do this with care...proper alignment and pressure of the heatsink on the chip die(s) is critical...especially too much pressure which can crack the chip and ruin your mobo!

 

orionz: please read the other posts in this thread. The GPU is a little lower than the CPU. This is why the thermal pad on the GPU is thicker than the paste on the CPU, and why it is necessary to use a copper shim if the GPU thermal pad is damaged.

 

Another way i found to reduce overall temps is to lower the "max processor state" in the power options. I set mine to 60%. Using this method, i was able to reduce my cpu and gpu temps to 81C max when playing gta iv with little or no performance diff.

EDIT: this was with an OC'd gpu.

Sent from my Xperia Z using Tapatalk

 

Hi everyone. i just cleaned mi vaio z1, i followed the video, i did some old paste cleaning and put a new one. after this, i turned on the vaio, but after a few seconds automatically shut down by itself, i realized that if i boot to windows quickly this don't happen. resuming, if i try to enter the bios the fan start to run really fast and turns off, but in windows environment this don't happen. can somebody guess whats going on, im quite nervous, this was my first time doing this with a notebook.

many thanks in advance.

 

Is the same thing happening with both speed and stamina modes?

 

That does seem a little odd. In Windows once you boot to it, can you verify what your CPU/GPU temperatures are? Maybe something is getting hot and not making contact, or getting a short somewhere.

 

I had the same problem after dropping my Z11 and, after replacing paste and cleaning everything the problem persisted. As soon as I powered on, the fans sounded like they were going to take flight and the cpu/gpu zoomed to 100 C and beyond, quickly causing a shut down. I then learned that the thermal pads also need replacing - either with like materials or with copper shims. I was never successful making this work, ended up making an ADH claim for the laptop and purchasing a new (open box) replacement in 2012 (see sig) which, ironically, I virtually never used - just couldn't imagine not owning a Z!. I hope someone wiser than I can help you resolve this. If not, PM me if you are interested in a good deal buying my Z13 as a replacement and perhaps parting out your Z1. Kills me to part with it now too, but it's just sitting in a bag, er, not gathering dust! (In all, I own too many laptops - 5 in addition to 3 in my sig!).

Best of luck.

 

Thanks for all the reply.

i have been struggling to discover what is the problem. i already found out that. the problem was the bad quality of the paste and that wasn't cleaned very well. i did it again, several times. the vaio worked quite well, but yesterday i realized that the Gpu possibly was damaged, i put a week ago the cooper shim in replacement of the rubber protector(broken), i followed all the steps. as explained in the several posts, but still when i tried to test the Gpu, the performance was quite bad, a lot of lag and freezes in not very exigent games. as a result, today i did everything again, clean, paste etcetera. now i have been testing the gpu, but the temps are over 100 (worse than before). at least, stamina is working quite decent (temps less than 53 using firefox).

i will continue trying to find out whats going on, perhaps i need to wait for the curing period for some hours. many thanks.

 

This is my first post on the forum.

Wanted to say thanks to everyone for their contribution on the thread, particularly to jollywombat and AGabi for the great disassembly pics, explanations & commentaries.

My question is the following:
Can i replace my non-working fan with a new one, while keeping the same heatsink connected to the motherboard?
Thus, not having to deal with any thermal paste work, and also saving a couple of bucks in the process.

Background history:
Since the purchase of my SONY VPC-Z1190S in 2010, the fan always had a rough noise from the get-go (typical Sony fan default).
But I decided to stick with it instead of sending it to Sony and wait a couple of weeks.
The fan lasted almost 5 yrs. And now, it has died completely.

Initially, i thought that i had to replace the whole cooling system (fan with heatsink), and eventually apply thermal paste in the necessary places.
But through my research, i found that the fan is sold separately form the heatsink.
For SONY VPCZ1 Fan MCF-528PAM05 | Fleming Mall
Sony Vaio VPCZ1 Series Cooling Fan MCF-528PAM05

So just wanted to know if i could make a nice clean fan replacement, without getting my hands too dirty.

