Jump to content
Corsair Community

high pitch sound may be from h80i


fifo

Recommended Posts

hi guys I need your help again.

 

I have evetything connected besides the graphics card. I start the system up, everything is ok for the first two kinutes approx, then I start hearing a high pitch sound and it seems to be coming from the cpu block, not 100% sure though, the only way I can make sure is by disconnecting the power to the h80i pump but I think this may fry my cpu. So what to do I have removed power to sll the hard drives and fan controller. eyooped each. fan in turn with my finger, noise is still there. I have pht my ear sfainst the axi760 to seeif its coil whine but it doesn't seem to be coming from that, it fets modt loudest when my ear is bear cou block.

what can I do to resolvr this? uts driving me nuts.

Link to comment
Share on other sites

Are there any fans connected to the H80i? Temporarily disconnect them and see if the noise goes away.

Have you tried a different SATA power cable?

Depending on the noise, it could possibly be air bubbles in the pump. Lay the case flat for a few hours/days and see if that helps.

Link to comment
Share on other sites

I have put my finger on the fan to stop it, the noise is still there. I have disconnected the top two exhaust fans as well. I am going to try to power the h80i alone from another psu, so only the h80 will be running every thing will br off. now if the source of noise is the h80i which it seems like, how does laying the case flat fix the problem. if there is air in the system giving the high pitch sound, how is that air going to escape, it is a sealed system. if their are air bubble isn't that a faulty cooler?
Link to comment
Share on other sites

if there is air in the system giving the high pitch sound, how is that air going to escape, it is a sealed system. if their are air bubble isn't that a faulty cooler?

 

The system isnt fully filled to allow for expansion.

this air get agitated and make air bubbles that get trapped in the impeller housing which produces a noise.Normally this air cavity is in the upper part of the radiator...

Link to comment
Share on other sites

well I don't think its the pump because when I press the reset button on the motherboard the noise stops and starts again after 2 minutes and only happens when their is a sata device connected and the arror code displays AE legacy boot event.

there is one thing I am worried about when switching psu I connected the sata cable which I thought was powering the h80i but wasn't thus I started the pc, I so the h80 fan was not running hence after 10 to 15 seconds I turned it off. now I am worried has thus fried or dameaged the cpu in some way? it seems to start up the same as before. if the pump is not running does it damage the cpu?

Link to comment
Share on other sites

thanks,

 

when you say the system has safeguards against overheating, you mean the h80i protects the cpu or the motherboard? if the h80i loses power or the pump fails for some reason can the h80i still cool the cpu (intel 5930k) at least to an acceptable level at idle (like below 60 c) like air coolers with the fan disconnected?

 

theres one strange thing I have noticed, to check whether this high pitch sound is coming from the psu I tried another axi 760 psu, now with this psu the light on the h80i (that lights up the text corsair) did not light up but the fan connected to the h80i was running and the optical and hard drives connected to the same sata ribbon sounded like their were running (the sort of sound optical and hard drives make) I don't know if the h80i pump was running, I hope so as I had the system on for a few minutes. now this time I did not get the AE error code (legacy boot event) after d6 (no graphics card) and no high pitched sound so I am puzzled. It would be nice if the psu was cause of the issue as it is easier to replace but I think the sata power from this psu is not working properly because the h80i light did not come on but the hard drives did. how do you explain this?

 

if the fan connected to the h80i is running does this mean the pump is also getting power ( basically does it use the same power line off the sata coonector)?

is the cpu block light powered of a different line or 5v?

I am trying to understand why the cpu block light is not coming on with this psu but the fan is where as on my other psu both fan and cpu block light come on. STRANGE!

Link to comment
Share on other sites

The motherboard's BIOS has the safeguard. Not the H80i. if the H80i stops working, it won't cool the CPU like an air cooler, the coolant in the H80i will just heat up.

 

If the pump has not failed, yes it will be powered. Even if the electronics for the fan controller and firmware for LINK is not working, chances are again, if the pump has not failed, it will most likely run. The pump is not controlled by the firmware.

 

Assuming the noise has now disappeared and the H80i is working normally. (Check the BIOS for the pump RPMS and CPU temperature) You can use the LINK software to turn the L.E.D.s back on.

Link to comment
Share on other sites

then how do you explain the light not coming on with the second psu but comes on with the first psu. I don't understand this?

if the h80i pump is not running, how much approx. will the cpu temp increase in 15 seconds at idle and how much would in increase in a few minutes?

Basically if the pump fails or power is not connected is it like running the cpu with no cooler at all?

Link to comment
Share on other sites

Have you tried a different SATA power cable and port on the PSU?

 

I have had an issue before with the H100i where the firmware did not want to "initialise" where when i would turn the computer on a L.E.D> flash will appear then turn off, same with the fans. Your issue may be related but i doubt it.

 

I do not own a X99 chip so estimating the approximate temperature increase over time can be either over exaggerated or plainly incorrect. Some chips may also run hotter than others which too would make estimating an approximate quite difficult.

 

AIO watercoolers and custom loops cool a CPU by having water passing through the block, if the pump fails the CPU it is not "running without a cooler" it will basically heat up the water in the loop. Similar to a air cooler if the fan fails to spin to move air, the heatsink (imagine the heatsink as the water) will heat up as well. Either way the system will shutdown when the temperature gets too hot.

Link to comment
Share on other sites

ok, so the cpu block will still provide some cooling without the pump running. I'm assuming this would at least keep the cpu below 60 c at idle during post or attempting to boot without an operating system where it just hangs.

the fans don't start and stop, but what I find strange is with the second psu the led does not turn on but the fan seems to run, where they both are powered by the two wire sata power connector of the h80i. the hard drives seem to make a running noise but the motherboard does not detect them. I wonder could this be due the sata voltage being low. I just hope it was running the pump when the fan was running, may be the led needs 12 v and cannot work with a voltage of 11v for example whereas the fan can, don't know about the pump though. These corsair axi psu are suppose to be very stable and have error detection. anyway I will try a different sata port on the psu this weekend as you suggested.

I have also raised the issue to corsair support.

Link to comment
Share on other sites

If the H80i is not running, keeping the CPU below 60C for a period of time will depend on how much heat is being put out by the chip. If it outputs quite a lot it will not keep it below 60 for a period of time.

 

Test the H80i one more time. Turn your computer on, enter the UEFI/BIOS, monitor the CPU temperature for a few minutes or leave it on, do something else and regularly come back to check the temperature. if the temperature does not change much, you are basically good to go and build your computer. For the hard drives not being detected by the motherboard, double check your SATA data cables. The H80i will still run on 11v, it's just not recommended as the pump is designed to run at maximum speed. (12v)

Link to comment
Share on other sites

Would the chip still put out quite a lot of heat during post or sata detection, so not running any operating system or software as there was no boot cd.

 

Basically what is bothering me is firstly when i changed the psu i connected the sata power cable which i though was connected to the pump, and saw the fan was not running so i quickly switched it off in 5 to 10 seconds. Now i don't think this generated much heat.

 

But when i did connect the correct sata cable and saw the h80i fan running, i thought the pump would be running as well and left it running for a few minutes but i did see that the led did not turn on, and the hard drives were not detected but did make a running noise. This is what has led me to think may be the pump was not running because the l.e.d.. did not turn on and the hard drives were not detected with this psu. The only thing i could think off is the voltage being under 12 v to cause this. If the fan was running but not at full speed is is it ok to assume that the pump is running but at lower speed even at 9 or 10V at which would provide enough idle cooling?

I also thought may be i could have connected the sata cable into the PCI!E port of the psu but that is not possible because the pcie port is 8 pin and the sata is 6 pin and i could not get it in when i tried to confirm, i think the clip is in the middle of both connectors. So you can only get the connector slight in on the right of the port but not enough to hold it or clip it flush to the port on the psu.

Link to comment
Share on other sites

It will depend on the voltage. More voltage = more heat. I don't think the voltage will vary during POST/UEFI unlike during idle in Windows. So it would probably output a decent amount of heat.

 

The fans not running full speed does not mean the pump is running at a lower voltage. The pump will always and will run at 12v unless any kind of voltage reducer is used. Be cautious of which connector plugs into which header. Forced connections or incorrect connections will not be good.

 

In my honest opinion, you are over complicating something really simple to check. Yes, i understand that you are worried that the H80i pump may not be running thus a possibility of damaging a expensive chip but in order to find out if the H80i pump is cooling the CPU, you need to let the computer do its POST checks then enter the UEFI to check the CPU temperatures. The time for this to occur should not be long and the temperature should not be very high at this point from a cold boot. Worst case scenario would probably be 70-80C range. Simply looking at the L.E.D. on the pump block not lit and HDs not being detected, i understand, can be concerning but in order to find out if the H80i is working you need to enter the UEFI to check the temperatures.

 

It is best to resolve one issue first before looking into the other. Attempting to resolve two issues in one time may be time consuming, frustrating and create a lot of anxiety which is not wanted.

Link to comment
Share on other sites

OK,

 

i will follow your advice and check this in the weekend,

first i will see what the temperatures are like in the bios with my current psu, then i will try the replacement psu, then i will install windows 7 and see if every thing is functioning OK and check for high pitched sounds from the motherboard or cpu.

 

But this high pitched sound when a sata device is connected with no boot CD seems strange, i will check if the high pitched sound appears again during windows 7 setup. I will also remove the PC piezo sounder and see if the sound still occurs, i am sure its not the sounder, i seem to get a feeling it might be a coil or cap on the sata controller circuitry or the motherboard power supply circuitry. You got any ideas of what this strange high pitched sound could be? I've tried disconnecting all the fans, the sound is still there.

Link to comment
Share on other sites

  • 2 weeks later...

hi,

 

firstly I think I have managed to get rid of the high pitched sound by ensuring the memory sticks are fitted flush into their sockets, their were two dims not lit because the dimm was not all the way in on one end of the dimm socket. these x99 mb only have clips on one end of the memory socket, so you have to make sure the other end is flush in.

 

now I have connected a graphics card, checked temps in the bios. The bios cpu temp started at 22c with an mb temp of 21c and case temp of 21c and slowly wentup and settled at 37/38 c, mb temp went upto 30c and case temp went up to 26.1c. my mb by default overclocks the core i7 5930k a bit from 3500mhz to 3750 mhz, but I adjusted the cpu ratio and got it down to 3500 mhz. the memory is running in xmp mode 2666 mhz

I installed Windows 7 and ran speed fan 4.5. see attached results.

 

Are these core temps ok( see attached screenshot )? I am a bit concerned about my seagate 2tb sshd hard drive temp.

this is with one radiator fan in push mode eith the side panel off the case.

 

now I am going to see what the temps are like with the other psu where the h80i l.e.d. did not come on to check whether the pump is running or not with that psu.

Systemp.png.20b432251c100f59a14b169c0614bbc5.png

Link to comment
Share on other sites

Those temps are fine for idle including the HD temp. I'd suggest you run a CPU stress test e.g. IntelburnTest Standard test to ensure the load temperatures are fine. The slight increase to 3.7Ghz sounds like turbo boost. If you don't want it you can disable turbo boost in the UEFI.

 

Temperature (°C)

Operating 0 to 60

Non-operating −40 to 70

Source: Seagate

 

If the HD gets too hot, relocate it to an area with good airflow or put a fan next to it.

Link to comment
Share on other sites

thanks toasted,

 

you know if one side of the dimm modules are not flush into the socket like how mine was before and you start the pc. does it cause any damage to the socket ot dimm module? Secondly while its the pc is on if you just move it a tiny bit from side to side and in to ensure its correctly seated which I did before when one of the led were not lit, but did not put enough pressure to push it in as I thought they were in properly, then turned the system off and pushed the module in properly. would this cause damage to the memory or socket? I can get the memory replaced?

 

you know where in the bios to turn the turbo boost off?

I can see a setting for it, st the moment I have the memory set to xmp profile 1

and the cpu mode set to xmp where options are auto/manual/xmp.

so my memory is running at 2666mhz but standard intel core i7 5930k is stated as 2133 on the intel website. Can it it run at 2666mhz, seems to be working ok do far installed windows.

when I disabled xmp for the memory the system kept resetting it self a few times anf then off. I then had to power it again and the bios reported overclock unsuccessful, just by disabling xmp weird. so I renabled it.

