200°C RAM! Overheating G Skill Trident Z Royal DDR4 3600mHz CL16 32GB (4x8GB)

[Zeal1320]

Yes, u read the title correctly. my gskill ram is overheating. Unless hwmonitor is telling lies? <- i doubt so..

 

Backstory, I was having this issue with my pc, where the PC will suddenly crash out of nowhere, like when idling on the desktop without any apps launched. it will still suddenly crash / bluescreen.
Error : clock_watchdog_timeout

 

The temperatures u see on hw monitor is not editing magic, real data from hwmonitor.

 

 

I was thinking of taking out the old thermal pads inside to replace with new ones, but im not confident on taking the heatsink apart.

I have seen some videos on taking the heatsink apart but it looks sketchy af.

if anyone has any suggestions, other than buying a new set of ram.. please shoutout here! TIA.

[DreamCat04]

2 minutes ago, Zeal1320 said:

Yes, u read the title correctly. my gskill ram is overheating. Unless hwmonitor is telling lies? <- i doubt so..

 

Backstory, I was having this issue with my pc, where the PC will suddenly crash out of nowhere, like when idling on the desktop without any apps launched. it will still suddenly crash / bluescreen.
Error : clock_watchdog_timeout

 

The temperatures u see on hw monitor is not editing magic, real data from hwmonitor.

 

 

I was thinking of taking out the old thermal pads inside to replace with new ones, but im not confident on taking the heatsink apart.

I have seen some videos on taking the heatsink apart but it looks sketchy af.

if anyone has any suggestions, other than buying a new set of ram.. please shoutout here! TIA.

HwMonitor loves to misread stuff from time to time. I doubt RAM would survive 240°C, it'd immediately die if it were at that temperature. HWMonitor has read around 500kW of CPU package power on my i7-9700 in the max column, which is plain impossible

[WereCat]

These reading don't make any sense. I doubt that DDR4 can run that hot ever and even if it could it would fail way before that.

Use Hwinfo64 instead.

[PDifolco]

8 minutes ago, Zeal1320 said:

Yes, u read the title correctly. my gskill ram is overheating. Unless hwmonitor is telling lies? <- i doubt so..

 

Backstory, I was having this issue with my pc, where the PC will suddenly crash out of nowhere, like when idling on the desktop without any apps launched. it will still suddenly crash / bluescreen.
Error : clock_watchdog_timeout

 

The temperatures u see on hw monitor is not editing magic, real data from hwmonitor.

 

 

I was thinking of taking out the old thermal pads inside to replace with new ones, but im not confident on taking the heatsink apart.

I have seen some videos on taking the heatsink apart but it looks sketchy af.

if anyone has any suggestions, other than buying a new set of ram.. please shoutout here! TIA.

Buggy readings, since 1 week or so HWMonitor shows me 150C CPU whenever I play Cyberpunk, should be an Overheat Quickhack from the game lol  

[DoctorNick]

12 minutes ago, Zeal1320 said:

Yes, u read the title correctly. my gskill ram is overheating. Unless hwmonitor is telling lies? <- i doubt so..

 

Backstory, I was having this issue with my pc, where the PC will suddenly crash out of nowhere, like when idling on the desktop without any apps launched. it will still suddenly crash / bluescreen.
Error : clock_watchdog_timeout

 

The temperatures u see on hw monitor is not editing magic, real data from hwmonitor.

 

 

I was thinking of taking out the old thermal pads inside to replace with new ones, but im not confident on taking the heatsink apart.

I have seen some videos on taking the heatsink apart but it looks sketchy af.

if anyone has any suggestions, other than buying a new set of ram.. please shoutout here! TIA.

 

 

Yes try hwinfo64 instead like @WereCat mentioned: https://www.hwinfo.com/download/

Choose sensors only when launching the program

[Daniel131105]

If it's 240° it will burn the whole ram and probably your PC but I doubt it is that hot

[Zeal1320]

Just downloaded HWinfo, so far max temps are around 40-50 under light load. Thanks for the recommendation guys!

however, im still not sure about the random crashes though 


 

Edit : hwinfo64 showing 175 degrees C........

 

[Zeal1320]

21 minutes ago, DoctorNick said:

Yes try hwinfo64 instead like @WereCat mentioned: https://www.hwinfo.com/download/

Choose sensors only when launching the program

 

23 minutes ago, PDifolco said:

Buggy readings, since 1 week or so HWMonitor shows me 150C CPU whenever I play Cyberpunk, should be an Overheat Quickhack from the game lol  

 

29 minutes ago, DreamCat04 said:

HwMonitor loves to misread stuff from time to time. I doubt RAM would survive 240°C, it'd immediately die if it were at that temperature. HWMonitor has read around 500kW of CPU package power on my i7-9700 in the max column, which is plain impossible

 

25 minutes ago, WereCat said:

These reading don't make any sense. I doubt that DDR4 can run that hot ever and even if it could it would fail way before that.

