Question i5-4690K running at 99C-100C constantly

[The_Shadow_630]

I recently purchased a MB and CPU and once getting it installed and up and running, I have discovered that the CPU temp is reading in the mid 90s in BIOS and upper 90s to 100 on idle once booted. This is a used setup bought from a private owner. MB is asrock z97 pro4 and the processor installed is an i5-4690K. Cooler is a Corsair H60.

I have put my hand on the water block as well as the radiator and neither feel particularly hot at all. I removed the CPU cooler and have cleaned off the thermal material that was on there as it only appeared to be making contact in 2 spots along the edge of the CPU. I have ordered some new thermal paste and intend to apply it and reseat the cooler to see if that helps, but I figured I would check with you all and see if you thought there could be anything else.

I have not put an excess amount of load on it, but it is throttling pretty heavily. Oddly enough SpeedFan seems to detect the temp at much much lower, effectively normal, levels whereas ThrottleStop and other software detects all four cores running at about 96-99C almost all the time. I've known SpeedFan to be a little off at times, but never this much. Also, before you ask, there is no overclocking of any kind going on with this at the moment.

 

[Quanticriver]

things i would check

make sure that the h60 is seated properly
i would do a bios reset to default (optimized) values

If possible use another cooler for reference (e.g. intel stock cooler)

 

[The_Shadow_630]

Thanks, my first go to was absolutely checking to make sure the block was seated well to the processor, which is why I ordered new thermal grease so I could reseat it. The BIOS settings all look pretty normal to me, but i can certainly reset them just to be sure I am not missing some random setting several layers down,

I guess my two biggest worries were 1) The H60 is actually not working (I say this as there is no RPM reading from the CPU Fan, which is where the pump is plugged in) or 2) the temp sensor on the CPU is malfunctioning and making it think that it is running that hot. I unfortunately do not have another LGA 1150 cooler laying around to test. I wish I could use the cooler I had on my AM3+ board on this one, but I don't think there is a way to actually secure it on there with the bolt out pattern being so very different between them.

 

[GameableXP]

I tried to make sense of the original post as best I could, but it seems like you don't even have any thermal paste on the CPU. This is definitely the reason for this problem, the reason the thermal grease is on there is to transfer the heat from the CPU to the cooler

 

[The_Shadow_630]

I tried to make sense of the original post as best I could, but it seems like you don't even have any thermal paste on the CPU. This is definitely the reason for this problem, the reason the thermal grease is on there is to transfer the heat from the CPU to the cooler

Let me clarify. The temps I was reporting were when the cooler was installed with the original thermal material that came with the cooler still on it. I literally took the entire unit, MB and all, without removing the cooler from the CPU, and transferred it into my case.

As part of trying to troubleshoot the cause of the excessively high temps, I removed the cooler from the CPU and discovered, based on the remains of the thermal material on the CPU, that the cooler was only making real contact with the CPU in 2 places on the outer edge. I have new thermal paste on order that should arrive today so that I can apply it, as I have cleaned off the CPU and the cooler, to see if the cause was there not being a good thermal transfer due to the seemingly minimal contact between the CPU and the cooler.

This post was to see if anyone could think of or suggest other reasons that the CPU may be running so very hot other than a bad thermal connection between the cooler and the CPU. That way, if current steps I am taking now do not work, I have next step things to check out. Hope that helps clear it up a bit for you.

 

[EndEffeKt_24]

If there is enough room in your case I would take a step back from that very mediocre and possibly broken aio and get a good aircooler.

 

[RobCrezz]

It will be either the contact of the cooler (which you have already looked at), or the pump not getting enough power (or the pump just broken).

If the pump is powered by the motherboard, you need to make sure the header for it is set to run at 100% or 12v - some board will think its a fan and try and run it slower to reduce noise, but that can stop the pump running entirely.

 

[The_Shadow_630]

If there is enough room in your case I would take a step back from that very mediocre and possibly broken aio and get a good aircooler.

There is plenty of space, and I have long been an air cool guy. In this case, the closed loop water cooling system just came with the system, so I was going to try and use it just to see, If I can get it to work with reasonable temps, I may keep it at least for a bit until I select a good air cooler that I am happy with the look and performance of. The case has a tempered glass side panel, so I want it to look good and perform well.

It will be either the contact of the cooler (which you have already looked at), or the pump not getting enough power (or the pump just broken).

