Results 1 to 8 of 8

Thread: G5 Hotter than normal; memory controller heatsink

  1. #1
    Join Date
    Dec 2006
    Posts
    81

    Default G5 Hotter than normal; memory controller heatsink

    Hi guys. My G5 dual 2.3 seems to be having some heat issues in the last couple days. I'm running Leopard 10.5.8. I will add that it feels hotter sitting next to it than it used to, but I don't know what the readouts were since I hadn't run Temperature Monitor until today.

    Temperature Monitor showed the graphics processor chip (nVidia 6600) alert at 168.8 F. THe memory controller heatsink was reading at 155.3 F, which was within limits.

    I had an ATI 9600 Pro Mac edition so I swapped. But now, the alert shows that the memory controller heatsink goes up to 170.15 F. When the nVidia was in use, the memory heat sink wasn't showing any alerts, but now it is.

    Currently the memory controller heat sink fluctuates between 164–170.15 degrees F.

    Here are my readouts:

    CPU A Die: 140.4 F
    CPU B Die: 142.2 F
    Drive Bay: 85.1 F
    Main Logic Board Backside: 120.7 F
    Memory Controller Heatsink: 168.125 F
    Processor Card A Ambient: 77.5 F
    Processor Card B Ambient: 77.9 F
    Backside: 121 F

    Here are Celsius readouts (about 10 minutes later):

    CPU A: 59.7 C
    CPU B: 59.9
    Drive Bay: 30 C
    Main Logic backside: 49.5
    Memory Controller Heatsink: 74.5 C - 73. 5 C
    Processor Card A: 25.2 C
    Processor Card B: 25.5 C

    My Fans are:

    CPU A Intake: 436 rpm
    CPU A Exhaust: 437 rpm
    CPU B Intake: 363 rpm
    CPU B Exhaust: 354 rpm
    Drive Bay: 10000
    Slot: 76 rpm
    Backside: 51 rpm

    Is this normal or cause for alarm, something I can solve, or needs to be serviced?

    Thank you.
    Last edited by macmac; 06-20-2010 at 07:36 PM.

  2. #2
    Join Date
    Jan 2002
    Location
    NW Montana
    Posts
    8,197

    Default

    Are those temps accurate? Have you used/tried any other apps to check accuracy. Like Hardware Monitor, iStat or several others some free, some pay...

    Do those graphics cards have any cooling fans? I never had a G5 and am busy as heck now, but myself or some one will comment.

  3. #3
    Join Date
    Mar 2001
    Location
    Virginia... where one Democrat CAN make a difference
    Posts
    2,929

    Default

    Those look about right to me but it's been a LONG time since my G4...
    Damien,

  4. #4
    Join Date
    Feb 2001
    Location
    on the landline, Mr. Smith
    Posts
    7,787

    Default

    I think that is about right, as I have a strong memory of G5s having the memory controller being hottest. I used to monitor a half dozen G5 xserves, and I remember being surprised about the memory controller, but they were all the same and never had an issue.

    Can't speak to the video card.
    "Imagine if every Thursday your shoes exploded if you tied them the usual way. This happens to us all the time with computers, and nobody thinks of complaining." -- Jef Raskin

  5. #5
    Join Date
    Dec 2006
    Posts
    81

    Default

    Thanks guys. Here's today:

    Hardware Monitor:

    With ATI Radeon 9600, I got a memory Controller Heatsink alert.

    Ran Apple Hardware Diagnosis twice: Error in DIMM #6 slot. So I removed it and its paired partner. Would that have anything to do with Memory Controller Heatsink alert?

    I cleaned and reinstalled the nVidia 6800 Ultra DDL (which is a heavy duty card) and now the Memory Controller Heatsink seems to be holding its own, but now I get a new alert for the Slot Combined, saying it has exceeded 10000 (is reading 17359). I assume it's talking about all 3 PCI slots (and sata), but what could be causing such an increase?

    The fans at idle are about 300, but they do kick in if I'm doing something. I had wondered if they weren't working properly but I guess they are.

