1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

Thermal pad or Thermal paste?

Discussion in 'Overclocking' started by Vin, Sep 4, 2003.

  1. Vin

    Vin Guest

    hey there,

    after some excessive amount of case modding to fit 4 additional
    suckholes/blowholes (3 in the side case blowing IN and 1 on top of case
    blowing OUT), i've finally managed to bring my case temperatures under
    control -- about 6C above the room ambient (ambient is 28C, case is 34C).
    But my case to cpu temperature delta has gone out of whack! previously with
    my case open and a floor fan blowing into the case, the cpu temperatures
    hovered about 18C above the case/system (when idle), but now it's a shocking
    24-25C above the case temperature (currently 58-59C when idle). Thus I've
    decided to chuck the retail AMD HSF and get myself a Thermalright SLK700
    cooler.

    Now that the backgrounder is done with, my actual question: should i use
    thermal paste instead of a thermal pad with my new cooler? I've read that
    AMD will void your CPU warranty if you use thermal paste/grease. Also would
    thermal paste require any maintenance, like removing, cleaning and reseating
    the cooler every few weeks?

    thanks,
    vin
     
    Vin, Sep 4, 2003
    #1
    1. Advertisements

  2. Vin

    Rick Guest

    IME the chances of actually using a cpu warranty are next to nil, so I
    wouldn't let that stop me from using thermal grease. You'll almost
    certainly get better cooling than with a pad.

    Most thermal greases don't require maintenance unless you dismount
    and remount the cpu (obviously).

    My $.02

    Rick
     
    Rick, Sep 4, 2003
    #2
    1. Advertisements

  3. Vin

    Pham Guest

    Use paste for best results. Commercial pads are said to be pretty good-
    but impractical for small purchases. Go with Arctic Silver Ceramic-
    best combination of performance and novice-safety.

    BTW-
    Some say it's bad etiquette to cross-post. Choose one group at a time.
     
    Pham, Sep 4, 2003
    #3
  4. Vin

    Alan Guest

    I agree with Rick, and I only use thermal grease. It lowers the temps
    substantially.
    A couple of weeks ago, I had to have a CPU replaced by AMD, they did so
    without any problems. AMD is an excellent company to deal with.

    Artic Silver has instructions on their site as to proper installation of
    thermal grease. Minimal amount spread to the thinnest film is the secret to
    best temps. I don't use Artic Silver though, rather I use some that came
    with a Dr. Thermal heatsink.

    Having said that, your CPU should not have risen more above the ambient temp
    of the case. Did you remove the Themal material, clean the residue with
    isopropol alcohol and replace it with a fresh one? I would also check your
    mounting of the heatsink and the clip tension.

    A rule of thumb is that with the heatsink rests at a similar temp above the
    case, and 24C is too high. Look for 10C when idle and 15C under 100% use,
    this will depend upon the effectiveness of your heatsink, and mounting.


    Alan
     
    Alan, Sep 4, 2003
    #4
  5. Vin

    Vin Guest

    i've read the thermal paste installation instructions on the AS3 web site.
    the thermal paste installation is quite detailed and helpful, but i'm quite
    confused about the cleaning of the cpu core/ceramic bit (cleaning the old
    heatsink doesnt apply to me, as i'm replacing my stock HSF with the
    thermalright 700)

    how do i safely and properly clean the thermal pad residue from the cpu
    core/ceramic? i checked with my local chemist, but he doesnt have any
    isopropyl alcohol with him. also it's mentioned that if i use any other
    cleaning agent, i'd still need to do a final cleaning with isopropyl
    alcohol -- are there any other commonly available cleaning agents that i can
    use?

    another thing, do you directly dab a tissue paper with a cleaning agent and
    then rub away the residue from the core/ceramic of the cpu? or would a
    cotton bud be preferable?

    thanks,
    vin
     
    Vin, Sep 4, 2003
    #5
  6. Vin

    Joe727 Guest

    You might want to take a look at these Processor and Heatsink Installation
    Videos available on the AMD site:

    http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_869_4348^6678,00.html

    Joe
     
    Joe727, Sep 4, 2003
    #6
  7. Vin

    Alan Guest

    First you need to scrape off the thermal material, DO NOT USE A METAL
    OBJECT, as this will scratch the surface. Plastic is better, and take your
    time to be very careful. You must not scratch the chip surface or the
    heatsink surface.

