M1530 Thermal grease, pad? worth it?

hey happy new year to y'all heres the thing so since dell replaced my mobo recently i kept getting high temps and they sent new heatsink with thermal pads and that kept my temps down, but as before the problem may come again so i got Arctic silver 5 thermal grease, im gonna stick now to the dells pads until i notice higher temps, so what do you suggest, switch to Artic silver? i understand thermal grease dries up, what about the thermal pads? they deteriorate? snce when i got my system i wasnt getting high temps but over two months it started going up and up and without thermal pad or grease gpu will hit 100 c... so what do you suggest in this case? thanks in advance!

P.S: i already now this will "void" my warrantly for those who will post: "that will void your warrantly" ^_^ did it in the past technician came didnt even noticed, or cared, and still have warrantly n_n

 

Jajaja.. I even change the color of my LCD lid (from Black to Red) and the technician didn't noticed!!!!

 

That's exterior cosmetics, same as scratching it with a crowbar, it doesn't void warranty.

 

technically, changing the lid requires disassembling the entire display stuff.... so technically playing around with the insides of your computer doesnt void warranty, so technically if you remove the heatsink, u don't reuse the thermal grease... so you put on new thermal grease.. so....

just do it if ur that anal.... and if you want to put a cooler machine on your lap (i usually use the sleeve as a cushion between my m1330 and my lap)

 

dramatically lowers temps during load. i just made a thread about it recently....

http://forum.notebookreview.com/showthread.php?t=337934

 

"without thermal pad or grease gpu will hit 100 c".

I would like to remind everyone that 100c is 212f and you now have the temperature of boiling water in your computer.

I would do something about this. Forget about what the manufacture posts as specs. That heat is going to equal a short lifespan for that chip.

I can tell you that I have seen a brand new heatsink for the M1530 and I see what Dell's mistake is. I don't think it's grease. The unit comes with these cheap pads. When I disassembled my heatsink I saw what these pads do. They melt and the contents spill over the sides of the chip where they act as nice insulators, further heating up the chip. There is no more of the pad on top of the chip and there are large voids (where you can see there was burning due to lack of heat compound) which look like air pockets.

Whether these things are pads or thick squares of crummy grease, I can't be sure. All that I am sure of is that it doesn't doe it's job and it ends up on the sides of the chips.

I applied Dynex silver paste from Best Buy and the results were:

Before
CPU low:55c load:75c high:80c
GPU low:65c load:88c high:94c

After
CPU low:45c load:65c high:67c
GPU low:54c load:75c high:77c

The results in video games and 3dmark were just as shocking as the temperature drops. No more downclocking = no more stuttering in games.

 

While I can see that a lot of people do have great success with changing the paste, I recently did it to my XPS 1530 and it has made no difference to the overall temps whatsoever. My CPU/GPU still idle at ~60-65 and reach 100 under load.

I'm always open to the suggestion that I made a mistake (it's not rocket science, however), but just keep in mind changing the paste isn't going to be the final fix for a more complicated cooling issue your laptop may have.

 

Changing the thermal paste will only have a noticeable difference if the existing paste is inadequate in some way (quality, application, etc).

Thermal paste is not some magic solution that lowers temps, all it does is try to achieve a better surface interface between the chip and heatsink.

So the question to ask is - is there a problem with your existing thermal paste? The answer may be yes, it may have been sloppily applied, or may be an inferior product. Or it could be perfectly acceptable and you may see almost NO difference by re-doing it yourself. This is why you see varied results from people here.

Also note that the success depends on the application. I have seen/read some reports of people who clearly had no idea what they were doing ( 'the temps were not as low as I hoped, so I applied some more paste'....!?) so if you decide to do it, make sure you know what you're doing, why you're doing it, and how to do it.

 

I'll have to repeat myself I guess:

Aparaently this is how all of the M1530's are done. I spoke to a tech about it. The only time real compound is used is when they come to replace your board or heatsink. That likely explains why people see temperature improvement after a tech has replaced their board or heatsink. The stock crud is removed by the tech, then heatsink compound is applied.

