Brand new build - hard crashing

[BradS550]

I received my custom build from PCS last week and I've run it about 3-4 times since setting it up. I haven't made any changes to it's stock settings (bios/overclocking etc) and I've had it hard crash on me twice. No warning, no error just instant crash. The screens go black (no output) but the fans and internal lights are still running. The fans stay at an increased but constant speed and long pressing the power button or hitting the reset button has no effect. The red cpu light is illuminated on the MOBO. i have to switch off the power on the PSU, then power back on as usual. Boots back into windows fine with no error message.

Both times it's happened, i have not been doing anything computationally Intensive. On the second occasion, i was running hwinfo to monitor the core temps (as i suspected cpu overheating/shutdown protection behaviour from the first crash). Cpu die temp was around 60⁰-70⁰.

Spec:
Case
LIAN LI LANCOOL II GAMING CASE
Processor (CPU)
AMD Ryzen 9 7950X 16 Core CPU (4.5GHz-5.7GHz/80MB CACHE/AM5)
Motherboard
GIGABYTE B650 GAMING X AX : (WIFI 6E, DDR5, USB 3.2) - ARGB Ready!
Memory (RAM)
32GB Corsair VENGEANCE DDR5 6000MHz (2 x 16GB)
Graphics Card
24GB NVIDIA GEFORCE RTX 4090 - HDMI, DP
1st M.2 SSD Drive
1TB SEAGATE FIRECUDA 530 GEN 4 PCIe NVMe (up to 7300MB/R, 6000MB/W)
Power Supply
CORSAIR 1000W RMe SERIES™ MODULAR 80 PLUS GOLD, ULTRA QUIET
Power Cable
1 x 1.5 Metre UK Power Cable (Kettle Lead)
Processor Cooling
PCS FrostFlow 240 Series RGB High Performance Liquid Cooler (AMD)
Thermal Paste
STANDARD THERMAL PASTE FOR SUFFICIENT COOLING
Extra Case Fans
1x 120mm Black Case Fan
Sound Card
ONBOARD 6 CHANNEL (5.1) HIGH DEF AUDIO (AS STANDARD)
Network Card
10/100/1000 GIGABIT LAN PORT
USB/Thunderbolt Options
MIN. 2 x USB 3.0 & 2 x USB 2.0 PORTS @ BACK PANEL + MIN. 2 FRONT PORTS
Operating System
Windows 11 Home 64 Bit


Please advise.

 

[Martinr36]

Have you apllied all the windowas updates including those under the optional updates tab

 

[BradS550]

Windows is up to date. 'optional' updates are optional by definition and should not be required to prevent a hard crash

 

[B4zookaw]

Yeah but optional updates often include driver updates, which are important for system stability.

What temp is the cpu at idle? And do you see any error events in the event viewer?

 

[ubuysa]

It's not unusual with a new build to find that items have moved slightly whilst in transit. It's not impossible that when things warm up, this could cause problems.

I would check that everything is fully seated; PCIe cards, RAM, M.2 drives and all cables.

Is that cooler big enough for that CPU I wonder?

 

[B4zookaw]

Is that cooler big enough for that CPU I wonder?

It’s not ideal, 280mm would be my min, but the 240 should be able to manage most usage

 

[BradS550]

Hi Ubuysa,

It's not unusual with a new build to find that items have moved slightly whilst in transit. It's not impossible that when things warm up, this could cause problems

Components are secured in place using latches and screws. They should not of moved a significant amount to the point where they are losing contact with their respective terminals.

I would check that everything is fully seated; PCIe cards, RAM, M.2 drives and all cables.

Everything seems securely seated and fastened. Why would the CPU LED be lit on the motherboard after the crash?

Is that cooler big enough for that CPU I wonder?

Surely if it wasn't, this would of been identified when PCS were testing the machine after it was built, and likely of experienced the same problems i'm now having.

I've been monitoring the cpu temps during usage and i've not seen them get above 95. which is well within operating temp for this cpu.

Thanks all for your contribution thus far

 

[SpyderTracks]

Hi Ubuysa,

Components are secured in place using latches and screws. They should not of moved a significant amount to the point where they are losing contact with their respective terminals.


Everything seems securely seated and fastened. Why would the CPU LED be lit on the motherboard after the crash?


Surely if it wasn't, this would of been identified when PCS were testing the machine after it was built, and likely of experienced the same problems i'm now having.

I've been monitoring the cpu temps during usage and i've not seen them get above 95. which is well within operating temp for this cpu.

Thanks all for your contribution thus far

It's going to be very difficult to offer any advice if you're constantly saying "it should be this way". The reality is things do come adrift in shipping, not everything is secured with screws.

