[Z590] Asus ROG Maximus XIII Hero - PCH High Temperature

[EncodeGR]

My PCH temperature is fairly high, which concerns me. I'm used to everything else in my system being pretty chilly, usually around 28-30 C. The airflow is also really good. The *only* thing that gets hot is the PCH, and i don't understand why.


-- Some Backround Info, Setup --

Intel Pentium G6400 4.0GHz (Lol, well... until i get my Rocket Lake chip)

MSI GTX 1070Ti Gaming X (Until i actually manage to find a 3080/3070, if ever...)

HyperX DDR4 32GB 3600Mhz XMP CL17 (4 x 8 GB)

Asus ROG Maximus XIII Hero (Z590)

Asus ROG Thor 850 Plat

Samsung 980 Pro M2. NVMe PCIe 4.0 500GB (installed in M2.2_2, cpu bound)

Noctua NH-U12A

Noctua NF-A14 PWM (x6)



-- Things i've tried --

• Updated Drivers. Updated ME FW. Updated BIOS.
  
• Inspected PCH heatsink and thermal pad and resit. No change.
  
• Removed M.2 NVMe drive and GPU. No significant change.
  
• Checked the PCH 1.8v VCCIN voltage, stable 1.8v.
  
• Made sure nothing is "overclocked", using XMP, or enhanced in any automatic way.
  
• Lowered VCCIO and VCCSA to 1.10v, since automatic was pushing both to 1.40 for some reason !! (CPU weak memory controller ?). No change.
  

** I need to mention i couldn't find the PCH Core Voltage, anywhere in the settings. No idea why.


-- PCH Temperature --

PCH temperature starts at 35-38 C, during a cold boot. During the next 5 minutes it climbs and stabilizes at 56 - 62 C, while IDLING.

And that's with a system without GPU, or M.2 NVMe drive. Just the CPU and its iGPU which stay at 24 C, cooled by a Noctua NH-U12A. With a fully assembled system and under load it goes up to 70 - 80 C.


Now... The only thing i have not tried yet, is to completely remove the plastic ROG-logo cover, on top of the PCH, as it may be severely suffocating the heatsink, restricting the airflow towards it.

That would explain the temps under load, but i'm not sure if that explains the initial fast-climbing temperatures while idling. And honestly, without the cover the LEDs underneath it are blinding, so the Aura lighting needs to be completely disabled. Could this be a flawed design ?? I doubt.

I don't understand what's wrong. I've seen people reviewing this mothereboard, while reporting PCH idle temps at 35C and 55C under load. I know that right now, it's not gonna cause any sort of damage. I am however concerned about the longevity and a possible degredation over time, thinking that my PCH is at 75-80 C while gaming.

Sorry for the long post, had to make sure everythong's covered.

Any help would be appreciated 🙂

-- HWInfo & HWMonitor -> 3 minutes after cold boot --
-- Bios Monitor -> 10 minutes after cold boot --

[sdmf74]

TasmaN wrote:
Hi, everyone! Found some options: if set options like on my screenshot temperature of PCH is around 45-50C at idle. By default all of them are disabled...so no need to set a fan toward it. Maybe, ASUS should look this part of UEFI...

I too am concerned about the longevity of my M13HERO with the PCH running at such high temps all the time but what is suggested above is a terrible idea! I can't believe nobody questioned the reason why these settings are dropping idle temps by 20-30c

By enabling ALL the power management features of the motherboard you are nuetering all your devices lol, you are essentially letting windows do power saving instead of your motherboard/bios doing them or disabling them. You are killing your pcie/M.2 SSD speeds and giving them a kind of wake up delay. Also Probably making your overclocks unstable, not too mention it affecting your expensive gpu etc etc etc.

I didn't pay $500+ for a good overclocking Mobo and expensive CPU just to make it (and all the devices connected to it) run like an entry level $150 one.
Heck if you want your CPU to run 30c+ cooler you can manually overclock it to 3200mhz and set a static voltage of about 1v too!

I believe Asus should have at least made the chipset heatsink with slits in it that run all the way through the heatsink to optimize our casefans airflow.
Also the reason why the chipset is running hot isn't because of PCIE 4.0
but because the chipset runs at double the speed of the Z490 and previous Intel chipsets.
On Z590, Intel has doubled the link speed, going from PCIe 3.0 x4 to x8. The jump to x8 effectively doubles the amount of bandwidth available for any chipset-connected devices (storage and networking for example). Additionally, the chipset now has native support for USB 3.2 Gen 2x2 (20 Gbps).

