The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    Cleaning heatpipe increased thermal performance (HPC 8510p))

    Discussion in 'HP' started by nakamoomin, Aug 13, 2008.

  1. nakamoomin

    nakamoomin Notebook Geek

    Reputations:
    54
    Messages:
    80
    Likes Received:
    5
    Trophy Points:
    16
    I made an interesting observation the other day. After reading up on heatpipes I decided to check the condition of my own. I have been troubled by some unusual heat issues in the past two months.

    I changed the thermal paste to OCZ Freeze which didn’t seem help one bit, so I was curious if the heatpipe itself might be the culprit. After what I’ve seen so far, I have to conclude that it was.

    When I changed the paste, a month or two ago, I noticed a small dent in the underside of the keyboard corresponding to where the heatpipe (HP) was located. I didn’t think much of it at the time, but later I’ve wondered if this contact between KB and HP was to blame (the aluminium KB cover leading heat to the internals of the PC rather than let the HS and fan take care of it). I bent the alu slightly to avoid contact, then reassembled. Thermal performance increased slightly.

    What I noticed this time was that the middle of the GPU heatpipe (between CPU and GPU) had gotten a darker tint; it was a dark blueish grey. Almost like metal that has been heated with a blowtorch.

    I’ve heard of ways of cleaning copper (removing oxide) with acidic cleaners, so I thought I’d give it a go. I decided to use acidic acid (household chemical a.k.a. synthetic vinegar). I dipped a Q-tip in the acid and started working the surface of the heat pipe. It is also possible to remove the HP and put it in a jar containing acidic acid or lemon juice, leaving only the HS out of the bath. This requires that you have a good cleaner for removing any residue from the contact surfaces (like 96% alcohol, acetone, isopropyl), which I didn’t.

    After half an hour of rubbing gently and swiping off with a clean Q-tip, the HP looked brand new. When running Thermal tests, I couldn’t believe my eyes. Light tasks (Opera + Songbird + Wifi) usually produced core temps of 60-61C with the fan running constantly on 70% or more. After the procedure it now hovers between 50 and 54C!!! That’s a 7-10C difference!

    Running TM Nations Forever would lead to massive heat buildup in the PC, noticeable under the lower, left hand side of the laptop, eventually leading to BSOD. Now it runs for hours with peak core temps of 70 (GPU) and 72 (CPU).

    After 3 min cooldown, temps are back to low 50s.

    Idle temps are mid 40s (low 40s on batt).
    This is even better than when it was new! How can this be?

    Here is the thing: A heat pipe system consists of a thermal connection (block) a heatsink (HS) and a heatpipe connecting the two. The liquid in the pipe evaporates at the block, and runs through the pipe to the HS, where the lower temp causes it to condensate to a liquid. A wick transports the liquid back to the hot side (block). Any contact with the pipe will cause a so called hot-spot (actually it is rather a cold spot, but more on this later). The contact point will be a bridge for the heat to exit, leading to condensation in the pipe. This will lead to a heat transfer (phase change gas -> liquid = thermal energy out). On the inside of the pipe this will be a colder point (a place to condense) but it will simultaneously lead to rather large amounts of heat being dumped on the outside of the pipe.
    Remember; the goal of the HP-assy is to make the block as cold as possible, while making the HS as hot as possible (thereby increasing efficiency of the fan/airflow)

    So, why didn’t my attempt to fix this by bending the alu away from the pipe work?

    The discolouration is the culprit. A matte object will easier absorb and emit thermal energy than a shiny one (matte = less reflection, shiny = more reflection). The matte dark area on the pipe continued to form a hot-spot (cold-spot) even after the contact between KB and HP was removed. This lead to dumping of heat inside the PC, leading to a lower temp at HS again leading to reduced efficiency. I was aware of the extreme need of “cleanness” on the inside of the tube in the manufacturing process, but not that the HP would be THAT sensitive.

    When checking my older laptops (Asus) I noticed that their heatpipes seem even shinier than my newly cleaned one (HP Compaq). I wonder if it is possible to increase performance even further with a shinier pipe…

    As to all of you who have read articles on cleaning heatsinks; Yes, a clean heatsink (no residue/dust on fins) will increase performance, but a matte surface will always be better at heat transfer than a shiny one (of the same material, of course), just keep that in mind.

    I hope this was helpful to at least some of you. What caused the contact between HP and KB in the first place, I don’t know. I suspect me pressing to hard on the keyboard (the HP is located directly under WASD… ;) ;) (HP Compaq 8510p)


    My conclusion/advice to people struggling with poor thermal performance is:

    1) Make sure the HP-assy does not make contact with other objects than the thermal design indicates

    2) Make sure HP is clean (preferably shiny)

    3) Make sure HS is clean (free of dust/residue)

    4) Check contact surface (block)/change thermal compound

    PS while typing this (on AC with WiFi, MS Office, Opera) core temps are 40 and 41 Power setup is “HP Optimized” (Balanced)
    .naka
     
  2. flipfire

    flipfire Moderately Boss

    Reputations:
    6,156
    Messages:
    11,214
    Likes Received:
    68
    Trophy Points:
    466
    Great finding. Ive noticed this improvment myself, but to a smaller degree

    I clean the heatpipes aswell using Articlean 1+2 (clean+polish) the same time i change cpu or thermal paste
     
  3. brianstretch

    brianstretch Notebook Virtuoso

    Reputations:
    441
    Messages:
    3,667
    Likes Received:
    0
    Trophy Points:
    105
    I'm good about cleaning the heatsink with Articlean but it never occurred to me to clean the heatpipes. Nice find!
     
