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

Discussion in 'Overclocking & Cooling' started by graysky, May 6, 2007.

  1. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    When my Ultra-120 X and I have to say I'm a little puzzled. The base where it should contact the heat spreader is not smooth at all, it's actually grooved! You can see a scratch which is where I gently ran my thumb nail over the surface; I could feel the rough edges.

    Have a look for yourself:
    [​IMG]

    Anyway, others encouraged me to lap it which I've never done before. After wrestling with the idea for a couple of days as well as reading many articles/guides, I decided to give it a go. $20 worth of sandpaper, a $2 piece of flat glass, and 4 hours of careful work (and sweat) later, I was left with a pretty darn flat HS. You can see by the pictures that this particular one was quite concave instead of being flat which isn't good for keeping contact between the HS and IHS of the CPU.

    [​IMG]

    Did it work you're probably wondering. The temp data as measured in speedfan.exe for a ~1 h x264 encode (uses all 4 cores with a CPU load of >99 %). I had speedfan log the temps (which it does every 3-4 seconds) and I averaged the whole data set per core for the 2nd pass of the 2-pass encode (the 2nd pass is the most CPU intensive). Room temp for both experiments was ~23 C.

    Code:
    Before lapping the HS:
    
    Core 0: 51.9
    Core 1: 51.4
    Core 2: 45.6
    Core 3: 45.6
    
    After lapping the HS:
    
    Core 0: 49.9
    Core 1: 49.4
    Core 2: 44.0
    Core 3: 44.4
    
    Delta:
    
    Core 0: 2.0
    Core 1: 2.0
    Core 2: 1.6
    Core 3: 1.2
    System specs: Q6600 @ 9x333=3.01 GHz (stock voltage), P5B-Deluxe in an Antec p182 case.
     
  2. covertash

    covertash Gawd

    Messages:
    988
    Joined:
    Apr 24, 2006
    Very nice! If you used Arctic Silver 5, after a few more days the paste should settle in and further reduce your temps by another degree or so. :cool:
     
  3. Eastcoasthandle

    Eastcoasthandle [H]ard|Gawd

    Messages:
    1,041
    Joined:
    Jan 27, 2006
    Very good results and good lap!
    I read a post of someone with a Big Typhoon where they put ramsinks on the other side of the base plate to reduce temps found here. I doubt if that will work in your case but worth mentioning.
     
  4. elyrly

    elyrly [H]ard|Gawd

    Messages:
    1,812
    Joined:
    Aug 19, 2005
    can you post set-up
     
  5. Commander Suzdal

    Commander Suzdal 2[H]4U

    Messages:
    2,134
    Joined:
    Jul 29, 2004
    A $22 investment (plus hours of research and labor) for a 2 degree improvement? I'd be not so happy... I've often wondered how much lapping helps if you use decent thermal paste, and now I wonder even more.
     
  6. dopee123

    dopee123 [H]ard|Gawd

    Messages:
    1,101
    Joined:
    May 19, 2006
    that's soooo sick. I'm going to try that. Thanks!
     
  7. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    Well, I believe AS5 is a decent paste. Those were the temps right after I installed, they have come down since then. I lapped my q6600 the next day and it gave the biggest bang-for-the-buck. With it too, the temps have come down over a period of the first 18h and now, several days later, they are continuing to drop as well.
     
  8. Commander Suzdal

    Commander Suzdal 2[H]4U

    Messages:
    2,134
    Joined:
    Jul 29, 2004
    Oh, yeah, AS5 is great, I just meant that as long as you had it in there filling in the grooves, would you see that big an improvement from lapping the grooves away? I guess it's getting better now as it "cures"? If it ends up being more like a 4-5 degree difference once it settles in, I could see how that would be worth it to you.

