cancel
Showing results for 
Search instead for 
Did you mean: 

Overheating chipset.

NotHarry
Level 7
I have had a lot of trouble with my new 'Asus rog strix z690-A Gaming wifi D4' motherboard. My latest problem is a chipset that idles at 60c+.
I want to strip the heatsink for chipset in order to reseat it but you have this daft piece of plastic over the heatsink that has a cable tie attached to it, and I don't know how to remove it.
Have any of you guys striped the northbridge heat sink from one of these boards, how do you get the piece of plastic off first?
1,695 Views
97 REPLIES 97

IT Troll wrote:
Ah yes, I can see that in your photo. Definitely no fibres like that on my factory pad, it was quite soft and spongy. The Thermalright pad I replaced it with is quite similar, although rated at 12.8 W/mK. I think the glued on acrylic cover was a significant factor with my motherboard.


Yeah I can imagine that was more tricky to remove if there was glue, Glad it all worked out in the end though, I've knocked off about 20c now since the mod, I could probably do more by using the Gelid Ultimate I have but the pas are stiffer I don't know how they'd react to the chipset and the heatsink but either way I am happy now doesn't go anywhere near the 70's like it did previously.

IT_Troll
Level 10
Jeez that forum upgrade took a long time. I wonder if the server chipset overheated during the process... 😉

reddeath
Level 9
Looking at the chipset temperatures of my ROG STRIX Z790-A GAMING WIFI D4: between 75 and 85 degrees.

When rendering in Davinci Resolve (on a NAS) the PCH temp raises up to 96 degrees! This is clearly not right.

Has anyone tried to do fix high temps on this board? What are the temps on other copies of this board?

I can't comment on your board specifically, but that does seem to be on the hot side. Still within limits though! Based on my experience, there are a few things you can do to lower temperatures:

1. Remove the PCH heatsink, lap it flat and replace the thermal pad with a good quality one of the correct thickness. Beware potential warranty issues. This is a good, but time consuming and higher risk approach. Expect a 10-15C temp reduction. I might do this, once my warranty expires. Find some images from other users and follow those if you want to do this.
2. Get a small (8cm) fan and point at towards your PCH heatsink . Expect a 10C reduction.
3. Lower your BIOS PCH voltages (in Tweakers Paradise). Expect a 2-3C reduction at mid-point voltages.
4. In advanced Power Options (in Windows) go to PCI Express/Link State Power Management and select moderate or maximum power savings. Expect a 5C reduction at maximum power savings.
5. Make sure you have good case airflow. My PCH creeps up by 5C if I turn off the case fans. Room temperature also make a difference, obviously.
6 The BIOS can make a difference as well.

NOTE that 3 and 4 can cause WHEA errors if you go too far, so use HWInfo to monitor for those. On my 690 Hero I can use middle PCH voltages (0.92V/0.76V) and moderate power savings in power options. Some people have used minimum PCH voltages without problems, but that gives me WHEA errors. Same with power savings. If I go for maximum I get errors, but moderate is fine.

For me, the small fan is a cheap, simple and effective solution and for most of the year my PCH now runs at 55-60C. Up to 65C in games. Get a speed controlled fan (or two) and plug into the motherboard. Set up for the best balance of noise/cooling using the BIOS or Armoury Crate. I do clean my PCH fan setup weekly as it's a dust magnet.

Hope that helps...
Z690 Hero, BIOS 3401, MEI 2345.5.3.0, ME Firmware 16.1.30.2361, 7000X Case, RM1000x PSU, i9 12900K, ASUS TUF OC 3090TI, 2 x 16GB Corsair RAM @ 5200MHz, Windows 11 Pro 23H2, Corsair H150i Elite AIO, 4x Corsair RGB fans, 3x M.2 NVME drives, 2x SATA SSDs, 2x SATA HDs.

reddeath
Level 9
Well, the chipset's heat spreader is directly behind the GPU, RTX 3080. CPU fans on the AIO are blowing from the side, but there will not be any airflow reaching the chipset, as there is no gap between the GPU and the chipset.

How can one replace the heat pads? Are there any guides that show this? Will the warranty be void if doing so?

Murph_9000
Level 14
This may be slightly controversial, but one thing about this is that it's a bad idea to buy one of Nvidia's own GPUs if slot mounting, better to get a Strix/TUF or any of the partner cards that has a reasonably standard cooler. The Strix & TUF coolers, for example, exhaust some of the GPU cooling air over the chipset when slot-mounted. That air is warm, but still cooler than the PCIe 4 chipsets, and does provide some cooling for them. Nvidia's own cooler design basically just blocks the chipset cooling, essentially a design flaw in their cards when the chipset will be under the GPU on the majority of mainstream boards.

Also, front mounting the GPU does make a big difference. The X570S in my Crosshair VIII Extreme runs about 5 to 10C cooler with the Strix GPU front mounted in the case.

