I'm having this same issue and no where on the internet or at local tech support places am I able to find a solution. I have an Asus Maximus VIII Formula motherboard, and attached to it I have a water pump on the water pump header, 2 Thermaltake controller boxes with three 120mm RGB fans each connected to the CPU and CPU_Opt header, and then 2 Thermaltake controller boxes with 3 120mm RGB fans each connected to additional CHA_1,2,3,4 headers--tried different headers at different times. Here is the problem:
When the computer powers up, all boxes are linked and all colors/speeds are synced. When the motherboard actually POSTs, about 5 or 6 seconds after powering on, it kills the LED on the fans/boxes connected to any fan header that isn't the CPU fan header and sometimes also temporarily kills the fan speed or permanently changes the control box to low speed (the red light on the control box vs. the blue) which is absolutely crazy considering that the control box connector cables from Thermaltake are supposed to sync the settings across all connected boxes--yet somehow, whatever is going on with these fan headers is cause the boxes to cut the LED and/or change settings on the physical box.
The craziest part of this problem is that even though the fan cuts for a split second, and the LED shuts itself off, if I turn the LED back on, even to saved color from the paused RGB mode, it still has the color saved (turquoise or purple or gold or some color that isn't a standard color with the "mode" button). So the boxes aren't resetting, they're just somehow not receiving enough power to run the fans at standard speed (not even full speed) and power the LED when the motherboard boots itself up, or the BIOS loads or something. It's making me so angry because I can't close my case since every single time I boot up, I have to disconnect the boxes where the fans cut off, change the settings to match the other fans, then reconnect them in case I want to sync the settings if I decide to change colors or something.
I have tried every permutation of the fans and boxes. All fans on all boxes on all headers. All the fans work perfectly and the control boxes work perfectly if they are plugged into CPU or CPU_Opt, and all fans almost consistently fail (a few outlier samples) if they're plugged into any CHA_ fan header AFTER the motherboard boots. So before the Asus/ROG logo appear on the screen, the fans work fine, when the motherboard "boots" or "POSTs" or whatever it does at the exact split second the logo appears on the monitor, then non-cpu header fans kill the LED and more often than not also screw up the speed setting on either their own boxes, or the CPU boxes they are connected to with the box connector cables from Tt.
This is so insanely frustrating, and I am 100% it has to do with the motherboard, but I can't explain why it's happening. All the motherboard fan headers are 4-PWM, all rated at 1A, which is more than enough for 3 fans, all the settings in the BIOS are the same, I've tried XMP on XMP off, overclocking the CPU not overclocking the CPU, less RAM (4 sticks down to 2 sticks and different combinations to make sure there isn't some "plays well together" issues), set all fans to PWM 100%, DC 100%, PWM 50%, 30%, 10%, every combination, and a number of other things. I can't be the only person in the world with multiple Riing 12 control boxes and more than 6 fans (or more fans that a CPU fan header will allow).
Last but not least, I also used MagLev fans from Corsair in an identical configuration but with a simple fan splitter rather than a control box (obviously, because they don't have them) and the motherboard seems to do the same thing but cutting power temporarily to the magnetic bearing, but it's much more difficult to observe.
So, long story short (maybe too late for that now) what setting am I overlooking that is causing my motherboard to cut the power to those fan headers below what is required for the Thermaltake control boxes to keep the LED lights lit and keep the fans running on the blue-light speed, rather than the red light speed, when they aren't connected to the CPU header. Is the CPU header getting priority for some reason? Is having the CPU, CPU_Opt, and W Pump headers all connected with the 3 fan (near 1A max output) config and the water pump (EK DDC pump) someone diverting power from other fan headers to prioritize and ensure that the CPU headers boot up with proper power supply? Can I connect the control boxes to the each other, leave the two CPU headers the way they are and then connect the remaining two boxes through a molex adapter to the PSU and then have all boxes controlled by the settings for the CPU fans? What do I do?
Please help and if this problem can somehow be solved, I will dedicate my life (or some of it) to spreading the word about whatever magically setting I'm overlooking so that people after me don't have to suffer through the misery of replugging in fans 2,000 times to make sure I've narrowed down the problem to such a specific issue that their can't be any confusion as to what the problem is, but still not be able to solve it.