If the pump is powered by the motherboard, you need to make sure the header for it is set to run at 100% or 12v - some board will think its a fan and try and run it slower to reduce noise, but that can stop the pump running entirely.

Thanks, I will take another look at the BIOS settings for the CPU Fan ports. I went in and set it to Maximum thinking that there might not be enough juice going to the pump, but it did not make a difference. However, if there is not good enough thermal transfer, it would have made little difference anyway. I just cannot imagine that the previous owner of this thing would have been able to work with it at all with it throttling so hard all the time. At these temps it is throttling back to under 1GHz, which is like taking a step back in time to some of my first computers with physical turbo buttons on them.

 

[RobCrezz]

Yeah It could just be that the pump has died completely. But try that first, it might just not be getting enough power to run the pump.

 

[EndEffeKt_24]

Tape the turbo button down then.
Should fix the issue. At least it did back in the day.

 

[The_Shadow_630]

Tape the turbo button down then.
Should fix the issue. At least it did back in the day.

LOL, where is that sarcasm punctuation when you need it.

I had the benefit of mine actually clicking down and locking in place for a grand total of 133MHz of processing power.

I will hopefully be getting to reseating the cooler this weekend (been a crazy week). I will test and let you all know the results once I get that done. Otherwise, I think I am going to order a good air cooler and install it. I hate that you cannot really tell if the water pump on these aio's is running or not so you cannot easily see where the source of the problem might be.

 

[The_Shadow_630]

So, applied new thermal compound and remounted the cooler. Good news is, when I got into the BIOS to check temp, I actually saw the temp at about 67C, but then watched it steadily rise to 94C. Loaded defaults in the BIOS and restarted. Once fully booted, I started Real Temp to see the temps back up to the same point they were before. However, after running for a bit, a moment of hope came when I saw them drop down to the mid to low 70s range and the CPU was running at full turbo speed. It stayed there for a bit, then went back up and throttled again.

Honestly, that leads me to believe that the pump is working intermittently and the drop in temp was when it kicked on. As of this writing, temps are back up in the 95-99C range steadily. Have ordered an ARTIC cooler that should come in tomorrow and hopefully be able to get these temps under control. Will keep you all updated and thank you for the help.

 

[RobCrezz]

Yeah sounds like your pump is kaput.

 

[Rodrigodrt]

So, applied new thermal compound and remounted the cooler. Good news is, when I got into the BIOS to check temp, I actually saw the temp at about 67C, but then watched it steadily rise to 94C. Loaded defaults in the BIOS and restarted. Once fully booted, I started Real Temp to see the temps back up to the same point they were before. However, after running for a bit, a moment of hope came when I saw them drop down to the mid to low 70s range and the CPU was running at full turbo speed. It stayed there for a bit, then went back up and throttled again.

Honestly, that leads me to believe that the pump is working intermittently and the drop in temp was when it kicked on. As of this writing, temps are back up in the 95-99C range steadily. Have ordered an ARTIC cooler that should come in tomorrow and hopefully be able to get these temps under control. Will keep you all updated and thank you for the help.

shizzle i was gonna blame the pump, ive a 4770k and had the same problem, my pump failed and to make sure it was that, I plugged it to one of the CPU_PWM or SYS_PWM headers, then on bios if they indicated 0rpm, then, its as dead as those mofos trying to storm area 51 next month.

 

[The_Shadow_630]

Yeah, moved it from the CPU1FAN port to the case fan port and still got the same result. Hopefully getting in my good old air cooler tomorrow and getting it mounted on will allow this thing to actually run at least at it's base clock and might even allow me to OC it a bit. The cooler I ordered is pretty beefy, so it should be able to easily mitigate the heat generated by this thing. Also, it is going to be able to face sideways to vent right out the back of the case with the assist of the case fan right next to it. I usually pull in from bottom and front and out through top and back.

 

[The_Shadow_630]

Well, air cooler installed, and temps dropped WAY down as anticipated. BIOS temp down to about 37C. Complete boot up and did a little stress testing on it. All four cores set to run at max clock at all times and a quick 10 minutes stress test, max temp only briefly reached 72C on any core. Those are temps I can certainly live with. Now just looks like it is time to throw out that useless aio and start enjoying my new build.

 

[theonerm2]

If it's not the pump it's definitely that the cpu needs delidded. I'd say it's the pump. Those AIO's fail often.