FX-9590 Temperatures too High with Factory FX-8150 Water Cooler

[ThE_MaD_ShOt]

Yea I would say temps are your problem. Your idling 10c higher then my 100% load temps. Them temps are way up there under load . Hell when my h100 died, and my temps went up to 62 my system would lock, bsod, or reboot itself. My idle temps where high like yours. You may have a failed 8150 aio.

 

[Ravenas]

socket temp is meaningless litterally doesn't matter I suspose it could be the VRM OCP kicking in and throttling back causing a bsod
give the vrm heatsink a feel and if you can't keep your finger on it for more then 3 seconds it COULD be getting a bit toasty under a 220W load
Unless your chip is defective it should't be crashing at 60C amd says 62c but you usually have a fair bit more then that before things start going wonkey
id point a fan at the VRM and see what happens with the lack of airflow due to the AIO its probly getting toasty

I can keep my finger on the heat sink its fine. The water going in to the heat sink is cold and the water going out is hot. Seems to be functioning properly.

 

[ThE_MaD_ShOt]

I can keep my finger on the heat sink its fine. The water going in to the heat sink is cold and the water going out is hot. Seems to be functioning properly.

My h100 felt the same way. Rad was warming up and all. Pump wasn't fully dead just wouldn't run full speed. I rma'd it and they sent the h100i I have and all has been great.

 

[Ravenas]

My h100 felt the same way. Rad was warming up and all. Pump wasn't fully dead just wouldn't run full speed. I rma'd it and they sent the h100i I have and all has been great.

Hmm maybe I should invest in a new water cooler... Where am I going to fit one of the giant water coolers? haha....

 

[ThE_MaD_ShOt]

You can go with a high end 120 aio in push pull or a 240 like a h100i and mount the rad on the side panel. Just need to run the system with the rad just propped up for a short while to make sure everything is good be fore mounting the rad to the panel. I am running mine in just push fashion. Or time for that new case you had in the back of your mind LOL


Have you tried putting your 8350 back in?

 

[Ravenas]

You can go with a high end 120 aio in push pull or a 240 like a h100i and mount the rad on the side panel. Just need to run the system with the rad just propped up for a short while to make sure everything is good be fore mounting the rad to the panel. I am running mine in just push fashion. Or time for that new case you had in the back of your mind LOL


Have you tried putting your 8350 back in?

No haven't tried anything with my 8350. I had to buy a new motherboard because my old MSI motherboard fried after using this processor with a BIOS lower than the required for a 9590. My CPU fan wouldn't work on that bad motherboard either so I wonder if it may have gotten my pump, although it says my pump is running fine.

 

[ThE_MaD_ShOt]

No haven't tried anything with my 8350. I had to buy a new motherboard because my old MSI motherboard fried after using this processor with a BIOS lower than the required for a 9590. My CPU fan wouldn't work on that bad motherboard either so I wonder if it may have gotten my pump, although it says my pump is running fine.

Hmm I would think it would fry the mobo. I would think that the most would happen would be it would post. The vrm circuit is there and all the power circuits are there regardless of the bios. Bios is just software/firmware to id the chip.

It could have hurt the pump.

 

[Ravenas]

Hmm I would think it would fry the mobo. I would think that the most would happen would be it would post. The vrm circuit is there and all the power circuits are there regardless of the bios. Bios is just software/firmware to id the chip.

Yes but when you see smoke and then inspect the motherboard and see a lot of black in places it shouldn't be, then it's time for a new mobo. The MSI motherboard I had didn't work with 4 different AM3+ processors that I tried. MSI gave me a refurbished one when I applied for a warranty.

I agree with it possibly causing damage to the pump. Maybe the pump is receiving voltage and the software thinks that it is operational, but it's actually not working.

