Clevo P150 and P170 cooling systems neglected by Clevo! (solutions here)

Yeah I can see now they cheaped out on the 7970m gpu cooler by making the brackets thinner. I guess that explains why users are putting extra pressure down to get cooler temperatures.
There is also something else different in those pictures. If you go to last picture at bottom you can see aluminium clip? connecting side brackets on 7970 whilst 485 are missing probably due to extra width on brackets.
Just opened my 6990m cooler and exactly same as 7970 cooler but running much cooler than 7970.

Sent from my GT-I9000 using Tapatalk 2

 

I notice several mentioning the fan not going to 100% until certain temps. You force the fan to 100% by clicking FN+1 notice I said 1 NOT F1.

This works on mine, but never use as I never get close to 90C. I have tried to make mine reach the infamous 90C, but can't get past 85C so far.

If that does not work on yours maybe / likely there are a few different versions of the bios/ec firmware. Which may also contribute to some having better temps.

Sent from my PC36100 using Tapatalk 2

 

Does your 1 key have an icon on it that suggests it has a function action? Because the Clevo P150EM doesn't have that. The 1 key is just a normal key for me. And I did try it last week it never did anything.

 

That then probably explains why users are getting high temperatures on their 7970 cards. Interesting whether this is hardware change that some EM users get the Fn +1 full speed fan whilst others get nothing.
Sent from my GT-I9000 using Tapatalk 2

 

i bet its a bios thing. whats your bios that has the fn+1

 

Assuming you use the thinest pad @ 0.5mm and that it crushes to 0.25mm, and assuming a die area of 500mm^2, your going to transfer 14W/K.

Assuming IC7 cures to a thickness of 0.05mm with same die area, you are going to transfer 45W/K.

I think IC7 wins.

Thickness of the TIM layer is the most critical factor no matter what you use.

 

Fn+1 does not work for mine and it's a Myth system.

 

I have a Sager who does their own bios generally as well and as far back as I can remember has always had the manual fan control.

My BIOS Versions and dates...

BIOS: 1.00.05LS1
EC: 1.00.06SA

 

Do you have the files for those anywhere? If so, would you mind uploading them to a site like upit.cc or Free File Hosting - Online Storage; Upload Mp3, Videos, Music. Backup Files or Gaming News | Game Reviews | Game Videos | Downloads |Mods | FileFront ? I can't seem to get a matching pair of SAGER (note, Sager, not Clevo) EC and BIOS anywhere.

 

I don't have a copy of those as they came with the system.

 

hmm this sager bios i would be careful about just flashing it, as there could be problems, talk to an expert like prema and ask what his opinion is. maybe he can tell us the difference between the two.

 

Yes I have read a few post where you have to flash to Sager a certain way as to not brick you system. It was a simple method posted by Pream, I believe. You can change between the various Clevo bios', but it is always at your own risk.

My previous system was a Clevo, but I switched to the Sager bios then because of the fan control.

 

Eh... whats the formula for calculating w/k then? I'm wayyyyyyyyyyy off lol. And thanks for this info!

I think its a better idea we give Clevo some time to fix this. As far as I'm aware, they are aware of this issue since many people complain of it.

 

Good News, I have just done two of the above mods but with a small change. I actually taped the fan and the radiators completely around top and bottom and to the end of the outside radiator.

Here were the numbers on two different P150em's with the Nvida 675M's.

Before Tape Mod
Laptop 1: At Idle 47C
Laptop 1: Kombustor Stress Test at 92C 6:30

After Tape Mod
Laptop 1: At Idle 42-41C (fluctuated)
Laptop 1: Kombustor Stress Test, 82C at 6:30 and 84C at 10:00

Before Tape Mod
Laptop 2: At Idle 46C
Laptop 2: Kombustor Stress Test, 91C

After Tape Mod and Artic Silver 5
Laptop 2: at Idle 41-40C fluctuated
Laptop 2: At 6:30 81C and at 10:30 83C

Basically a -10C or a 18degree Fahrenheit drop in temp.

 

Allright, after much research for my next purchase of the best TIM available relative to longevity, heat transfer ability, and efficiency degradation from over-application to fill large gaps (which is needed for this situation with an uneven heatsink) I have come back in a full circle to and chosen IC Diamond TIM.

Why?

- Since we have a problem with a flexible uneven mount, IC Diamond is the best because it is unique in the fact that it has been proven that it does not suffer in efficiency from thicker application than normal like other TIMs do. It is much more tolerant to larger gaps that typical TIMs can't fill efficiently. Now, this doesn't mean pour a bucket of the stuff on, but you can easily use a pea size with what you'd normally apply a rice grain, I.E. grease which depends on being very thin for optimal operation. I need thick TIM, especially for this uneven heatstink!

- Once cured, IC Diamond is a very strong-gripped heat efficient sludge that will seal the GPU to the heatsink firmly, and really put a leash that untamed heatstink. This will ensure the best chances of sustained closure even if the weight of the heatsink is being strained.


Also, for those who have well made even heatsink surfaces, I find that Phobya HE Grease Extreme is pretty much the best performing TIM paste available (even if by just 1 degree) and also the easiest to apply. It is consistently a top performer in many reviews across the internet. Ive also noticed a pattern in all of the tests that have Phobya HE Greast is that it stays in the top 3 even more consistently performing than the other tops like Tuniq, Artic Silver, GelID, etc. The others bounce around a lot in the top 10 and don't ever seem to stay in a single place, which the HE Grease does.



I will post temps when I get it in the mail and applied optimally

 

That or figure out the dimensions for a copper shim

 

Sounds like a great prospect for those willing to do that, but (stupid erased comment here)

 

Not sure what you're referring to, but I just meant basically putting down paste, then a thin square of copper, paste, then sandwhich it down under the heatsink.

HP DV9000 Series Heatsink Mod - YouTube

Not the best example as that guy is a bit odd, but it shows the idea. Would help with heat transfer and make sure you have the right fit from the heatsink (pressure) if you suspect it to be loose..

 

Ah I see. Well, personally to me that seems to be high risk for a mess, knowing myself. Lol. I'll try simple and straight forward first.

 

You put washer only on these two corners ?

 

You multiply by the area and divide by the thickness. Not necessarily intuitive but: then you can interpret the unit of W/K as the power transfer through the TIM when there is a temperature gradient of 1°K established across the TIM.

In theory of course... Given the problem you have outline in this thread, I would not be surprised if someone were to reports better results with a thermal pad (vs. unsuccessful attemps with thermal grease).

This is interesting and would explain a lot of widely varying results people obtain with various compounds.

It's really unfortunate that aftermarker TIM products like these are required as workarounds for ancient design flaws.

 

OK so online it is said that the AMD 7970M has a die surface of 212 mm²

So X * 212 mm² / T = W/K

I'm not seeing what X is supposed to be though. What is it?

 

X is the "thermal conductivity" constant of the material in units of W/m·K (Watts per meter-Kelvin).

So you had mentionned 7 W/m·K for the thermal pad, and 4.5 W/m·K for ICD7; these would be "X" in this case.

 

Correct. The heatsink has only 2 available ring tabs in opposite corners to use. Without the main tabs on, it looks just like I drew in the diagram.

Oh, right. OK thanks

 

did not wok fo me

 

Awaiting TIM to arrive... hopefully it comes today

 

Hi Guys!

I recently received a NP9150 with the 7970m from xoticpc. I'm getting very high temps with almost everything I run with the 7970m.

BF3 temps get up to 90c with a thermaltake cooler supporting the whole laptop.

I had the IC diamond with the package and I think I wasted money on that. So I repasted with thermaltake tg-2. Still getting high temps. I also did the foiled method, and it helped a bit. But still quite high because I get 85-88c on bf3 and i'm not in full screen mode.

I thought that I botched my paste job, so I did it again with the same thermal paste. Still no improvements. MSI kumbustor test brings my laptop to 90c in 1 minute with max fans. (fn+1)

Will try with a different brand this afternoon. prolimatech pk-1. Hopefully this works out better. Oh, and I have constant ambiant temps at 27c because I work in an environmentally controlled office. Lots of servers and stuff.

I'm thinking that it's a driver issue. but I might be wrong because diablo 3 gets really hot at 85c and it's not even graphics intensive.

I'm just sharing my experience here and I will try some of your solutions. Will post later if new paste is actually better. Or if my card just runs really hot..

EDIT: okay, just repasted with a prolimatech PK-1. Temps were very good. Was able to get the card to 84-85c and fluctuating in 10min kumbustor burn in test. Paste method was dot in the middle, put in the heatsink, no pressure applied, let the paste distribute itself by tightening the screws. So far so good. Lemme know if you have more suggestions in dropping the temps.. =)

 

Why not get an ic diamond? Oh and I am still waiting for my laptop so I can mod the vent port >.>

 

The 2 biggest issues discovered are that the heatsink isn't being pressured enough, and the lid of the notebook restricts airflow greatly. It sounds like those are the 2 things you didn't try... so... try them?

 

guys if u can get a spare bottom cover try my mod!!it dropped my temps 7-8 celsius.

http://forum.notebookreview.com/sager-clevo/619377-p151hm1-np8130-cooling-mod.html

 

Beefsticks, excellent troubleshooting/writeup. +rep.

My thoughts on the spring clips were also that they seemed a bit "wimpy" for lack of a better term. I want a strong force pushing down on that heatsink, there should be little to no temperature variation with any kind of hand pressure that can be applied to the heatsink. That is clearly an indication of an improperly designed or built heatsink.

Secondly, you are spot on with the fan grille on the bottom. I see this on a lot of laptops and first thing I always want to do is cut it out completely. It's like breathing through a sock. They could just design in an X pattern in the plastic even But having tiny little holes and blocking part of the fan is not helping anything.

My system is running fairly cool. Other than the tape mod, at this point in time I'll likely not do anything else. Eventually I'll probably take a dremel to the bottom panel though and open it up a bit.

