lapped my ultra-120 extreme (pics and temp results)

Page 4 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

See my post above.
 

oops, sorry I just replied, didn't read through all the posts. 😉 But still wouldn't using nickel increase the price too? I would rather have cheaper price and better conductivity than worry about a little oxidization (look at most HSF manufactures, ie. Zalman 9xxxx, which are ure copper). I see why they would do it on the heat pipes but why the base?
 

Yes, having little->no exposure to oxygen will definitely decrease the rate of oxidation. However, in it's factory packaging it is exposed to everything, which is the reason I speculated they use nickel. I would also like to point out that nickel plating looks pretty, much like chrome, and while it does not help in cooling, your average person that doesn't know anything about heat transfer will think it performs well because it's "shiny and smooth", regardless of surface flatness or composition.

Yes, since nickel is currently trading at ~$13/lb and copper is at ~$4/lb. But the mass of nickel used in plating is very small, because it is such a thin layer. I'd guess that a 745g TRU is <5% nickel by weight (~40g = 1.4oz = ~$1.15), so the cost would not be greatly influenced by the plating.
 

Absolutely. Thermal interface material should always be used when mating surfaces. However, after lapping will probably require less to be used for ideal heat transfer.
 



Hey whats up KyleSTL! I see you are still regulating on this thread. Nice work. In reply to above, quite the opposite if you have silicon based thermal paste in contact with copper. It actually causes oxidation. You have to be careful with thermal paste selection if you have exposed copper heatpipes or a copper base. Xigmatek makes a huge mistake by including silicon based thermal compound with their heat sinks (which have exposed copper heatpipes at the base). That is a big no no. The proper thermal compound to use with copper has non-silicon based organic oils as their medium. Arctic Silver 5 is a good example with their use of the ester oils.

http://www.arcticsilver.com/PDF/as5msds.pdf
 
I lapped my Q6600 a week ago, dropped the temp only 4.3 Celsius on average on all 4 cores.

I did not use any liquid, 300grit sand paper, then 600 then 1200...I took a piece of glass out from a large frame and used that as the flat surface which I taped the sand paper too.

I did not take the time as others here to make it incredibly shiny but I did have a reflection and took pics too cause im a nerd.

I intend to take the processor off again and make it perfect, also I am going to do the waterblock which doesn’t seem concave in any way…
 
SANDING ADVICE: (If your a perfectionist)

I have some advice to people who want the perfect mirror finish, or are perfectionists when it comes to the sanding process. When you sand go diagonally both ways (forming an 'X') and than sand vertically and horizontally (making an '+'). It does not need to be in this order or any order but the directions should be sanded equally and switched frequently to ensure a smooth finish. Going in one direction up and down may provide something which seems smooth but it really isn't, because if you touch it going a different direction you noticed you just formed ridges now. So making an 'X' and '+' is generally a good way to ensure this doesn't happen.

And try to avoid sanding it by just holding the sand paper, your fingers will create small groves in it generally.

I'm pretty sure this won't make any performance difference, but this is just a little something I thought some of you may want to know. And in case your curious how I know this, I use to do body work on cars, and this is how we would have to sand or else the car would look horrible after painting it.
 

Because gold does not conduct heat as well as copper (or silver). It however, conducts electricity the best out of the three. And cost, duh!
 
I just finished... Took a lot of time!

Started with 400grit and finished on 1000.

my.php
[/URL]
[/img]
my.php
[/URL]
[/img]

I definitely saw the concave. It took forever with 400grit to get rid of it. I put a huge sheet of paper on the table and taped it down. Used a soapy sponge for the wet. It worked great.

It was super hot today so as I expected I dont see any temp differences. I even lapped the CPU... Oh well.. maybe the thermalpaste will need a break in! 36c to start 36c to finish.
 
i've been reading about lapping and overclocking a lot lately and after reading all this i've decided to lap my cpu & possibly my hsf. the reason why i say i "might" lap the hsf is because i'm using the stock hsf which isn't bad at all imo has a MUCH bigger block than typical aftermarket hsf's. it is about 2x2inches. i have no clue why it is so big but anyhow it works well so i'll probably just lap my athlon 64x2 4600+. i will try and post pics and stuff later on. wish me luck! =D
 


See if your hsf is flat by using a razors edge on it, if it rocks, or there is a gap then it ain't flat and will only benifit from lapping that too. If you've got it all out of your case anyway it's worth the short amount of time just to check.
 
So is it worth taking the gamble to purchase an efficient Thermalright Ultra 120 eXtreme cooler? I don't have any plans on lapping.
 
My non-lapped Thermalright Ultra 120 eXtreme cooler working fine on my as yet non OC core i7 920.

It is a very dependable cooler with its 1600rpm fan

 


As I am reading tru this I'm thinking the same thing.

I just bought a CoolerMaster V8 will receive it this weekend and after spending $65.00 on it I'm sure as #@$% not going to lap it..... lol
 


It's to give an anal retentive person somthing to do?

lol.

I think they were trying to get a smooth non concaved service.