  4. 2.0

    2.0 Former NBR Macro-Mod®

    Reputations:
    13,368
    Messages:
    7,741
    Likes Received:
    1,022
    Trophy Points:
    331
    Excellent observation and write up.
     
  5. nakamoomin

    nakamoomin Notebook Geek

    Reputations:
    54
    Messages:
    80
    Likes Received:
    5
    Trophy Points:
    16
    Thanks guys!
    I thought it would be worth sharing. Probably many others who wonder why their themal performance degrades over time.

    I'l see if I can get some pictures posted to make a "how-to" on the subject.
    (my cam was out of reach when I did this)

    If anyone has alternatives to the acidic acid, I'm all ears.

    Here are some screens of the temps for those interested:

    IDLE
    idle.jpg
    IDLE

    PEAK
    peak.jpg
    PEAK

    3 min after PEAK
    3min.jpg
    3 min after PEAK

    BROWSING/OFFICE
    browsing.jpg
    BROWSING/OFFICE

    Enjoy!
     
  6. yummer

    yummer Notebook Enthusiast

    Reputations:
    0
    Messages:
    27
    Likes Received:
    0
    Trophy Points:
    5
    FYI, there is no gas/liquid inside of a standard heatpipe. The function of a heatpipe is derived from the heat conductivity of the metal only.
     
  7. nakamoomin

    nakamoomin Notebook Geek

    Reputations:
    54
    Messages:
    80
    Likes Received:
    5
    Trophy Points:
    16
    I'm afraid you are mistaken...

    http://en.wikipedia.org/wiki/Heat_pipe

    this is why they are so efficient.

    :)

    best regards,
    .naka
     
  8. 2.0

    2.0 Former NBR Macro-Mod®

    Reputations:
    13,368
    Messages:
    7,741
    Likes Received:
    1,022
    Trophy Points:
    331
  9. nakamoomin

    nakamoomin Notebook Geek

    Reputations:
    54
    Messages:
    80
    Likes Received:
    5
    Trophy Points:
    16
    Yes, it is horrible what manufacturers will do to make a produck look "stylish"..
    I bet the copper HS contrasted the black chassis and therfore was painted.

    Or someone thought covering the HS in black would increase thermal performance. Fortunately HP didn't make the same mistake with the HP Comp 8510p/w.

    I have nothing bad to say about stock thermal performance (as long as you don't push the KB onto the HP. :) )
     
  10. 2.0

    2.0 Former NBR Macro-Mod®

    Reputations:
    13,368
    Messages:
    7,741
    Likes Received:
    1,022
    Trophy Points:
    331
    Well, I'm going to give it a go on both the pipe and the fins on my TX2500.

    Once again, thanks for taking the time to post your finding.
     
  11. nakamoomin

    nakamoomin Notebook Geek

    Reputations:
    54
    Messages:
    80
    Likes Received:
    5
    Trophy Points:
    16
    I'm not sure if you've gotten around to it yet, but when you do, thermals would be appreciated. :)

    good luck.
     
  12. 2.0

    2.0 Former NBR Macro-Mod®

    Reputations:
    13,368
    Messages:
    7,741
    Likes Received:
    1,022
    Trophy Points:
    331
    LOL, yeah it's on the top on my to do list bust seems to keep getting bumped. I think I'll give it a go this weekend.
     
  13. 2.0

    2.0 Former NBR Macro-Mod®

    Reputations:
    13,368
    Messages:
    7,741
    Likes Received:
    1,022
    Trophy Points:
    331
    Ok so I did it yesterday. This was done on a TX2500 with ZM-80 2.1GHz AMD Turioun ultra. Cleaned the heat pipe and also cleaned off the black paint from the radiator fins using 91% isopropyl alcohol, flattened Q-tips and a jeweler's screwdriver. I also applied Arctic Ceramique to the CPU.

    Ambient temp 25C.

    Idle on Balanced and performance were unchanged before and after.

    But...

    running W prime

    WPrime Quicktest.

    powermode Top temp: ( Before/After)

    AC Balanced: 65c/59C
    AC Performance: 89c/81c

    Wprime Long test took 20mins.

    AC Performance: 103-105C/91-92C

    Now that's a nice improvement. A good 11-13degC drop in top temp. Now I'll consider playing games on the TX2500.

    Thanks again for the tip! Worked well enough. :)
     
  14. highlandsun

    highlandsun Notebook Evangelist

    Reputations:
    66
    Messages:
    615
    Likes Received:
    6
    Trophy Points:
    31
    Cool, I may have to try this on my dv5z. Right now it idles at 60C...
     
  15. xhepera

    xhepera Notebook Consultant

    Reputations:
    36
    Messages:
    117
    Likes Received:
    0
    Trophy Points:
    30
    Thanks for the tip! I have a minor nit to pick though. Instead of "acidic acid" I think you mean "acetic acid" yes?
     
  16. nakamoomin

    nakamoomin Notebook Geek

    Reputations:
    54
    Messages:
    80
    Likes Received:
    5
    Trophy Points:
    16
    hehe of course! a typo there.

    @ 2.0
    Wow! great results! Congrats!

    Mine is still going strong. I've opened it up to check for oxide-layer buildup, but its still as clean as a whistle. :)

    Idles well under 50, low 50s for browsisng/MS Office.
    Tops out at 62 (C) during gaming.