 

Yeah I've looked for spares online but specifically for the P150EM and I couldn't find any. I don't know what phrase to use to find them or if the notebook is even old enough to have the effect of spares being released for sale. Do other models share the exact same lid? Can you link me to spares for sale? That'd be great

Yes. Perfect description with few words. I used it for the description lol


And thanks for the rep! ... LOL it more than doubled mine

 

ask ur reseller to buy a spare bottom cover.i think they will help u!!

 

I could never picture myself cutting out parts of my bottom plate, but I could see sitting down with a bunch of patience, a hobby knife, and the bottom plate to continue to open up some of those closed "vent" holes and maybe work at widening the rest. I'd probably practice that on something else to see if it could be made well enough to not look cut up and tacky. Basically, taking lots of time to slowly cut and scrape it away vs just trying to cut or hack them out! Using the knife upside down and scoring/cutting lots of times vs cutting with the edge can yield some nice results. I've used this technique before while building scale car models. Sometimes I'll open make the cars have opening doors even when the model didn't call for it. Requires lots of patience and custom work but its doable!

 

we can downclock if You dont need full power and drop 6-8C
I try that in Witchr 2 and still was over 30fps

 

Hi!

Thank you Beefsticks! I tried bending the arms of the heatsink (where you put in the screws) upwards to create more pressure. Repasted it. Now I'm getting very good temps. running diablo 3 highest settings with 66-67c max. Before the repaste I was getting 80-85.

Also tried bf3 on ultra settings I get to around 71-73c. Max fans (fn + 1) have not yen been enabled, If I do I get it down to around 68-69c.

Assassins creed 3 gets to around 68c. and with max fans I get it down to 63-64c. All in all i'm good and happy! =)

Edit: Ambiant temp while doing all these gaming were about 30-31c. And I used a thermaltake Massive 23GT as a base. And I applied the Foil method. =) Got myself an extra roll of aluminum tape now lying in my room. And I dont know what to do with it anymore.. =)

 

I think this would be a safe bet for those looking online for washers; I just made my order: http://www.amazon.com/dp/B004K1FDI4/ref=biss_dp_t_asn
From the picture the OP's washer seems a bit larger than 7mm (.276") outside diameter and a bit larger than 3mm (.118") inside diameter, which should be almost exactly the same as the washers in this linked pack on amazon. I am assuming that 0.032-0.040" thickness means that half are .032 (same as OP's) while half are .040, still a good thickness.

 

Those look good. Although, you'll have 96-98 leftover washers

 

Not sure if this was mentioned in this thread before. But when you say "Weight pulling this side up" on the right side of the GPU heatsink, it's actually in reverse when you put your notebook right side up. If you know what I mean, when the notebook is right side up the left side of GPU heatsink is the one that's being pulled off the GPU. (left/right with respect to the picture)

 

Yeah maverick and I walked through that... I thought I mentioned it in an edit?

Edit: I guess its not quite as apparent as I thought. I kinda just worked in "Now do it right side up" but didn't elaborate much on why. I made it more obvious, thanks.

 

Getting a bit confused with the washer deal at the moment. So you need to put washers on both corners with the tabs? And how much does this actually help. I ran Kombustor to test this and pushing on the corner only dropped the temperature 1, maybe 2 degrees. It's hard to tell what drops are caused by the pressure and which ones are caused by having the cover off.

I finally did the foil mod earlier today after getting my hands on a roll. Temperatures in BF3 dropped from 81-83 C to about 74-77 C. Good deal.

 

it sounds to me like your system already has good heatsink pressure so you wont have the same benefits others are seeing. people with bad pressure see 84+ after the foil mod

 

i put washers but my temp was same in furmark 90C
in games is 75-85C it all depent what game I play

 

well i think anything under 85 is fine...
just to be an the safe side i can use any aluminum foil i can pick up in a grocery store, or do i need some kind of special one?

 

The 7970m can handle up to 105 degrees celsius before it borks, so it's 15 away from that limit.

 

You need foil with an adhesive backing on it. I'm not sure if they sell adhesive aluminum foil in grocery stores (I've never actually looked).

I went to Home Depot and found a roll of aluminum foil tape for about $8, not including tax. It's this one actually. 150 ft long, although you only need a piece about 3 inches long to tape everything up.

 

where d'ya hear that? 105 seems a bit much imo...

 

I heard it from my builder. It seems reasonable as the max temps of most CPUs are the same.

90C may not be very good in the longterm but the cooling system can keep it under control at least.

 

It all depends on where the uneven pressure is located normally on your GPU, and where you need to apply the washers to counter act that pressure. For me, it was only needed on one corner, and I got -10C from it. The opposite corner only gave me -1C and if I pushed too hard it'd start to actually heat up. That told me theres a big problem with uneven pressure on mine. It shouldnt be hard to figure out which drops are from pressure since the pressure testing shows immediate results. Just give it 5 minutes for temps to stabilize after you've taken the lid off. If temps only dropped 1-2C no matter where you pushed, your cooling heatsink has been applied more evenly than many others'.

 

Yeah, I let the temperatures stabilize after taking the cover off. It stabilized to about 70 C, after pressure on the one you got -10 C with it was fluctuating between 68 and 69 C. I didn't try any of the other corners though.