Murph_9000 wrote:
This may be slightly controversial, but one thing about this is that it's a bad idea to buy one of Nvidia's own GPUs if slot mounting, better to get a Strix/TUF or any of the partner cards that has a reasonably standard cooler. The Strix & TUF coolers, for example, exhaust some of the GPU cooling air over the chipset when slot-mounted. That air is warm, but still cooler than the PCIe 4 chipsets, and does provide some cooling for them. Nvidia's own cooler design basically just blocks the chipset cooling, essentially a design flaw in their cards when the chipset will be under the GPU on the majority of mainstream boards.

Also, front mounting the GPU does make a big difference. The X570S in my Crosshair VIII Extreme runs about 5 to 10C cooler with the Strix GPU front mounted in the case.


Here is a picture of the card I have from the side that faces the PCH heatsink. It seems to be open and presumably air blows through it on the chipset. At least on part of it.

https://hexus.net/media/uploaded/2020/10/6030ddee-47b9-4b05-8c6f-0456991b89fd.jpg

More pictures visible in the article here:

https://hexus.net/tech/reviews/graphics/145987-inno3d-geforce-rtx-3080-ichill-x3/

If the GPU fans are at 0 RPM, due to low load, the PCH temps hover around 80+ in idle. Going above 90 when PC is under load for a while.
Sometimes when the GPU fans run at 1300+ RMP, due to some light GPU usage, or a bug in fan management, the PCH temps hover at 70+. This could imply that there is some airflow from the GPU to the PCH, but the GPU must be cool in order for the flow to help a bit.

It is indeed worrysome for me that everything in the whole PC is under 40 degrees in idle, except of the PCH which is at 80 😞

Spent some time to upgrade the CPU from i9-12900 to i9-13900.

Took the opportunity to improve the PCH temperature by doing the following:

Adjust power plan to “Maximum Power Savings�
PCH temperature dropped from 80 degrees Celsius to 57 degrees Celsius

Replace PCH thermal pad and remove plastic decorative piece from the heat sink
The temperature has dropped from 57 degrees Celsius to 46 degrees Celsius.

96837

96836


Note that all the MVME slots are fully installed with SSDs.

PauloPanda wrote:
Spent some time to upgrade the CPU from i9-12900 to i9-13900.

Took the opportunity to improve the PCH temperature by doing the following:

Adjust power plan to “Maximum Power Savings�
PCH temperature dropped from 80 degrees Celsius to 57 degrees Celsius

Replace PCH thermal pad and remove plastic decorative piece from the heat sink
The temperature has dropped from 57 degrees Celsius to 46 degrees Celsius.

96837

96836


Note that all the MVME slots are fully installed with SSDs.


What does the Maximum Power Saving power plan do for the performance of the PC?
Do you have any pictures of how it looks like after removing the plastic covers?

Hello guys. Two weeks ago i got an Asus Z690-E strix. I also bought an I7-13700K and 64GB DDR5 6000 ram. I've used intel since 2006, and this chipset was the hottest i ever owned. It really didn't make any sense, since supposedly it has the same TDP many other intel chipsets. In fact, it has the same 6W tdp as my Z170 board.


Anyways, this board was running consistently into the high 70's low 80's in idle. I know that the silicon is designed to run at those temps. But the balls holding the chipset will last a lot less with such drastic changes in temperature.

So i decided to optimize a little bit my setup in order to drastically lower the temperature.

1. Turn on my GPU fans. I know i know. Having a 100% silent GPU is the best feature. But having my GPU cooking at 55-59° near the chipset all the time didn't help. I used MSI afterburner and set my fans to 30% (around 790RPM). This keeps my GPU under 35°. And with the stock voltage and pad it dropped the temps from the mid 80's into the low 70's.

2. Undervolt the chipset. This made a massive difference. I went full deep and set the 1.05V to 0.8V. And the 0.8V to 0.6V. I know that not all the boards will do this and be stable. I was lucky to be able to use such undervolt with no issues at all. This dropped the temps from the high 70's into the high 60's. But i still thought i could do better.

3. Replace the GARBAGE thermal pad. I really dont know how in the world ASUS cheaps out on this component, specially in a board that costs more than 350 USD. Replacing the pad reduced the temps from the low 70's - high 60's in idle to 55°-60° in idle. And the max recorded temp recorded since i installed the pad is 69° while gaming (im using a RTX 3090).

This is a massive difference. from 84°-86° in idle to 55°-60° in idle. And the maximum temperature from 94° to 69° (nice). You should also consider it's summer in Argentina right now. And every day we are having +30° weather. Right now my room is at 29.8° and the chipset is running at 62°.

The pad replacement is relatively easy to perform. You have to remove 5 screws and all the heatsink comes off. I used Gelid Gp-ultimate Tp-gp04-r-d 120x20x2mm thermal pad. It's for M.2 ssds but you can cut it to fit the chipset. And 2mm seems perfect for replacement.

97193

97194

97195

97196

97197

These are the temps after almost 30 hours of uptime. I played CSGO and Rocket League. Both CPU and GPU got stressed and the chipset performed great. I'm also running three M.2 SSDS.