LIVE!!! Calling all XM chip, 3930K/3960X and ultrabook owners! Heat-Spring TIMs for testing!

UPDATED: I made a form for everyone to fill out so I do not have to try to find all the info over 4 different PMs and 4 different emails. This will streamline the process so I can see everything in 1 email and send you samples as fast as possible!
Everything you need to do is in Requirements spoiler. I have made this now as easy as possible for both parties. Also check the document section for all the important info on how to apply Heat-Spring(still working on this one. I did find all my photos. Just going to put it in a word file) and how to use and read data on Prescale pressure paper.
I also added a checklist in a word file to make it as easy as possible.
I also added privacy info on bottom of page

All of the testing and it's data will be submitted to Indium Corporation since they are giving us few samples to get started with. I want this is be as scientific as possible with as little margin of error. I want these results to be accurate so I will create a how to apply (TIM) and further instructions on what is expected so we are all turning in results that are consistent, factual, and impartial.

Background

Spoiler

I have been talking to Jim at Indium Corporation about buying an extremely small batch of TIMs about 10 to test in several models of laptops but after talking to him he is going to give us SOME FREE SAMPLES!!! I am looking for XM chip owners that can overvolt and overclock that use ICD or anything that is considered one of the best TIMs on market. I want to compare the performance when idling and burn test like Intel burn test and F@H. I will elaborate on what is expected below. Also I want to try these TIMs in ultrabooks that are known to run hot. I need experienced users that are willing to log and take detailed notes and turn them into spread sheets. I will mail the TIMs if I pick you for testing. These TIMs are extremely expensive so please be careful with them. I won't be able to afford to buy more/get more free samples. They range from $2.5-$15 apiece for small batches depending on size. So please do not waste Indium Corporations money or mine if I pick you. Even though Indium Corporation is giving us free samples I am still spending probably $100-200 shipping them to everyone. Also I WILL be trying 2-4 desktops. The 2-4 desktops that I want to try are the 3930K and 3960X. I want to use these because they have the most heat production and are best ones to do an extreme test on them. The hotter it runs the better for the test. The difference between a 4.5w m/k and a 87w m/k is little when you are working with below 60C or at least from my experience with TIMs. It is ~$15 dollars for a single 2x2 piece with an order of 300 dollars (20 pieces) so I don’t want to waste a single piece.

The people I am looking for are:
3930K and 3960X owners (2-4 pieces)
9xx XM 29xx XM and 39xx XM owners (20-40 pieces)
People with ultrabooks that run hot ASUS Prime? Sager 3610QM and such. (5-10 pieces)


Note I am putting my money on that only people will experience enough of a difference to justify the cost are those with ultrabooks that have poor cooling and those with XM chips that are pushing extreme temps but we will see AKA anyone above 60C should see a noticeable difference...at least that is my guess. I tested this in my netbook and it was a bust because of the crap heatsink. I know it won't work if you have bad contact and low pressure. I tried some pressure paper and it didn't even read lol.

Spoiler

Word "Heat-Spring Checklist" file. Basically everything below but in a very easy to follow checklist. Feel free and print it off if that is easier for you.
Details of requirements to get in on free samples.
1. You send me "Heat Spring Application Form". Name it as "Heat Spring Application Form Username" Example, "Heat Spring Application Form HopelesslyFaithful"
2. You must pay for shipping if you are out of the United States. (Canada/Mexico will be considered case by case depending on costs to ship.)
3. You have to have said chips or a special interesting situation to try it
4. You have to be willing to overclock to the max with voltage increases. (looking for people who are able to push 80-100C in system)
5. Collect detailed temps and max temps, clocks, and voltages with current TIM and post HWinfo CSV file (ambient temps are a plus but not required. A general idea would be nice.)
6. Collect the exact same detailed temps and max temps, clocks, and voltages with Heat-Spring TIM and post HWinfo CSV file.
7. If effective push clocks and voltages to new max level with Heat-Spring TIM and post HWinfo CSV file.
-The point of 5, 6, and 7 is to see how much the temps drop and then see how much more speed you can get until you hit old temps.
8. I will do for you…. but I would like if you did it because it would save me a lot of time. Create a single graph comparing results.
Optional 9. Run F@H when not using computer to monitor long term temps and see how well it works and lasts and post HWinfo CSV file
-The point of this is to see how well Heat-Spring TIM holds over long term. This TIM should never have to be replaced.
10. I will complete on my own with my samples. I will be testing how well the material works with reusing it. As long as you don't crinkle/fold it. You should be able to reapply it. The material is a very fine sheet. It should stick to your copper heatsink...maybe permanently. As in you have to lightly scrape it off-indium diffuses with copper. Heat-Spring has a special coating to prevent this but it still sticks from what I have been told/understand.
11. I will do on my own which is make a complete graph comparing all results.
12. Take pictures and detailed notes of system specs and cooling system. I want to know what systems it was in and how it performed. I will have data accessible for each test for review and for combined results. If you can make a Power Point type show on your laptop that would be great!
13.I would like everyone to make a Power Point slide show of your system with results. I posted an example of my netbook so follow that. I'll make a new one with my other systems that'll give you a better idea later.

HWinfo tutorial requirements
Remember use file naming format listed below in spoiler.-* denotes 100% mandatory.
What I want to be recorded is a separate .csv file that has 5 mins of idle with current TIM and Heat-Spring.*
What I want to be recorded is a separate .csv file that has 1-4 week of usage with with current TIM and Heat-Spring. (I would like people to do long term test with both but if you refuse to do so let me know. We will work something out.)
What I want to be recorded is a separate .csv file that has minimum of 15-60 mins of burn testing using Intel burn test and F@H with current TIM and Heat-Spring.*
What I want to be recorded is a separate .csv file that has 5 mins of idle with Heat-Spring using new max clocks and voltages if Heat-Spring is effective.*
What I want to be recorded is a separate .csv file that has minimum of 15-60 mins of burn testing using Intel burn test and F@H with Heat-Spring using new max clocks and voltages if Heat-Spring is effective.*
I want idles, burn test, and long term tests done before I mail you samples. Along with PPS or something put together with description/pics of machine and cooling system.*
Also note I changed it. I want burn tests with F@H/Prime95 which ever you can do and Intel Burn test. Intel burn test destroys CPUs and is a great test but not 100% consistent, which is why I want both. Now if you are running 80C+ with Prime/F@H do not do Intel Burn test because you will hit TJMax! Test it with Heat-Spring after you change it if it works.

**Also when I say make at least 15 min run please make sure its at least 15 mins at max temp. It can take a computer a good 5-10 mins depending on ambient temps and cooling system to reach "max temps" so i need a good 15 mins at max temps to make a good graph for comparision. I just wanted to clarify that because i had to have someone remake a log because it was not consistent. Thanks and hope this clears everything up!**

Required Documents to be Filled out:
"Heat Spring Application Form HopelesslyFaithful"-example

"Heat Spring Application Form"-blank

HWinfo tutorial THANKS to STiHiaL!!!!-minor edits by myself

Spoiler

HWinfo logging to file with format .CSV which can be read with Microsoft Excel, and maybe it can be read with similar program like Openoffice/Libreoffice. Unclaimed data can be deleted or hidden.

1. You need to download HWInfo ( HWiNFO, HWiNFO32 & HWiNFO64 - Hardware Information and Analysis Tools)

2. Starting HWInfo and opening Sensors screen:


3. Configuring Scan Interval (changing value in ms / 1000 ms = 1 s) and Sensors Display (while disabling/hiding unneeded sensors)
Set recording to 2 second intervals so 2,000ms. Record everything because for me when I disable stuff it still records so don't bother disabling anything. It is fairly easy to work with it in excel.



4. Start logging and type the name of logfile


5. Press Stop logging (when you finish logging) to save the logfile.

Name file with "USERNAME" "CPU" "TIM" "dates of test".csv Example, "HOPELESSLYFAITHFUL 920XM ICD 20120901 20121001.csv"

Myself and anyone else who is going to help will turn log files to charts/graphs

What is Heat-Spring?
It is a TIM from Indium Corporation. It is made out of Indium and has a thermal conductivity of 87w m/k from what I have read and been told. It comes as a thin sheet and is very malleable. I will update further later and provide as much data as I have. Most is from Indium Corporation since they are one of the only companies that test and make stuff with it.

Pros:
High thermal conductivity - 87w m/k. 19 times higher than ICD. 10 times higher than MX-4. 10 times better than AS-5 (AS-5 is 8.7w m/k????? better than ICD and equal to MX-4????)
Very low thermal resistance 40PSI+ (highly dependent on pressure will include image later. Image is in page 1-3 of thread)
Reusable - it "should" stick to copper heatsink after applied and burned in. Gives the ability to switch out CPUs and other benefits like saved time because you do not need to reapply.
Durability - It has showed in other tests to have virtually no loss in thermal resistance and thermal conductivity after long term use. Which means in the laptop and computer industry apply once and forget. This is largely do to from my understand that it is just pure Indium and a sheet of TIM not a paste so no worries of drying out or air bubbles.
????
????

Cons:
Expensive
Delicate - if it gets folded it will stick together and be unusable and (easy to tear???)
Conductive - not a big deal because there is no dripping or run off since it is a solid piece
Custom size - You have to order or cut to the exact size of CPU which may lead to waste if not taken into account.
???



If you know of a pro and/or con please PM or post and I'll add and double check it.

Die sizes/IHS size:
9xxXM - .5x.875" (1/2x7/8")
29xxXM - 12 x 24mm???
39xxXM - 10 x 22mm???
39xxK/X - 38mmx38mm (1.5x1.5")

Still need testers!!!-I have already mailed two sets of samples out but still have plenty left!

Current Testers: This list is not guaranteed until I finish the list of requirements and instructions and everyone agrees to it. So keep that in mind and still apply. I will be picking the most qualified people. People in US are preferred due to shipping costs. If you live in another country and want to pay shipping let me know we can work it out. I think I am getting enough pieces for 2-4 desktops and 20-40 laptops.
Apu71 (3610QM and 680m Sager NP9150)
Myself (920XM and 3720qm)
Reborn (920xm and 2 2920xm?)
????
????
????
????
????
????
????
????
????

People who said they want to but have not gotten back to me with address and test logs
eKretz (920XM)
extide (3930K)
STiHiaL (920XM)
SVL7????
DR650SE???
Jimbo???
Mobius 1-if repaste with ICD does not get better results than stock TIM


UPDATE: on a side note does anyone have a 3770K that has the garbage TIM in it and wants to try Heat-Spring? That might be interesting.

Graphs and Documentation
Please note all this documentation currently comes from Indium Corporation. It is not independent research or was independent research and is now owned/used by Indium Corporation. Some info down here I find a bit confusing and If anyone can better elaborate on it I would appreciate it! My biggest question/concern is the grease they use....what does 2 or 3 mil mean? How was that test run and how does lets say 3 mil affect bulk thermal resistance? Does 3 mil make it look worse than 1 mil? How does all that apply to TIMs for computers? I want to understand this data to make sure we have a good idea of this info and want to make sure we are not reading charts full of data that is lets say curved to show certain results
safety sheet of H-S

Prescale Ultra-low Film Documents/Directions
https://www.dropbox.com/s/3feb1xjrjxkkdzj/Prescale Ultra-low film documents 20121110232058.pdf

Safety Sheet
Dropbox
https://www.dropbox.com/s/8jq7ktherljg050/Indium Corporation safty sheet 20120913130753.pdf

Video of Heat-Spring
Overview of Indium Corporation's Heat-Spring® Thermal Interface Material

Visit this page for all content below and more.
Compressible TIM (SMA-TIM) Heat-Spring® - Thermal Interface Materials



1500 hour 90C bake test
http://documents.indium.com/qdynamo/download.php?docid=482

Graphs:

Spoiler

Glossary
I did not know what BLT meant...thought it was a sandwich...haha just kidding. Googled and found out it is Bond Line Thickness. Found this page with tons of terms. If you think there are more terms to add post/PM me and I'll add them.

NOMENCLATURE

Spoiler

Rcs - Contact resistance between two bare solids
RcTIM - Contact resistance of an ideal TIM
H - Hardness
P - Pressure
m - Mean asperity slope
kTIM - Thermal conductivity of the TIM
kp - Thermal conductivity of particles (fillers)
km - Thermal conductivity of the polymer matrix
RTIM - Thermal resistance of TIM (same as impedance)
BLT - Bond line thickness
Rb - Thermal boundary resistance
Rc - Contact resistance of TIM
DF - Density factor
Rjc - Junction to case thermal resistance
K - Consistency index in (5)
r - Radius of the substrate
C - Empirical constant in (7)
Rbulk - Bulk thermal resistance
Ea - Activation energy
A - Acceleration factor
G - Shear modulus
Refer to document - Storage shear modulus
Refer to document - Loss shear modulus

Greek - Refer to document. Forum does not support most of the characters.

Source-Ravi Prasher
http://weble.upc.es/ifsin/Block5/paper_proc7.pdf


If you contributed info and helped make this happen please PM me so I can add you to the list of contributors for credit!!!
STiHiaL-HWinfo tutorial
xxxxx
xxxxx
xxxxx
xxxxx



Archive

Spoiler

Note I am getting 9 2x2 samples but if anyone wants to do a large test of 50-100 people please read below!

Now if a ton of people want to try this and are willing to chip-in like 5 bucks or so for me to buy 20 2x2 pieces. I am willing to buy a full test kit worth but I need to know if people are willing to send me the few bucks for a full kit. I don't have $300 to fund a full kit....since I am still trying to sell my laptop and jobless. Now if someone buys my laptop...then I won't mind so much forking 300 bucks up for a test kit. So think of it if you buy it you are donating $300-$500 for me to buy a full kit and pay for shipping all the pieces to everyone.

Also since I am only able to fork out about 50-100 bucks for samples. If anyone wanted to just chip in a large chunk instead of their one piece that would be cool. I would be willing right now to chip in max 100 bucks if others chipped in 50-100 to reach the ~$300 for a full kit to test. Up to you guys.

Privacy Statement
I value your privacy very much so I take as many precaution as possible. I will never give out your address or real name. I will sensor everything I submit so that no private information is transferred to anyone. Your real name, address, and paypal accounts are all kept in a truecrypt file container that is triple encrypted so no one can ever get access to it but myself. For all testing charts only your NBR name will be used.

 

Well if you/they are right, it basicly have 10x better heat conductivity than the paste that is being used today (MX-4/AS 5). Why am I sceptical and why haven`t the desktop community jumped all over this?

Page 6 says the same though, 86 w/mk
http://www.dfrsolutions.com/uploads/white-papers/CPV.pdf

 

Sorry, but its not gonna work. Conductivity does not equal capacity.

 

largely due to cost and is only usable in extreme cases. The hotter something is the bigger the gap gets in TIMs. So if your running at 80-100C this should dramatically help. It is largely used in servers and testing. You can re-use the TIM because it sticks to the copper heatsink after you use it. So you can pull out the CPU and put a new one in and just re-tighten it down. This TIM should also never have to be repalced. ICD/AS-5/MX-4 start to wear out over time...this should not.

 

Meaning paste have the best capacity but worst conductivity but still is ahead compared to a product that have bad capacity and good conductivity?

Care to elaborate?

 

this is also really good not due to just conductivity but also due to very low resistance. Also more pressure you have th better it is. For this to be optimal you need like 40PSI and up if i remember. I'll post all the documentation i have gathered over the last 6 months

 

Well it sounds pretty interesting though. Looks to me that you need atleast 40PSI to equal thermal paste and gets better the more pressure you have on the metal foil (makes sense, more pressure, better contact between CPU/GPU and heatsink). Question though is how much pressure you can get with the screws over the heatsink over the CPU/GPU in notebooks..

Compressible TIM (SMA-TIM) Heat-Spring® - Thermal Interface Materials

 

in other words "crank those darn screws down!"

too bad we can't measure pressure in CPUs :/ would be great info to have!

 

Pressure shouldn't be a problem. According to the testing that Innovation Cooling has done, the notebook heatsinks they tested will press down at around 60-80psi on average, so if there is a tangible advantage for this TIM, it should be apparent.

I wouldn't mind chipping in a little and trying this out, if enough people join in and it's cheap enough.

 

There should be some random datapoints on contact force for various laptops in the following thread:

http://forum.notebookreview.com/hardware-components-aftermarket-upgrades/584682-ic-diamond-24-giveaway-reliability-survey-10.html#post8751039

http://forum.notebookreview.com/hardware-components-aftermarket-upgrades/584682-ic-diamond-24-giveaway-reliability-survey-9.html#post8676252

And here (concerns mostly desktop scenarios though):
TechPowerup IC Diamond Results Thread

 

so you're saying it could potentially work better than a tooth paste ?

 

I would love to see this being tested - should prove very interesting no matter which way the final results fall.

 

i and many others have made countless statements on how the testing of that one place is not even remotely accurate. I refuse to even humor such non-sense any further

everyone else that posted some data on pressure and links i'll add and log all of that later when i got more time to work on this

 

Er, thermal paste does not have thermal capacity, is all I'm saying. Your cooling system can still fill up with heat, so to speak.

 

Thermal paste is only meant to fill in the gaps between the chip and heat spreader, making heat transfer to the heatsink more efficient by eliminating air gaps. If you have too much thermal paste it can actually block heat and hinder the cooling system.
I didn't read whatever post your commenting on, but just thought I'd throw that in there.

 

he was saying there is a limit to what a cooling system can handle whether water pipes/radiator or heatpipes and fans. I am curious about that because i get that and dont get that for several reason. I am going to be vague and brief because i don't have time.

How can a heatpipe have a total limit of heat possible to transfer when all it does is lets say "collect heat and move it" to a heatsink and fan that blows it.
The hotter it gets the more heat a fan can blow off because of the difference in temps from ambient and heatsinks.

I know I have botched that explanation/question but I'll work on it later. Maybe someone got it and can elaborate for me.

EDIT: no one is up for testing? Also can anyone tell me or show me where it says what sizes the dies are in laptops for each XM chip. need to figure out what sizes to get whether it is better to get bigger chunks that i cut on my own or get smaller pieces for each chip.

 

Im in for 2. 3930k and 3920xm.

Ivy Bridge Mobile die is ~0.5"x1.5", fairly small as it is bare die.
Socket 2011/3930k is a LOT bigger, like 1.75-2" square or so as there is a heatspreader here.

 

exactly how big is the IHS on the 3930K? I need to know so I can get a big enough piece and how to cut it. I'll have to custom cut every CPU so at MOST give an extra .1 inch is measurement if you are using a ruler for the 3930K give it an EXTRA 1/16-1/8 extra in size. let me know how much extra you are including because I'll use that to judge how I want to cut your piece

 

This sounds like a awesome project and i hope we can all get some new data to read up on.

Currently im with the IC Diamond test & im also only using a i7 2720QM so no real reason to apply here.
Now if i only had a overclocked i7 2960XM.

Sent from my GT-I9300 using Tapatalk 2

 

Ok, I will have to measure it then.

 

thanks I appreciate it. Would you also check the 3920XM? that should be the exact same size as all other IB systems right? If so than I got the 9xxXM system covered becuase i own a 920xm ES but i still need to find out how big the SB is

 

Yeah, all the mobile Ivy Bridge Quad's use the same die.

 

Right, so yknyong got 80PSI with his 820QM, which means combined with this heat spring, it should be far superior to regular paste?

 

I believe the radiator is the only piece designed to have capacity, the heatpipes are mainly meant for conductivity as well. There are definitely all kinds of equations were missing out on here but you can also just think about it. The radiator has capacity and a rate of dissipation with the fan and total surface area of the radiator. I think the rate of dissipation is the true limiting factor.

I'm not sure if the capacity can "bottleneck" the cooling system or if you can improve a cooling system almost endlessly with more airflow. I feel like someone smarter than myself should jump in here.

 

Pipes do have limiting factor otherwise there wouldn't be 2 different heatsinks o VAIO F1x for I7 quad and I5 dual CPUs. Heatsink for I7 has 2 heat-pipes.

Pay attention that for desktop CPU you need to give 2 different pieces of this stuff.

 

That's only if you de-lid your CPU. I'm guessing most people won't bother doing that.

 

Then there is no point to test this stuff on them.

I still remember article about new Ivy or Bridge desktop CPUs having real heat troubles because of cheap and bad thermal grease exactly between that metal shield and chip

Ivy Bridge's heat problem is indeed caused by Intel's TIM choice :: TweakTown USA Edition

Intel's Ivy Bridge Core i7 3770K Overheating Issue Detailed

 

Most other Intel CPU's with heatspreaders (at least high end ones) including i7 2500/2600/2700, all LGA2011, higher end core 2 series, all LGA1366 are soldered to the heatspreader as opposed to them using thermal paste. For the soldered CPU's heat transfer between DIE and IHS is not an issue.

 

I know that. It is said in the article that it was a surprise to see a grease there

Most likely 3930K/3960X are really soldered then no need to de-lid. I just don't know the list of processors which doesn't have solder while 3770K looked similar to 3930K.

JEE, IS IT ONLY ME WHO HAS UNBELIEVABLY SLOW LOAD PAGES ON NBR RIGHT NOW, INCLUDING THIS TREAD??!
EDIT: Never mind. feel faster now

 

It is IB that they changed to use grease and not solder. All SB and earlier (3930/3960 are SB 32nm) are soldered. This is why I am not doing IB because of that issue. When they solder the CPU it has a very high thermal conductivity. I forget the numbers but I think it is less than Heat-Spring. I would have to find out what solder it was to check what the thermal conductivity it is. If it is lead I think it is already ~50? I dont know I can email Jim (he is one of their engineers and used to test all types of solders) and ask him if anyone knows the solder.

EDIT: btw It would be interesting to see a test if this is better than solder


EDIT:I emailed Jim asking him if he had any experience/knowledge about the types of solder. I defiantely have my homework to do and work cut out for me with making a guide and so forth...kind mad I didn't reserve a few posts lol oh well.

@Cloudfire that means it should work better but we will have to wait and see. Trying to keep an impartial and lets wait and see before we make any comments mind set.

Also how do you test such a thing? I would be interested in trying a couple things. I wonder if extide wouldn't mind trying with his desktop slowly tightening over an hour to see how the temps vary and how much pressure affect it. would make a very interesting graph if that was testable and measurable to see how much pressure affect it and where is the minimum and were is the point where the improvements are negligible. I thought I remember Jim telling me anything over 40 PSI is good and once you break 50/60PSI it has marginal effect but that might have been me just pondering it after we had a conversation. Don't want to put words in his mouth and miss represent him.

 

I afraid it will be complicated to make a good pressure in the area between processor's die and metal shield as there is no screws there... Don't recommend to bother.

I think many people just either still do not know about this theme or afraid to waste their IC Diamond they already have as grease is not reapplyable. While indium stuff they will have to give back.

Just my opinion. Thread must become more popular.

 

the section where i bolded makes no sense to me. They get to keep it for good if i send the stuff to them to test. They get to be one of the few to test it in return for work Also they get FAME!!! for being awesome testers or shame if they take it and dont do the testing grrrrr

Also ICD is not that expensive. it is $20-25 for a 24 carat tube that'll last a long time. Maybe a dollar or two per application...not very expensive. This is probably 2 times more expensive not sure how many times you get to ally with ICD in laptop CPUs. Desktop application this is very expensive 10-15 bucks to do one heatsink.

Anandtech(I think) was the one who replaced the junk TIM with ICD and had I think in the 10s of C drop in temps. I assume all they did was slap some in and hold it in place where they put heatsink over it. Not sure but It would be interesting to see someone try this with Heat-Spring

EDIT: Updated some layouts and added some copy and font changes for organization.

Also Jim replied it is conductive. But since it is a solid piece that really isn't that big of an issue. (no dripping or run off)

 

Bump need testers!!! I might have svl and conkers from TI waiting for responses from them and if they accept terms. Come on guys a unique opportunity is at hand!

 

Maybe you will manually search people who has 920xm 2920xm CPUs in their sig? and then ask them via PM?

After vBulleting4.2 I personally started to check for new threads 5 times rarely and check like 1 or 2 pages. Maybe it is a trend?

 

bump need testers! I have it on hand! he sent me 9 pieces instead of 4 so i got plenty for many XM chips the clevo/sager netbook and other ultrabooks with bad cooling. I'll be updating everything today or tomorrow again with further instructions and a cleaner and more elaborate terms of testing!!!

 

I measured the 3930k heat spreader the other night, it is 1.5" square, so 1.5" x 1.5", almost exactly. I have some pics of it next to a tape measure on my dslr, I will get them off later.

 

nah your good i trust you. I'll send you a 1.5x1.5 inch piece. After I updated terms and instructions I will start the mail out process. I will have you agree to terms in a post in this thread once I finish before I mail out. The post of agreeing is for "honor" purposes in an effort that if you blow off the testing people will know it

I might send you a piece for the 3920xm too since i got extra if you want to give me some basic info on the results. for your laptop. Could you get the 3920xm size for me?

 

There is a Canadian Guy whose temps are VERY high. like 90s watching YouTube. Also on idle his temps are 60. So it kinda VERY hot for a that kind of laptop.
But his CPU is not XM. It must и 1-st Gen Quad I7. 740QM I guess. He is about to order (or have just ordered thermal grease).

His fan is fully clean, no dust what so ever. So it is either fan is dying (it works on 6000 rpm. Some people had 10000 speed on such laptops) or bad thermal paste.
If it is fan problem than it would be interesting to find out how indium helps in this situation.

I have no idea if you need such a test or not, but just decided to inform. This is his profile http://forum.notebookreview.com/member.php?u=346830

 

http://forum.notebookreview.com/sony/686206-cpu-running-over-85-degrees-c-idle-vaio-f-series.html

this it?

Also I plan on PMing people with XM chips sometime this week and hunt some people down.

going to PM DR because i think he will help

 

Руку is his post. http://forum.notebookreview.com/son...s-owners-thread-part-6-a-263.html#post8818871

Who is DR?

 

I have a q9200 but would be very interested in how this works out. I'd love to use these if they really work...............

 

What is max GDP and clock and temp can u get?

 

I'm quite interested to help out with this--I've got an HP Pavilion dv6t with the Ivy Bridge 3610QM. This processor doesn't heat much on other laptops, but this is HP, so...

Before I repasted with ICD, I idled at ~55C and topped at about 95 when I gamed. With ICD I idle at ~48-~50C, but still top out high that's why I used ThrottleStop to throttle the CPU earlier than the processor's own throttling temp.

Seems like the cooling system on this laptop is fairly thin for a 45W processor. Or just not designed properly, since it seems that heat from the GPU can affect CPU heat and vice-versa.

I can test, but I'm from the Philippines. ) I can shoulder the shipping in case, if you're up for the effort. )

 

As long as u send through PayPal shipping and agree to terms we can do it. Post screen shots of ur temp. Did u crank down on the screws? ICD works best as do all TIMs under heavy pressure. Reapply and tighten screws as far as u can without stripping to see if that changes it. Get back to me after u try that. We need to make sure all prior examples are perfect pastes also refer to my garbage drawing on how to apply ICD in my cooling thread in Sig. Going to bed check this tomorrow

 

Can't seem to see the drawing. :/

I tightened the screws down ultra-tight, I even thought I snapped something because it creaked a bit. Anyway, I'll reopen it tonight and retighten the screws, in case. Then I'll pore through everything you need for me once my sched frees up. By when do you need everything btw?

I also put ICD differently, I think. Since Ivy Bridge bare dies are thinner and longer, I pasted a thin, long line of ICD on the center. This method consistently covers the entire die and minimizes spillover so...yeah.

 

Yea that's how u apply ICD. If u think u broke something last time don't tighten anymore...you might have stripped it. If u think u tighten that must than ur fine or in trouble if it is broken. Just see if anything is broken. Also if u remove and re apply see if u did put enough. First time I did it I was a hair short. Just make sure mounting isn't broken because if it is ur screwed

 

Nope, it's good. Checked just a few hours ago. Mounting was okay. ICD was spread fairly well and adequately. That's just really how my temps normally go.

 

I'm actually interested but i don't have XM series CPU. I could only provide moderate temperature testing at range around 70C-80C on couple laptops if you want.

 

Why only CPUs? Why not GPUs?

 

I will have to take the laptop apart again to measure it. Don't have time to do that right now but I should be able to squeeze it in sometime over the next few days.