Asus ROG X670E Hero Chipset Temperature

[Orodruin]

I have an Asus ROG X670E Hero motherboard. When I check the chipset temperatures from HWInfo64 and Armory Crate, I get the following result;

Armoury Crate
Chipset 1: 70 Degrees
Chipset 2: 74 Degrees

HWInfo64
Chipset 1: 74 Degrees
Chipset 2: 78 Degrees

Are these temperatures normal?
I want to know the silicon temperature limit and what is the maximum degree that is dangerous.

I can change the thermal pads of the chipsets, but I don't want the card to be out of warranty, so I won't do it for now.

My power profile is set to medium in the power settings.
I thought about choosing the Maximum Power Saving option, but will I encounter USB power issues or similar things if I choose this setting?

 

[Assimilator]

Those temperatures are fine; Armory Crate would warn you if not.

 

[Bill_Bright]

If they are in 74°F and 78°F, then they are fine. If they are 74°C and 78°C, then in my opinion, they are too high - not dangerously high, but too high just the same - PARTICULARLY because the system most likely is operating at near idle conditions while simply monitoring temps with those apps.

So, if your readings are in Celsius, then I would look at your case cooling to make sure you have good front to back air flowing through the case, that the case interior is clean of heat-trapping dust, and your cable management ensures cables are routed and tied back to minimize impacting air flow.

 

[9087125]

Do you have a fairly powerful big GPU? If so that is most likely the culprit. Mine were doing the same thing, cooking at ~72C at 25C ambient on idle., totally unacceptable. Turns out it was the GPU, even on idle it gives off a fair amount of heat (50-70watts irrc) and it's own fans are off, and with the case fans at very low the chipsets are getting cooked. So I proved this was the cause as I put a small fan on the gpu or just had the GPU fans going a bit and the chipsets soon went down to ~54C. So i had the choice of cranking up the case fans alot, or the gpu fans alot, or adding a fan in there somewhere. The truly incompetent people at Zotac didn't have the brains to think about this and how one might want to keep the fans spinning but without spinning too fast (noise), so they made it so 30% is the minimum which is very loud.

Didnt bother with a makeshift heatsink, its just too much work and too inefficient. a flat plate design to fit in there would have terrible heat dissipation and a lot of weight.

If it aint the GPU then put a fan in there somewhere, the chipsets having just a bit of conduction with the breeze is going to do wonders, even if that recirculates around in the case.

As far as Silicon's safe limits go, by itself it can take easily 125C all day everyday, its just intrinsic silicon with some dopant impurity ions. But an IC is more than that, it has many metal paths and delicate insulators wedged in between the terminals and gate that operate on the nano scale that if stressed could easily leak current which will mess with the logic gates, what about the thermal cycles? no science or data on these ICs I know of. The main degradation pathway is always quoted as electromigration which results usually at the ends of the metals where there is a sharp difference in conductivity at the interfaces, here current densities are high. Heat makes it worse but there is no accurate analogue for a real IC that i know of, i.e its not very reliable. CPUs are on a whole nuther level, they are the result of billions of rnd, so they can take some heat and are quite good at it, but some janky chipset in a cheapo mobo?? all bets off! Id be much more concerned about thermal stress from heat cycles, thats what keeps me up at night. A CPU can usually sit at ~125C for many many months without issue, from the very scant literature ICs are the same, designed to have a life expectancy of anywhere from 10-20years sitting at fairly high temps. Once the IC gets over ~100C the reliability diminishes exponentially, but I don't know how well that translates to an IC in a motherboard as those are rather poor analogues. From what I have learned over the years in electronics engineering its hard to say, but it's well know quality components can handle 105C continuous, but should never go above 90C, thats where reliability plummets, so ~70C is really pushing it imo. Again thermal stress is the big unknown.

 

[TheLostSwede]

That seems really hot, my X670E Aorus Master is at 46.5 and 58.3 degrees C in HWInfo64.