 

Short answer: no. You have to take the heatsink off the motherboard before you can take the fan rotor off the shaft. If you are patient with this, you might get it off without damaging the thermal pad on the gpu.

You might find that the fan is just jammed solid. Taking it off the shaft and cleaning the shaft and the well, and then lubricating the shaft with light grease might well give you another few years.

Handle the fan gently - the blades are brittle and will break if you try to bend them.

 

Thanks for quick answer anytimer.

I'm going to proceed cautiously. I will try to update if anything interesting.

 

Hi. Did someone have succeed using fujipoly thermal pads?
I'm planning using a 0.5mm 11w-mk for CPU and 1mm 6w-mk (high compression) for GPU.

 

Hi everyone,

Thank you all for the advice on this topic. It's been a great help!

My CPU was overheating at max load, and when I opened up my VPCZ19ME, it was plain to see that the paste had perished. On my first attempt at repasting I installed a Phobya thermal pad to the GPU. Needless to say (now having read through this and another thread) my GPU was in the 90s on idle! I think that was the moment I was most thankful for the for the hybrid graphics.

So, I knew I had to go down the copper shim route, but according to this post the gap isn't quite 0.5mm, and using a shim that thick may lift the heatsink off the Intel chips. I was able to source 10 0.4mm shims from China (ebay) and am pleased to say that all is running cooler. Idle temps aren't as low as some people got them, but my GPU maxed out on FurMark doesn't exceed 90 degrees.

Anyway, I now have 8 (I'm saving one just in case) 0.4mm copper shims spare if anyone wants one. I'm based in the UK so am happy to send in an envelope 2nd class to a UK address free of charge (as my way of extending the good deeds done here). Just PM me and let me know.

Cheers!

 

Interesting thread, let me post my problem here, maybe someone will help me.

So, I won't give up on my latpop still its 3 years old and for now performing great, all except temps....

So for the last two days i was playing with it, and finally managed to open it and do a thermal paste change. Way back someone told me that only CPU is in direct contact with Thermal Paste but it looks like CPU and GPU both are covered with Thermal Paste.

So I had small amount of Cooler Master TP from my 212+evo PC CPU fan and i used it, after i cleaned previous paste. But one thing bothered me. Around copper headsink, there's black paper like material around headsink, its nost sticky, no nothing, just standing there, so i decided not to touch it.

After a while, I managed to do everything and i assembled it. But when I turned on, temps were exactly the same or maybe even worse.

Idling around 55C and reaching 71C on 60% of processor usage (limited under battery options). One more interesting thing is that there's temperature spike. For example its 35c then it goes up to 50, then goes back to 43, and slowly stabilizes itself at around 50-55c idling on desktop with 750mhz cpu speed. Simple video reproduction boost temps to 65c, chrome and other stuff too, around 65-70.

I checked on the internet and I saw couple of people were bragging how their laptop temps dropped to 30c idle after thermal paste change + heatsink cleaning.

So, what I am doing wrong? I was never using my laptop for gaming, its just Chrome, Microsoft Word and Skype... and I'm keeping it mostly on my work desk. I'm not using ded card but its idling at 55c after changing paste.

Any help would be appreciated.

 

Before you go charging into this operation, and risk breaking ribbons etc, try cleaning the fan from on top first. You only need to lift off the keyboard by removing the screws beneath and detaching the keyboard ribbon (carefully). Very quick and simple.

I used a soft toothbrush dipped in isopropyl alcohol, a vacuum cleaner, and blew into it while holding the fan still (gently). Try to brush the dust up and out rather than knock it down. Some is going to get trapped down there, you just want to minimize it.

That was enough for me to quiet the whole thing down to like brand new, and didn't expose me to anywhere near the risk associated with removing the motherboard and heatsink.

 

If you're worried about trashing the keyboard ribbon, you can turn the keyboard over (towards you) without removing the ribbon. That gives you enough access to do the cleaning ^ this way.

