High MOSFET and VRM temperatures

[Nikola_36]

Hello guys, it has been brought to my attention that these temperatures could be causing the game Battlefield 1 to stutter.

http://imgur.com/a/4ECsT

MOBO: Asus H81m-K
CPU: i5 4590 (never hits 60C)
CPU cooler: Arctic Freezer i32 (a forum member told me that tower coolers offer poor cooling for VRMS/MOSFETs.

Sorry for poor understanding of this. Could the sensors be wrong? These temps are like this 24/7, and vary by only 1 degree. Should I cool them? Are they dangerously high? They have been like this for more than a year I belive (I didnt pay attention before than)

 

[Nikola_36]

I have found this forum thread helpfull, maybe this is what`s happening in my case also

https://arstechnica.com/civis/viewtopic.php?f=8&t=1329201

 

[Sedivy]

The 122C just means sensor is off. The 112 one is a bit more worrysome as is the 45C one when you cpu is in low 30s. I would in fact be inclined to think something is going on with your vrm. If you reboot and go into bios and check the temperature sensors from there, are the mobo sensors all in 30s or are there any higher ones?

 

[Nikola_36]

The 122C just means sensor is off. The 112 one is a bit more worrysome as is the 45C one when you cpu is in low 30s. I would in fact be inclined to think something is going on with your vrm. If you reboot and go into bios and check the temperature sensors from there, are the mobo sensors all in 30s or are there any higher ones?

Oh, I think the 45C one is the sensor next to the GPU (or maybe the RAM), because its always the same as the GPU temperature.

I will now enter BIOS and reply soon!

 

[Nikola_36]

I have searched for any temps shown in BIOS, and all I could find were the CPU temps, sitting at 36C which is fine.

 

[Sedivy]

That's odd. All the sensors that your HWMonitor is finding should be showing up in bios. You already knew cpu temp was ok so that's not helping. Ok how about this, open up your chassis (the side of it), and start your game. When things get hot on the HWMonitor, ground yourself first (touch your hand to metal chassis), then carefully hold your hand above the motherboard (try and not actually touch anything) and move it around. See if you can find any area where air above it is very hot. If there is indeed a 112C hotspot, it should be immediately obvious. Even 45 might be obviously warming the air above it.

 

[Nikola_36]

That's odd. All the sensors that your HWMonitor is finding should be showing up in bios. You already knew cpu temp was ok so that's not helping. Ok how about this, open up your chassis (the side of it), and start your game. When things get hot on the HWMonitor, ground yourself first (touch your hand to metal chassis), then carefully hold your hand above the motherboard (try and not actually touch anything) and move it around. See if you can find any area where air above it is very hot. If there is indeed a 112C hotspot, it should be immediately obvious. Even 45 might be obviously warming the air above it.

I did as instructed, and as far as I could I moved my arm above everything that looked like VRM or MOSFET and found no extreme temperatures (as far as my hand temperature receptors are effective and sensitive).

There could perhaps be a few degres hotter, above the motherboard but not 40/50 - because I belive I would be able to notice that. AND - maybe its important. These 122 and 112 temps are even present when I turn the computer off after a long period of rest. They immidiately kick in, I never saw them on lower values.

 

[Sedivy]

Yeah that's a false reading then, don't stress over it.
I've heard a lot of reports of BF1 stuttering so that might be due to something entirely different or even the game itself. What ram are you running with? What video card? Any issues with other games?

 

[Nikola_36]

Yeah that's a false reading then, don't stress over it.
I've heard a lot of reports of BF1 stuttering so that might be due to something entirely different or even the game itself. What ram are you running with?

I feared that Ram could be the problem, but I ran it fine without stuttering until recently. (15/20 days ago)

I know its a mistake, but I am running 8gb Kingston 1600mhz 1.35v and 4gb Kingston 1600mhz 1.5v. I belive both are at 1.5v in order to operate together.

 

[Karadjgne]

Temp software should be taken with a grain of salt. Every vendor, model, actual mobo is designed slightly differently from any other, and will have different addresses associated with the sensors. It's not uncommon for Hwmonitor to get the Northbridge /Southbridge addresses as a sensor, and ppl freak out due to high temps. SpeedFan even reports 1 temp as 255°C and another at -125°C, both of which are physically impossible in a normal pc. The sensors are not bust or bunk, SpeedFan is picking up on an address for a sensor, when in fact they are different components altogether that don't have a thermal strip to read from.

Oh, btw, jfyi the rated temp on mosfets is 125°C, max safe temp is @85-90°C for continuous use. If a mosfet was seeing continuous 112/122°C temps, it'd be quite short lived.

 

[Nikola_36]

Temp software should be taken with a grain of salt. Every vendor, model, actual mobo is designed slightly differently from any other, and will have different addresses associated with the sensors. It's not uncommon for Hwmonitor to get the Northbridge /Southbridge addresses as a sensor, and ppl freak out due to high temps. SpeedFan even reports 1 temp as 255°C and another at -125°C, both of which are physically impossible in a normal pc. The sensors are not bust or bunk, SpeedFan is picking up on an address for a sensor, when in fact they are different components altogether that don't have a thermal strip to read from.

Thank you for the explanation!

It did cross my mind that after a year this 122C constant temperature would have some effect. ^^