 

will prime 95 be ok for a stress test?

 

I have a fan next to the hd but it was turnef off I will turn it on snd see if it makes a difference.

Link to comment
Share on other sites

If a RAM module is not flush with the socket, i would check and make sure the slot is clear of any debris and make sure the module is being mounted properly according to the motherboard's user manual. If you cannot find the cause of the issue, check the module. Does it fit fine in other slots?

 

If you cannot find the issue. Remove the module and the other one connected on the other side for now and check your motherboard's manual for ram installation using X amount of modules. If you are 100% sure the socket is preventing the module from sitting flush get the motherboard replaced by your seller.

 

Refer to your motherboard's user manual for UEFI navigation. You will probably have to go to Advanced mode to see and change the setting.

 

You can use Prime95 to stress test the CPU but be advised the temperature may be higher than other stress testers.

Link to comment
Share on other sites

hi,

 

I think you misunderstood my response about installing the ram modules, I have got them inserted flush into thr socket now after applying a bit of of pressure on the other side that has no clip. the led on the ram is bright now as well whereas before it was dimm. these ram sockets only have a clip on one side unlike the older motherboards thst have clips on both sides, so the only way you can clip it is to ensure the ram is fully inserted. these new x99 mb have a clip on one end only so you can clip it without the other end fully inserted, so I had to apply a bit of pressure on the other end this time and its detected the full 32gb.

my question was about whethe any damage to the socket ot dimm module could be caused when I started the pc when I did not know the dimm modules were fully inserted before? Secondly while its the pc is on if you just move it a tiny bit from side to side and in to ensure its correctly seated which I did before when one of the led were not lit, but did not put enough pressure to push it in as I thought they were in properly, then turned the system off and pushed the module in properly. would this cause damage to the memory or socket? I can get the memory replaced?*

 

I will doenload the intelburn test that you suggested, as I don't want to fry my cpu with prime 95.

 

you know if the ibtel 5930k can run at xmp memory speed of 2666mhz?

thanks for your support!

 

hi,

Link to comment
Share on other sites

It is unlikely you caused any damage by failing to get the modules completely plugged in. It's a disruption of current, not an overabundance of it. Anything is possible I suppose, but if you had damaged it you would not be up and running now.

 

Your chip and all the X99 motherboards are supposed to be able to handle at least 2666 and most up to 3000 at this point. Further enhancements will surely come in time. Understand the XMP profile on your memory is a set of timings encoded into the DIMM so you don't have to enter the values, but you are overclocking the memory when running anything except 2133MHz. The XMP is a factory overclock of your memory. The more things you overclock, the larger the chance of instability. You may need to evaluate whether an increased CPU speed or memory speed is more important to you. Or possibly low level overclocks of both. It can be a tedious affair. Your motherboard manufacturer probably has a website where exact settings are discussed.

Link to comment
Share on other sites

when I switched the pc on the bios first chose xmp profile 2 where the voltage is 1.35v and speed 2800 mhz, I then chose profile 1 where the memory voltage is 1.2v and speed is 2666mhz which is the rated speed of the dimm. I wish the motherboard would start off with low speed settings instead of high speed settings. I did have the xmp switch on the mb enabled but did disable it later on, but the motherboard still starts in xmp mode. I hope the 1.35v dimm v has not damaged or reduced the life of my ddr4 corsair dominator platinum 4 x 8gb sticks. the bios has chosen the timings as well, I guess I better check that as well. anybody familiar with the Asus x99 deluxe motherboard bios? is there a optimize default setting that selects the correct settings of your hardware. my gigabyte ep45 ud3lr motherboard had such a setting and picked the correct timings for my ram.

 

I have ran the intel burn test standard test, first set the no of times to 10(180 seconds), followed by 30 (389.25 seconds) and then followed by another 30 ( approx 389 secs)

so after about 15 mins or so the cpu temp went to 43 c and the core temps went up to 64 to 70c see attached files. now the core temps jump to mid 55c every 2 or 3 seconds and go back to 64 to 69c may be due to each test coming to an end or something and the next one starting, I don't know, do you guys get this as well on the core temps.

 

Are these temps ok? the core temps seem a bit high to me why the big difference between the cpu temp and the core temps?

temp1.png.210531871defe98a9e1054c6ae3655e2.png

temp2.png.7855cb2a9c4931694c4c39f2d3cfe6b4.png

temp3.png.12bfedf97d4ca1a32c51a4a132c28bba.png

suc1.png.6eb1678291a4cc6467e1f76267e51b65.png

Link to comment
Share on other sites

Yes, this was this staring to sound like Asus. Load up your BIOS. On the "E-Z" version look to the left side. There is a drop down menu labeled "XMP". Select disabled. This will force the memory to load at it's natural setting of 2133. It will also automatically reduce your BCLK speed from 125 to 100 (you won't see this on the screen). The 100 BLCK is also the natural base clock speed and will likely be more stable while using less power at idle. Your previously measured CPU speeds will change as well. This is normal. You are now back at the natural settings for your chip and memory.

 

If for some reason after making the changes, CPU-Z or whatever you are measuring your memory speeds still says 2666(1333), you will have to change the memory default in the advanced BIOS. The easiest way to do revert all of your settings back to beginning is to: 1)Enter the advanced BIOS from the E-Z BIOS screen. 2) Select the very last column with the save and exits. 3) Select load optimal defaults. The screen will flash lightly. 4) Right below select save and exit. This should return you to normal.

 

If you have been making changes with AI Suite 3 to the TPU, you will have to undue those first using the software. Open AI Suite. Select the TPU box. Bottom right, select default. It will ask you for confirmation and notify you to restart. Select yes. Let it restart fully. Then restart again and go back to the BIOS if necessary. I am not sure if you moved the physical TPU or XMP switches on the motherboard itself. If you have, put them back in the default position - full left. The manual has pictures if you are unsure. Asus gives you multiple ways to overclock the memory and processor, but the switches are the least accessible and and offer no further tuning options. Combining overclocks through the 3 methods (switches/BIOS/AI Suite) will strongly increase your chances of problems.

 

Intel Burn test runs a series of short tests. It is normal for your temperature line graph to have smooth rolling hills. It's hard to make evaluations of your temperature without very specific system information, but you are not in the danger zone.

Link to comment
Share on other sites

thanks for the response, are my core temps ok (65 - 70) ? the cpu temp was 43 c which is quite different from the core temp. why is this?

 

I currently have all mb switches to the left hence disabled. but thinking of enaply epu the power saving switch to save power, don't know what it actually does.

in the bios I have xmp enabled and I have set the cpu clock ratio from 30 to 28, the bios has chosen a bclk of 125 . so I get 3.5 ghz for cpuwhich is the rated speed for 5930k and 2666mhz for memory which is the advertised speed for the ram. is this memory speed ok for the processor or should I decrease it to 2133mhz?

if I disable xmp in the bios which is what I did last time but did not disable it for the processor. so pc kept restarting snd turned off. when I turned it on it said unsuccessful overclock or something and started up with xmp disabled but the cpu clock was now 3 ghz (below rated 3.5 ghz) and ram 2133 so I reenabled xmp. what is xmp exactly extreme memory profile, I thought it was an overclock profile but only for me memory speed not cpu speed. as long as the xmp profile 1 does not harm my cpu or memory and if the 5930k can work at that memory speed then I don't mine the xmp. the bios chooses bclk of 125 when xmp is enabled I can change it to 100 if its required. is it ok at 125. I've had no problems so far, I will have to check the timings and voltage with the datasheet of the ram.

also the bios reports a target chache speed of 3000mhz, is this right? or what should it be?

Link to comment
Share on other sites

The cpu temp was 43 c which is quite different from the core temp. why is this?

 

The ASUS X99-DELUXE reports the CPU temperature via PECI rather than CPUTIN and I suspect SpeedFan needs fixing to report CPU PECI. What do HWiNFO and my SIV report and how do these compare to SpeedFan? What is the package temperature?

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...