 

I just received an VPC-Z122GX from a friend in order to attempt a repair. The machine currently only runs for about 10 minutes, fan spins way up blowing scorching air,then shuts down on its own.

Am I looking at dried up thermal paste or is there something deeper with the machine? He did tell me several years ago he spilt coffee on it once, but I opened it up it doesn't look like it actually got in there, no moulding or stains on the board. Blew the fan with canned air, a cloud of dust came out but didn't seem to cure it.

 

@ToniCipriani sounds like a pad/paste replacement case to me. Try runing Open Hardware Monitor to see which component is overheating. Does the same thing happen when in stamina mode? If so, then it's likely the CPU, if not then it's likely the GPU. If CPU, then you can throttle the laptop power in Windows Advanced Power Settings until you get the paste/shim, usually only 10% reduction will prevent an overheat.

 

Problem is we can't even get the machine running long enough to launch a hardware monitor, or for me to use an offline password tool to get in. Friend can't remember the password, he only uses the fingerprint.

It is however better in Stamina mode, but eventually the same thing happens and it shuts down.

 

If the air coming out is hot, the thermal paste is probably fine. If you can feel the hot air 4 to 6 inches away from the vent, the fan and air passages are fine too. You will eventually need to clean the fan and air passages - the laptop should be able to run under full load without shutting down, e.g. while playing games - but that is not your immediate problem.

You need to find out the CPU usage and what process is causing the massive load. Task Manager should give you a clue (remember to click 'show processes from all users). ProcessExplorer from Sysinternals is probably the best.

Once you have found the process(es) causing the high CPU usage and hence the heating, you can figure out what to do about it. Could be malware; could be a badly configured driver or software; could be some useless bloatware. Ask the guys here for help if you get stuck.

 

It's not a process. I tried booting from a live USB Linux (Ubuntu 14.04.1 LTS) and leaving it in the BIOS, it still heats up like crazy. Only when booted into Windows (at the login screen) and in Stamina mode it's marginally better.

EDIT: On second try after re-reading your post, I'm leaning once again on the thermal material. At 4-6 inches away, air feels cool.

 

Silly question here. I just got some Fujipoly 0.5mm pads (6W/mK) to replace the GPU pad, but I loaned out my thermal grease to a friend and he didn't give it back to me yet. Out of laziness is it OK to use the pad for both the CPU and GPU, or is it going to be too thick and lift the heatsink? And I just realized the pad actually has slightly better thermal performance than my grease (Arctic MX-2, 5.4W/mK), I misplaced my AS5.

 

Absolutely NOT OK. The CPU and GPU are at different levels in the VPCZ1, which is why the GPU pad is much thicker than what one would normally expect. I would advise you to wait for the grease, or get something else locally. I have found that even the cheap generic white thermal paste gives decent results on the CPU - never needed to use anything else. The GPU is the tricky animal, but you have that covered.

 

Hi @TonyCipriani. I used a 0.5mm Phobya XT pad with a conductivity of 7W/mk on my GPU first, and it just didn't do the job. The GPU overheated almost straight away. I guess it's the thickness that matters. Paste will only be a few microns thick, so that rating will suffice, but not for the pad. Also, you want the CPU paste to be as thin as possible, so I wouldn't recommend a pad.


Sent from my iPhone using Tapatalk

 

Just looked it up - a copper shim has a thermal conductivity of 385W/mK

 

Yes, but wouldn't you still be limited by the conductivity of the paste in a way? But I guess since it's only for the irregularity on the surface so actual conductivity is higher.

 

The paste is only there to fill the microscopic pores on the surface. The rating of the copper will be reduced at each point of contact, but the paste helps minimise the impact.


Sent from my iPhone using Tapatalk

 

And now that makes me wonder what that original pad from Sony was actually rated. I do know the pads go up to like 17W/mK, but not copper shim high.

