Will66 Posted December 17, 2020 Share Posted December 17, 2020 Hi, I have just installed an h100i xt to replace an old h100i i used to have in the machine. I have icue installed and only one fan is running. I am using noctua fans instead of the provided ones and have plugged them into the splitter coming from the pump and then the pump into the cpu fan header. Have i done something wrong? My old h100i worked fine before but i did run the cpu fans off the cpu fan header on a splitter and ran the pump of a cpu opt header. Weirdly when i updated the pump firmware in icue it ramped both fans up to full speed during installation but when i completed it went back to only making one spin. Any help would be great thanks! Link to comment Share on other sites More sharing options...
c-attack Posted December 17, 2020 Share Posted December 17, 2020 Is that a Noctua NF-F12 Industrial 2000 rpm PWM IP67? Unfortunately it is a general PWM compatibility issue. Some of the Corsair controllers have trouble with select European designed models like the Noctua Industrial series, BeQuiet Silent Wings 2/3 PWM, and EK Vardar fans. It also varies a bit by controller. I can run the Noctua Industrial 2000 rpm PWM on my Commander Pro and Pro series AIO. The Platinum/XT controller seems to be the most particular and someone reported they could not get the Chromax to comply — one that was working on all others. Varies in the fan series too. Stay away from the Industrial 3000 rpm versions on all controllers (and for current load reasons). EK Vardar ER series are hopeless on all Corsair controllers, yet their industrial looking 120x38mm Meltemi runs just fine on my Commander. You’ll need to move the F12s to the MB for control. It might be possible to wash the PWM signal through a PWM hub, but I am not sure anyone has tried and I don’t have a Platinum/XT to run the test. Link to comment Share on other sites More sharing options...
Will66 Posted December 17, 2020 Author Share Posted December 17, 2020 Thanks for the insight. I hope they look into a fix for this but at least i can switch to mobo for fan control which i used to do anyway! Link to comment Share on other sites More sharing options...
c-attack Posted December 17, 2020 Share Posted December 17, 2020 It's a physical issue, so this would be something that changes as new hardware comes in. However, since the issue is limited to a select regional series of fans, I am not sure where the fault lies. If every one followed the spec, then PWM on one fan should be the same as another. However even within the brand one may work and the other does not, suggesting differences there as well. Unfortunately this is something we learn as users try a particular combination. I have about 70 different BeQuiet, Noctua, and EK fans, so it came up rather early on. Link to comment Share on other sites More sharing options...
Will66 Posted December 17, 2020 Author Share Posted December 17, 2020 Yeah should be an industry standard spec in this day and age! Do you think if i put both fans onto a y splitter and then into the pumps fan connector they would both run or should i just go ahead and use the mobo? Thanks Link to comment Share on other sites More sharing options...
c-attack Posted December 17, 2020 Share Posted December 17, 2020 It would be the same on the Y-splitter. The Platinum/XT coolers have two independent fan headers so that "splitter" coming off the pump is really Fan channel #1 and fan channel #2. The controller seems to have difficult rectifying the signal and it generally gets stuck at a random rpm after the boot up at max. That means that you are likely to see different results each time, rather than fan #2 doesn't spin. It is kind of funny they are doing different things. Generally when I have tested this, it has been in pairs or triples and they all do the same thing -- run 800 rpm regardless of instruction, no spin, or stuck at max. You could try the splitter, but I would first check to see if you really have control over the one fan. Performance Tab +. A custom curve with graph will appear bellow. Trying using the fixed RPM setting to move it up and down and see if the fan responds. Link to comment Share on other sites More sharing options...
Will66 Posted December 17, 2020 Author Share Posted December 17, 2020 Doing that it seems i have full control of the fan plus interestingly the second one comes on and bounces around the rpms from 0 to the defined number. When i set a lower custom rpm the fan that always spins did as it was told and the other one would constantly sound like its turning on and off very erratic! Then when i use the presets balanced or quiet only the one fan spins. However if i change the preset to extreme then the other fan after about 7-8 seconds turns on and goes full speed. Very odd behaviour lmao Link to comment Share on other sites More sharing options...
c-attack Posted December 17, 2020 Share Posted December 17, 2020 On the XT cooler, do you see PWM % or does it allow you to input a fixed RPM? A large part of the issue seems to revolve around "RPM vs PWM %" control. Some Corsair controllers like the Commander Pro use RPM as the control. Others use PWM %. This can lead to different results with different fans, positive or negative. It may be trying to match 'RPM=X true/false?' when on the preset curves, but when using fixed PWM % is adjust the speed in a different way. I suspect if you program your own fan curve (which most people should), it would return to erratic behavior. You can use it in fixed mode, setting a quiet 600 rpm or whatever on the desktop and then something tolerable for when you know you are going to be at extended load. That requires a bit of management. You may prefer the MB option, but see if your board have some fan hysteresis (spin up/down delays) to ease the fan ramping. CPU temps is going to shift incessantly. The fans do not need to do this. Slow and steady will continue to expel heat from the coolant. It won't do anything for the CPU/socket temp which is conductively cooled. Link to comment Share on other sites More sharing options...
Recommended Posts