Strange shutdown then no power to motherboard

[FusionGuy]

I was shutting down my PC and all of a sudden everything powered off meaning no status LEDs on motherboard (Maximus Extreme XI) or on GPU (Strix 2080 Ti OC). I tried to start it up again but nothing happened. I unplugged/plugged in the power supply and tried again but nothing. All this happened after I was using my new Corsair RGB mouse and keyboard but everything worked as it should at that time. I thought maybe the mouse/keyboard caused it. I plugged in my old wireless mouse/keyboard and tried again. I turned on the PSU and all status LED's came on. The PC started up but my RGB lighting on the Evolv X was not being recognized in AURA. I thought maybe the RGB controller got fried or something. I left everything as it was getting late and tried again the next day. I started up the PC with the new keyboard/mouse and everything seems to be working. I don't know if this was just a coincidence or what. I never experienced anything like this before so I don't know what the cause could of been. Would it be the PSU or mobo that was preventing it to power on.  This is a brand new build too. Any ideas?

[Mindovermaster Global Moderator]

I smell the PSU. Do you have a spare to test it out?

[FusionGuy]

No I don't. But the thing is everything is working now. So even if I did what would I test? Its a brand new Seasonic Prime Titanium 850W.

[farmeunit]

PSU is always the first thing I test.  Even if it's new.  Seasonic is usually pretty good, so it's unlikely, but I have had even new parts be faulty.  It is an odd coincidence that the keyboard/mouse are the only differences, though.

 

I have had more PSUs fail than most other components except hard drives.  Even if you have an old smaller wattage PSU, just to test with, that would be great. 

[FusionGuy]

This system is about a month old now. No issues at all until I connected the new mouse/keyboard that day. I already had iCUE installed for RGB fans and just connected the mouse/keyboard and it was recognized. I would say it was operating for about 3 hours after the new keyboard/mouse was connected, then it shut down.

I'm scratching my head on this one.

[Mindovermaster Global Moderator]

New or not, system components can die at any time. Slightly or heavily used. I had a few RAM sticks that were DOA (Dead On Arrival).

 

Get a new PSU and get it checked out. Because a PSU can harm your whole system if it is defective.

[DevTech]

2 hours ago, FusionGuy said:

This system is about a month old now. No issues at all until I connected the new mouse/keyboard that day. I already had iCUE installed for RGB fans and just connected the mouse/keyboard and it was recognized. I would say it was operating for about 3 hours after the new keyboard/mouse was connected, then it shut down.

I'm scratching my head on this one.

 

11 minutes ago, Mindovermaster said:

New or not, system components can die at any time. Slightly or heavily used. I had a few RAM sticks that were DOA (Dead On Arrival).

 

Get a new PSU and get it checked out. Because a PSU can harm your whole system if it is defective.

It is a Seasonic PSU, one of the top PSU you can buy in the same league as the SuperFlower PSU used in EVGA.

 

So unlikely, it is far from the first thing to consider.

 

I would start with looking for a BIOS update for the motherboard and a Firmware update on the GPU and any other controllers that have firmware. That's what it smells like to me if we are using noses telepathically connected over the internet...

 

 

 

[Mindovermaster Global Moderator]

3 minutes ago, DevTech said:

 

It is a Seasonic PSU, one of the top PSU you can buy in the same league as the SuperFlower PSU used in EVGA.

 

So unlikely, it is far from the first thing to consider.

 

I would start with looking for a BIOS update for the motherboard and a Firmware update on the GPU and any other controllers that have firmware. That's what it smells like to me if we are using noses telepathically connected over the internet...

 

 

 

Great or not, they can all fail at one point or another. Or, rather, get a voltage imbalance, therefore blowing up your computer.

[DevTech]

1 minute ago, Mindovermaster said:

Great or not, they can all fail at one point or another. Or, rather, get a voltage imbalance, therefore blowing up your computer.

They could also spontaneously burst into flames...

 

The situation described is much more of a fit to some low level firmware based software detecting something wrong with the motherboard and doing a panic shutdown. Probably false detection due to a software bug which is why I suggested update the crap out of stuff which still has some number of bugs due to its recent roll-out..

 

 

[FusionGuy]

9 hours ago, Riva said:

Same thing happened to me while gaming last week. Powered up the PC and it died again like 2-3 times in a row for no apparent reason. Then it just worked ok ever since.

Did you notice if your motherboard and or GPU status LED's also turned off?

[DevTech]

11 hours ago, Riva said:

Same thing happened to me while gaming last week. Powered up the PC and it died again like 2-3 times in a row for no apparent reason. Then it just worked ok ever since.

 

2 hours ago, FusionGuy said:

Did you notice if your motherboard and or GPU status LED's also turned off?

Other things to try:

 

1. If you have a water cooled CPU, make sure you have a fan blowing air onto the power regulators around the CPU socket. What works best is a side mounted fan directing cool outside air directly on that area, but anything at all is vital.

 

2. In BIOS, make sure "Load Line Calibration" is set to ON or Max Level and the number of "power phases" is set to the highest number

 

3. unlikely and almost impossible to see would be a slight damage to one of the socket pins when installing the CPU.