    What I wonder to myself, I guess, is like this example:

    Say I have the CoreDuo and the retail HSF. I could lap the heatspreader and the HSF, or I could take the time and money and buy an aftermarket cooler. Option 2 would probably lower my temps a lot more for a lot less trouble and only a little more cash. If I wanted ultimate cooling I could do both of course, but if I had already lowered my load temps by 10 degrees C with a better HSF, lapping to gain another 2 or 3 wouldn't seem necessary to me. But that's just me, ol' lazy baztad that I am!:)
     
  9. Aelfgeft

    Aelfgeft 2[H]4U

    Messages:
    2,343
    Joined:
    Jun 8, 2005
    The temp decreases honestly seem insignificant. This is not a criticism, just a disappointment in the actual results. My idle temp went from 38C to 33C over the course of a couple days, non-lapped, I assume due to the AS5 taking, since the room temps were the same. I always still admire the work people put into getting that extra degree. It can be worth it on here. :D
     
  10. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    @SC: I'd say get rid of that stock cooler ASAP. The Ultra-120 Extreme is what I'd recommend (I love mine) and a quality 120 mm fan for it. Lap it and your CPU and I'm sure you'll be happy.
     
  11. hegulator

    hegulator Limp Gawd

    Messages:
    302
    Joined:
    Jan 27, 2005
    I think some of the "dissapointing results" people have after lapping is due to how they apply their AS5 after lapping. On a non-lapped CPU, you'll want a "decent" ammount of thermal compound to fill in all peaks and valleys and unevenness. When I say a decent ammount, I mean what the thermal compound makers typically reccomend.

    With a lapped CPU, there is _much_ less space to fill, therefore you'll want to use _much_ less thermal compound. Just enough to kinda "glaze" the area should be more than enough. If you're using the same ammount of thermal compound, you're not getting yourself very far, IMO.

    You can see 1-2 degree varriances also just based on how well you mount your HSF to your CPU.

    Also... 20 bucks for sandpaper?!?!!! WTF? I picked up a "variety pack" of wet/dry 200-1000 grit at fleet farm for like 8 bucks? If that?
     
  12. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    I thought that was true as well, but when I applied too little paste (less than I used for my unlapped setup my temps were through the roof... so bad in fact that I thought I ruined the HS because I took off too much material. Turns out I wasn't using enough TIM, which goes against the internet buzz about TIM and using too much. No offense to you, hegulator, but I think 1/2 the stories you read online aren't true and are just people spreading the same story they read and accepted as factual to others, who in turn pas it on and on.

    Here is a pic of the amount I used which gave great temps:
    [​IMG]
     
  13. young wing

    young wing 2[H]4U

    Messages:
    2,552
    Joined:
    Jan 21, 2007
    Looks like a really good looking waste of time to me. Sorry but I don't see any big temp drops either..:eek:
     
  14. killingfield1975

    killingfield1975 Gawd

    Messages:
    636
    Joined:
    Sep 18, 2006
    meh for 2 dgrees i dont think i would do it but you got to have him a thumbs up for trying it though.
     
  15. killingfield1975

    killingfield1975 Gawd

    Messages:
    636
    Joined:
    Sep 18, 2006

    dude i splatter a crap load of arctic silver onto my cpu. i dont use that thin film method. i found that using a good amount the cpu and the heatsink will press the rest of it out anyway. ive seen it soo many times where i applied a shitload vs a regulard load and the rest of the gunk just gets dump outside of the heatsink.

    but overall my method is a little different

    i apply paste on both heatsink and cpu then massage it out to a thin film where it looks very cloudly on the surface. next i apply a dot of thermal paste to the cpu about the tip of an 3.5mm jack. then i spread it out to roughly the size of a eraser tip and slap the heatsink on. this method works for me EVERYTIME.
     
  16. unclewebb

    unclewebb [H]ard|Gawd

    Messages:
    1,799
    Joined:
    Jun 21, 2006
  17. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    Thanks for the concern, but it's actually correct as show, this is a Q6600 and the cores are rotated 90° relative to a C2D.
     
  18. unclewebb

    unclewebb [H]ard|Gawd

    Messages:
    1,799
    Joined:
    Jun 21, 2006
  19. W6FO

    W6FO Gawd

    Messages:
    608
    Joined:
    Dec 12, 2005
    fux ya'll

    I'm gonna go lap my Celery II 850 and then OC da sum bitch to 3.8 jigs and there will be fuxin icicles formin when I'm dun.

    werd
     
  20. cornelious0_0

    cornelious0_0 [H]ardForum Junkie

    Messages:
    12,783
    Joined:
    Apr 6, 2003
    I'd personally still lap the heatsink for a 2C drop...it's still a gain, but a lot of the temperature difference can come down to thermal paste application, i'll agree with that.

