[SOLVED] dont have CPU_OPT, PUMP or AIO_PUMP

[zentenon]

hey i order a nzxt kraken m22, and i dont have CPU_OPT, PUMP or AIO_PUMP on my motherboard. my motherboard is a Gigabyte h310 m ds2 2.0, this mother only has cpu fan and sys fan please help and answer me where i can put my pump header into.

 

[NightHawkRMX]

It really doesn't matter what header you plug it into, as long as it is configured correctly in bios.

Others may disagree, but I would plug the pump into the CPU fan header and set it to max speed in bios. Plug the aio fan into the sys fan header and set the profile dependent on CPU temp.

 

[zentenon]

thanks bro i appreciate that.

 

[CountMike]

thanks bro i appreciate that.

I always do it opposite but what's most important is that pump runs full speed.

 

[rubix_1011]

I also recommend AIO pump speed to always be 100%. If you want anything with a fan curve, let PWM manage the radiator fans...that's more applicable than having the AIO pump fluctuate.

 

[zentenon]

I always do it opposite but what's most important is that pump runs full speed.

thanks bro

 

[zentenon]

I also recommend AIO pump speed to always be 100%. If you want anything with a fan curve, let PWM manage the radiator fans...that's more applicable than having the AIO pump fluctuate.

thank you bro

 

[Paperdoc]

You have only one CPU_FAN header, which should control all cooling of the CPU chip and uses for this the temperatures sensor built into the CPU chip itself. Then you have one SYS_FAN header that uses a different temp sensor built into the mobo, and this header should be used only for the case ventilation fans.

So, how to connect both the pump and the fan of the Kraken system to a single CPU_FAN header? There is a good way that uses a quirk of the designs of 3- and 4-pin fans, and all you need is a simple device called a 4-pin SPLITTER. Here's an example of a two-pack of one with two outputs

https://www.amazon.com/JBtek-Black-Sleeved-Splitter-Converter/dp/B01EF9OI0O/ref=sr_1_4?crid=10TAVN9GDP2R1&dchild=1&keywords=fan+splitter+4+pin&qid=1605062798&sprefix=fan+splitter,aps,183&sr=8-4

It has one input female (with holes) connector to plug into your mobo CPU_FAN header. It has two male (with pins) outputs to plug your fan and pump into. It has no other connections. Do NOT get a fan HUB, which too many sellers mis-label as a Splitter also. A HUB has one extra "arm" that must plug into a power output from the PSU, and you do NOT want one of those for this job.

Look at the close-up photo of the two outputs. You will see that one of them is missing Pin #3. Now, the mobo header can only deal with the speed signal coming back to it from ONE fan, so the Splitter sends back ONE speed signal (from the output with ALL 4 pins) and ignores the other. This becomes important for your system because besides providing power and speed control, the CPU_FAN header monitors the speed signal it gets for FAILURE - that is, no speed signal. In an AIO system, the PUMP is the critical device whose failure could cause huge problems, so you must plug your PUMP unit into the output with all 4 pins. Then it can be monitored for failure. Plug your rad FAN into the output with one pin missing. The mobo will never be able to tell you the FAN speed, but you can check from time to time to make sure it is still working, and it WILL run and will have its speed controlled by the CPU_FAN header.