 

4. also unlikely, would be uneven clamp down on the water block or cpu cooler permitting a "hot spot" on the CPU chip 

 

5. excessive power draw on the PCIe bus due to GPU out of balance - would be solved by a GPU firmware update which you have already done by now, right?

 

 

[Mindovermaster Global Moderator]

11 minutes ago, DevTech said:

1. If you have a water cooled CPU, make sure you have a fan blowing air onto the power regulators around the CPU socket. What works best is a side mounted fan directing cool outside air directly on that area, but anything at all is vital.

I never ever heard of that... Care to explain that?

[DevTech]

10 minutes ago, Mindovermaster said:

I never ever heard of that... Care to explain that?

If you have a 125 watt CPU, you need to cool that down with a heask sink and then either air blowing through the heatsink or water flowing through it.

 

If you have an air cooler, then the air also blows past the power regulator chips for the CPU, providing the required cooling.

 

If your CPU is producing 125 watts of heat, all that enery is passing through the power chips that take in 12 volts from the PSU and convert it to 1 volt at the CPU - thats 125 AMPS! (crank my auto engine baby) so you really need to cool down those chips or you risk a CPU Glitch!

 

With a water cooling block that has become so popular these days with the very convenient integrated cooler kits, there is NO left over air flow to cool those massive regulator chips! So, blow some nice cool air on them from a side mounted fan!

 

 

[Mindovermaster Global Moderator]

6 minutes ago, DevTech said:

If you have a 125 watt CPU, you need to cool that down with a heask sink and then either air blowing through the heatsink or water flowing through it.

 

If you have an air cooler, then the air also blows past the power regulator chips for the CPU, providing the required cooling.

 

If your CPU is producing 125 watts of heat, all that enery is passing through the power chips that take in 12 volts from the PSU and convert it to 1 volt at the CPU - thats 125 AMPS! (crank my auto engine baby) so you really need to cool down those chips or you risk a CPU Glitch!

 

With a water cooling block that has become so popular these days with the very convenient integrated cooler kits, there is NO left over air flow to cool those massive regulator chips! So, blow some nice cool air on them from a side mounted fan!

 

 

Well, I always had clear air from the front go to the back, with no or little obstruction. (front bays are removed)

[FusionGuy]

30 minutes ago, DevTech said:

 

Other things to try:

 

1. If you have a water cooled CPU, make sure you have a fan blowing air onto the power regulators around the CPU socket. What works best is a side mounted fan directing cool outside air directly on that area, but anything at all is vital.

 

2. In BIOS, make sure "Load Line Calibration" is set to ON or Max Level and the number of "power phases" is set to the highest number

 

3. unlikely and almost impossible to see would be a slight damage to one of the socket pins when installing the CPU.

 

4. also unlikely, would be uneven clamp down on the water block or cpu cooler permitting a "hot spot" on the CPU chip 

 

5. excessive power draw on the PCIe bus due to GPU out of balance - would be solved by a GPU firmware update which you have already done by now, right?

 

 

GPU Tweak didn't notify me of a firmware update. The mobo was updated today.

[DevTech]

1 minute ago, Mindovermaster said:

Well, I always had clear air from the front go to the back, with no or little obstruction. (front bays are removed)

In the old days when water cooling was like a science project, details like this would not be forgotten.

 

But those handy water cool packages are the default these days and rightly so, but there are 2 "newbie" mistakes that can be made:

 

1. forgetting that airflow to the CPU power circuits

 

2. blowing hot air inside the case through the water cool radiators instead of blowing cool air form the outside inwards.

 

As you have pointed out there is often other air flow that mitigates the problem, but high-end CPUs with the most powerful heat generating GPUs will be prime candidates for a bit of extra attention to detail...

 

Also, one should never be confident on airflow from a fan over a distance more than about 6 inches since air has a will of it's own once it is set free to roam where it wants...

 

 

[Mindovermaster Global Moderator]

57 minutes ago, DevTech said:

In the old days when water cooling was like a science project, details like this would not be forgotten.

 

But those handy water cool packages are the default these days and rightly so, but there are 2 "newbie" mistakes that can be made:

 

1. forgetting that airflow to the CPU power circuits

 

2. blowing hot air inside the case through the water cool radiators instead of blowing cool air form the outside inwards.

 

As you have pointed out there is often other air flow that mitigates the problem, but high-end CPUs with the most powerful heat generating GPUs will be prime candidates for a bit of extra attention to detail...

 

Also, one should never be confident on airflow from a fan over a distance more than about 6 inches since air has a will of it's own once it is set free to roam where it wants...

 

 

Well, yes, but rather I had water or not, I ALWAYS had cool air coming from the front. Ever since I had my CM HAF cases...

 

So I guess this is why I never had to worry about it. 

[techbeck]

Try looking in to the power strip?  Maybe you have a faulty outlet.

[conna]

I doubt it was the Seasonic.  It's all I use in the last 10 years of client builds and not 1 has failed or caused issues.   I would trust Seasonic to run my life support system if I had to.