No one is talking about "magic solutions". If you even read my post, you would see the dramatic difference in temperature reduction from proper compound application.

I can assure you that something is wrong. Either you have smothered the chips with compound, poor quality compound was used, or there is uneven contact with the chips and the heat plates of the heatsink. It could also be a fan problem. If this is done correctly you will never see such high temps again.

There are no hidden variables here. The 8600m GT consumes 20 - 23 watts at full load in every chip. That means that the cooler has to disipate the same amount of heat in every M1530. There is no variation other than the speed of your CPU, and even then the temperature delta between chips is negligable.

The only variable that exists is the three different revisions of the heatsink. I have the worst model (the one with aluminum heat plates) so those temps that I list are good reference numbers for what you can expect if you properly do the mod. I will be receiving the version with copper heat plates soon and after testing I will post the difference between the performance of both heatsinks.

Edit: Another possible flaw in the heatsink assembly would be a hole or crack somewhere in the heatpipe. This would definitely create out of control temps.

 

Im not even replying to your post, I would have quoted it if I was.

Im replying to the OP with some general comments....

 

My mistake. Sorry bout that. You made good points anyway about production quality and user error during compound replacement.

 

This is going to be the case most of the time (quality/application). Dell has a production issue with the bond between the heat plates and the chips. For every person that has cleaned everything and replaced the heatsink compound, and provided there is no additional problem, there is an overwhelming success rate.

In my case the CPU max dropped 13c and the GPU max dropped 17c. That is a dramatic decrease in temps and it was soley a result of compound replacement. Consistently, other M1530 owners are reporting the same results with properly performing this mod.

To the OP - yes it is absolutely worth it to replace the compound if it is the only problem.

Edit: I forgot to add that I also undervolted. It definitely makes a difference for temperatures.

 

Ok i understand about the appliation, but now that dell replaced my heatsink and thermal pads, im getting good temps.... i use to get 97 in the gpu, now i get

cpu idle: 38 high : 56
gpu idle: 58 - 60 high 75 - overclocked 78
any thoughts about this temps?

so the question is... i think this temps are ok, cpu is undervolted, the thermal pads are gonna keep it that way or are gonna deteriorate, cause when i got my computer i had normal temps but in two months, they started going up and up.... so bought the Artic silverin case this happens again
i had mysystem running for two mroe months in this extrem temps, and chip didnt broke...

Dude i agree with u, the heatsink is all messed up, is not the proper design there should be a copper shared by a cpu and chipset and another copper alone for the gpu, besides i think the old heatsink that had cooper contacts with the gpu and chipset were there best, now the cpu only gets coppr and other two get aluminium.... about the chip lifetime, i think is a high performance chip, i mean is heavy graphics processor and in order to work it needs proper cooling wich is no provided by dell, the same would happen to a Intel processor withoud thermal paste, and proper heatsink, but for example my T8100 is 45 nm process fabrication, has some features and use less power to operate so it wont go to thse temps, so fast, but it will get to those temps, mine once did hit 105c, aother thing i noticed was that new heatsink fan runs a lo faster, my old fan probably wasnt doing its work... didnt run at proper rpms... this one... really kicks... and u can see my temps on top of the post... they are the same or even better than yours... the thing is that i hope this is constant and wont get worse again... about the nvidia chips i wouldnt worry if mine ran at extreme temps for two months... that shoul be fixed... thank u for the responses! tell me more ^_^

 

You are correct about the need for separate heat pipes. This shared design is indeed a problem since the CPU and the GPU can pass off heat to each other. They really should be separate.

Your present temps are right on par with mine and we have the same heatsink (aluminum heat plates). Watch the temps for a week or so while the thermal compound 'cures'. If your temps are going to go northward again it likely will happen soon. I used cheap Dynex Silver Paste on mine and it works beautifully. I would expect that as long as the tech did a good enough job to bring your temps down to where they are now, the tech did a sound job and I would not expect your temps to rise. You have to use really poor quality paste for it to evacuate the area between the heat plate and the chip.

