Any harm in running fans below specification?

[pantsaregood]

I've got a few Noctua NF-A15s on my heatsink. They're PWM fans rated between 300-1200 RPM. While playing around with the PWM duty cycle settings in my BIOS, I noticed these fans will actually reliably spin up at about 150 RPM.

At this speed, they keep my CPU at an acceptable temperature and they're dead silent. Is there any harm in running them below the advertised specification?

 

[Sheep221]

I've got a few Noctua NF-A15s on my heatsink. They're PWM fans rated between 300-1200 RPM. While playing around with the PWM duty cycle settings in my BIOS, I noticed these fans will actually reliably spin up at about 150 RPM.

At this speed, they keep my CPU at an acceptable temperature and they're dead silent. Is there any harm in running them below the advertised specification?

Be aware of the fact that they may spin at 150 if they were slowed down from higher RPM, it may happen that you will turn on your computer next time and fans won't start because voltage that sets them at 150 won't be able to turn the fan into motion. Technically nothing will happen to fans if they will be operating at ~150 RPM but they will likely won't turn on at next boot and if you don't notice yourself you may end up running your CPU without fans spinning, this actually happened to me once and I was running full load CPU for 2 days without fan being turned on and it sat at some 80C, I found that out accidentally and fixed the issue.
So if nothing else, extra caution.

 

[pantsaregood]

They aren't running at low RPm by voltage, it's from PWM. They will soon up from being completely stalled too.

 

[UsandThem]

I can't imagine being able to hear them at all under 500 RPM inside a case on a heatsink. I only start to hear mine (barely) when they are 650+ RPM (Fractal Design R5)

At 900 RPM they only produce 13.8 dB. If they cool your CPU at only 150 RPM (pretty amazing I might add), run them at 300 RPM so you know they will actually start up.

http://www.dvtests.com/noctua-nf-a15-pwm-review/

 

[VirtualLarry]

Be aware of the fact that they may spin at 150 if they were slowed down from higher RPM, it may happen that you will turn on your computer next time and fans won't start because voltage that sets them at 150 won't be able to turn the fan into motion. Technically nothing will happen to fans if they will be operating at ~150 RPM but they will likely won't turn on at next boot and if you don't notice yourself you may end up running your CPU without fans spinning, this actually happened to me once and I was running full load CPU for 2 days without fan being turned on and it sat at some 80C, I found that out accidentally and fixed the issue.
So if nothing else, extra caution.

I had something similar happen with one of my 1045T Thuban CPUs in my ASRock Extreme4 990FX rig. Was using an OCZ Vendetta cooler (92mm heat-pipe), because it used the standard AMD mounting bracket. Apparently, the fan bearing got clogged with dust, and died.

I was doing distributed-computing, and CoreTemp recorded a max temp in excess of 80C, but the thing kept on running! It didn't crash.

When I discovered that a few days later, I removed that heatsink and put on a regular AMD 4-heatpipe stock heatsink (the 125W one, even though my Thuban was a 95W model), and downclocked my CPU down to stock.

I don't use that rig very much any more, but I bought some Arctic Cooling FDB 92mm replacement fans, so my OCZ Vendetta is good to go again.

Oh yeah, I had it OCed to 3.51Ghz or so, using bus-clock OC. (1045T isn't multi-unlocked.)
Stock voltage, amazingly enough. 1.325V or so.

 

[ClockHound]

Be aware of the fact that they may spin at 150 if they were slowed down from higher RPM, it may happen that you will turn on your computer next time and fans won't start because voltage that sets them at 150 won't be able to turn the fan into motion. Technically nothing will happen to fans if they will be operating at ~150 RPM but they will likely won't turn on at next boot and if you don't notice yourself you may end up running your CPU without fans spinning, this actually happened to me once and I was running full load CPU for 2 days without fan being turned on and it sat at some 80C, I found that out accidentally and fixed the issue.
So if nothing else, extra caution.

What motherboard was this and voltage or pwm control?

Most (modern) motherboards hit all the fans with 100% PWM or 12V at power on before switching to saved fan profiles. If using an old manual voltage controller might an issue. Or an older gigabyte board - all my old gigabyte boards had the worst fan control, next to some server boards.

Dependent on the board, the BIOS will spin the fans up before entering/exiting sleep.

The beauty of PWM is that the fan is always getting 12V. The PWM controller on the fan then interprets the duty cycle from 0% to 100%. With Noctua PWM fans even if they spin down and stop, as soon as the PWM signal is above startup threshold (14% - 20%) they will start spinning.

OP, don't worry about these old voltage controlled fan tales. If your motherboard offers fine grained PWM cycle and you have verified the min duty cycle to keep the fans spinning, no worries. Just as long as you're happy with the idle temps. Could add 2% to allow for dust accumulation.

A15s on my D15 spin down to 262rpm at idle. Never had a single startup issue with my Asus X58 Sabertooth. I can hear a difference between 262 and 500rpm at 1 meter. Subtle, but in a quiet room, that's a nice quiet difference. ;-)

 

[Sheep221]

What motherboard was this and voltage or pwm control?

Most (modern) motherboards hit all the fans with 100% PWM or 12V at power on before switching to saved fan profiles. If using an old manual voltage controller might an issue. Or an older gigabyte board - all my old gigabyte boards had the worst fan control, next to some server boards.

Dependent on the board, the BIOS will spin the fans up before entering/exiting sleep.

The beauty of PWM is that the fan is always getting 12V. The PWM controller on the fan then interprets the duty cycle from 0% to 100%. With Noctua PWM fans even if they spin down and stop, as soon as the PWM signal is above startup threshold (14% - 20%) they will start spinning.

OP, don't worry about these old voltage controlled fan tales. If your motherboard offers fine grained PWM cycle and you have verified the min duty cycle to keep the fans spinning, no worries. Just as long as you're happy with the idle temps. Could add 2% to allow for dust accumulation.

A15s on my D15 spin down to 262rpm at idle. Never had a single startup issue with my Asus X58 Sabertooth. I can hear a difference between 262 and 500rpm at 1 meter. Subtle, but in a quiet room, that's a nice quiet difference. ;-)

Just like the VirtualLarry, I had ASrock board of that time just with P67 chipset - I use it to this day, they are pretty old these days and at least mine, does not turn fans to 100% at boot anyhow, and it happened several times also with case fans not turning on when booting up at all. Silent profile gets activated instantly. PWM should prevent that but over the years I found out that PWM is working very differently for each Cooler and mobo combo, and I also agree that gigabyte boards do have terrible (if any) fan control, I still own H55 and H61 based gigabytes and they are really bad at this. Hopefully they changed something since then, but I owned like 5 gigabyte boards and every one of them had issues of various degree. Not the problem with their GPUs tho. I am aware of the 100% fans at boot but I had to warn the OP anyway, not knowing what board he might be using.

 

[pantsaregood]

Thanks for the input everyone!

The reason I was concerned with fan speed actually has nothing to do with cooling. I was more concerned with balance becoming more of an issue as the fan speed became extremely low. Usually lower RPM places less wear on bearings, but I was concerned that abnormally low speeds could effectively warp the fan due to balancing issues.

 

[UsandThem]

Thanks for the input everyone!

The reason I was concerned with fan speed actually has nothing to do with cooling. I was more concerned with balance becoming more of an issue as the fan speed became extremely low. Usually lower RPM places less wear on bearings, but I was concerned that abnormally low speeds could effectively warp the fan due to balancing issues.

Probably the best thing to do if you want to know for sure what would happen to the Noctua fans over time by running them under 300 RPM is to contact Noctua technical support. They are very good at responding to inquiries. I'm sure there has to be a good reason they specifically list 300 RPM as the minimum on that fan. They are very open about their fan specifications.

 

[ClockHound]

Probably the best thing to do if you want to know for sure what would happen to the Noctua fans over time by running them under 300 RPM is to contact Noctua technical support. They are very good at responding to inquiries. I'm sure there has to be a good reason they specifically list 300 RPM as the minimum on that fan. They are very open about their fan specifications.

And that spec is stated as 300rpm +/- 20% - so one could receive an A15 that spins down to 240rpm and still be in spec. In my experience with Noctua, they meet or exceed their specs. In the OP's case, his fans have exceeded the specs.

Whatever you do, Mr. GoodPants, don't contact Noctua support, they will want to repatriate your friction-defying skunkworks models and replace them with other not-quite-as-frictionless 'within spec' models. :biggrin: