Shutdowns at startup, CPU lights on.

[dpeter]

Hello, since i moved my system from my old case to my new case i have been suffering from shutdowns a few seconds/minutes after powering up for the first time in the day. Usually i could reboot and all is good but today the pc refused to boot and the CPU light on the mobo was on. I noticed that my AIO fans were not spinning during this boot error. Before i moved my system from my old case to new case i noticed that the pwm function on my aio fans were broken(system reporting 0 rpm) and so i switched it to dc and bought a pwm fan hub to check if it was the splitter fault that this was happenning, after moving the case and using the fan hub the pwm problem persisted. So when my cpu light stayed on i tought it might be the fan and changed back from the fan hub to the original splitter, i also checked cable connections, especially the cpu and it seemed ok. After that when i pressed the power button my Motherboard basically blinked when i tried to power it on and after 2-3 times pressing the button, it did power but CPU light was still on, so i long pressed power button and restarted the system and it worked again, i really don't know what is happening. I know my english is kinda bad so i apologize for that.

[Mavendow]

In my most recent build this occurred due to rapid fluctuations on the 12v and 3.3v rails, which I discovered running HWiNFO one of the times I managed to get it to boot. My PSU was successfully RMA'd and I've had no problems since. In a past build (2010s) it was a faulty graphics card which later fried nearly everything - including the house's wall socket. In a very long ago build (2000s) my PC did this for a few months before the Gigabyte motherboard exploded and caught fire. These experiences, combined with the fact your crashes are progressive, sounds like electrically faulty hardware.

 

Best back up your data before you do anything else; preferably by removing the drive containing your data rather than trying to boot it again.

[dpeter]

9 minutes ago, Mavendow said:

In my most recent build this occurred due to rapid fluctuations on the 12v and 3.3v rails, which I discovered running HWiNFO one of the times I managed to get it to boot. My PSU was successfully RMA'd and I've had no problems since. In a past build (2010s) it was a faulty graphics card which later fried nearly everything - including the house's wall socket. In a very long ago build (2000s) my PC did this for a few months before the Gigabyte motherboard exploded and caught fire. These experiences, combined with the fact your crashes are progressive, sounds like electrically faulty hardware.

 

Best back up your data before you do anything else; preferably by removing the drive containing your data rather than trying to boot it again.

My GPU is fine, i know that. My 12V rail is fluctuating between 12.096V and 12.130V and my 3.3V rail is fluctuating between 3.296 and 3.299. So i think im good on theses aspects, but i agree that it's most likely hardware failure but i dont know which one. As i said my GPU is fine, my CPU also works flawlessly when system boots, so i don't know what it could be other than maybe my mobo(which is 2 months old) or my PSU(just completed a year). And after i opened it up i tried rebooting a few times and all of those it booted perfectly fine with no shutdowns so far.

[Mavendow]

Depends on the rate of fluctuation. In my case the fluctuation was small but very rapid. 3.3v sounds alright but I'd double-check the 12v. Many websites tell you it's fine if the fluctuation is small but (apparently) that isn't true since my PSU manufacturer initiated an RMA when I told them about the rate of fluctuation.

 

I will also note that, on my current PSU, my 12v is rock solid. Not even a 0.001 change since I opened HWiNFO. Not all PSUs will necessarily have such good voltage control, but I say this just to ensure you double-check.

 

You may be able to tell it's a voltage problem if your CPU core voltage jumps seemingly out of lockstep with clock speed. Prior to replacing the PSU my CPU was throttling up and down so I locked the clock speed. But the voltage kept jumping around and then declining to unreasonably low values anyway. Post PSU replacement this stopped.

 

As for telling whether it's the PSU or something like a VRM, that depends on your hardware. I determined it was my PSU because my old mobo had voltage probes for PSU and VRM. VRM was relatively stable while PSU was oscillating faster than HWiNFO could poll (HWiNFO Settings->General->polling period). I could tell it was oscillating too fast because it was changing erratically. What this means is the changes were out of lockstep with the poll rate, and since it would occasionally not change during a poll period, that meant it was changing faster than could be polled. I could, however, find poll rates where it changed steadily, which meant the voltage itself wasn't erratic as it was changing at some multiple of my chosen poll rate.

 

My new motherboard doesn't have two probes. So if you have this situation you may have to figure it out based on swapping hardware. Maybe someone here has a better method, I honestly don't know.

 

In the GPU case, that one was not obvious as all voltages seemed normal. The only indication was the shutdowns and random orange fuzzy lines on the screen. You did not mention this but it wasn't clear to me either because the fuzzy lines were not very noticeable until it got very close to failing. When the thing finally failed it somehow shorted every single hot I tested to ground and neutral. Even the case fans, surge strip, and wall socket lol. Baffling, that one.

 

In the motherboard (non-VRM) case, that Gigabyte's crashes were accompanied by memory and volume read errors in the Windows Event Viewer. I didn't test voltages back then 'cuz I had not yet taken any repair classes.

 

Oh, and one more thing. The likely reason it doesn't crash after some minutes of uptime is heat. Once components heat up their electrical properties change.

[dpeter]

1 hour ago, Mavendow said:

Depends on the rate of fluctuation. In my case the fluctuation was small but very rapid. 3.3v sounds alright but I'd double-check the 12v. Many websites tell you it's fine if the fluctuation is small but (apparently) that isn't true since my PSU manufacturer initiated an RMA when I told them about the rate of fluctuation.

 

I will also note that, on my current PSU, my 12v is rock solid. Not even a 0.001 change since I opened HWiNFO. Not all PSUs will necessarily have such good voltage control, but I say this just to ensure you double-check.

 

You may be able to tell it's a voltage problem if your CPU core voltage jumps seemingly out of lockstep with clock speed. Prior to replacing the PSU my CPU was throttling up and down so I locked the clock speed. But the voltage kept jumping around and then declining to unreasonably low values anyway. Post PSU replacement this stopped.

 

As for telling whether it's the PSU or something like a VRM, that depends on your hardware. I determined it was my PSU because my old mobo had voltage probes for PSU and VRM. VRM was relatively stable while PSU was oscillating faster than HWiNFO could poll (HWiNFO Settings->General->polling period). I could tell it was oscillating too fast because it was changing erratically. What this means is the changes were out of lockstep with the poll rate, and since it would occasionally not change during a poll period, that meant it was changing faster than could be polled. I could, however, find poll rates where it changed steadily, which meant the voltage itself wasn't erratic as it was changing at some multiple of my chosen poll rate.

 

My new motherboard doesn't have two probes. So if you have this situation you may have to figure it out based on swapping hardware. Maybe someone here has a better method, I honestly don't know.

 

In the GPU case, that one was not obvious as all voltages seemed normal. The only indication was the shutdowns and random orange fuzzy lines on the screen. You did not mention this but it wasn't clear to me either because the fuzzy lines were not very noticeable until it got very close to failing. When the thing finally failed it somehow shorted every single hot I tested to ground and neutral. Even the case fans, surge strip, and wall socket lol. Baffling, that one.

 

In the motherboard (non-VRM) case, that Gigabyte's crashes were accompanied by memory and volume read errors in the Windows Event Viewer. I didn't test voltages back then 'cuz I had not yet taken any repair classes.

 

Oh, and one more thing. The likely reason it doesn't crash after some minutes of uptime is heat. Once components heat up their electrical properties change.

I just checked HWINFO again and my 12V fluctuates every 4-5s between 12.144V to 12.168V so i don't think there is a problem with that AFAIK since its only fluctuating about 0.024V which is probably safe and still within the safe voltage limits(11.8-12.4V), but i will certainly keep an eye out for that since that is my third motherboard...(First one had energy issues, got a replacement that was stuttering a lot after a few months and rma confirmed issue and now i have the b550 tomahawk from msi) . About the VRM core i don't know which one is it so im attaching a screenshot from HWInfo and hoping you could tell me what it is. It is not fluctuating fast enough to be faster than the poll rate. I got no fuzzy lines happening here, my GPU works as expected so far. And i don't see anything noteworthy other than the Kernel-Power on the event viewer(will attach screenshot too).