Now comes the "quirk" I mentioned. The PUMP is intended to run full speed all the time, and only the rad fan should have its speed altered by the CPU_FAN header. Because of the differences in design of 3-pin fans and 4-pin fans, if you plug a 3-pin fan into a mobo header that is using the 4-pin PWM Mode to control its fan, that 3-pin fan will always run full speed. Well, the PUMP of your system is wired just like a 3-pin fan, and we WANT it to run full speed all the time. When you plug it into the Splitter output with all 4 pins, oddly it will NOT connect with Pin #4 but that is exactly what should be, and the pump acts like a 3-pin fan and runs full speed. But the FAN from the rad, plugged into the other output arm, IS of the new 4-pin PWM design and it WILL be speed-controlled when connected this way. To make this work, though, you MUST make an adjustment in the configuration of the CPU_FAN header in BIOS Setup. See your manual, p. 24-25, where it details how to adjust each of the two mobo fan headers. Select the CPU_FAN header to adjust. Set Fan Speed Control to Normal so it will use it pre-programmed automatic speed control system. For Fan Control Use Temperature, set it to CPU (IF you have a choice - often you don't on the CPU_FAN header). For Fan Control Mode be SURE it is set to PWM (not to Auto) so the header does use that Mode and send out the proper signals. St Fan Stop to Disabled so it cannot stop, and set Fan Fail Warning to Enabled so it WILL warn you if the pump fails. Wehn you have your adjustments made, use Esc to return to the Main Menu and then use F10 to get to the Exit menu, p. 36. There choose SAVE and EXIT to save your settings and reboot.

As an additional aside, when you are setting your CPU_FAN header configuration, you also can go back and choose the SYS_FAN header to configure that. It should be set to use the Motherboard temperature sensor, instead of the CPU internal sensor.

When you use a Splitter to connect both the pump and the rad fan in this manner to the CPU_FAN header and set that header to use the new PWM Mode, the pump will receive full power at all times and run full speed as it is designed to do, while the rad fan will receive proper power but also a PWM signal to control its speed according to the internal temperature of the CPU chip. That's how the Kraken system is supposed to work. Your mobo will tell you that the "CPU Fan Speed" is constant at about 3000 rpm, but that is really the pump. It will never tell you the actual speed of the rad fan, but that does NOT affect its ability to control that speed properly. Meanwhile you are free to use the SYS_FAN header for power and speed control of case ventilation fans.

 

[zentenon]

THANK YOU SO MUCH

 

[zentenon]

Now i have another problem i dont have any usb 2.0 intenal connector in mobo my only internal usb is busy with the case. what i should do? i dont use my 2.0 in my case

 

[Paperdoc]

What you need is called a USB2 Header Splitter. Here is one that can convert one mobo USB2 header into four - two new headers on each of the two circuit boards.

https://www.newegg.com/black-en-lab...sb_header splitter-_-9SIACJF8K33841-_-Product

A mobo USB2 header actually has two USB2 ports on it. Each of those can transfer data at the USB2 speed, and can supply current to an attached device (if needed) up to the port limit of 0.5 A. This Splitter will allow two ports to become 4 ports. The way it is wired indicates that, on each board, the left new port on one board is in parallel with the left new port on the other board, and those two share one original mobo port. Likewise for the right-hand new ports. That means that each pair of new ports shares the resources of one original mobo port, so you can't quite expect them to be fully USB2 compliant. However, if you do not plan to use any USB2 devices that need lots of power, it will work. I expect that whatever thing you have in your case that needs to connect to a USB2 header does not actually use a high data transfer rate, so sharing that with another device probably won't make much difference at all to data transfer.

 

[zentenon]

thank you bro i mean that because this time i will connect my liquid coleer in the headers and i will buy that accesorie thank you

 

[Karadjgne]

Me personally, I'd bypass the motherboard altogether for the pump and go directly to the psu via sata/molex to fan adapter. Then use cpu_fan header for the fan and sys_fan header for the case fans via splitter.

Reason being, you have no idea how much amperage draw is on that pump at full loads. It's designed to go to a header by itself, not share with a fan. Direct from psu, it'll run full speeds constantly, the fan operated by the cpu temp, the case fans run by the motherboard sensor.

Only disadvantage being Cam won't show pump speeds, just take it for granted it's at full speed, whatever that might be.

 

[zentenon]

Me personally, I'd bypass the motherboard altogether for the pump and go directly to the psu via sata/molex to fan adapter. Then use cpu_fan header for the fan and sys_fan header for the case fans via splitter.

Reason being, you have no idea how much amperage draw is on that pump at full loads. It's designed to go to a header by itself, not share with a fan. Direct from psu, it'll run full speeds constantly, the fan operated by the cpu temp, the case fans run by the motherboard sensor.

Only disadvantage being Cam won't show pump speeds, just take it for granted it's at full speed, whatever that might be.

ok but which cable is the one that i need to connect direct from psu

 

[Paperdoc]

What Karadjgne says is correct - the NZXT website does not show any spec for the fan current draw. Most such fans do not draw a high current. The limit for the CPU_FAN header is 1.0 A total, and the AER P120 rad fan is spec'd at 0.32 A max. If you want to do it that way, here's an adapter example that converts one Molex 4-pin power output from the PSU to a single 3-pin fan power supply connector, plus "replaces" the "used" Molex output with another for use by another device.

https://www.amazon.com/C2G-Cables-Go-27078-Pass-Through/dp/B00007FGUD/ref=sr_1_2?crid=13MQI2RLRTOJJ&dchild=1&keywords=molex+fan+adapter+4+pin&qid=1605193708&sprefix=Molex+fan+adapter,aps,172&sr=8-2

You should understand that this has an impact on the failure detection arrangement. Doing it that way will mean that there is no way for the PUMP speed signal to reach the mobo header, so it cannot be monitored for failure. If you plug the rad fan into the CPU_FAN header, the FAN speed will be shown there and monitored for failure.

Personally, I think it more important to monitor the PUMP speed for failure, and that is really hard to do without feeding it to the mobo. The pump normally operates quietly with little vibration, so examining it yourself manually for whether or not it is running is tough. There is a third option if you want to be sure the pump gets lots of power with no risk to the CPU_FAN header, AND has its speed monitored for failure. Get a small fan HUB, something like this
from Silverstone

https://www.amazon.com/SilverStone-Technology-Silverstone-Splitter-SST-CPF04-USA/dp/B07N3HP8S5/ref=sr_1_3?crid=32CX2BZOIMM2W&dchild=1&keywords=fan+hub+4+pin&qid=1605194451&sprefix=fan+hub,aps,187&sr=8-3

What distinguishes this device from the Splitter I suggested is that it also has an extra cable "arm" that must plug into a SATA power output from the PSU to get all the power for its load (the pump and the rad fan) from the PSU and none from the CPU_FAN header. It still connects to that header to pick up the PWM control signal (and the header must still be configured to PWM Mode) and to return to the header the speed of ONE device - the one that is plugged into the specially marked output port. So you plug the PUMP into that marked port and its speed can be monitored by the host CPU_FAN header. Plug the rad fan into the next Hub output port. If you do it this way, the control and failure monitoring and speed displays will be the same as I suggested with a Splitter, but the POWER source for the pump and fan will not be the CPU_FAN mobo header; it will be power direct from the PSU, as Karadjgne has recommended.

 

[Karadjgne]

Only bad part about pump/fan on the same splitter/hub is that there's only a single control source. When the cpu heats up, the pump speeds up and the fan speeds up the same amount. Since pwm is a %, speeds differences won't matter, but everything will be the same % regardless.

There is no ideal situation, that's one of the drawbacks of mATX, limited headers.

 

[Paperdoc]

Actually, that is not a problem with this common situation. What we are discussing is connecting both the pump and the rad fan(s) of an AIO cooling system to the same mobo header signal set - whether by Splitter or Hub. The secret is that the PUMP is designed as a 3-pin fan is, and its speed is controlled only by changing the VOLTAGE supplied on its Pin #2. The FAN, on the other hand, is a new PWM design 4-pin fan. It receives from Pin #2 the full 12 VDC constantly, and also receives on Pin #4 the PWM signal used by a chip inside the fan to modify current flow from that 12 VDC supply though the motor windings to change its speed. So, if the mobo header is using the new PWM Mode to control the "fan(s)" connected to it, both the pump and the fan motor receive a constant 12 VDC power supply on their Pin #2, whereas the PWM signal on Pin #4 is NOT received by the 3-pin pump (which has no special chip to use that signal, anyway) but IS received by the 4-pin fan. So the two devices respond differently. The PUMP always runs full speed because it always gets that 12 VDC supply, but the fan speed will be altered by the PWM signal it uses.

