A few questions about installed water cooler.

Archer27

Commendable
Feb 22, 2016
97
0
1,640
So tonight I installed my water cooler(Lepa Aquachanger 240) and for my first liquid cooling install I think I did a decent job.

I removed the CPU from it's socket to clean the old paste off more effectively and am powering the pump directly from the power supply via Sata/Peripirhal to Molex .

Upon booting up after the change I got a few messages:

"Overclock failed"
"Asus anti-surge was triggered to protect system from unstable power supply"
"Keyboard"
"New Processor Installed"

After a reboot, these messages went away and haven't come back. Is it because I reseated the CPU and added the pump and 2 radiator fans to the electrical load? Is it normal? Should I be concerned?

Power supply is stable, voltages read what they are supposed to. VCCIN has actually increased from 1690 to 1740 at idle however actual core voltage remains roughly the same as before.

Ran Short FFT in Prime 95 for 10 mins, stable with core temps maxing out at 58c...much better than before.
 
Solution
Did you check to see if your bios settings changed. Generally when you get messages like that, the bios reverts to default. That may be why the errors didn't come back after a reboot. If your settings did not revert and are still configured for your overclock and everything is working normally, I wouldn't be too concerned. It's probably due to the bios rechecking everything and resetting the hardware tables due to the power being disconnected.

Unless you get further errors, your settings are different or your overclock is gone, I wouldn't be too worried about it.
Did you check to see if your bios settings changed. Generally when you get messages like that, the bios reverts to default. That may be why the errors didn't come back after a reboot. If your settings did not revert and are still configured for your overclock and everything is working normally, I wouldn't be too concerned. It's probably due to the bios rechecking everything and resetting the hardware tables due to the power being disconnected.

Unless you get further errors, your settings are different or your overclock is gone, I wouldn't be too worried about it.
 
Solution


I went the otherway, I put my fans on the CPU_FAN, and my pump on the CPU_OPT. In actual fact though, it shouldn't matter, the BIOS controls both of these headers using the CPU temp as the reference temperature.
 

Archer27

Commendable
Feb 22, 2016
97
0
1,640


No settings reverted, everything was as it was before. During the Prime 95 test would errors or BSOD's shown up if something was amiss?

 

Archer27

Commendable
Feb 22, 2016
97
0
1,640


Ah I see. I may have to make that tweak in that case. Thank you! So the messages on boot I got shouldn't be an issue?

 
Well I suggest to revisit your motherboard connection manual and your watercooling connection manual because depending on the motherboard and cooler you might have to connect the pump to either cpu_fan or cpu_opt_in (the secondary cpu fan option that the higher end boards have). So do that, reseat the cables and do a CMOS reset and configure the bios settings from the beginning.
 


Yeah, it doesn't really matter either way, except, you should be more concerned with the pump not working than you are with the radiator fans not running, as the CPU will overheat much more quickly without a pump than it will without fans, and if there is no RPM signal on the CPU_FAN header, the system SHOULD shut down, unless you have that feature disabled in the BIOS. I would use the CPU_FAN header for pumps in every case.
 

Archer27

Commendable
Feb 22, 2016
97
0
1,640


Based on my Asus Manual (Z97-E) I don't have the opt port.
Currently all fan ports are taken.
The install manual for the watercooler recommends either direct to the power supply or CPUFan for the pump.
The rad fans it says use any fan control port.
Problem is, it would mess up my fan setup if I plugged the pump into a fan port as I only have 4 including the CPU fan

 

Archer27

Commendable
Feb 22, 2016
97
0
1,640


I feel you misunderstand. As of this moment, the pump is plugged into the power supply(via supplied molex adapter), the rad fans are plugged into CPU fan(via supplied splitter), and the chassis fans are plugged into fan 1, 2, 3 as they should.
To give you an idea, the case I have is http://www.newegg.com/Product/Product.aspx?Item=N82E16811147244

The two front fans take up two ports respectively. Perhaps put a splitter in for them?

 
It's not a matter of "enough power", it's a matter of, case fan headers react to different sensors, and don't tend to have the necessary response to CPU core temp that the CPU fan header has. The CPU could be overheating, but the fan would still be going way too slow because the sensor table used for the case fans wouldn't increase the supplied voltage, thus increasing the fan speed, until far too late. Don't even consider using a case fan header for a pump or CPU air cooler fan.

Since your board has limited options, your best bet will be to continue using the molex power for the pump and leave the radiator fans connected to the CPU fan header, or do get a PWM hub that has a separate molex or SATA power source and can power several components off the CPU fan header signal. That's what I would recommend.

http://www.amazon.com/SilverStone-Technology-All-Black-CPF04/dp/B00VNW556I

 
I understood but what I meant was - put the pump in the cpu_fan, put the rads in the chassis fans, put the two front fans in the PSU directly, and put the rear fan in the 3rd chassis fan (assuming your 3rd fan is a rear fan). This will do the job because this way the motherboard will control the pump and the rad fans (which are what is really important) while the chassis fans are not that important and the font fans will just run at 100% constantly (plugged in the PSU) which shouldn't be a problem and you would want cool air to go inside the case all the time.
 

Archer27

Commendable
Feb 22, 2016
97
0
1,640


That PWM controller hub is pretty cool. I'm seriously considering going that route. Thanks!
The pump is rated at 50,000 hours MTFB which is the equivalent of roughly 5 years runtime so I'm not all that concerned about failure anytime soon.

 
It will be loud as hell with the case fans connected directly to the PSU. I would advise against that unless you don't mind it sounding like the attack of the killer fans the whole time you're using your computer. I'd spend the 11 dollars and get the pwm hub. That's the best solution to the dilemma, because then you can keep the case fans on the chassis headers where they belong, and both the pump and radiator fans will be controlled via the CPU PWM signal, but will have no issues with power limitations or over current problems like they could if you connected them all to a single fan header.

Pump, radiator and case fans would all be managed that way, and would only get loud under full sustained loads, but would probably never get anywhere near as loud as just two fans connected directly the the PSU with a full 12v would be.
 

Archer27

Commendable
Feb 22, 2016
97
0
1,640

Exactly, I'd rather not have my rig sound like a jetplane lol. Having all fans controlled with PWM is the only way to go, I agree.
 
I've got to second the not putting the fans directly to the PSU. Maybe in the short term until you get a fan controller or fan hub, but not as a long term solution. I can't imagine running any fan (except especially quiet fans) at full speed the entire time the computer was on.
 
Yes, it drives me nuts even when my secondary case fans come on after I reach a predetermined temp and are full speed just for a second or two to get them going and then wind down to the speed assigned by the Grid+ v2 CAM software profile. Which is another option as well, perhaps even a better one.

The NZXT Grid+ v2 has six 3 pin fan headers and can control up to eight fans by using a splitter on any two of the six headers. You can precisely control the speed and curve of each individual fan (Or have one profile that is assigned to two fans if you're using a splitter for two of them) using the CAM software utility. It will work with 3 or 4 pin fans, although if you're using more than two four pin fans you'll need to get a couple of extra 4 pin to 3 pin adapter cables which are like two bucks each.

I'm using the Grid+ v2 along with the Hue+ lighting system, and both are phenomenal. There is really nothing else like them on the market yet.

One of the great things about this setup is the fact that if you have a lot of case fans, but don't want them ALL always running, just when they are actually needed, you can configure some of them to actually be off until temperatures reach a spec determined by you, and will then turn on and go immediately full speed for a second to get them going (Some fans don't like starting at very low voltages) and then quickly wind down to the speed you have assigned them to for whatever temp they are currently at. Which brings up another point that makes the Grid+ v2 far superior to other fan controllers and hubs.

They use the system sensors to monitor and control fan speeds, not stupid, inaccurate temp probes or estimated readings. The unit connects to one of your motherboard's internal USB 2.0 headers and takes it's thermal sensor readings directly from the same information you'd see in the bios or via a monitoring utility in windows. Nice. It's also rather small and mounts wherever you want inside the case, so you don't need a drive bay and you also don't need to put your hands on the computer every time you want to make a change, you just open the utility and with a few clicks, changes are done.

 

Archer27

Commendable
Feb 22, 2016
97
0
1,640
So an update, everything seems to be running fine EXCEPT games which will drop frames like nobody's business.

I have checked quite a few things already as well as run tests. First, my specs:

Intel I5 4460
Asus Z97-E
NVIDIA GTX 960 SSC 2GB
16GB RAM

What I have tried:

1. Ran Prime 95, both Large and Small FFT while monitoring temps, voltage, frequency etc. Combined time of 40 mins. CPU seems stable and never goes over 58c.

2. Ran GPU-Z, card is running at 16x all the time, card is seated properly in PCI-e slot and temps are ok.

3. While playing, ran MSI afterburner, cpu usage as well as GPU usage is fine as well as temps. However, still got FPS drop on multiple games...steam and otherwise.

4. Uninstalled the new Nvidia Driver(364.51) as well as tried rolling back to the previous one, no difference.

5. Ran Heaven benchmark....ironiclly my benchmark improved as well as my fps on this compared to before..kinda contradicts the momentary drop to 5 FPS I got in a game tonight!

A brief history of changes made in the past few days before the drop:

1. Ran windows update(Vulkan API installed itself during this time)

2. Ran CCleaner registry cleaner...later system restored due to issues related to my AV caused by it

3. Installed the water cooler/reapplied paste/reseated CPU

4. Installed that new NVIDIA driver


I'm at a loss....and a bit upset. This is a new machine that I just built and was getting good fps until the above things were done.

Any suggestions?
 

Archer27

Commendable
Feb 22, 2016
97
0
1,640


I always run at 1080p max settings on guild wars 2, MGS V, Street fighter V etc to name a few. The fps drop is across the board
 
Have you tried 361.91? The last few drivers have had issues.

I also suggest using DDU to completely remove the old drivers before installing the new ones. Particularly if you think you are having problems with your previous drivers.

The procedure for using DDU, is to remove the drivers via Programs and Features, and uninstall everything nVidia. You may be asked to reboot after some of the components. Once all the drivers are removed, boot into Safe Mode and run DDU and clean for nVidia drivers. Then reboot into Normal Mode and install the new drivers.
 

Archer27

Commendable
Feb 22, 2016
97
0
1,640


I have not(362 was working just fine but when I rolled back to it it too was having the same issuestho), nor have I tryed the method you have suggested yet. Will do and will get back with the results.

EDIT: also if it makes any difference, I am on windows 7 pro 64 bit
 
This:

http://www.tomshardware.com/faq/id-2767677/clean-graphics-driver-install-windows.html


And if that doesn't work, despite the PITA it might entail, I'd say start fresh. Reset BIOS to optimal default settings, do a CLEAN install of Windows (Not just a reinstall, refresh, reset or rollback, but a clean install, including deleting all partitions on drive 0 boot drive and recreating the boot partition. Be sure to do this with all other drives disconnected so there are no problems with previous boot partitions that may be on other drives that have had Windows installed on them in the past, and are hidden):

http://www.sevenforums.com/tutorials/1649-clean-install-windows-7-a.html



And then reinstall all your applications. I hope it goes without saying that you should back up any important documents, files, settings and make sure you know your product key, before doing the clean install so you can reactivate Windows afterwards.