[SOLVED] Is Stopping and Starting constantly safe for CPU Fan Longetivity

[edo101]

Hi guys, so my last build was an i7-930 and I cooled it with a Silver Arrow SB-E. That fan was constantly spinning and would spin up when temps got higher but it never stopped spinning. This was back in the bios days. back from 2010. The rig still runs today but as some of you know by now, I upgraded to a 10850K last week.

As such, I moved to UEFI along with it an NH-14s. In my Z490 Taichi, I didn't want to set the fan to full speed so I customized the fan curve to spin as temps went up. However I guess mobos and fans these days have a quiet mode? My issue is that the NH-14s stops and starts quiet frequently when PC is idle or i'm just surfing the web. I wanted to know if this is safe for the fan? If not, how can I make it spin all the time but also not spin high unless absolutely needed. I'd like to preserve this cooler's life. I also have an AIO Artic Freezer II 360MM coming in (which I am still debating if I'll use or just switch to air with either the NH-14s or DH-15. My goal is 5.0 all core on the 10850K with halt states and C states to lower the voltage and clock when I am not doing something streanous.

With the AIO as well, I anticpate this behavior will happen with its fans as well. so I want to know if this is safe or not, for any CPU fans. Wether it be air or AIO

 

[jay32267]

It is totally safe for the fan/motor.
They are designed to run this way.
....and there really is no added wear due to starting and stopping.

 

[edo101]

It is totally safe for the fan/motor.
They are designed to run this way.
....and there really is no added wear due to starting and stopping.

How do you know this? @jay32267

 

[USAFRet]

Yes, it is fine. They are designed for this.

 

[Phaaze88]

A static speed - even a 100% setting - is easier on the motor(be it fans or liquid pumps) than a constantly changing one.
I know this from reading up on it. Auto settings aren't always better.
Manufacturer's design the products around 100% operation - you being able to set it how you want it is a bonus of sorts.

 

[edo101]

A static speed - even a 100% setting - is easier on the motor(be it fans or liquid pumps) than a constantly changing one.
I know this from reading up on it. Auto settings aren't always better.
Manufacturer's design the products around 100% operation - you being able to set it how you want it is a bonus of sorts.

Yeah once I get time this weekend, I will manually set my OC so the CPU is boosting less with higher voltages

@Phaaze88 I figured starting and stopping adds resistance to the fan's motors which would wear them down? Especially if they do this like 100 times a day? The cooler is Noctua and uses the https://noctua.at/en/products/fan/nf-a15-pwm 140mm fan. This is doing this behavior with this fan curve:

 

[Phaaze88]

It's inevitable. If you get a few years out of it, that's still good, yea?

 

[Deleted member 2720853]

By the time your fan even begins thinking about dying, you'll have already replaced it/gotten a new build entirely.

 

[edo101]

It's inevitable. If you get a few years out of it, that's still good, yea?

By the time your fan even begins thinking about dying, you'll have already replaced it/gotten a new build entirely.

Well I moved my fan from CPU_fan2 to CPU-fan1 and it no longer stops. It follows the curve to a T

 

[Phaaze88]

Faulty header?
Are you using any 3rd party software that can control your fans outside of bios?

 

[edo101]

Faulty header?
Are you using any 3rd party software that can control your fans outside of bios?

Not that I can think off. I have some HW monitoring tools though but nothing controlling them. Someone said they saw some DC control on my fan_2 cpu settings? They mentioned some DC stuff. I don't know. I'd hate if this pin is faulty. Would mean, undoing this build and going for replacement @Phaaze88

The only thing is the stop issue. Other than that, the fan ramps up and the speed monitoring shows as such

 

[Phaaze88]

That starts going over my head - the DC to PWM stuff. DC should be for 3pin operations, I thought?

 

[Paperdoc]

Your issue may well be linked to the CPU vs DC thing, so let me explain. These are two different methods by which a mobo header can set the speed of a motor. A 3-pin fan can ONLY be controlled by changing the VOLTAGE supplied to the fan from Pin #2. It can range from 12 VDC for full speed down to 5 VDC for minimum speed - anything lower may cause the fan to stall, and it will not re-start until the voltage is raised a volt or two. This is called Voltage Control Mode or DC Mode. The ideal way to control a 4-pin fan (which your NH-14S cooler system has) is the newer PWM Mode, in which Pin #2 always supplies 12 VDC and the PWM signal from Pin #4 is used internally by the fan motor to modify that power supply. That is called PWM Mode. However, if you set the mobo header to run in voltage Control Mode (aka DC Mode) for your 4-pin fan, it WILL have its speed controlled this way by acting as if it were an older-style 3-pin fan.

