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.

    Max VRM Temp + Help for Identification?

    Discussion in 'Hardware Components and Aftermarket Upgrades' started by Che0063, Dec 13, 2019.

  1. Che0063

    Che0063 Notebook Evangelist

    Reputations:
    341
    Messages:
    368
    Likes Received:
    478
    Trophy Points:
    76
    I've been running my 2in1 tablet (m3-7Y30) with a 14W TDP for almost a year now, thinking that everything was fine and dandy. I had limited my CPU temp to 90C deliberately in Intel DPTF via BIOS settings, and so I assumed that everything would be fine.

    Today when I decided to open up HWInfo, I got a surprise, because although my CPU was perfectly fine at around 90C, my VRMs routinely stayed at 110C.

    [​IMG]

    I've run my system for prolonged periods at its 14W TDP, and I assumed it was fine because I never get VR Thermal alerts in ThrottleStop. CPU always runs at full 2.4GHz turbo. Nothing ever throttles except the GPU (intentionally set in the BIOS - throttle iGPU before CPU) as shown by TS:

    [​IMG]


    That's all great and I'm happy to get full combined CPU plus iGPU performance, but the VRM temps are really concerning me - I've seen it go up to 120C when a heatwave swept over us a few days back. I know MOSFETs can handle up to 150C, but I feel uncomfortable with this temperature - any advice? Is it fine or should I work to reduce the temp?

    I can set in the BIOS for the entire system to throttle back if the VRMs reach a certain temperature; though stranglely by default VRM throttling is disabled - I'm assuming it's because my OEM intended the VRMs to only have to deal with 4.5W, and I'm now running them at over 3x their rated power.

    Also what does everybody else set their PROCHOT temps to? I've set it to 90C just to be safe.

    I'm having trouble identifying where the VRMs are - I've attached close-up images of my laptop's (tiny) motherboard, can anybody help? I'm assuming they aren't heatsinked, given the high temperatures
    https://imgur.com/gallery/8At0WzJ

    EDIT: This laptop is entirely passively cooled, so I can't feasibly redirect air flow. The laptop case is so thin I haven't found a single laptop fan that can fit in there.
     
  2. jotm

    jotm Notebook Evangelist

    Reputations:
    347
    Messages:
    480
    Likes Received:
    87
    Trophy Points:
    41
    That's an interesting BIOS you've got there. Is it a custom one for that laptop?

    Yeah, VRMs and the South Bridge seem to be ignored by everyone from manufacturers to overclockers. I remember reading some laptops would throttle when the SB/PCH would overheat.

    I would not want any kind of chip working at 95+ continuously.
     
    Papusan likes this.
  3. Che0063

    Che0063 Notebook Evangelist

    Reputations:
    341
    Messages:
    368
    Likes Received:
    478
    Trophy Points:
    76
    Annotation 2019-12-14 125143.png
    Interestingly Intel PCH seems to be completely under control.

    Fully unlocked BIOS out of the factory:
    XIsio.jpg bISO.jpg
     
  4. unclewebb

    unclewebb ThrottleStop Author

    Reputations:
    7,815
    Messages:
    6,414
    Likes Received:
    6,728
    Trophy Points:
    681
    The vast majority of CPUs that Intel has manufactured during the last decade have the throttling temperature set to 100°C. Some are a little less and some are a little more. During the 3rd Gen, Intel's magic number was 105°C. Intel seems to consider a peak core temperature under 100°C to be safe. Here are the two signals that go active when an Intel CPU gets hot.

    [​IMG]

    A peak core temperature up to 100°C seems perfectly fine. Keep in mind that the temperature sensor is tiny. As soon as you move away from this hotspot, the temperature drops rapidly.

    Your HWiNFO reported VR Temp definitely looks HOT but no one in any forum is going to know for sure how hot is too hot. If your tablet is important to you and you want to be on the safe side, it would be a good idea to reduce your turbo power limits during any heatwave.

    One thing I noticed is that you have set the thermal throttling temp to 90°C in the bios but your screenshot shows that your CPU is at 90°C but the CPU speed is not throttling at all. It also shows a peak temperature of 92°C. Usually if it reaches the thermal throttling temperature, even for a millisecond, a check mark will appear in the TS - PROCHOT box to confirm throttling. The multiplier will also instantly start reporting slightly less than a perfect 24.00. In my opinion, the reduced bios temp is not working. Try monitoring the iGPU speed with GPU-Z. Maybe the iGPU is throttling at 90°C but the CPU does not seem to be.

    Congrats on using TS to get the most out of that tablet. :)

    Edit - Here is a reference pic of some CPU thermal throttling from my archives!

    [​IMG]

    I was doing some testing of the TS - PowerCut feature. The CPU is reporting low power consumption to prevent power limit throttling. This was working great until thermal throttling kicked in. THERMAL is red in the Limit Reasons CORE column. You can see that the multiplier is slightly less than 34.00 which is what it normally runs at when 4 cores are active and it is not throttling. The check mark in the PROCHOT box also confirms thermal throttling along with the 99°C reported peak core temperature.
     
    Last edited: Dec 15, 2019
    Che0063 likes this.