    I really need to try the single line of paste across the cores, as I've always just put on a dab in the center and rubbed in across the entire heatspreader with saran wrap on my finger until there's a fairly thin film/layer. Probably not wroth redoing it with the stock cooler, but I'll definitely be lapping both my Ultra-120 Extreme and HR-05 along with the CPU's IHS when the time comes.

    Warranty....what warranty? :D :p
     
  21. HybReD1234

    HybReD1234 Limp Gawd

    Messages:
    177
    Joined:
    Aug 20, 2006
    Need an update on that second picture.
     
  22. tissimo

    tissimo Limp Gawd

    Messages:
    318
    Joined:
    Dec 8, 2007
    My ultima 90 and the same poorly machined base. I will be lapping it and probably the IHS in the next couple weeks.. Hopfully get me that 3.6 oc while staying under 60*c :)
     
  23. mucker

    mucker Limp Gawd

    Messages:
    188
    Joined:
    May 6, 2006
    My Ultra 120 Extreme sucks. The base was scuffed with a very small nick just off center brand new out of the box. My Q6600 temps can vary 10C between cores and that really bites.

    The same Q6600 with my plain Ultra 120 is much better with only a 3 degree C delta between cores and the temps are slightly better because one core on the extreme wants to run hot.

    The clamping mechanism on the Ultra 120 is much better than the Extreme. The Extreme clamp is horrendous.

    I am very dissappointed in the Extreme's performance. The lack of pressure unfortunately puts this superior designed sink on par with it's older brother.

    I am talking with Thermalright right now, but I am just getting noise back from them thus far. I'll take a refund or a swap for a plain 120.

    The guy wanted me to download TAT......sheesh.....:rolleyes:
     
  24. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    mucker - sounds like you got a bad sample... the ultra-120 ex rocks for me after lapping anyway. You can also add a washer between the bracket and the HS if you need additional pressure.
     
  25. mucker

    mucker Limp Gawd

    Messages:
    188
    Joined:
    May 6, 2006
    Well, when I buy something new, I expect it to perform as advertised directly out of the box without any modification. I am not interested in lapping or applying washers. I simply want it to work, just like the Ultra-120.

    I sent Thermalright pics of the base where a small nick is obvious, just off center. There are other scuff marks that could not be cleaned out of the surface. Brand new product?

    Anyway Thermalright was NO HELP AT ALL. The tech was talking down to me like I didn't know jack shit, he said "no RMA for you". It was like he wasn't listening at all.

    There is an obvious flaw with my sink (un-even base or clamping). I can't believe how piss poor the clamping is. The sink is superior to the Ultra, but the clamping system fuxors the whole deal. What a shame for such a pricey sink. Oh yeah, did I mention that THERMALRIGHT SUPPORT SUCKS SHIT!
     
  26. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    I can understand your frustration.. I shared it when I learned just how convex mine was too (see my original post on lapping it). I also understand how an expensive product should just work out-of-the-box too and I'd call their tech department back and ask calmly to speak to a supervisor or manager and explain your problem and insist on an RMA/exchange. That said, quads are by design concave and this HS is by design convex, if you want the best performance and optimal heat transfer, lap both of them; the u-120 ex is superior to the u-120 when lapped.

    You can read my two posts on lapping in here, or just read the article that combined both their content along w/ an intro/conclusion over at techarp here.

    The real question you have to answer to yourself is are you more concerned with the principal of the matter and holding TR accountable or are you more concerned with having a cool running PC?
     
  27. kaluminati

    kaluminati Gawd

    Messages:
    579
    Joined:
    Dec 11, 2004
    Too be honest your results are pretty insignificant. I gotta say there is a "margin of error" in any of these kinda tweaks.

    Your room temp could have gone down a bit. You could have screwed down the heatsink a little tighter this time around. You could have applied the thermal grease a little better this time.

    I have mine lapped as well, but I lapped it soon as I got it so I have no control temps =(
     
  28. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    @kal - the room temp was well controlled (thermostat is in room and set to a constant value), I will agree with you about variations due to seating, but if you look at my CPU lapping thread wherein I have mated the lapped HS with the lapped CPU and measured the temp differences, they are certainly statistically significant.

    [​IMG]
     
  29. NiViK

    NiViK Limp Gawd

    Messages:
    284
    Joined:
    Sep 12, 2000
    It would be interesting to see what the unlapped difference was after 300 hours