But one thing that I can't stress enough is that if your CPU or GPU passes 90c you should investigate immediately. Those temperatures are not normal, and not healthy for the chips. You have to disregard the manufacturers data about maximum safe temps. It only takes one transistor to pop on your chip and afterward it will never be the same, or it will fail all together.

The Nvidia GPU's will not run stable and you will experience repeated downclocking when the chip gets hotter than 85c. It only takes a little temp reduction (down to 80c or less) to correct this. This tells us that the real safe operating temp for the 8600m gt is 80c or less. But it also tells us that 80c is the roof. We really want to be 10c lower than that at 70c or lower.

When Nvidia says that the 8600m gt will operate safely even in the 90c + range, what they are saying is that the chip will continue to operate. What they do not come out and say is that the chip will indeed downclock. This is certainly not the behavior we're looking for.

I can't wait for my copper heat plate heatsink to arrive. After testing that unit, we will have some answers as to the difference in performance between the heatsink revisions.

Since you have undervolted and had the heatsink reinstalled you are doing the best that you can do without heavy mods to the heatsink assembly. I would say that you're in the clear.

 

This goes back to the age old question of 'is there something wrong with my XPX 1530'. There's about 10 pages worth of them here.

Rephrasing my original; I replaced the dried up thermal grease and my abnormally high temps are no lower. The CPU still down clocks and the palm rest still burns.

 

"Rephrasing my original; I replaced the dried up thermal grease and my abnormally high temps are no lower. The CPU still down clocks and the palm rest still burns."

I'm going to say that either the surface of the heatplates is not flush with the chips or the heatpipe has a hole or crack. This assumes that your heatsink assembly is clean and has a properly operating fan.

It would have to be a REALLY bad job of compound application to produce the temps that you speak of.

 

If I may ask, what tool/utility is most commonly used in this thread to gauge temperatures.

I just got my XPS and see it as a long-term investment worth maximizing the life cycle on. I'd rather give it a thorough assessment now than find fault with it later.

 

Here are 4 good choices - RMclock (you really need this one), Notebook Hardware Control, GPU-Z, and ATItool. Hardware Monitor is a 5th choice and is good because it is easy to generate log files, but it is not free.

You will need these tools in order to post temps on this board. Everyone here has sort of set them as the standard. We can all trust the temperature readings produced by the mentioned software.

 

yeah i think im safe or now... but i was safe before and took two month for the temps to raise... so i ll just wait and when i see any temp raise i ll apply the thermal Arctic Silver to the chips... my gpu speaking o downclocking... wouldnt downcock until reaching 96 degrees... once hitted 96 games FPS will drop ! but.... i wouldnt take more than 1 minute running a 3d application to get to those temps, i can say that even, playing a mpeg 2 video... in power dvd at 8000 kb/s the gpu reached 87 c! can u imagine? just playing a video... maybe for the Pure video decoding, but still, makes sense that it gets hot, is a high performance chip.... nad has nno heatspreader like some Pentium 4 Northwood that had headspreader and that made contact wih the heatsink, i have a pentium 4 running without thermal paste or compound and runs at very fine temps, this gpu has a manufacturin process of 80 nm, so that explains what it gets hot, faster than the 45 nm, T8100 intel cpu... anyway how come that there are some people that playing and stressing wont get past 60 c! so is mid 70's and 80? high? i had a Compaq pentium 3 800 mhz, was like 13 years old XD and running at 80 degrees idle, and hitting 100c almost always when browsing or whatever, and it survived years that way... so new chips should handle better right? and gpu, wich runs hotter than cpu, should be ready for this temps.... i mean physically, if 13 year old cpu...... 180 nm cpu, new chips, should handle much better.... right?

lets design some new heatsink for the dell m1530... with so many threads... we can be millionaires ... XD

 

If your GPU makes it to 96 before downclocking then you may have a really strong chip. I can't be certain, but according to my tests mine would downclock somewhere between 85c and 90c.