 

[FreedomEclipse]

That seems really hot, my X670E Aorus Master is at 46.5 and 58.3 degrees C in HWInfo64.

can confirm

 

[Bill_Bright]

A CPU can usually sit at ~125C for many many months without issue

Huh? Sorry but not even. That is a blanket statement and like all blanket statements, its not correct (and yes, I note the irony in that statement).

While there may be some CPUs designed for personal computer use that can tolerate those temps for a short period of time, very few could for "many many months". Those designed for automotive use that sit inside engine compartments are specifically designed and formulated for such extreme use. But typical PC and laptop CPUs definitely are not.

As far as Silicon's safe limits go, by itself it can take easily 125C all day everyday

True. But "silicon" is an element, Si with an atomic number of 14. And it has melting point of 1414°C. So yeah, heat is not a problem with silicon.

Silicone, however, is a synthetic substance made up of silicon, oxygen and other elements, most typically carbon and hydrogen. In some formulas, the silicone has a high heat resistance and is often used in kitchen cookware or even exhaust gaskets as adhesive sealants in automobile engines.

However, the silicone formulas used in CPUs is very different from a silicone frying pan or oven mitt and no way can a CPU sit at 125°C for "many many months" "without issue". In fact, many will likely be permanently damaged if allowed to reach and/or sit at 125° for more than a few seconds! And for that reason, thermal protection features are integrated into PC and laptop CPUs just to prevent that from happening.

You say a CPU can sit at~125°C for many many months "without issue", but then say,

Once the IC gets over ~100C the reliability diminishes exponentially

should never go above 90C, thats where reliability plummets

Are those not issues?

As seen by this article, CPUs typically have a TJMax rating set by the manufacturer. Note below (my bold underline added),

TJMax (Thermal Junction Maximum) value, which is a specification that represents the maximum operating temperature of your CPU as rated by its manufacturer. A CPU that crosses or is anywhere near its TJMax temperature is considered to be overheating. That's because its transistors become unreliable if the CPU overheats and stays near its TJMax temperature for a long time.

The TJMax value also acts as an indicator for the CPU to initiate its safety mechanisms and lower its clock speed and voltage to prevent overheating that may cause permanent damage.

Also noted in that article,

Intel CPUs tend to have TJMax values of around 100 to 110 degree C, whereas AMD CPUs typically have a TJMax of around 85 to 95 degree C.

That clearly indicates there are many, if not most, CPUs that cannot tolerate temperatures even approaching ~125C° and especially not for a long period of time.

In fact, for many electronic devices, including solid state and IC devices, are operated "near" their upper temperature limits for extended periods of time, that increases the aging of the device. For this reason, we need to ensure our electronics are operated comfortably within its published, normal operating temperature range. Cool as possible is not necessary but keeping it properly cooled is essential to achieve normal life expectancies.

~70C is really pushing it imo.

There we agree - though, for me personally, when my CPU temps hit and sit above 60°C for more than a few seconds, that is an indicator for me that I am overdue in cleaning my case air filters.

 

[Orodruin]

I have an Asus Tuf Rtx 4070ti super graphics card. When the card is idle, that is, when the graphics card is not under load, the fans do not work. There are 3 120mm fans on the front of my case. But the fans are running at 400rpm. I had it muted so it wouldn't make noise.

@TheLostSwede
@FreedomEclipse
@9087125
@Bill_Bright
@Assimilator

 

[Bill_Bright]

Do you have any exhaust fans in back (in addition to the PSU fan)? If not, you may be creating too much over or positive pressure inside the case. You want a little over pressure to prevent air (and dust) from entering through the cracks, ports and optical drives. But too much can create pockets of stagnant, heated air. So a good "flow" through the case is desired. If no fan in back, you might try adding one, or maybe moving one of the front to the back.

 

[oxrufiioxo]

Mine sit in the 40-50c range depending on what I'm doing and thats with a 4090 basically sitting right on top of them on a X670E Aorus Pro X.

Not gonna say what's too high or not but that seems high to me it's been a while but I've worked with a couple X670E hero's and don't remember them breaking 60C.

 

[Bill_Bright]

If you remove the side panel and blast a desk fan into the case interior and the temps drop to a comfortable temperature, then your case cooling needs attention.

