Max fan actually increases temperature

Posted by PusheenHater@reddit | buildapc | View on Reddit | 22 comments

In the BIOS I have all my fans set to Silent. All fans run around 500 rpm.
CPU temperature is 35 C.

For one of the exhaust fans, I test it by setting it to Max. It goes at 1500 rpm.
The CPU temperature actually instantly jumps to 37 C.

I set the fan back to Silent. CPU temperature instantly drops back to 35 C.

I repeat the steps over and over and it's consistent:
Max = 37 C
Silent = 35 C
The increases and decreases are instant.

Why is this? It's not like maxing out the fan draws that much more power.

[-]

bblzd_2@reddit

Some good theories. I just want to add that idle temperatures are meaningless, might as well turn the fans off completely and keep it quiet.

[-]

A PC fan isnt capable of reducing the pressure so much to change the CPU temperature. What's probably happening is airflow inside the case is being redirected so much that warm air from something (probably GPU) is going to the CPU heatsink.

[-]

apoetofnowords@reddit

Yup. Something like running a fan in a warm room. Cold air is at the bottom, hot air at the top. But the fan just mixes it and makes you even warmer. It's case airflow pattern issue.

[-]

pythonic_dude@reddit

Between the limited space inside pc cases and the power with which any modern fan blows, the effect of convection is within statistical error.

[-]

Unicorn_puke@reddit

Yep it's more airflow in, out and direction of travel. It's also a lot more about volume of air and temperature of the air going in for how a system will keep cool

[-]

Excellent_Occasion26@reddit

Are we willing to consider that changing the BIOS setting may also have changed the allowable current draw at the CPU ?

[-]

the_gamer_guy56@reddit

Ive seen enough. Direct die cool your sound chip. Its the only way.

[-]

mere_iguana@reddit

Turbulence is a hell of a drug!

[-]

Hrmerder@reddit

I would say your fan directions are wrong or you have mixed fans some strong some weak and the weak ones are negating the strong ones positives

[-]

lazyhustlermusic@reddit

This is dumb margin of error crap bro

[-]

kind_bros_hate_nazis@reddit

So he should delid the RAM is what you're saying

[-]

lazyhustlermusic@reddit

Individually reassemble each transistor by handed in a suspended gel solution.

It’s the only way

[-]

DZCreeper@reddit

That is due to airflow pattern, not power consumption.

Your CPU cooler is a pressure restriction, so the rear fan will naturally draw air from around the cooler. Effectively pulling cool front intake air away before your CPU cooler can use it.

Typically you should run the rear exhaust fan below the RPM of your CPU cooler fan. That way it assists the natural airflow.

[-]

PusheenHater@reddit (OP)

This makes sense but I've never heard of this before. If you research positive vs negative air pressure for PC cases they always say it's the same performance except there's more dust in negative.
I wonder why no one mentions this?

[-]

DZCreeper@reddit

That is referring to the overall case pressure. Assuming you have equal flow volume running positive vs negative pressure only changes where the dust accumulates.

The CPU heatsink adds a significant amount of restriction, your rear case fan by itself cannot efficiently pull air through it. The fan mounted on the heatsink directly does this job better.

[-]

daanos60@reddit

Because usually it's the amount of fans and not fan speed that causes the negative/positive pressure

[-]

fray_bentos11@reddit

Many laptops like fan settings mode to power limits. For example, high fan/performance/ultimatemode definitely does this on Asus laptops. Your temps increase at the higher fan mode because the higher fans cannot dissipate the addition heat. The GPU and/or CPU will often thermal throttle instead to keep things safe. This is why I undervolt CPU/GPU whenever possible.

[-]