 

OK it appears it did the trick. Went out and got some thermal paste (Antec Diamond Nano 7) and low and behold, no more shutting down. Maybe the CPU paste being drier and harder than cement had something to do with it.

And seems like the thermal pad worked for the GPU. Running a GPU stress test right now with OCCT, fan speed went right up but no shutting down or freezing. Temperatures hover around 57/60C for the CPU cores, seems OK, but one seems a bit high, TMPIN0 is like 80+C. Is this the GPU?

And thanks everyone for the help.

 

Hey guys! Just wanted to show my appreciation for the guides posted here which helped me perfectly restore my VAIO Z13 which was having heat issues and shutting itself off even when I were just trying to install a clean Windows 10.

Therefore I took apart my VAIO replaced the entire fan unit, applied IC Diamond 24 Carat used a 0.5mm*15mm*15mm Copper shim even though my GPU Pad was intact, and apparently I removed what seem to be the plastic of a Double Sided Tape on the VRAMs and installed everything back.

The first time I plugged in the power supply, the laptop turned itself on and booted into Windows 7 fine, however after I shut it down through Windows I couldn't turn it back on and i panicked. But I took everything apart again and with only the display connected to the motherboard I plug in the power again and it turned on fine, also allowed me to check that the new fan unit was working fine as damn it was so quiet I couldn't determine if its working or not without looking at it.

Previously I didn't even bother recording accurately the temps when it was idling or when it hit full throttle as it wouldn't even make it there. Now the Temps are recording at estimated 45-40 degrees when idling and 80-85 degrees when running games. I am really really happy with the results.

Given the production date of the VAIO Z1 series, I believe many people will have issues with the fan and heat sink unit as well as the thermal paste not working as it did before. Hence, I decided to post the websites where I got my parts from, however, to whomever who is going to use these links, I would like to say that I was wary when I ordered the parts as it is clearly from a China seller and was prepared to accept if the unit turns out to be an imitation. Upon receiving the fan unit, I have to say I could hardly doubt its originality, but hopefully I weren't just lucky and that whoever orders from them would have the same experience as me. The copper shims aren't perfect but they are dirt cheap and they do the job as intended.

If linking such links are not allowed please kindly remove them. Thanks.

VPCZ1 Heatsink + Fan Unit Combo : http://www.aliexpress.com/item/Bran...Z1-Z1-Heat-Sink-MCF-528PAM05/32241605716.html
IC Diamond 24 Carat 4.8gram Thermal Compound : http://www.amazon.com/gp/product/B0042IBAOG?psc=1&redirect=true&ref_=oh_aui_detailpage_o02_s00
Copper Shim (Chinese Website) : http://world.taobao.com/item/17617464062.htm?fromSite=main&spm=2014.12193480.0.0
ArtiClean 60ml Kit (Makes removing Old Thermal Compound a piece of cake) : http://www.amazon.com/gp/product/B0007TOR08?psc=1&redirect=true&ref_=oh_aui_detailpage_o02_s00

Total Costs: ~160USD (Shipping to Singapore address)
Satisfaction of reviving my VAIO Z: Priceless.

edit* Forgot to mention the most important message, Thanks again everyone who shared their experiences and hopefully I could provide the same amount of help to the next person and much as the guys did for me before.

 

This is the issue I have with my notebook: under full load (cpu+gpu to 100%) it shuts down unexpectedly, faster and always in the summer, and a bit less in the winter.

I've started already the following thread, but will duplicate here so subscribers can help me to diagnose where the problem could be:
Sony Vaio VPCZ13 powering off on high CPU load (thermal paste burnt)?

----------------------------------------------------------------
has been working perfectly since 2011 or so, until a few weeks ago. When I have the power plan to performance, and the Maximum processor state to 100%, it randomly shuts off several times a day, specially when I put a high load on the processor, which can be doing easily opening several Chrome tabs.

However, if I set the Maximum processor state to 70% or lower for example, I never ever have a shutdown, either using the integrated graphics card or the nVidia 330m.
----------------------------------------------------------------

1. Yes, under load the fan speed increases.

2. Yes, with two cores at 100% use the air is coming out of the vent and I can feel it slightly warm, at say 7 inches (19 cm) in a straight line from the exhaust grid.

3. The air blowing out of the vent is hot, I'll post more temperatures information next.

----------------------------------------------------------------

Ok there's a simple spreadsheet of the temperatures. The cores and gpu temperatures are measured with CPUID HWMonitor exactly just before shutting down. The exhaust vent temperature was measured with a K-type temperature probe for a multimeter.

Notebook: Sony Vaio VPCZ13M9
CPU: Intel Core i5-460M
GPU: NVidia GeForce GT 330m
BIOS > Advanced > Platform Thermal Configuration > Throttle On Temperature: 93 ºC (default)

( TEMPERATURES\ STRESS STATE)... || ......... IDLE .........|| ... CPU+GPU STRESS
ROOM TEMPERATURE..........................|| ..18 ºC / 64 ºF.................................
CORE #0..................................................|| ..65 ºC / 149 ºC.. || ...98 ºC / 208 ºF
CORE #1..................................................|| ..66 ºC / 151 ºF... || ...104 ºC /219 ºF
GPU..........................................................|| ..66 ºC / 151 ºF....|| ...93 ºC / 199 ºF
VENT EXHAUST (COPPER LEVEL)......|| ..59 ºC / 138 ºF...|| ...74 ºC / 165 ºF

If I select the optimus switch to Stamina (use integrated graphics), the vaio computer may shut down but much less frequently.

 

Thanks for this. I fashioned my own copper shim from an older laptop heatsink and a lot of filing and sanding. Also noticed the fins on my heatsink were split at a few points so I've patched it up with aluminium tape and cleaned and oiled the fan. Temperatures sitting around 45C idle and 85C under load.

I'm curious, why has there been no mention of the chipset during all this chat about the VPCZ1 heatsink? I'm aware it wont (or shouldn't) reach the temperatures the the CPU and GPU, but from what I can see it plays a large role in the alignment of the heatsink (particularly over the CPU).

It's on the other side of motherrboard but makes contact with the bracket that retains the heatsinks, through another thermal pad (seemingly the same white pad used on the CPU, not the thicker grey one on the GPU).

I might be wrong about this, but if one was to remove the white chipset pad from the heatsink bracket I imagine the alignment of the heatsink over the CPU would be affected, further complicating the issue with the shim on the Nvidia GPU.

Any thoughts? I'm surprsed it's not been mentioned yet, which makes me think I'm either over complicating things, or my model( VPCZ11Z1E) is slightly different.

 

timwjohn,

I haven't made enough posts to use the PM function yet, but would be very interested in one of your extra copper shims. My VPCZ1190X has temps well over 100 degrees and is constantly shutting down due to overheating. Is there another way to reach you for details?

Cheers,
Eric

 

Edit the balanced power plan and in Processor power management set the maximum processor state to less than 100%, for both battery and plugged in. That way you will prevent your computer from shutting down since you are throttling the cpu.

 

Thanks, Paloseco. I will definitely give that a try while I continue researching a longer-term fix. Seems like changing the thermal paste will be the ultimate solution.

 

I'm afraid of breaking something if replace the thermal paste. Probably I will look for a newer computer like the Dell XPS 15 with 4k display or something like that.

 

Well, I'm in the same boat in the sense that if I can't fix this machine I will end up replacing it. So I figure there's nothing to lose by trying first to renew the thermal paste. In addition to this thread (which is excellent), I found the following helpful video, which shows the disassembly of a Z11:

The XPS 15 you mention looks pretty sweet. I wouldn't consider it a direct replacement for the VPC-Z, however, since it is nearly a pound heavier, 1.5" wider, and 1" deeper. The VPC-Z is primarily my wife's computer and she loves how small and handy it is.