One of the advantages of the newer PWM design is that it can run the fan slower than you could do with a similar unit designed for Voltage Control only. Similarly, it can be re-started at a lower speed. So you can set up a 4-pin fan connected to a header using the new PWM Mode and run it successfully with those settings for minimum speed. BUT if you then change the MODE setting of the header to use the older DC Mode, that same fan way well stall at the lowest speed setting that used to work before, and it will take a bigger voltage rise to get it to re-start.

My suspicion is that, without realizing it, you have configured your two CPU_FAN1 and ...2 headers to use different MODE settings. So when you run that fan under PWM Mode it works fine and never stalls. But when you re-connect it to the other header that is using DC Mode, it is eventually told to run so slow that it does stall, and then it will not re-start until the temperatures rise enough to raise its speed control Voltage high enough. In that case, merely changing the configuration of BOTH fan headers to PWM Mode will avoid this issue, no natter where it is plugged in.

Personally, I agree with you that it is better not to have the fan motor starting and stopping all the time - that is a stress on the motor. But in this case there's a hidden factor that may be more important. When a small DC motor stalls because it is fed a low voltage, it does NOT stop drawing current. Some current keeps running through the windings and causing motor heating until the supply voltage rises enough to re-start. So the stalled motor is not the same as a stopped motor with no current running through it.

 

[edo101]

That starts going over my head - the DC to PWM stuff. DC should be for 3pin operations, I thought?

Your issue may well be linked to the CPU vs DC thing, so let me explain. These are two different methods by which a mobo header can set the speed of a motor. A 3-pin fan can ONLY be controlled by changing the VOLTAGE supplied to the fan from Pin #2. It can range from 12 VDC for full speed down to 5 VDC for minimum speed - anything lower may cause the fan to stall, and it will not re-start until the voltage is raised a volt or two. This is called Voltage Control Mode or DC Mode. The ideal way to control a 4-pin fan (which your NH-14S cooler system has) is the newer PWM Mode, in which Pin #2 always supplies 12 VDC and the PWM signal from Pin #4 is used internally by the fan motor to modify that power supply. That is called PWM Mode. However, if you set the mobo header to run in voltage Control Mode (aka DC Mode) for your 4-pin fan, it WILL have its speed controlled this way by acting as if it were an older-style 3-pin fan.

One of the advantages of the newer PWM design is that it can run the fan slower than you could do with a similar unit designed for Voltage Control only. Similarly, it can be re-started at a lower speed. So you can set up a 4-pin fan connected to a header using the new PWM Mode and run it successfully with those settings for minimum speed. BUT if you then change the MODE setting of the header to use the older DC Mode, that same fan way well stall at the lowest speed setting that used to work before, and it will take a bigger voltage rise to get it to re-start.

My suspicion is that, without realizing it, you have configured your two CPU_FAN1 and ...2 headers to use different MODE settings. So when you run that fan under PWM Mode it works fine and never stalls. But when you re-connect it to the other header that is using DC Mode, it is eventually told to run so slow that it does stall, and then it will not re-start until the temperatures rise enough to raise its speed control Voltage high enough. In that case, merely changing the configuration of BOTH fan headers to PWM Mode will avoid this issue, no natter where it is plugged in.

Personally, I agree with you that it is better not to have the fan motor starting and stopping all the time - that is a stress on the motor. But in this case there's a hidden factor that may be more important. When a small DC motor stalls because it is fed a low voltage, it does NOT stop drawing current. Some current keeps running through the windings and causing motor heating until the supply voltage rises enough to re-start. So the stalled motor is not the same as a stopped motor with no current running through it.

Yep turns out I need to spend more time with my mobo. There is a section for Auto control under CPU Fan 2. I have moved it to PWM and set the curve like CPU fan 1. Everything works as it should now.

Thanks guys. I just prefer the fan spinning all the time cause the start stop thing worries me. Physically a fan will have more resistance starting and stopping constantly which has to i feel cause more wear than necessary. CPU fan 2 in its advanced settings was also set to silent mode. So I have corrected it now. thanks guys

 

[rubix_1011]

This is a good time to get familiar with these settings and work with your BIOS fan curves. Often, larger air coolers run very quietly, so the ability to set those curves to run consistently without impact to noise level is beneficial for longevity and performance. Smaller, 120mm fan coolers might need a more dynamic curve to assist with noise levels at lower thermal loads, but in the end, it is all preference. Some people like the idea of near silence while others would rather opt for the performance and ability to cool effectively regardless of noise level.

 

[jay32267]

How do you know this? @jay32267

I used to work for General Motors on electric motors and fans.
Starting and stopping an electric motor generally doesn't add any significant wear.

 

[Paperdoc]

Glad yo hear you have found and adjusted configuration items, and it's all working properly now. Thanks for the update.