[Mavendow]

Re: VRM I don't see it there so presumably your mobo doesn't have that probe. Iirc it was pretty clearly labeled VRM.

Re: PSU 4-5s oscillation is absolutely fine. Also note that the 12v probe typically has a large margin of error so that range may not even be the full range your mobo is seeing from the PSU and may instead be slight fluctuations from power draw. If there is a power problem between the PSU and the mobo, then it's the mobo, because no way would all the rails show such steady voltage if the PSU were the one feeding bad power.

Re: GPU I think we can rule this one out entirely, then. You have a 3700x which does not afaik suffer from the DWM/nvidia issues of the 2000 series and earlier.

 

You didn't mention checking CPU VCore voltage; I was best able to check this with CPU-Z rather than HWiNFO. Since it seems your voltages are stable from the PSU, perhaps we might see a difference here. I'm learning on this one also.

 

I notice in the event viewer you have A LOT of errors I don't like when I'm fixing computers... Fking DCOM, took me hours to figure out. They're mostly stuff that slows down execution or makes software hang rather than anything critical. That said, those Event 51 Disk warnings, that's a hardware write error which is only normal for removable drives.

 

Event viewer lists drives in a way that doesn't correspond to the drive letter clearly - it's NOT what's shown in Disk Management, some people think it is. Thankfully I've found how to get that in the past; just run the following powershell script. Note that PS execution is disabled by default so just paste the text into notepad, name it whatever.ps1, right click for "edit with powershell" to get the built-in script editor which also has the option to run the script.

# Build System Assembly in order to call Kernel32:QueryDosDevice. 
   $DynAssembly = New-Object System.Reflection.AssemblyName('SysUtils')
   $AssemblyBuilder = [AppDomain]::CurrentDomain.DefineDynamicAssembly($DynAssembly, [Reflection.Emit.AssemblyBuilderAccess]::Run)
   $ModuleBuilder = $AssemblyBuilder.DefineDynamicModule('SysUtils', $False)
 
   # Define [Kernel32]::QueryDosDevice method
   $TypeBuilder = $ModuleBuilder.DefineType('Kernel32', 'Public, Class')
   $PInvokeMethod = $TypeBuilder.DefinePInvokeMethod('QueryDosDevice', 'kernel32.dll', ([Reflection.MethodAttributes]::Public -bor [Reflection.MethodAttributes]::Static), [Reflection.CallingConventions]::Standard, [UInt32], [Type[]]@([String], [Text.StringBuilder], [UInt32]), [Runtime.InteropServices.CallingConvention]::Winapi, [Runtime.InteropServices.CharSet]::Auto)
   $DllImportConstructor = [Runtime.InteropServices.DllImportAttribute].GetConstructor(@([String]))
   $SetLastError = [Runtime.InteropServices.DllImportAttribute].GetField('SetLastError')
   $SetLastErrorCustomAttribute = New-Object Reflection.Emit.CustomAttributeBuilder($DllImportConstructor, @('kernel32.dll'), [Reflection.FieldInfo[]]@($SetLastError), @($true))
   $PInvokeMethod.SetCustomAttribute($SetLastErrorCustomAttribute)
   $Kernel32 = $TypeBuilder.CreateType()
 
   $Max = 65536
   $StringBuilder = New-Object System.Text.StringBuilder($Max)
 
   Get-WmiObject Win32_Volume | ? { $_.DriveLetter } | % {
       $ReturnLength = $Kernel32::QueryDosDevice($_.DriveLetter, $StringBuilder, $Max)
 
       if ($ReturnLength)
       {
           $DriveMapping = @{
               DriveLetter = $_.DriveLetter
               DevicePath = $StringBuilder.ToString()
           }
 
           New-Object PSObject -Property $DriveMapping
       }
   }

 

If it's not a hard drive then you can probably ignore it, or if you like, turn off paging for non-permanent drives so the system doesn't try to page a USB stick (System->Advanced System Settings->Performance->Settings...->Advanced->Virtual Memory->Change). If it IS a hard drive that may signal a disk or SATA problem. The latter can of course be caused by electrical malfunction on the motherboard, as I mentioned earlier.