    I'm new to the temperature readout interpretations. I don't want to neglect the computer if it is indeed in danger, which is why I want to know if this is cause for concern where I shouldn't be using it!! or just normal temperature fluctuations.

    Here are my current readouts with Hardware Monitor:

    2010-06-21 21:48:51 -0400

    CPU A Die Temperature: 62.7℃
    CPU B Die Temperature: 62.4℃
    Drive Bay: 26.0℃
    Graphics processor case: 53.0℃
    Graphics processor chip: 74.0℃
    Main Logic Board Backside: 47.1℃
    Memory Controller Heatsink: 72.9℃
    Processor Card A Ambient: 38.5℃
    Processor Card B Ambient: 36.0℃
    CPU A Voltage: 1.2769 V
    CPU B Voltage: 1.2744 V
    CPU A Current: 23.19 A
    CPU B Current: 22.95 A
    Processor Card A 12V Supply Current: 2.82 A
    Processor Card B 12V Supply Current: 2.73 A
    Slot 12V: 8208.0
    Slot 3.3V: 1393.0
    Slot 5V: 7693.0
    Slot combined: 17408.0
    CPU A Exhaust Fan: 670 RPM
    CPU A Intake Fan: 649 RPM
    CPU B Exhaust Fan: 583 RPM
    CPU B Intake Fan: 565 RPM
    Drive Bay: 1000 RPM
    Main Logic Board Backside: 20.0 %
    PCI Slot Fan: 49.0 %
    Graphics processor fan: 32.0 %

    iStatPro is showing about the same.

    I took out the front intake fan to clean; the rear exhaust looks pretty clean from my flashlight in there. How is the proper way to clean these?

    Thank you so much.
    Last edited by macmac; 06-21-2010 at 08:30 PM.

  6. #6
    Join Date
    Dec 2006
    Posts
    81

    Default

    ADDENDUM: I've learned that the default alarm trigger values set in Hardware Monitor are not meant to be used as actual values; they are merely placeholders for the user to enter their own. So I've been basing my concern needlessly, perhaps? Although the computer did seem to be spewing hotter air than normal, which made me investigate this. I made some changes (took out error ram according to Hardware Diagnostic, relocated computer to get better ventilation, etc. So now I just need numbers so that I know where I really stand.

    So, since I have no idea what the real values ought to be, is there a list somewhere / or what do you have yours set at, since you know what you are doing?

    Thank you!

  7. #7
    Join Date
    Jan 2002
    Location
    NW Montana
    Posts
    8,197

    Default

    Check the manufacturer's recommended operating temperatures. Some equipment seems to run hotter than I'd assume or want but many seem OK at 50 C or even - 60 C according to the company that makes it. Check it out.

    Cooler is better. How much airflow do you have around your tower?

  8. #8
    Join Date
    Dec 2006
    Posts
    81

    Default

    The computer is on a 1" shelf board so that it can slide out easily from underneath the desk when in use, making the back of the computer about 3 feet from the wall / back of desk.

    The front and sides are open with no obstructions around them.

    I'm referring to the threshold values that a person would enter in the Hardware Monitor preferences so as to trigger the alerts. I don't know that Apple or mfg list all those items, and I couldn't find reference to them. I would like to know what the normal values for these items are, particularly for the G5 2.3 dual (early 2005), or a general ballpark to know if I'm in safe territory:

    CPU A Die Temperature
    CPU B Die Temperature
    Drive Bay
    Graphics processor case
    Graphics processor chip
    Main Logic Board Backside
    Memory Controller Heatsink
    Processor Card A Ambient
    Processor Card B Ambient
    CPU A Voltage
    CPU B Voltage
    CPU A Current
    CPU B Current
    Processor Card A 12V Supply Current
    Processor Card B 12V Supply Current
    Slot 12V
    Slot 3.3V
    Slot 5V
    Slot combined
    CPU A Exhaust Fan
    CPU A Intake Fan
    CPU B Exhaust Fan
    CPU B Intake Fan
    Drive Bay
    Main Logic Board Backside
    PCI Slot Fan
    Graphics processor fan

    Thanks.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •