Jump to content
Corsair Community

Recommended Posts

Posted

So, I'm in a sticky situation. I bought two Noctua 3k (AF-14) fans after reading that PWM was the only requirement and only one of them seems to be working properly when connected to Corsair AIO splitter. Both fans are identical model.

 

Both fans work flawlessly at any RPM when connected to the motherboard.

 

Only one fan works properly (anywhere between 0-3000 RPM) when both are connected to AIO splitter. The other fan will only work at 0 RPM or 3000 RPM (100%).

 

Only one fan works properly when connected in isolation to AIO splitter.

The other fan when connected in isolation will again, only work at 0% or 100%.

 

I've tried reinstalling iCUE, force reinstalling AIO pump firmware, resetting all profiles and extensively configuring in iCUE over several days, bios update, different motherboard, seriously you guys are my last resort.

 

I've read that Corsair has had incompatibilities with Noctua AF-14, but it seems weird to me that one fan works fine and the other will only work at 0% or 100%?

 

Honestly, what can I do at this point? I bought these aftermarket fans after reading that the only compatibility requirement was PWM. Are they paperweights or has a Corsair rep said anything about patches?

Posted

Specifically it has issues with the NF-A14 and F12 Industrial 3000 rpm versions. The 2000 rpm industrial versions work fine and I have used both across several coolers and controllers for years. This is a little different than the other incompatibilities where it is only the one very specific variant. The 3000 rpm version has a different motor with an 800 rpm m in speed and a very large current draw. It seems likely that is not the only difference with its siblings.

 

My first suggestion would be to send them back and pick up the 2000 rpm version or the 1500 rpm Chromax versions. Both are more than enough. There are no AIO radiators that really benefit from massively high fan speeds and once you sail past 1800 rpm, it's just going to be noise with no additional cooling benefit.

 

That may not be possible, so option two would be to run them from the MB controller. Using CPU temp is not ideal, but if you have a newer MB with a 10K thermistor probe (2 prong), you can run that inexpensive wire up to the exhaust side of the radiator. Exhaust temp will essentially be the same as coolant temp and you can then get a better control source.

 

If you wanted to be thorough, you could temporarily disconnect the Noctua from the splitter and run it out the door to the stock ML140 fans. No need to unbolt anything. Plug both in and see if they behave normally. You also can swap the two Noctua fans on the splitter to see if the behavior follows the fan or stays on the wire, but that is more curiosity rather than a solution.

Posted
Specifically it has issues with the NF-A14 and F12 Industrial 3000 rpm versions. The 2000 rpm industrial versions work fine and I have used both across several coolers and controllers for years. This is a little different than the other incompatibilities where it is only the one very specific variant. The 3000 rpm version has a different motor with an 800 rpm m in speed and a very large current draw. It seems likely that is not the only difference with its siblings.

 

My first suggestion would be to send them back and pick up the 2000 rpm version or the 1500 rpm Chromax versions. Both are more than enough. There are no AIO radiators that really benefit from massively high fan speeds and once you sail past 1800 rpm, it's just going to be noise with no additional cooling benefit.

 

That may not be possible, so option two would be to run them from the MB controller. Using CPU temp is not ideal, but if you have a newer MB with a 10K thermistor probe (2 prong), you can run that inexpensive wire up to the exhaust side of the radiator. Exhaust temp will essentially be the same as coolant temp and you can then get a better control source.

 

If you wanted to be thorough, you could temporarily disconnect the Noctua from the splitter and run it out the door to the stock ML140 fans. No need to unbolt anything. Plug both in and see if they behave normally. You also can swap the two Noctua fans on the splitter to see if the behavior follows the fan or stays on the wire, but that is more curiosity rather than a solution.

 

Thank you for taking your time explaining the issue, I appreciate it. I actually enjoy the noise profile of these fans, I know I'm a weirdo but I just love the powerful sounds they produce, kinda' like the noise of a powerful engine.

 

As for the reduced cooling above certain airflow, again I thank you for mentioning it but this was an intentional decision as I'm using the powerful airflow through the radiator to provide a better intake for my GPU.

 

In order to satiate your curiosity, I actually performed the original ML140 fan test as you requested and they work fine. I'm not gutsy enough to switch them while live though. I already tried swapping the Noctua fans on the splitter as per my original troubleshooting, but that had no effect, the same fan worked and the same fan didn't run below max. Even when running only one of the noctua fans on the splitter, the same fan worked and the other didn't. They both still work flawlessly when on Motherboard.

 

 

I do however have a very recent board (x570) with two thermal headers and considered somehow running the fans off of the motherboard, but couldn't think of a decent temperature source. Thank you for suggesting I use the exhaust from the radiator, I might try doing that.

×
×
  • Create New...