 

If it is a permanent disk, easiest way to check whether its the disk or motherboard is by checking the disk's SMART attributes. HWiNFO can do this too. If the SMART attributes look fine then the paging error may either be the mobo, UEFI misconfiguration, or a corrupt OS installation. However, due to the stated timing of the crashes, only the first case is likely. The latter is impossible because that only occurs if there's a problem with delayed startup service for some hardware device. That would be listed around boot time. It seems your network card's driver registry is corrupt or misconfigured, but that should never cause a crash, the card just wouldn't work if it was a problem.

[dpeter]

1 minute ago, Mavendow said:

Re: VRM I don't see it there so presumably your mobo doesn't have that probe. Iirc it was pretty clearly labeled VRM.

Re: PSU 4-5s oscillation is absolutely fine. Also note that the 12v probe typically has a large margin of error so that range may not even be the full range your mobo is seeing from the PSU and may instead be slight fluctuations from power draw. If there is a power problem between the PSU and the mobo, then it's the mobo, because no way would all the rails show such steady voltage if the PSU were the one feeding bad power.

Re: GPU I think we can rule this one out entirely, then. You have a 3700x which does not afaik suffer from the DWM/nvidia issues of the 2000 series and earlier.

 

You didn't mention checking CPU VCore voltage; I was best able to check this with CPU-Z rather than HWiNFO. Since it seems your voltages are stable from the PSU, perhaps we might see a difference here. I'm learning on this one also.

 

I notice in the event viewer you have A LOT of errors I don't like when I'm fixing computers... Fking DCOM, took me hours to figure out. They're mostly stuff that slows down execution or makes software hang rather than anything critical. That said, those Event 51 Disk warnings, that's a hardware write error which is only normal for removable drives.

 

Event viewer lists drives in a way that doesn't correspond to the drive letter clearly - it is NOT what's shown in Disk Management, though some people think it is. Thankfully I've found to get that in the past; just run the following powershell script in an administrator PS window. Note that PS execution is disabled by default so just paste the text into notepad, name it whatever.ps1, right click for "edit with powershell" to get the built-in script editor which also has the option to run the script.

# Build System Assembly in order to call Kernel32:QueryDosDevice. 
   $DynAssembly = New-Object System.Reflection.AssemblyName('SysUtils')
   $AssemblyBuilder = [AppDomain]::CurrentDomain.DefineDynamicAssembly($DynAssembly, [Reflection.Emit.AssemblyBuilderAccess]::Run)
   $ModuleBuilder = $AssemblyBuilder.DefineDynamicModule('SysUtils', $False)
 
   # Define [Kernel32]::QueryDosDevice method
   $TypeBuilder = $ModuleBuilder.DefineType('Kernel32', 'Public, Class')
   $PInvokeMethod = $TypeBuilder.DefinePInvokeMethod('QueryDosDevice', 'kernel32.dll', ([Reflection.MethodAttributes]::Public -bor [Reflection.MethodAttributes]::Static), [Reflection.CallingConventions]::Standard, [UInt32], [Type[]]@([String], [Text.StringBuilder], [UInt32]), [Runtime.InteropServices.CallingConvention]::Winapi, [Runtime.InteropServices.CharSet]::Auto)
   $DllImportConstructor = [Runtime.InteropServices.DllImportAttribute].GetConstructor(@([String]))
   $SetLastError = [Runtime.InteropServices.DllImportAttribute].GetField('SetLastError')
   $SetLastErrorCustomAttribute = New-Object Reflection.Emit.CustomAttributeBuilder($DllImportConstructor, @('kernel32.dll'), [Reflection.FieldInfo[]]@($SetLastError), @($true))
   $PInvokeMethod.SetCustomAttribute($SetLastErrorCustomAttribute)
   $Kernel32 = $TypeBuilder.CreateType()
 
   $Max = 65536
   $StringBuilder = New-Object System.Text.StringBuilder($Max)
 
   Get-WmiObject Win32_Volume | ? { $_.DriveLetter } | % {
       $ReturnLength = $Kernel32::QueryDosDevice($_.DriveLetter, $StringBuilder, $Max)
 
       if ($ReturnLength)
       {
           $DriveMapping = @{
               DriveLetter = $_.DriveLetter
               DevicePath = $StringBuilder.ToString()
           }
 
           New-Object PSObject -Property $DriveMapping
       }
   }

 

If it's not a hard drive then you can probably ignore it, or if you like, turn off paging for non-permanent drives so the system doesn't try to page a USB stick (System->Advanced System Settings->Performance->Settings...->Advanced->Virtual Memory->Change). If it IS a hard drive that may signal a disk or SATA problem. The latter can of course be caused by electrical malfunction on the motherboard, as I mentioned earlier.

 

If it is a permanent disk, easiest way to check whether its the disk or motherboard is by checking the disk's SMART attributes. HWiNFO can do this too. If the SMART attributes look fine then the paging error may either be the mobo, UEFI misconfiguration, or a corrupt OS installation. However, due to the stated timing of the crashes, only the first case is likely. The latter is impossible because that only occurs if there's a problem with delayed startup service for some hardware device. That would be listed around boot time. It seems your network card's driver registry is corrupt or misconfigured, but that should never cause a crash, the card just wouldn't work if it was a problem.

Since the pc is working fine since i changed from the fan hub to the splitter i think that was it, it was a really cheap one and i think the mobo just turned on the cpu light when it couldnt turn on the fans, the only thing that does not fit in this theory is the random shutdown at startup... I will try this fix for the event viewer erros later, since im planning to re-install windows and erase everything later, thank you.

[Mavendow]

2 minutes ago, dpeter said:

Since the pc is working fine since i changed from the fan hub to the splitter i think that was it, it was a really cheap one and i think the mobo just turned on the cpu light when it couldnt turn on the fans, the only thing that does not fit in this theory is the random shutdown at startup... I will try this fix for the event viewer erros later, since im planning to re-install windows and erase everything later, thank you.

Okay, cool. The random shutdown is what I was trying to target but if you're satisfied with the status quo then I am too. Best of luck.

[dpeter]

1 minute ago, Mavendow said:

Okay, cool. The random shutdown is what I was trying to target but if you're satisfied with the status quo then I am too. Best of luck.

I mean, the random shutdown is probably the cause of the cpu light, for both of these issues to occur it only makes sense to me that they are the same, the random shutdown is not really random, it occurs when i first turn the pc on when i start the day, i turn it on, log myself in and go get some breakfast and when i return its off. The cpu light started today and it kinda sounds like the issue is progressing. I don't want to think it's my motherboard since it's my third one and im already pissed at these kind of problems, but i think it's the only thing i can target here.

[Mavendow]

8 minutes ago, dpeter said:

I mean, the random shutdown is probably the cause of the cpu light, for both of these issues to occur it only makes sense to me that they are the same, the random shutdown is not really random, it occurs when i first turn the pc on when i start the day, i turn it on, log myself in and go get some breakfast and when i return its off. The cpu light started today and it kinda sounds like the issue is progressing. I don't want to think it's my motherboard since it's my third one and im already pissed at these kind of problems, but i think it's the only thing i can target here.

It's possible the disk is electrically flawed; that's what I believe you should rule out before you go for yet another mobo. That said, the fact the hub didn't work points to a problem on the motherboard (as you mention) since a preexisting electrical problem is implied and it's unlikely a new piece of hardware also has an electrical problem.

[dpeter]

2 minutes ago, Mavendow said:

It's possible the disk is electrically flawed; that's what I believe you should rule out before you go for yet another mobo. That said, the fact the hub didn't work points to a problem on the motherboard (as you mention) since a preexisting electrical problem is implied and it's unlikely a new piece of hardware also has an electrical problem.

It's not that the hub did not work, i noticed that when the cpu light was on the AIO fans were not spinning, during the normal boot i noticed that they only start to spin after the cpu check is done. Maybe it was the hub, maybe it wasn't. And by the disk, you mean my hard drive? I have 2 NVMes and 1 HDD

[Mavendow]

1 hour ago, dpeter said:

It's not that the hub did not work, i noticed that when the cpu light was on the AIO fans were not spinning, during the normal boot i noticed that they only start to spin after the cpu check is done. Maybe it was the hub, maybe it wasn't. And by the disk, you mean my hard drive? I have 2 NVMes and 1 HDD

Aight. In this case I said "disk" to refer to any storage medium. A drive can be comprised of multiple physical disks... but we also refer to the physical device as "hard drives." Eh.

 

I also thought of something else: it's possible your PC is going into sleep mode due to inactivity and then crashing. I had a computer long ago that did not tolerate certain sleep states for whatever reason. The sleep mode that activated after __ mins of inactivity was not one which crashed it, but every PC is different. You mentioned "going to breakfast" so I wanted to bring this up. I doubt it, but I'd rather you be aware than mislead you to buy yet more hardware.

[dpeter]

7 minutes ago, Mavendow said:

Aight. In this case I said "disk" to refer to any storage medium. A drive can be comprised of multiple physical disks... but we also refer to the physical device as "hard drives." Eh.

 

I also thought of something else: it's possible your PC is going into sleep mode due to inactivity and then crashing. I had a computer long ago that did not tolerate certain sleep states for whatever reason. The sleep mode that activated after __ mins of inactivity was not one which crashed it, but every PC is different. You mentioned "going to breakfast" so I wanted to bring this up. I doubt it, but I'd rather you be aware than mislead you to buy yet more hardware.

Yeah everytime the pc should be asleep but was shutdown.Tomorrow im gonna sit right through it to see what happens haha

[Mavendow]

18 hours ago, dpeter said:

Yeah everytime the pc should be asleep but was shutdown.Tomorrow im gonna sit right through it to see what happens haha

Alright! That'd be great if it's remedied simply by disabling the offending sleep state in the UEFI, or if that's not an option, by removing it from the Power Configuration in Windows. This hadn't occurred to me in many years, so if this is the culprit I will have to update my knowledge. xD

[dpeter]

1 hour ago, Mavendow said:

Alright! That'd be great if it's remedied simply by disabling the offending sleep state in the UEFI, or if that's not an option, by removing it from the Power Configuration in Windows. This hadn't occurred to me in many years, so if this is the culprit I will have to update my knowledge. xD

Well, today there was nothing so good sign i guess, and i noticed that in windows i have it set to neve sleep, it just turns the screen off after 10min

[Mavendow]

6 hours ago, dpeter said:

Well, today there was nothing so good sign i guess, and i noticed that in windows i have it set to neve sleep, it just turns the screen off after 10min

Aww. Well, that leaves the electrical issue. I've tried to think of any other possible culprits, but I got nothin'. So I think we've narrowed it to your motherboard or one of your permanent disks. Pending any new developments I'll leave it up to you.

[dpeter]

Today everything continues to be ok. Im starting to think it could be the fan hub that i changed for the splitter, it explains why the cpu light was on(no cpu fan detected) and it was the only thing that i changed so far, the only way that this explains the shutdowns is if it failed while the pc was on, so the cpu overheated and system got shutdown i guess?

[brickson98]

This is a bit out of pocket, but if it's not power delivery issues or actually fan related (which it really sounds like it is) you could try reseating your cpu if your issues come back. I mean, reseating your cpu power and atx power connectors might not be a bad idea at that point as well. But this is likely super off base, because from what I'm reading in the thread, it sounds like you've already solved it with the fan hub/splitter situation. 

[dpeter]

2 minutes ago, brickson98 said:

This is a bit out of pocket, but if it's not power delivery issues or actually fan related (which it really sounds like it is) you could try reseating your cpu if your issues come back. I mean, reseating your cpu power and atx power connectors might not be a bad idea at that point as well. But this is likely super off base, because from what I'm reading in the thread, it sounds like you've already solved it with the fan hub/splitter situation. 