 

[ThE_MaD_ShOt]

Yes but when you see smoke and then inspect the motherboard and see a lot of black in places it shouldn't be, then it's time for a new mobo. The MSI motherboard I had didn't work with 4 different AM3+ processors that I tried. MSI gave me a refurbished one when I applied for a warranty.

Oh sorry I may have worded that correctly. I wasn't saying that the mobo wasn't fried. I think there may have been something else that contributed to it like a weak vrm circuit of something. Hell I have one Msi board here I have to Rma due to a dead Pci-e socket.

 

[newtekie1]

The VRMs are overheating, IMO. One of the issues with switching to liquid cooling is the amount of airflow over the VRM area is significantly reduced. The motherboard only has an 8 phase VRM, so they are going to be pretty stressed with a 9590.

 

[Ravenas]

The VRMs are overheating, IMO.

Is it possible to get accurate temperature readings from them?

 

[thebluebumblebee]

Are you really running a FX-9590 and a 290X in a Cooler Master Elite 335? That case is made for maybe a 200 watt system, not the (220-CPU + 300-GPU) that you have. Also, there's no cutout behind the motherboard so I would think that heat would build up there.
Do you have the power to the pump plugged into the motherboard or to the PSU?

 

[Ravenas]

Are you really running a FX-9590 and a 290X in a Cooler Master Elite 335? That case is made for maybe a 200 watt system, not the (220-CPU + 300-GPU) that you have. Also, there's no cutout behind the motherboard so I would think that heat would build up there.
Do you have the power to the pump plugged into the motherboard or to the PSU?

The power is plugged in to my motherboard via CPU fan and USB. The fans are working properly on the water pump and the pump shows it is working via software.

 

[OneMoar]

@ThE_MaD_ShOt hes running a 9590 his chip uses twice the power as yours and runs at least 30% hotter read the thread ffs 60C @load from a 9570 using the AMD AIO is pretty dam acceptable I would almost say its perfectly fine given the chip and the cooling hes running
his vrm's are most likely cooking them selves to death from the load he needs more cooling on them
@op take the side panel off and point a box fan in there and see if that changes any thing
and a word of warning even tho asus claims support for the 9570 that doesn't mean jackshit the 9570 is a pos with a huge swing in power consumption and heat output between any given sample so YMMV

 

[newtekie1]

Is it possible to get accurate temperature readings from them?

No, there might be a temp sensor in the board near them, but it won't be accurate. Best option would be to point a IR thermometer at them when the machine is under load. Aim for the grey boxes under the heatsink, not the heatsink itself.

You can also try to point a fan at the vrm/socket area and see if it helps with the crashing.

Though, IMO, your board just isn't built for the high TDP of the 9590, even if ASUS says it technically supports it. I'd try lowering the CPU core a few notches to help take some stress off the VRM.

 

[Ravenas]

@ThE_MaD_ShOt hes running a 9590 his chip uses twice the power as yours and runs at least 30% hotter read the thread ffs 60C @load from a 9570 using the AMD AIO is pretty dam acceptable I would almost say its perfectly fine given the chip and the cooling hes running
his vrm's are most likely cooking them selves to death from the load he needs more cooling on them
@op take the side panel off and point a box fan in there and see if that changes any thing

Wouldn't the motherboard temperature displayed give me a indication on how the VRMs are in regards to temperature?

 

[OneMoar]

Wouldn't the motherboard temperature displayed give me a indication on how the VRMs are in regards to temperature?

no that board doesn't have a VRM sensor that I am aware of

 

[Ravenas]

No, there might be a temp sensor in the board near them, but it won't be accurate. Best option would be to point a IR thermometer at them when the machine is under load. Aim for the grey boxes under the heatsink, not the heatsink itself.

You can also try to point a fan at the vrm/socket area and see if it helps with the crashing.

Though, IMO, your board just isn't built for the high TDP of the 9590, even if ASUS says it technically supports it. I'd try lowering the CPU core a few notches to help take some stress off the VRM.

