Question CPU Socket Temperature (Ryzen 5 5600x)

Jan 27, 2021
33
2
35
0
Some threads that I have read suggest resetting the CMOS, but that is not an option for me right now. Everything on the computer is brand new with the exception of the GPU due to how hard it is to get one right now. I had a heck of a time trying to get the PC to post whenever I built it. After a week of pouring through forums and trying this or that, purchasing a super cheap GPU (GT 1030) allowed me to get into BIOS and change it from UEFI to CSM. That did the trick and allowed my older GPU to work. I returned that cheap GPU, so if I do a full BIOS/CMOS reset, I'm afraid that I'll be locked out of getting back in until I could use a different GPU to set it to CSM once again.
 
Some threads that I have read suggest resetting the CMOS, but that is not an option for me right now. Everything on the computer is brand new with the exception of the GPU due to how hard it is to get one right now. I had a heck of a time trying to get the PC to post whenever I built it. After a week of pouring through forums and trying this or that, purchasing a super cheap GPU (GT 1030) allowed me to get into BIOS and change it from UEFI to CSM. That did the trick and allowed my older GPU to work. I returned that cheap GPU, so if I do a full BIOS/CMOS reset, I'm afraid that I'll be locked out of getting back in until I could use a different GPU to set it to CSM once again.
What makes resetting CMOS a non-option? If you never reset CMOS, not even after first building it, that would have been an excellent reason you had troubles getting it to POST.

What's your question about socket temperature?
 
Reactions: kurdtnz
Jan 27, 2021
33
2
35
0
What makes resetting CMOS a non-option? If you never reset CMOS, not even after first building it, that would have been an excellent reason you had troubles getting it to POST.

What's your question about socket temperature?
I tried to be as detailed as I could be without giving you a book :). I'll explain a little further. When I first built the PC I could not get a visual on screen. All I got was no display detected. I knew the GPU worked because I could pull it out and plug it back into my old machine and it worked just fine. All I would get on the new machine was a series of beeps, which I found out was a video card error beep. I think it was one long beep followed by two short beeps, but I don't recall now. That didn't make sense since the GPU would work in my old machine.

On another thread someone asked if I had swapped the system from UEFI to CSM, which I had not because I could never get anything to display, so there was no way that I could change any settings. While I was doing all of this testing trying to get the system to boot, I did reset the CMOS (many times lol). It turns out that the fix was quite simple...my graphics card was just old and it requires the legacy CSM boot method, but the motherboard was set to UEFI by default. Whenever I purchased the cheap GT 1030 GPU and put it in, I was able to get into the BIOS and change the boot method to CSM. I swapped the cards out once again and it booted right up with my old GPU.

The PC has been working for a few weeks now. No issues whatsoever. However, when I rebooted the machine yesterday, I opened the Dragon Center monitoring window that I typically leave up just to keep an eye on things and I was immediately alarmed to see that the socket temperature was reading 186C (366F). I assumed it was a bad reading or else something would have surely stopped working if that was the case. I rebooted again and went into BIOS to check the temperatures there. The CPU socket and the motherboard temperatures there show as 0C (32F). I'll show what I see in various monitors below:



Another issue that I saw is that the fans sensors don't appear to be showing their speeds any more. You can see CPU and Pump (nothing connected), but Sys1-3 are on the next page sitting at 0 as well. Those were reporting a speed prior to this reboot. What is even more odd is that even though the CPU fan is showing no RPM's at the bottom of this image, it does show that the fan is at 20% in the gauge at the top. This gauge does go up when I change the profile from silent to either balanced or extreme performance.

 
Jan 27, 2021
33
2
35
0
What is your full system specs? What kind of temperature are you getting and what software are you using to read the temps?
See my comment to drea.drechsler regarding programs and temperatures.

As far as specs, these are the components that I am using:

PCPartPicker Part List

Type|Item
:----|:----|:----
CPU | AMD Ryzen 5 5600X 3.7 GHz 6-Core Processor
CPU Cooler | Scythe Mugen 5 Rev. B 51.17 CFM CPU Cooler
Motherboard | MSI MAG B550M MORTAR WIFI Micro ATX AM4 Motherboard
Memory | Silicon Power XPOWER Turbine RGB 16 GB (2 x 8 GB) DDR4-3200 CL16 Memory
Storage | Samsung 970 Evo 500 GB M.2-2280 NVME Solid State Drive
Storage | Western Digital Caviar Blue 1 TB 3.5" 7200RPM Internal Hard Drive
Video Card | XFX Radeon R9 280 3 GB DD Video Card
Case | darkFlash DLM 21 MicroATX Mid Tower Case
Power Supply | SeaSonic FOCUS Plus Gold 650 W 80+ Gold Certified Fully Modular ATX Power Supply
Case Fan x2 sets | Cooler Master SickleFlow 62 CFM 120 mm Fans 3-Pack

The graphics card that I am using is an older Radeon R9 280.
 