That being said I'm open to other solutions to cool the chipset but ASPM is NOT an option IMO.

[blukas69]

sdmf74 wrote:


I believe Asus should have at least made the chipset heatsink with slits in it that run all the way through the heatsink to optimize our casefans airflow.
Also the reason why the chipset is running hot isn't because of PCIE 4.0
but because the chipset runs at double the speed of the Z490 and previous Intel chipsets.
On Z590, Intel has doubled the link speed, going from PCIe 3.0 x4 to x8. The jump to x8 effectively doubles the amount of bandwidth available for any chipset-connected devices (storage and networking for example). Additionally, the chipset now has native support for USB 3.2 Gen 2x2 (20 Gbps).

That being said I'm open to other solutions to cool the chipset but ASPM is NOT an option IMO.

Great explanation I was not aware of that, Makes perfect sense.

[singamer]

TasmaN wrote:
Hi, everyone! Found some options: if set options like on my screenshot temperature of PCH is around 45-50C at idle. By default all of them are disabled...so no need to set a fan toward it. Maybe, ASUS should look this part of UEFI...

I tried your settings and my chipset temp dropped from about 75C to about 70C! A significant drop. Now I am wondering what I have compromised to get this reduction in chipset temp. Any idea?

[corb06]

Thanks for the solution! Dropped from 81-82 idle to 62-64. 

[Zeroed85]

I think the PCH temp has been improved with Thunderbolt enabled in beta BIOS 0909. It was once 80C with Thunderbolt enabled and now 65-70C on BIOS 0909.

[PeterK77]

I have the same MOBO but populated with i7-11700K. PCH has 70-80C during idle load (ambient temperature is 30C). So I disassembled plastic cover above PCH heat sink. Temperature went down approx. by 10C. And when I added another 80x80 noctua ventilator against PCH, temperature went down by 5-6C. So, now I see temperature 56-60C instead of 80C. I am also afraid about longevity of MOBO. Here, there is web page where temperature is reported, which is way lower.

https://www.techpowerup.com/review/asus-rog-maximus-xiii-hero/15.html

[sdmf74]

Something else to keep in mind. I was just going over my notes and when I first got this motherboard/CPU I had all the APSM bios settings enabled by accident
(forgot to disable them) and I was getting hundreds of whea errors in the event viewer So if you are gonna restrict your cpu/chipset with power mangagement settings be sure to check your logs for WHEA errors. If you have them then disable all APSM settings in bios.

Keep in mind im not 100% sure this was the cause but the errors did go away after disabling APSM, some other things to check if you are getting these types of errors:

1. Raise Vcore Voltage (this was not the cause in my case but usually is the culprit for this type of error)
2. Increase Dram Voltage slightly (even if using XMP)
3. Lower System Agent and VCCIO memOC Voltages ( dont set these to auto voltages, find the lowest possible voltages for these as you would with vcore w/out sacrificing stability)
System Agent has its own LLC setting now. I found that setting SA LLC to level 3 & setting SA voltage as low as possible kept it closest to what its set to in bios (least vdroop).

With 3600mhz 16 16 16 36 ram 32gb kit overclocked to 3733mhz I was able to get them down to:
SA-1.347v
VCCIO mem oc-1.310v

I have even seen Z590 OC guides recomending to leave these voltages set to AUTO when as high as 1.56v or more! Bad idea IMO

[tflor]

Hi all

Just purchased the motherboard last week and, to my dismay, I'm also experiencing high PCH temperatures with my Maximus XIII Hero!

I've been using Asus boards since Pentium II times (ASUS P2B), and Asus Hero boards since 4790K and never have I ever seen PCH temps so high!

I was shocked when I had seen it and thought to myself - surely they forgot to put a thermal pad between the PCH and the heatsink so took the PCH heatsink apart and checked, but no, thermal pad is there, and it is of good quality too!

BIOS updated to the latest available (1007), MEI updated to the latest available (15.0.23.1706 V5), all drivers in latest versions.
CPU is not even overclocked (XMP applied though), VCCIO and VCCSA both manually set to 1.1V

It is clear to me that Asus botched the PCH heatsink design, and worse yet, they covered what is already a crappy heatsink with a plastic cover! I'm sure RMA's for this board will be through the roof! I've already seen my PC BSOD multiple times and I've only had it running for a few days.

