• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Cpu low usage but high temp

My third gen Asetek cooler dried out in two years
A sample-size-of-one anecdotal experience does not render moot the whole point. Until Man can create perfection 100% of the time, there will always be exceptions and units that fail prematurely.

There are many of AiO products that have already provided many years of service and still going strong. While I admit this is also anecdotal, I personally have not come across a factory made AiO cooler, that was not obviously damaged, that had leaked coolant. I have, however, seen several where the impellor bearings had seized or were grinding suggesting seizing up was not too far off.
 
A sample-size-of-one anecdotal experience does not render moot the whole point. Until Man can create perfection 100% of the time, there will always be exceptions and units that fail prematurely.

There are many of AiO products that have already provided many years of service and still going strong. While I admit this is also anecdotal, I personally have not come across a factory made AiO cooler, that was not obviously damaged, that had leaked coolant. I have, however, seen several where the impellor bearings had seized or were grinding suggesting seizing up was not too far off.
To be certain, when I said “dried out,” I meant evaporated, not leaked. I’ve seen many CLC units do this, just typically not as quickly as in my use case.

All I mean is that the OP should remove the unit and check for sloshing, particularly if there are no funny pump noises and the mount isn’t in question. :oops:
 
To be certain, when I said “dried out,” I meant evaporated, not leaked.
I understood that. But drying out via evaporation is still due to leakage. It just indicates it is leaking so slowly, the coolant has time to evaporate before it has time to drip and pool at the bottom of the case, or spew all over the motherboard.

And sure, it does not hurt to make sure the system is not full of air. Too bad more (all) of those systems don't use clear tubing.
 
I first have a question ... you said the images were provided so we could see what you are experiencing ... 90 / 100 C but the images don' reflect that condition. I would sugges you d/l and run HWiNFO, run "sensors only", then do whatever it is that creates the condition. After closing the game .... look at HWiNFO screen and take a screenie if the core temps. speeds and voltages

If people are honest ... there's two reasons to use an CLC type AIO:

1. You just don't like the aesthetics of an air cooler.

2. You get to tell your friends you have water cooling.

The don't cool significantly better than moderately priced air coolers, more often they do not, and are significantly louder

Yes there are a small amount of AIOs that can cool a CPU to a lower temperature than an air cooler. At the top are the OLC type with copper rads and the 280 / 360mm jobs, but the few that do require extreme speed fans to do so .... Id love to see them compared with air coolers w/ 2000 rpm fans that these AIOs need. How do they compare with both at a non-ear rattling 1200 rpm ? How do they both compare with 2000 rpm fans ? There is just 1 AIO (EK 360 Phoenix w/ copper rads and flow 10x that of a typical CLC) tested by TPU OC (AIDA 64 test used for comparison) that beats the best air coolers by more than 2C. The $175 H150 pro w/ 2000 rpm fans ties the $55 Scythe Fuma 2 w/ 1200 rpm fans which is 3 dbA quieter. The H100i Pro beats the air cooler by 1 C but is 13 dbA (2.5 times as) loud.

If you want water cooling but don't want to tackle the assembly of a custom loop, you are best served with an OLC type of AIO with copper rads which provides some visual indication that water is flowing. The Swiftech H240 X3 / H360 X3 have copper rads and a reservoir which is a visual indicator that the pump is working. In addition, none of the problems associated with galvanic corrosion that you get with CLC type units. And while it doesn't address the "is my pump working" problem (snapped impeller shaft, air lock ) you can add a GFX card block or extra rads if ya like.

You can also address air locking quite easily with the swiftech units. If installed vertically in the front of the case, the radiator side port will be on top and makes a great bleeder valve. On builds we do we will install a G-1/4 valve and use it as a bleeder / fill valve here and periodicaly you can just let air escape and top off w/ coolant of add corrosion inhibitors, algaecides. If installed on top of the case, tilt the PC back rear side down resting the top rear corner on 1 or more texbooks enough so that you can leave power plugged in. I just use small $15 12 V PSU that looks like a laptop brick and plug in the pump only. Let it run for an hour and than bleed and top off if necessary.

I have done this tilt thing twice with CLC type coolers ... it did work but, solution was only temporary .... the problem returned and the users replaced the CLCs. An yes, I have seen AIO failures ... (pump shaft, pump fried (dry) ... My son built a box with a Swiftech OLC; it failed and was replaced under warranty after 4+ years but he wanted to go all out RQB in his next build, set the Swiftech replacement aside and used a Corsair CLC to have "consistent RGB control". Six months later, "RGB has gotten old" and the additional noise is bother some ... I have the feeling next time I visit, I might see the Swiftech back in

So what can you do ?

1. Have you installed the fans in accordance with Swiftech's instructions ? ... For best performance rad fans must blow in. They call it "The law of Thermodynamics" for a reason. Contrary to what we learned in 8th grade about hot air rising ... the earth doesn't have a giant fan in the sky. This won't make huge difference but if your ambient is 23C and your internal case air is 28C which do you think will cool better ? With cooant at 33C, 23air (Delta T = 10) takes twice the heat away from your CPU as 28C air (Delta T = 5) ... again, thermodynamics, it's a law, The most common reaction is that the interior case temps will rise .... for the sake of argument lets say it does ... so what ?

a) Proper case design means that you air is turning over between 1 and 2 times a second.
b) What component are you worried about that are in the exhaust path way ? Do the usage requirements of those components state problems in environments between 20 and 30 C ?
c) Your CPU is at 90 - 100C ... are any other components in this "worry zone"

2. Where is your case ? Under a desk rear against wall ? You rear case grilles and vented slot covers are important part of your cases coolomg design. If you have 2 fans on top of your case blowing out and a ear blowing out and 3 in front blowing in are you "balanced ? Not f you have intake air filters, especially ones clogged with dust, a secondary effect of installing AIO fane contrary to Corsair's written instructions is that you have more air blowing out than in ... This creates negative case pressure, and while dust intake is a negative component of this, what is more worrisome is the temperature of the air bringing it in. With case rear backed up against a wall ... your 600 watt PSI and 275 watt GFX card are exhausting heat into the space which that make up air is coming in thru the rear case grilles and slot covers. You can test for air recyling wih fake smoke, the kind that installers use to test smoke detectors ... aka '"smoke in a can" ($12 on Amazon) .... blow some behind the case in the area of the case grille / slot covers and if it blows away from the case you fine ... if you see smoke inside, you exhaust heat is being recycled.

3, Check software, use a utility or check BIOS for pump speed indication .... be aware that a sheared pump shaft will show motor spinning but pump may not.

4. What is case ventilation ? Air turnover ? Take your fan's advertised CFM and multiple by 50% to get something close to the real number. Multiply by the number of fans ....

4 fans x 50% x cfm = say 120 cfm

Now look up your case volume ... say 1 .2 CF

That would mean you turning case air over 1.67 times a second

5. Even core temps indicate a uniform mount but that doesn't **always** mean proper. No TIM could result in even temps as could not taking the protective plastic off the cooler before installing. If you did the mount, we can rule these out.
 
Back
Top