This only started to happen after i changed my case(i also updated my bios a day prior), so the most likely probabilty is that during the transfer of my build something was loose or did something wrong, pc components don't just start to do this out of nothing.

[brickson98]

1 minute ago, dpeter said:

This only started to happen after i changed my case(i also updated my bios a day prior), so the most likely probabilty is that during the transfer of my build something was loose or did something wrong, pc components don't just start to do this out of nothing.

Yeah, connections can get bumped and what not when you move components. Hence why I mentioned reseating the cpu if your issues continue. But, I wouldn't go as far as to say "pc components don't just start to do this out of nothing." I've seen some weird coincidences at work. You'll have an issue that is seemingly connected to a recent software or hardware change, but in all reality is completely unrelated and just so happened to appear after the change was made. Of course, it's extremely unlikely, but it's far from impossible that hardware went bad in the change to the new case. Especially if it's solder joint related. For example, if you have an already weak solder joint, moving it around when transferring the hardware could be enough for it to come completely disconnected. 

But it sounds like you've got it solved with the fan splitter/hub situation, right?

[dpeter]

1 minute ago, brickson98 said:

Yeah, connections can get bumped and what not when you move components. Hence why I mentioned reseating the cpu if your issues continue. But, I wouldn't go as far as to say "pc components don't just start to do this out of nothing." I've seen some weird coincidences at work. You'll have an issue that is seemingly connected to a recent software or hardware change, but in all reality is completely unrelated and just so happened to appear after the change was made. Of course, it's extremely unlikely, but it's far from impossible that hardware went bad in the change to the new case. Especially if it's solder joint related. For example, if you have an already weak solder joint, moving it around when transferring the hardware could be enough for it to come completely disconnected. 

But it sounds like you've got it solved with the fan splitter/hub situation, right?

So far so good, as for the cpu i did not take it out of the motherboard during the transfer so i don't think it's that it's the cause.

[brickson98]

20 minutes ago, dpeter said:

So far so good, as for the cpu i did not take it out of the motherboard during the transfer so i don't think it's that it's the cause.

Seeing as you resolved it with the fan stuff, I'd agree. But sometimes just moving it around causes the heatsink or water cooling block to shift pressure ever so slightly and cause contact issues with the cpu pins. Just something to keep in mind for future troubleshooting. Computers can be little trolls like that sometimes lol. 

 

Don't forget to mark your solution so people see the thread as solved.

[dpeter]

3 minutes ago, brickson98 said:

Seeing as you resolved it with the fan stuff, I'd agree. But sometimes just moving it around causes the heatsink or water cooling block to shift pressure ever so slightly and cause contact issues with the cpu pins. Just something to keep in mind for future troubleshooting. Computers can be little trolls like that sometimes lol. 

 

Don't forget to mark your solution so people see the thread as solved.

I will give it a few days just to confirm the issue is truly gone.

[dpeter]

Well, CPU light problem returned, did a few tests and confirmed its not CPU fan related, that fix was just a placebo it seems. A few days prior to rebuilding i updated my BIOS could that be it? I tried clearing CMOS and no luck, can i downgrade to an older BIOS using the bios flashback feature?

[dpeter]

I have a theory of what happened, at my first post i mentioned that the pwm functionality of my AIO fans was not working properly and i thought it was the fans, i have the wraith prism cooler that came with my cpu and tested the pc with it, no luck. I believe that my motherboard cpu fan header fried for some reason(2A limit , 2 0.4A fans connected to it) and now the board is having trouble powering on too, some electrical issue is happening here and its not the first time i have a fried board in this setup. Im activating the RMA for the board that is clearly bad, and im talking with Corsair Support and they said i could send my PSU to them so they can properly test it and see if my psu is the bad guy, if it is they will send me a new one. I thank you guys for all the help here but i think this issue is now closed. Fried board, possible defective psu killing boards.