Luckily I got my board from Amazon and after speaking to their customer service, they agreed to take the board back.
A replacement is on order - and guess what, it isn't ASUS board for the first time in years for me!

Overall, I must admit, I am very disappointed because, other than the crappy PCH heatsink design, it's one of the best equipped and engineered Z590 boards for the 11th gen Intel platform.

I'm actually wondering, how the hell did this board pass QA checks???

Oh, and to those ASUS PR guys saying that all is good and within spec - if so, why bother with a heatsink, surely there is no need for it? And it costs money to make, so might as well try and cut the cost and not include one at all!
Maybe you should actually forward the concerns of your loyal user base to your engineering department, not try to make a bad situation go away with a bit of PR bull$h1t.
And certainly learn from it to make sure mistakes like this don't happen!

Below screenshot of my temps, with the plastic PCH cover removed, after about an hour of sitting idle at the desktop.

[sdmf74]

Please read the thread, if you had you would know that the reason for high PCH temps is not a defect, it is because Asus chose NOT to make the chipset PCI gen 4 but rather PCI gen 3 but double the speed of previous generations! So the chipset is running at x8 instead of x4. This creates more heat. They prolly did this cause they didnt want the extra cost & noise of adding a fan to the chipset like AMD did. They also added a ton of other features to the chipset like dual thunderbolt, dual Nics, extra M.2 slots etc. etc.
I agree though that they could have put slits (for better airflow) in the chipset heatsink.

If you exchange you're motherboard the only thing you are accomplishing is screwing yourself out of a quick & easy exchange in the event you encounter an "actual" defect in the next few weeks AND you risk screwing the rest of us out of great return/exchange policies by Amazon because people abuse their policies by returning perfectly fine hardware. You are just gonna get another motherboard with the same exact Non-issue.

If you want to lower your chipset temperatures you can:
1. Get a better case & fans with better airflow (Caselabs is coming back, maybe invest in a nice case)
2. Disable some of the chipset hardware you are not using through the bios, ie One of the Nics (chances are you are not using both of them). Disable any SataIII & M.2 & PCIe ports you have nothing plugged into.
3. Dont be a noise snob & turn up your fans, they are there for a reason.

I was able to get my idle PCh temps down to... well currently it is sitting at 59c which is perfectly acceptable. I have all my APSM power saving features disabled.

[tflor]

Hey sdmf74

Thanks for trying to be helpful, but I'm acutely familiar with the architecture of Z590 chipset - it wasn't ASUS' choice; ASUS don't design chipsets. It was Intel's decision to switch the DMI link speed from PCI-E 3.0 x4 to PCI-E 3.0 x8. As ASUS was engineering the motherboard, this should've been taken into account and PCH heatsink design beefed up to ensure the PCH is being cooled properly.

I have read the thread through and through, ahead of writing up my 1st post, and as I'm sure you can agree, solutions posted here so far are half measures and not permanent fixes.

The only two permanent solutions to this problem that I can envisage are:
1) ASUS release new BIOS version, with PCH undervolted (unlikely as might cause instability; unless the PCH is overvolted in the current BIOS version)
2) ASUS release a revised PCH heatsink and agree to supply at no additional cost, to all who purchased the board, and pull the existing inventory from the channel to apply the revised heatsink (extremely unlikely, as this would mean lost revenue and reputational damage - this would basically be ASUS admitting that this is a PCH heatsink design flaw)

I'm unwilling to be applying half measures to limit the connectivity options which I paid for (and it isn't a cheap mobo we're talking about here), increasing fan speeds, buying new fans or cases - just to accommodate the product that should not have had this design flaw in the first place. It is your right of course to disagree with me and do all those things, if you prefer.

I sincerely hope ASUS will own up to the problem and fix it, but given the choice between entrusting to a soul-less company and returning what I consider to be a flawed product, I choose the latter.

And finally - I have returned my flawed Maximus XIII Hero motherboard, exercising my right of return, as any sane consumer would, and I would highly encourage everyone who can, to do the same. I purchased an ASUS MAXIMUS XIII Apex instead (best motherboard I could find with an actual PCH heatsink), and here are my temps (note that I didn't buy a new case or fans - neither of which I need; I didn't increase the fan speeds, nor did I apply any other shenanigans):