Jan 27, 2021
33
2
35
0
It seems like the other sensors are going haywire today. You can clearly see in the previous image that I posted that the CPU core was reading at 35.2C, but now it's just pegged at 10C. I even booted up a game, which would typically push it up to around 40-50C, but it still reads out at 10C.

 
Jan 27, 2021
33
2
35
0
What makes resetting CMOS a non-option? If you never reset CMOS, not even after first building it, that would have been an excellent reason you had troubles getting it to POST.

What's your question about socket temperature?
To clarify a little further, after I was done with the cheap GPU I just returned it, so I no longer have a second GPU option. I'm afraid that if I reset the CMOS then it'll reset the motherboard back to UEFI and I will no longer be able to use my GPU since it requires CSM.
 
To clarify a little further, after I was done with the cheap GPU I just returned it, so I no longer have a second GPU option. I'm afraid that if I reset the CMOS then it'll reset the motherboard back to UEFI and I will no longer be able to use my GPU since it requires CSM.
I think you're mistaken as you can set BIOS to CSM mode at any time; just spam the DEL key when it's booting up to go into BIOS and do so.

I never figured an R9 280 to old of a GPU to not work with UEFI. I've an HD 7770 that works great...or as great as it ever did.

You definitely got something strange with those temperature readings and I don't think it's the hardware. You might think about removing the dragon center utility, that may be part of the problem. Your system may also be a candidate for fresh/clean install of Windows. It may seem harsh but it's lot easier than chasing down strange problems.
 
Last edited:
Jan 27, 2021
33
2
35
0
I think you're mistaken as you can set BIOS to CSM mode at any time; just spam the DEL key when it's booting up to go into BIOS and do so.

I never figured an R9 280 to old of a GPU to not work with UEFI. I've an HD 7770 that works great...or as great as it ever did.

You definitely got something strange with those temperature readings and I don't think it's the hardware. You might think about removing the dragon center utility, that may be part of the problem. Your system may also be a candidate for fresh/clean install of Windows. It may seem harsh but it's lot easier than chasing down strange problems.
I did try that whenever I was having issues getting in. I spammed DEL and pretty much every single F key on my keyboard, every time I would just end up getting the GPU error code beeps. I can't even remember the amount of times I rebooted, but I'd try a different key every time and I reset the CMOS quite a few times. I even pulled the battery out and held the power button for over a minute and left it alone over night just to make sure that there was nothing left in there and then tried the next day...no luck. I was not able to get anything to display to my monitors at all. Whenever I powered my machine up the monitors would activate (buttons and options would light up), but then I'd just get a "no display detected" message on screen and they'd shut down.

The first time I popped in the GT 1030 it booted right into BIOS without issue. The only thing I touched was the UEFI to CSM toggle switch and I shut down. I popped my old GPU in and it booted right up too.

I am using a Radeon R9 280, which is from like 2014. I am going to get an RTX 3070 whenever I'm able to get my hands on one.
 
Jan 27, 2021
33
2
35
0
Any ideas for things to troubleshoot? Could re-seating the cooler have an impact? I could try that if it could potentially help. I just don't know if that would even be a potential fix.
 
Jan 27, 2021
33
2
35
0
All,

So I just wanted to circle back provide an update for those interested in learning more about this type of issue.

Another piece of this build that I had been working on was the fan arrangement and RGB configuration in my case. I am using 6 CoolerMaster Sickleflow RGB fans, and an issue I was having was correctly connecting the proper amount of fans per header. Due to some distance and cord length issues, I ended up using the 3x1 splitters that came with the fans (the fans came in 2 packs of 3 fans each) to plug 3 fans each into 2 of the sysfan headers. I later learned that these headers have a max draw of 1A, and the fans plugged in had a max pull of 1.11A, so I ordered another splitter. I now have the fans plugged in with a 2x2x2 configuration between the 3 sysfan headers that I have on my motherboard, which puts me well below the max draw limit of each header.

As soon as I fixed the fans like this and booted up, all of my sensors were working again. All temperatures started to show temperatures that were going up and down based on load, and all fans were showing their RPM measurements that also fluctuated with load. All seems well now.

For conversation sake, would the fact that I was essentially exceeding the max draw of my sysfan headers leading to some sort of power short or something? The fans were well within the limit if you just look at the rated current, but the safety current is what put me at 1.11A (effectively .11 over the draw limit). I know it's virtually impossible for anyone to come to a certain conclusion without looking at the system themselves, but I'm curious to hear input from others on what may have been happening.

Thanks!
 

ASK THE COMMUNITY