    I have Isopropyl to clean remaining residue and remove any oils, so that is
    what I now use. I cut small pieces of paper towel and use them, (oil free
    and doesn't leave particles)

    Before that I used a very slight amount of cleaning fluid (varsol type)
    which is absolutely not recommended. However I was extremely careful and
    made sure there was little to no residue left. Of course this is an oil
    base, so it is not perfect, but it did work.

    I have read in a newsgroup of people using things like goop-off and orange
    citrus stuff, I believe these are not recommended either.
    Others have used nail polish remover or acetone (they are same base), I
    absolutely say NO to that, acetone will melt certain resins and damage the
    surface.

    Hope this helps,

    Alan
     
    Alan, Sep 4, 2003
    #7
  8. Hi,

    I used to think that also for many years, but as long as it is *on topic* I
    belive crossposting is ok now!

    Wayne ][
     
    Wayne Youngman, Sep 4, 2003
    #8
  9. Vin

    Strontium Guest

    Sorry Wayne, my server doesn't see this one, or I've already killfiled
    them...

    -
    Wayne Youngman stood up at show-n-tell, in
    bj7aoq$3c9$, and said:

    And, some say it's bad etiquette to use toilet paper. Who you choose to
    listen to, is your choice SFB.


     
    Strontium, Sep 4, 2003
    #9
  10. A dissenting opinion.
    You've made significant changes to your case airflow, and achieved a
    reduction in *case* temps with increased CPU temp. My first thought is that
    you've mounted a fan(s) where cool air is blowing directly onto the
    motherboard "case" temperature diode, while disrupting proper flow of cool
    air to the HSF area. If you've managed to create a dead air area around your
    CPU (quite possible if you've added a lot of inflow without adequate exhaust
    near the CPU/HSF), it's your mods that have caused the CPU temp increase.
    I'm offering this because you imply that you did nothing with your current
    HSF while doing the case mods, correct? If that's true, there's no reason to
    believe anything *but* the case mods affected the CPU temperature. If I'm
    right (happens occasionally), you're going to be disappointed with the new
    HSF, as no HSF can work efficiently if it's being fed hot air.
     
    Peter van der Goes, Sep 4, 2003
    #10
  11. Vin

    Vin Guest

    hi peter,

    i can't really pinpoint any reason for the increased cpu temps since the
    case mods were done after a lot of research -- my case currently has 1 rear
    exhaust fan, 1 psu exhaust, 1 front/bottom intake, 3 side panel
    intake/suckholes (1 bang opposite the cpu cooler with a duct, 1 blowing air
    onto the agp slot, and 1 towards the hdd area on the bottom/front) and 1 top
    blowhole exhausting hot air. The positions were chosen carefully and i've
    also tried to balance the intake and outtake fans.

    Thus i can't really understand why the temp delta (cpu to case) is much
    higher than before (24C vs 18C previously). but do bear in mind, that the
    actual CPU temperature is about the same now with the case closed as it was
    with the case open and a huge floor fan blowing into it -- it's just the
    cpu-case delta that has gone out of whack, which could suggest the heatsink
    ain't doing it's job well, even though the case is cooler now.

    vin
     
    Vin, Sep 4, 2003
    #11
  12. Vin

    Rick Guest

    I think you missed Pete's point: cooler temps/increased airflow in
    a case does not necessarily mean cooler temps/increased airflow
    around a cpu. The additional fans might have created a dead spot
    where airflow is needed most.

    Blowholes on top of a case aren't as effective as blowholes near
    or behind the cpu/agp area. Does your case have a fan cutout
    under the psu exhaust? If not, get one that does.

    Rick
     
    Rick, Sep 4, 2003
    #12
  13. Vin

    Phrederik Guest

    after some excessive amount of case modding to fit 4 additional
    The answer is PASTE! Leave it there unless you remove your heatsink.

    My question is this...

    WHY would you be punching holes all over your case if you didn't know the
    simplest thing about cooling your CPU?
     
    Phrederik, Sep 4, 2003
    #13
  14. Vin

    kony Guest

    With your configuration there aren't any significant airflow problems
    around the CPU as others have speculated, the system is responding
    _NORMALLY_ to the changes. You see the correct "delta (cpu to case)
    because you aren't significantly reducing case temp at this point, you
    are blowing a stream of air onto the chip taking the temp reading
    (probably the southbridge). You would now need a external temp sensor
    to take accurate reading, though it's not really necessary, the case
    now has plenty of ventilation. The stock heatsink should be more than
    enough to keep the system stable from a heat perspective unless the
    room ambient temp is excessively high, which from your earlier temp
    report it isn't. The primary reason to replace the heatsink at this
    point would be to reduce noise, though the other fans might be louder
    anyway.


    Dave
     
    kony, Sep 4, 2003
    #14
  15. Vin

    Vin Guest

    hi dave,

    so what you're telling me in a nutshell is that my case temperatures haven't
    really dropped, they're about the same or maybe even 1-2C higher than when i
    had a floor fan blowing into the case -- it's just so happens that the mobo
    sensor reporting the case temp now has better airflow around it and is
    'misreporting' the _GENERAL_ case temperature?

    thanks,
    vin

    ps: i'm not expecting any miraculous reduction in the cpu temps with the
    slk700, but when i installed my athlon xp a few months back, i used the
    default thermal pad and from what i've been reading, all things being equal
    thermal paste can definitely drop temperatures by 3-5C. Also the slk700
    reportedly has a much better thermal resistance than the stock HSF, so i'll
    let you know whether it made any difference or not come sunday :p
     
    Vin, Sep 4, 2003
    #15
  16. Vin

    kony Guest

    It's not misreporting the ambient case temp necessarily, but being a
    sensor that takes the temp of that chip, it was never an accurate temp
    sensor to begin with. I'm not really sure who started the trend of
    calling it the "system" temp, and futher extending that to refer to
    case ambient temp, but this notion is now so widespread...

    Given that the chip creates heat internally, yet is being cooled by
    lower-temp outside air, and that you have plenty of airflow in the
    case, it's most likely that sensor is now registering nearer to
    ambient case temp than ever before, but likely the reading is still
    slightly higher, partially because the chip isn't 'sunk to anything
    but mostly because even if it were, the heat source is in the
    immediate vicinity of the sensor.


    The SLK700 is definitely a better heatsink, though I'm not going to
    hazzard a guess as to what the temp drop might be, but it should be
    noticeable and/or quieter depending on the fan. I'm not fond of those
    TMD fans but it's partially because an earlier revison was prone to
    short-out, frying fan headers. Unfortunately I don't recall how to
    tell which are the earlier revisions with this problem.


    Dave
     
    kony, Sep 4, 2003
    #16
  17. I agree with Peter in that your temperature readings, and the conclusions drawn,
    are inconsistent with physics. In particular, one, or both, of the "case"
    readings must not be actual case temperature (at least around the CPU socket).

    Regardless of how good, or bad, the heatsink is, it didn't change and if case
    temperature truly went down then the CPU temperature would go down the same
    amount. I.E. The CPU heat IS going to 'flow' with the only question being what
    temperature the CPU has to rise to in order to force it and, since the thermal
    interface/ heatsink did not change, the temperature differential across it due
    to that flow will be the same. So, as case temp varies, the CPU temperature will
    follow in direct lockstep (assuming the same CPU loading/heat output). The fact
    that it didn't change indicates SOME temperature reading is not what you think
    it is.

    What would help is to get an independent temperature probe and measure the
    actual case temperature in a number of spots to see what is really going on in
    there, and without that the rest is mainly guessing, but it's not unheard of for
    the supposed 'case temp' to be influenced (heated) by the motherboard it's
    mounted on (plus their location is not conducive to good airflow) and Peter's
    suggestion that your additional fans, or one of them, are simply cooling IT, the
    temperature sensor, more (so it reads lower) is a reasonable guess. The part
    that I'm not so sure also follows is the airflow around the CPU socket being
    disrupted (Don't get me wrong, his comment about case airflow disruption is
    correct. It's just that he didn't have your follow up message giving CPU temps
    to go on). Since you say the CPU temps are about the same as with the case sides
    off, and a large fan blowing on it, I'd say the CPU socket airflow is just fine,
    that the apparent 'case temp' difference is simply the motherboard sensor
    being cooled more, and that your multi-fan hurricane setup is moving air as
    well, or more, as the case open observation (with the difference being air
    directly impinging on the case temp sensor, or the motherboard around it,
    causing it to read lower).

    Having said that, if you want lower CPU temps then your conclusion to get a
    better heatsink is probably correct since it seems apparent that 'case flow' is
    not the problem. It's just that you may also find that your 'high' case temp was
    the motherboard sensor reading high and that adding some, or all, of those case
    fans was not really needed.
     
    David Maynard, Sep 5, 2003
    #17
  18. You can't draw those conclusions from the data. What we know is that the reading
    from a temperature sensor you don't know anything about (such as were it's
    located) is lower but that the CPU temp differential to it is higher.

    Case temp may have gone down some but the amount (around the CPU) would be
    reflected in the CPU temperature reading. I don't recall, however, seeing you
    post same conditions, case closed, before and after numbers so we don't know how
    much, if any, things changed under similar conditions (what I recall is a
    comparison to case OPEN with a large fan blowing on it). I don't know from what
    data you come up with the possible 1 to 2 C rise.

    Without 'same conditions' data you cannot draw definite conclusions.

    What we have are two temperature sensor readings, from a sensor you THINK (why?)
    is 'case temp', at what would seem, and I stress SEEM, to be (nearly) the same
    real (leaving aside, for the moment, possible dead spots and inside temperature
    differentials) case temperature. Why would you presume the first one was 'right'
    with the current one 'misreporting'? With no knowledge of what the REAL case
    temperature is, and especially with an apparent measurement inconsistency, why
    would you presume EITHER of them is 'right'?

    Bringing back the 'aside', case temp where? Around the CPU socket? Around the
    hard drive? Around the motherboard near the southbridge? Around the video card?
    Maybe they're similar and maybe they're not.

    If you 'really want to know' what the temperatures are then you need to get
    something you know accurately measures temperature and measure the temperatures
    in the places you want to know about. If all you want is to reduce CPU temps
    then the data suggests you need a better heatsink as it appears 'more case flow'
    is not affecting it.

    Once you have a better heatsink you could try playing around with the case fans
    to see which ones 'make a difference' to your temp readings and perhaps reduce
    the noise.
     
    David Maynard, Sep 5, 2003
    #18
  19. Vin

    Vin Guest

    hi david,

    maybe all those case fans are a bit excessive, but after 3-months of having
    a 350mm floorfan beside my left ear, my 'multi-fan hurrircane' setup is much
    much quieter than the humming noise i can still experience in my head :)

    and the real motivation behind all these mods was my 1-yr daughter has
    started flinging stuff across the room now, and a few days back she holed
    the tv remote into my then open computer case!

    best,
    vin
     
    Vin, Sep 5, 2003
    #19
  20. Vin

    AnthonyR Guest

    What do you mean the chemist didn't have alcohol?
    Do you not have rubbing alcohol in your house? here in the states every
    house has a bottle, its used to clean wounds and stuff.
    What do you use for scraped knee's?
    AnthonyR.
     
    AnthonyR, Sep 5, 2003
    #20
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.