"has nno heatspreader like some Pentium 4 Northwood that had headspreader and that made contact wih the heatsink"

I think that this is the number 1 reason why these chips are having problems. But I will do my best to see what we can do about this. After I test the copper heat plate revision of the heatsink and record the results, I'm may try adding a tad more copper to it. I would like to know if the temps will further decrease by doing so.

There is one conclusion here, and that is the fact that the notebook OEM's have gotten out of control. For a few years now we have all been handed these ridiculously hot machines. These temperatures make the unit unreliable and uncomfortable to handle. If I had to guess at why, I would say that the first reason is cost and the second reason is weight.

I believe that they are trying to cut weight in these notebooks anywhere that they can, and it's at the cost of a hot machine. It's not a trade that is working in our favor.

 

Yes, there is some truth to that. Aluminium is about 35% less thermally conductive than copper. It is also the most abundant metal found in Earth making its cost cheaper. Though before Dell makes any modifications to the thermal system, they need to go through alot of formulas and calculations under the chip providers recommendations. I believe they dropped the ball on the TIM.

Some thermal pads are made out of parafin wax that melts into place and fills up the micro gaps. Sometimes it melts too much that it oozes to the side (from heat then applied pressure) and reducing contact surface to the heatsink insufficient after long term use.

How thick is Dells thermal pad anyway?

HWMonitor is free. Also use the quote button.

 

I have HWMonitor and it is not free. It's nagware. Constant popups reminding you to pay and register. In fact mine has expired and no longer will start.

If you know of a link to a free one, please advise.

 

what? the pro is paid man... but the HW monitor... is free i have it if u want it... complete temps of components, no pop ups, no registering... anyway u can try here http://www.cpuid.com/hwmonitor.php go down and choose either 64 or 32 bits and ready to go...

 

Thanks very much JAV1563! And thank you flipfire for pointing out the discrepancy.

I like this one much better than what I was using before. So I am mistaken and HWMonitor is actually free.

Edit: I see my mistake now. I had downloaded the 30 day trial and did not realize there was the
free version.

 

hahaha u are welcome! now speaking of faulty heatsinks, thermal paste and the thing i pointed earlier in the post, about the gpu in the m1530 having a separate copper heatsink, look at the new Studio XPS 16 review and look how they redesigned the heatsink, u can see the pic here:

http://www.notebookreview.com/shared/picture.asp?f=39915

see how processor and chipset share a cooper "verga" (i dont know how u call tha in english XD) and the gpu got itself copper stuff? so theres no transfer of heat from cpu or chipset to the gpu... exactly how i said itshould be in the m1530 o0! anyway with my temps not going anywhere past 80 in the gpu seems like a fair temp for this gpu, with people running their gpu at 90... hopefully this chip is as strong as las one, that lasted and kept going at 100 c for two months XD


P.S sorry i know is outside the context but why i have Rep power: 2 and i got less posts than slowdown? what does it mean

 

No problem slowdown, the regular version doesnt let you put the temps on the taskbar though =(

JAV: For every 150days of membership you get 1 rep power.

Rep power is used to give out reputation, which in turn gives you green boxes every 100points. Check out your userCP.

 

Thanks Flipfire! by the way my undervolted cpu is thanks to ur guide! maybe using, rmclock and rivatunner he can have the temp in the taskbar? o_0 anyway i have it always minimized or when i feel the notebook hot, i run it just to check, or when im gaming so it records, my max temps while gaming! ...

 

Ahhh, well, after all heatsink and pipe talking (wich i found very interesting) i have a new prob for u guys that could sound like all other probs, temps..., but in my case whenever cpu (is my wild guess, monitoring with everest ultimate) reaches 85 or a little more my system shutsdown itself, it has happened to me for the fifth time this week and is driving me nuts, why i cant have very high temps frying my lappy and, well, frying my legs, but NO, instead it only shutsdown... what a pain in the arse huh?..., idle im having right now on gpu 56, cpu´s 48-50, i have no overclocking and the 179'48 nvidia drivers, so no clue there, my xps m1530 has the cooper heatsink cuz i ordered like 4 months and a half ago my XPS m1530, so im LOST out here on the shutdown prob, im gonna try the thermal paste apply and the undervolt, but for god sake i really dont want my system to blackout one more time playing FEAR 2 demo at medium settings..., any help is right welcome since ur experts on m1530 lappies as im seeing on these 3 pages, THX IN ADVANCE!

(i also have to say that i hear a "clik clak" sound coming from the area where cpu and gpu are, someone told me before that it could be the hard drive (at bios settings ive choose at the hard drive area "performance", not caring about a noisy hard drive..., perhaps that explains it, i have win vista 32 bit, bios A12), so it could be the hard drive too, but again, im lost here, pls help me out ...

*cpu voltage is at 1.3..
**hard drive temps are at 33

PS: I HAVE THE T-5750 CORE 2 DUO AND THE 8600M GT CARD WITH 2 GIGS OF RAM..

 

hmmm...

My CPU is reaching temps of about 92C and its not shutting off, and my GPU is getting 88C, and still dell insists on replacing my heatsink * again* sigh..... but at least i know that if my lappy doesnt turn on one day ITS DELL's FAULT!!! i hope your reading this *techie*

 

thats weird... i have personally reached 105 on cpu and 110 in gpu.... and still this gear keeps running.... when i got thermal pads removed by myself.... thought i wont need them, now i know i do, so u have the lastest bios?? maybe u got some problem with the temperature sensor... when u read 85 maybe the machine is reading 120 c for some issue.... in the board, try get the motherboard replaced by dell that should handle the problem if keep having the problem then the problem is the cpu... and remember theres other components that might cause the shut down, maybe the powercord.. who knows.... call dell and tell them about shutdowns and high temps, and they should send new motherboard, make sure they include new heatsink, to handle heat issues and because thermal pads will be dried up when they remove the heatsink.

Right now im running with arctic silver and with extreme overclocking my temps wont go past 65 on cpu and 85 in gpu... without overclock... wont g past 80 on gpu.... and 60 on cpu...

still u should keep working if u get cpu to 85.

JaV!

 

I second Jav's comment, you should call dell, tell them that your computer is shutting down while gaming, they should replace your mobo and probably also your heatsink...

those shutdowns happened to me but only while OCing, i figured my hardware was running unstable

dont make it your problem, when its clearly dell's responsibility to do something about it

Best of luck!

 

its very strange, i think its the gpu cuz ive done couple of stress tests to cpu and ram only, cpu reached 85°, it last like 20 mins or so and nothing happened, all cool, also it seems the shutdown is kinda random cuz ive stressed my lappy running fear 2 demo along with other graphic apps like nvidia´s models, etc, editing some photos, multitasking, i even ran out of ram cuz of the many apps i was running, and NOTHING HAPPENED..., and when i was just playing today at my table when i have dinner (nothing else opened or happening) it shutted down two times..., so ive just checked temps and it seems to happend whenever gpu reaches 82 or more, and that... i think its not normal at all, cuz u guys have taken ur card to the limit values like 90 or even 110° and everything seems ok (altough ur gpus are melting) but all is cool if u know what i mean, so i will take this advice u guys are telling me, the only thing that i will hate will be that id be without pc or my lappy for a while...,

 

Just as an aside, there's always this argument about not opening up and doing things to the computer because it'llvoid the warrantly. It came up when I said I'd re-done the thermal paste on my XPS 13. But what technically is the point where you've done too much and your warranty is void?

When I had the old grainy screen problem with an M1530, I told Dell I'd be happy to replace it myself instead of them sending a tech, and they happily sent me a replacement screen to replace myself. Now, it may have voided my warrantly if say I messed it up and broke something, but they never mentioned anything about the warranty.