It does the same thing if I lock the voltage at 1.4, 1.45, and 1.49. The motherboard won't even post if I give it 1.5. It just seems very unusual that the crash always occurs at 60 degrees C on the sensor, which makes me think its the CPU and not the VRMS. I will try a box fan.

 

[vega22]

so you got it free you say?

with any explanation for them wanting to off load it?

 

[Ravenas]

so you got it free you say?

with any explanation for them wanting to off load it?

It was brand new through my company rewards points.

 

[de.das.dude]

60C should not be a problem for an FX vishera...
the problem lies either with the motherboard or something else.

using the motherboard utility from OEM to measure is the best way to get accurate temps as no one can calliberate better than the manufacturer themselves.

have you made sure that CPU thermal throttling is set off in the BIOS? its set to 60C by default on a lot of motherboards. even the first settings other than auto on mine is 60C.


anyway, i have seen mine cross 60C a few times without problems. Also unless you are over volting it, VRMs should not fail that miserably.
even though it says 220W i am sure that is the MAX TDP of the processor beyond which the processor itself would probably melt.


my Fx 8320 does boot at 1.5V i managed to get 5GHz with 1.5V, not stable though, but it booted and was stable enough for me to take screenies and do a CPU-Z validation.


i suspect it is something to do with the motherboard..

 

[ThE_MaD_ShOt]

@ThE_MaD_ShOt hes running a 9590 his chip uses twice the power as yours and runs at least 30% hotter read the thread ffs 60C @load from a 9570 using the AMD AIO is pretty dam acceptable I would almost say its perfectly fine given the chip and the cooling hes running
his vrm's are most likely cooking them selves to death from the load he needs more cooling on them
@op take the side panel off and point a box fan in there and see if that changes any thing
and a word of warning even tho asus claims support for the 9570 that doesn't mean jackshit the 9570 is a pos with a huge swing in power consumption and heat output between any given sample so YMMV

But I am running mine oc'd so my temps and power usage has went up also over stock settings. And I am on a Giga 990fxa-Ud3. His ch-v formula z is supposed to be far superior to mine. Also isn't the 9590 just a binned version of my chip?

I would be running mine at higher clocks but I crunch 24/7 so I want to keep the temps down while keeping stability priority one. I don't need to erroring out wu's.

 

[Ravenas]

60C should not be a problem for an FX vishera...
the problem lies either with the motherboard or something else.

using the motherboard utility from OEM to measure is the best way to get accurate temps as no one can calliberate better than the manufacturer themselves.

have you made sure that CPU thermal throttling is set off in the BIOS? its set to 60C by default on a lot of motherboards. even the first settings other than auto on mine is 60C.


anyway, i have seen mine cross 60C a few times without problems. Also unless you are over volting it, VRMs should not fail that miserably.
even though it says 220W i am sure that is the MAX TDP of the processor beyond which the processor itself would probably melt.


my Fx 8320 does boot at 1.5V i managed to get 5GHz with 1.5V, not stable though, but it booted and was stable enough for me to take screenies and do a CPU-Z validation.


i suspect it is something to do with the motherboard..

Is the an acronym in the bios for the thermal throttling?

 

[Toothless]

But I am running mine oc'd so my temps and power usage has went up also over stock settings. And I am on a Giga 990fxa-Ud3. His ch-v formula z is supposed to be far superior to mine. Also isn't the 9590 just a binned version of my chip?

I would be running mine at higher clocks but I crunch 24/7 so I want to keep the temps down while keeping stability priority one. I don't need to erroring out wu's.

9590 is a super-duper overclocked/overvolted version of the 8350. All that heat and power consumption must be fun to play with.

 

[de.das.dude]

9590 is a super-duper overclocked/overvolted version of the 8350. All that heat and power consumption must be fun to play with.

dont think its factory overlclocked. its just cherry picked 8350s that can do a good clock.

Is the an acronym in the bios for the thermal throttling?

lolwut?