Use Hwinfo64 instead.

Maybe its really time for new ram? or just new thermal pads..

 

 

 

[WereCat]

4 minutes ago, Zeal1320 said:

 

 

 

Maybe its really time for new ram? or just new thermal pads..

 

 

 

That can still be just a buggy sensor on the RAM. Even without cooling RAM will not run this hot.

For your crashing, have you ran any RAM test?

[Zeal1320]

1 minute ago, WereCat said:

That can still be just a buggy sensor on the RAM. Even without cooling RAM will not run this hot.

For your crashing, have you ran any RAM test?

have not done any ram test, can u recommend me one?

[WereCat]

10 minutes ago, Zeal1320 said:

have not done any ram test, can u recommend me one?

I prefer RAM Test from Karhu SW  https://www.karhusoftware.com/ramtest/#introduction but it's paid.... advantage is that it's significantly faster than the free Memtest https://www.memtest86.com

Let it run for at least 2h or 4 full passes... there must be 0 errors. I mean, if you get error sooner you know there is an issue with RAM so you can stop the test... but that does not automatically means the RAM is faulty, maybe just bad auto configuration of timings from motherboard.

[Zeal1320]

7 minutes ago, WereCat said:

I prefer RAM Test from Karhu SW  https://www.karhusoftware.com/ramtest/#introduction but it's paid.... advantage is that it's significantly faster than the free Memtest https://www.memtest86.com

Let it run for at least 2h or 4 full passes... there must be 0 errors. I mean, if you get error sooner you know there is an issue with RAM so you can stop the test... but that does not automatically means the RAM is faulty, maybe just bad auto configuration of timings from motherboard.

Thank you very much @WereCat, will try out memtest and update on the results later.

[PDifolco]

42 minutes ago, Zeal1320 said:

Just downloaded HWinfo, so far max temps are around 40-50 under light load. Thanks for the recommendation guys!

however, im still not sure about the random crashes though 


 

Edit : hwinfo64 showing 175 degrees C........

 

What's your RAM speed ? I suspect 4 sticks @3600 on a 5800X is prone to crashing

 

[Zeal1320]

12 minutes ago, PDifolco said:

What's your RAM speed ? I suspect 4 sticks @3600 on a 5800X is prone to crashing

 

 

3600mHz @ CL 16-19-19-39

[PDifolco]

8 minutes ago, Zeal1320 said:

 

3600mHz @ CL 16-19-19-39

This may be the cause of crashes, AM4 IMC are notorious for being unstable with 4 sticks at high speed, try 3200MHz or loosen the timings (CL18)

[Somerandomtechyboi]

1 hour ago, Zeal1320 said:

 

 

 

Maybe its really time for new ram? or just new thermal pads..

 

 

 

buggy ram sensor just ignore

 

as for crashing/instability issues ill just summon @RONOTHAN##

 

1 hour ago, DreamCat04 said:

HwMonitor loves to misread stuff from time to time. I doubt RAM would survive 240°C, it'd immediately die if it were at that temperature. HWMonitor has read around 500kW of CPU package power on my i7-9700 in the max column, which is plain impossible

 

1 hour ago, Daniel131105 said:

If it's 240° it will burn the whole ram and probably your PC but I doubt it is that hot

not sure what kinda temps theyll survive being powered on but if powered off 175c isnt gonna do anything to em and you need to hit somewhere around 200-220c to do something to the rams and that is to take them off the pcb with a bga rework station or even a heatgun would also work albiet you do have to follow a reflow curve so the pcb doesnt delaminate (something to do with pcb layers seperating if heated or cooled down too fast especially if not preheated)

 

no they wont die as theyre powered off and this is about their designed max temp since you gotta solder the rams onto the pcb somehow, they dont just magically attatch themselves to the pcb yk

 

and most rams are pretty voltage tolerant usually topping out voltage scaling wise before even degrading, some just straight up dont degrade even past 2v, so i wouldnt be surprised if pushing past 100c powered on would do nothing but destabilize the rams

[Zeal1320]

Loosening the timings and lowering the clock speed helped, now there is no crashes or stuttering.

And the temperatures are below 50°C under load.

Thanks everyone for their input on this topic!