 

[Karadjgne]

And the cpu will only activate its low/zero rpm warning / shutdown procedure if the Fan fails, since that's what's being read by the PWM signal. If the pump fails, nothing changes other than amperage load on the header. Intrinsically exactly the same as direct psu power hookup.

Only difference realistically is 1 way uses the header to supply the 12v, uncontrolled, unmonitored, and the other way uses the psu. Same results.

Cam will know if the pump fails, either way as it gets rpm signals and speed settings via the USB, not the header, so pump speeds will still be variable.

 

[zentenon]

thank you all guys my liquid coller functions perfect!

 

[Karadjgne]

Naturally. Paperdoc's a wizard when it comes to stuff 'inside the box'. My general direction is thinking of stuff 'outside the box'.

Enjoy your pc 😊, glad it all worked out.

 

[Paperdoc]

Glad to hear it's working. On the failure detection comments, detection is by the speed signal the header receives. That is why my notes emphasize that the PUMP is the device that should be plugged into the ONLY output of a Splitter or Hub that sends to the host header the speed signal of its device. That is the output that has ALL FOUR pins, or is otherwise marked. (Other outputs skip Pin #3, which is how the speed signal is comunicated.) If you do connect that way, then the PUMP is the device being monitored for failure.

 

[Karadjgne]

True, however, the pump is DC voltage speed controlled and the fan is pwm. If you set the output for pwm, neither will be controlled, if you set the output as DC voltage, both will be controlled, to a point. Pwm fans are not designed for lower voltages.

So it's best to run the pump as the 3pin, the fan as 4pin, and rely on Cam to supply pump failure warnings instead of the cpu.

Sorry if it seems like argument, but there really is no perfect solution when dealing with mixed pwm/DC and limited headers.

 

[Paperdoc]

I think you misunderstand. Maybe it's the missing Pin #3 that is confusing.

The output from a mobo header using PWM Mode is this:
Pin #1 - Ground
Pin #2 +12 VDC constant
Pin #3 Speed signal returning from the motor to header
Pin #4 PWM signal from header to motor

A Splitter (a Hub does this same thing) will feed all the signals on Pins 1, 2 and 4 unchanged to those SAME pins of its several outputs. (Well, a HUB gets the +12 VDC supply from the PSU instead of from the mobo fan header.) Now consider what each device plugged into those outputs "sees" and does.

A 3-pin fan (or a Pump wired like a 3-pin fan) gets Ground and a constant 12 VDC power supply from Pins 1 and 2. So it always runs full speed, as MOST AIO pumps are designed to do. It returns its speed signal using Pin #3; IF that pump is plugged into the output arm that HAS a Pin #3 to make the connection, that speed will be relayed back to the mobo header for speed measurement and display and failure monitoring. The fan makes NO connection to Pin #4 for the PWM signal, but it could not use it anyway because it has no chip to do so.

A 4-pin fan gets Ground and a constant 12 VDC supply from Pins 1 and 2, also. I returns its speed via Pin #3 to the Splitter, BUT if the fan is plugged into the output arm with NO Pin #3, that signal goes nowhere. The fan gets the PWM signal from Pin #4 and its chip uses that to modify current flowing from the 12VDC supply (Pin #2) through the windings to achieve speed adjustment. This is exactly how a PWM 4-pin fan should be operated, except for the lack of sending its speed signal back to the mobo.

By using a Splitter (or a Hub) fed from a mobo header in PWM Mode to send out these signals to those two different types of device, we get the result that the 3-pin device (the pump) acts exactly like a 3-pin fan plugged into a 4-pin header in PWM Mode will do - run full speed all the time. AND the 4-pin device (the fan) does exactly what it would do if it were connected directly to that 4-pin header operating in PWM Mode - its speed IS under mobo control.