 

[Orodruin]

If you remove the side panel and blast a desk fan into the case interior and the temps drop to a comfortable temperature, then your case cooling needs attention.

Here is the layout of my case. In the photo is my old motherboard. Now I have an x670e hero instead.

3 fans in the front. Radiator on top. 1 exhaust fan in the back. I have dust filters. I am using a Corsair 5000x.
I will increase the fan speed and try it out.
Let's see what the result is.

Mine sit in the 40-50c range depending on what I'm doing and thats with a 4090 basically sitting right on top of them on a X670E Aorus Pro X.

Not gonna say what's too high or not but that seems high to me it's been a while but I've worked with a couple X670E hero's and don't remember them breaking 60C.

Your temperatures seem very good. Let's see what kind of solution I can create. Hopefully I can lower them a little more.

 

[Bill_Bright]

And are the radiator fans intake or exhaust. Blowhole (top) fans typically exhaust EXCEPT when used to cool radiators, then they often are intake. If you have those 3 as intake plus 3 more intakes in front, that is 6 intakes with only 1 exhaust (PSU in its own chamber does not count as exhaust). Too many intakes may be your problem.

 

[TheLostSwede]

Can you get some airflow under the GPU, so it blows over the PCB of the motherboard?

My board has an external sensor next to chipset 1 as well, which is the one under the graphics card, but still the cooler one of the two...
Even after an hour or so of gaming, I only see a difference of 1-2 degrees C.

 

[Orodruin]

And are the radiator fans intake or exhaust. Blowhole (top) fans typically exhaust EXCEPT when used to cool radiators, then they often are intake. If you have those 3 as intake plus 3 more intakes in front, that is 6 intakes with only 1 exhaust (PSU in its own chamber does not count as exhaust). Too many intakes may be your problem.

When I turned the 3 fans on the front of my case to a slightly higher speed, I lowered the chipsets from 74 and 78 degrees to 71 degrees. Then, when I changed the pcie power setting from the power settings to power saving mode, my temperatures dropped to 64-65 degrees.

I still wanted to learn the thermal limit. Because the temperatures seemed high when I saw 78 degrees.

Can you get some airflow under the GPU, so it blows over the PCB of the motherboard?

My board has an external sensor next to chipset 1 as well, which is the one under the graphics card, but still the cooler one of the two...
Even after an hour or so of gaming, I only see a difference of 1-2 degrees C.

View attachment 370883

I increased the speed of the fans in the front of my case. Both chipsets dropped to 71 degrees. I changed the PCIe power setting to power saving mode. They dropped from 71 degrees to 64 degrees.

I'm wondering if I should replace the thermal pads on the chipsets with much better thermal pads.

 

[TheLostSwede]

I increased the speed of the fans in the front of my case. Both chipsets dropped to 71 degrees. I changed the PCIe power setting to power saving mode. They dropped from 71 degrees to 64 degrees.

I'm wondering if I should replace the thermal pads on the chipsets with much better thermal pads.

It could just be a somewhat poorly designed heatsink. Changing thermal pads might help, just make sure you use a similar thickness.

 

[Hankieroseman]

I have an Asus ROG X670E Hero motherboard. When I check the chipset temperatures from HWInfo64 and Armory Crate, I get the following result;

Armoury Crate
Chipset 1: 70 Degrees
Chipset 2: 74 Degrees

HWInfo64
Chipset 1: 74 Degrees
Chipset 2: 78 Degrees

Are these temperatures normal?
I want to know the silicon temperature limit and what is the maximum degree that is dangerous.

I can change the thermal pads of the chipsets, but I don't want the card to be out of warranty, so I won't do it for now.

My power profile is set to medium in the power settings.
I thought about choosing the Maximum Power Saving option, but will I encounter USB power issues or similar things if I choose this setting?

That's about where my 170W 7950X, X670 Extreme board runs when in the sim flying. 2 years now. Cooler is be quiet! Silent Loop 2 360. I hear it will run 95C all day long. Built for it.