 

There's also the Dell XPS 13 which is comparable to the Vaio Z13. However, doesn't have discrete graphics card (and not sure if 4K display even). The XPS is on the same level as the Macbook, depending on your taste.

 

It is not terribly difficult to change the paste. I encounter a lot of dust, so I need to clean the fan etc. every few months. Naturally, that entails changing the paste as well. The main points to keep in mind are:
1. Remember where the screws go. Lay them out on a sheet of paper in the pattern that they are mounted in. Scan and print the bottom and place the screws on that.
2. Remember the order in which you remove the various hardware modules. Some of them overlap, and won't go back in unless you put them back in reverse order. Note the dressing of the wires, especially the ones from the wifi module.
3. Fan blades are hard plastic - will break before they bend.
4. If you do it slowly and carefully, the GPU thermal pad is reusable about 3 or 4 times. Eventually it tears and then you need a replacement pad or the copper shim.
5. The old (original) thermal paste on the CPU will be rock hard. Will take a lot of hard work to remove if you don't have the special thermal paste remover fluid. Do not scratch the shiny surface of the CPU, or the flat surfaces of the heat sink.
6. Even the generic white or grey thermal grease works a lot better than one would think. I haven't bothered buying the specialty thermal pastes. If you do a lot of gaming, etc. you might get another 4-5 degrees with Arctic Silver or similar, but if you need to do it often, like I do because of the dust, then a grease that is easy to remove is recommended.
7. The black cloth adhesive tape used to seal the gaps between the heatsink and the case and PCB will now be dry. Replace.
8. If you unplugged the fan, remember to plug it back in before closing things up. Been there; done that.

 

Thanks for the tip. That's definitely looks like a good replacement for the VCP-Z1 - slightly shorter, narrower, and lighter, with a display that's just as large. I don't think we'd miss the discrete graphics card. We do a lot of streaming video, but definitely no gaming in this household. You're right there's no 4K display, but I think that'd be lost on a screen of this size anyway.

 

Thank you, anytimer.
1. That's smart, simple advice. I've had the bottom off a few times and would have benefited from that idea before!
4. Do you use a generic thermal pad for the GPU, or something specific? That sounds like it might be easier than the copper shim, if you've found it to work just as well.
6. It hadn't occurred to me that a generic thermal paste might be easier to remove in the future. Good tip.
7. What have you used for the cloth adhesive tape? I'm imagining black electrician's tape, but suspect that's not the same...
Cheers!

 

4. I'm now using a copper shim. The first few times I did this, the GPU thermal pad was in good enough shape to be resused. Then it tore beyond hope. The pads for the GPU are available as spares from Sony, but naturally, they're not cheap.
7. I use medical sticking plaster - the stuff they use for bandages etc. LOL. You can use anything you like - it is just to close the gaps so the air from the fan doesn't leak and is forced to go through the cooling fins and out the vent. Leaving the gaps un-taped will just reduce the cooling efficiency - not a problem unless you work in a warm environment. I expect spongy adhesive tape will be fine.

 

Thanks again, anytimer.
4. Sounds like a copper shim is the best long-term solution, then.
7. Hah - medical tape. Sounds perfectly fit to purpose!
Cheers,
Eric

 

After replacing the thermal paste and clean up the dust, my Z1 is working as it should.

My question is: If the fan is now working, able to blow heat air out, keeping the full loading peak temp at 100C, do I still need to lubricate the fan motor?

If yes, what material I can buy?

Thanks

 

100 ºC is extremely high, to the point of shutting down the computer. Is that the best you can get? I use Maximum CPU State to keep it at 95 ºC or so.

 

It has been improved a lot.

For last few hours, sometimes idle, sometimes have loading, frequently turbo boost to 3GHz
min 33C, Max 98C
Average around 55-65C

Even playing 10bit HEVC 1080P video is max below 100C no frame skipping, fan high speed

I tried getting a fan cool pad, but no use at all, running the same light internet browsing is 50-55C fan low speed