And optional updates are required as they are system drivers.

 

[BradS550]

Fair point.

Excuse my frustration. I've waited twice the time i was quoted for manufacture, only to receive a rig that has failed twice within the first few hours of usage.

 

[BradS550]

Has anyone got any info/specs on the cpu thermal protection procedure for the mobo? Interested to know if this is what's being invoked each time. And why the power button and reset button would have no effect.

 

[sck451]

Has anyone got any info/specs on the cpu thermal protection procedure for the mobo? Interested to know if this is what's being invoked each time. And why the power button and reset button would have no effect.

Could easily be something with the motherboard, though I guess it could also be a defective CPU. It's not exactly the highest spec motherboard (I think it's the cheapest AM5 board PCS sells...) but it shouldn't have this kind of failure. Either of these should have been caught in QC before it was shipped. I would suggest checking the power connections once more (especially the 12-pin CPU connector in the top left of the board and the 24-pin ATX connector on the right), but if it happens again I would contact PCS for support.

 

[Scott]

I think your cooler is a massive limitation here. If it's sitting at 95°C I would say that's far too hot for typical usage. For 95 I'd be expecting you to be running every core at max, if that's not the case then it's highly likely that you're seeing the 100C TJmax and the system is hard cutting to protect itself, and thus leaving on the CPU error LED.

With that being said, the cooler should be enough for light to moderate use without too much hassle, so I would suggest there's an issue somewhere.

If the Pump wasn't running (disconnected in shipping) or something of that ilk then it could definitely explain the circumstances.

The easy way to try and replicate the situation is by running a torture test such as prime 95. If your cooling isn't up to snuff it'll find that out for you pretty quickly.

 

[BradS550]

Yeah thanks for the input all.

I'll definitely be reaching out to PCS tech support.

Please note tho that normal operation barely gets the system above 75⁰. I only managed to get it to a maximum of 95⁰ by stress testing it with prime 95 and that did not trigger a crash.

seeing the 100C TJmax and the system is hard cutting to protect itself, and thus leaving on the CPU error LED.

Is 100⁰c the temp that the B650 invokes cpu protection shut down mode?

 

[Scott]

Yeah thanks for the input all.

I'll definitely be reaching out to PCS tech support.

Please note tho that normal operation barely gets the system above 75⁰. I only managed to get it to a maximum of 95⁰ by stress testing it with prime 95 and that did not trigger a crash.


Is 100⁰c the temp that the B650 invokes cpu protection shut down mode?

Yes, it will be thermal throttling before that though. If Prime 95 only got to 95 then that suggests your temps are OK. 75 for normal use is still quite warm but within acceptable limits I would say.

 

[BradS550]

Yes, it will be thermal throttling before that though.

Can you post where you're getting the info for the B650. I had a poke about and couldn't find any info on the limits of the board.

Also, whilst 95 degrees is hot for a cpu - i only managed get it there by using prime95 and these CPU's notoriously run hot. as spec'ed by AMD themselves - 95 is fine for this cpu. The first time it crashed, i was browsing the web. The second time i was browsing the steam store.

So would the best course of action be to contact PCS technical support?

 

[Martinr36]

Can you post where you're getting the info for the B650. I had a poke about and couldn't find any info on the limits of the board.

Also, whilst 95 degrees is hot for a cpu - i only managed get it there by using prime95 and these CPU's notoriously run hot. as spec'ed by AMD themselves - 95 is fine for this cpu. The first time it crashed, i was browsing the web. The second time i was browsing the steam store.

So would the best course of action be to contact PCS technical support?

Yeah, I think I would

 

[Scott]

Can you post where you're getting the info for the B650. I had a poke about and couldn't find any info on the limits of the board.

Also, whilst 95 degrees is hot for a cpu - i only managed get it there by using prime95 and these CPU's notoriously run hot. as spec'ed by AMD themselves - 95 is fine for this cpu. The first time it crashed, i was browsing the web. The second time i was browsing the steam store.

So would the best course of action be to contact PCS technical support?

I think you may be getting mixed up. The TJmax is a CPU design limitation. It's the limit that AMD have put on the chip. The 95C that you are seeing in Prime95 is the CPU hitting this limit and then being power throttled to stop it from going over.

It's likely you will see thermal throttling before this, but I'm unsure exactly when that line gets crossed. Info on your chip here:

https://www.amd.com/en/product/12151

 

[Scott]

As for next steps. I would keep testing it to see if you can re-create the issue. It's always better to send it back knowing that PCS can re-create your circumstances. The last thing you want is to send in the PC, for them not to be able to get the reset to happen, send it back and then it happens again.