Stay away from silvertsone all-in-ones

[KenjiUmino]

7 hours ago, FloRolf said:

you DO know how a watercooler works though, do you?

i do know that the water temperature should equalize throughout ALL the parts, that means temp at the cpu block should be the same as the temp at the radiator, give or take a degree or two - thit is how it should be, right? 

 

but when the tube from cpu block to rad is HOT while the rad and the tube that comes back from the rad isn't even WARM but the CPU is not getting cooled at all cooking away at over 75 celsius doing nothing then something clearly is off and the heat can't spread like it's supposed to.

 

in systems where the cooled components have a much much greater heat output than a CPU, like a car engine, the coolant comes into the radiator way hotter and a massive drop in collant temperature at the radiator is normal but on a cpu the temp difference between inlet and outlet should not be as extreme as it was with the silverstone AIO.

 

this is what i learned so far - am i missing something ?

 

the thing is also that this temperature difference between tubes wasn't there at all when the AIO was new and it was able to cool the CPU with absolutely no problem.

 

but it went from "absolutely no problem even under full prime load" to "already overheating when the computer sits idle in the bios only showing the hardware monitor page" 

[flibberdipper]

32 minutes ago, KenjiUmino said:

i do know that the water temperature should equalize throughout ALL the parts, that means temp at the cpu block should be the same as the temp at the radiator, give or take a degree or two - thit is how it should be, right? 

That's not how it works at all. One tub HAS to be hotter because otherwise you'd have literally zero thermal transfer.

[KenjiUmino]

27 minutes ago, tmcclelland455 said:

That's not how it works at all. One tub HAS to be hotter because otherwise you'd have literally zero thermal transfer.

then what is all that talk about how order of components does not matter because the temperature is equal across the whole loop ?

 

if it would be "normal" for one tube to get hot AF while the other stays cold the order of components WOULD matter because you would not want the already hot water from one cooling block going into another cooling block. 

 

you would literaly need to go rad -> block -> rad -> block

 

but every watercooling guy on youtube keeps telling me that order of components does not matter ... how come? 

[xHadrian]

8 minutes ago, KenjiUmino said:

then what is all that talk about how order of components does not matter because the temperature is equal across the whole loop ?

 

if it would be "normal" for one tube to get hot AF while the other stays cold the order of components WOULD matter because you would not want the already hot water from one cooling block going into another cooling block. 

 

you would literaly need to go rad -> block -> rad -> block

 

but every watercooling guy on youtube keeps telling me that order of components does not matter ... how come? 

Having one tube hot while the other is cold is a sign that your pump is dead. Nothing more. I doubt you could even notice the difference between the "hotter" tube and the colder one with an ideally working AIO.

[KenjiUmino]

Just now, xHadrian said:

Having one tube hot while the other is cold is a sign that your pump is dead. Nothing more. I doubt you could even notice the difference between the "hotter" tube and the colder one with an ideally working AIO.

that is EXACTLY what i was trying to explain ... the temp difference between components should not be more than a degree or two ... not enough to notice a difference by simply touching the tubing

[For Science!]

19 minutes ago, KenjiUmino said:

-snip-

56 minutes ago, tmcclelland455 said:

-snip-

 

I would say OP is correct, usually in AIOs if one tube is significantly hotter than the other, this is a strong indicator of a dead/dying pump. Asetek pumps struggle a lot once air managed to go into the inlet and can not overcome the airlock to re-hydrate itself (afterall, it is an water pump and not an air pump). So if through evaporation, enough air gets into the pump, then it will no longer function until you overcome the airlock.

 

Given that your standard 240 mm radiators (slim, like AIOs) hold about 120 ml of liquid, and if we assume the flow rate to by 0.1 gpm (400 ml per minute) for the Asetek pump, then you can calculate that any given segment of water will go through the whole loop every 20 or so seconds. And so in an functional AIO the temperature of the whole water system gradually increases until the heat output from the CPU matches the dissipation efficiency of the radiator + fan.

 

As OP then correctly says, this is why loop order does not have a significant impact on component temperatures.

 

 

[For Science!]

1 hour ago, tmcclelland455 said:

-snip-

This is a very nice visual representation of the above statement, you can see the 2 tubes coming off the AIO develop a yellow glow at equivalent speed, meaning both the water coming in and out of the CPU block is gradually heating up and that the water temps are equilibrium through the system very quickly.

[flibberdipper]

1 hour ago, For Science! said:

This is a very nice visual representation of the above statement, you can see the 2 tubes coming off the AIO develop a yellow glow at equivalent speed, meaning both the water coming in and out of the CPU block is gradually heating up and that the water temps are equilibrium through the system very quickly.

Sure as tits isn't the case with any AIO I've come in contact with (which is a pretty small sample size all consisting of Asetek designs). Almost always one tube is a bit warmer than the other. But I'd imagine that'd have more to to with the pump just not moving liquid fast enough + massive heat dumps to actually make it hit an actual equilibrium in a reasonable amount of time and maybe the way shtuff is oriented (I think Steve from GN touched on it with one of the recent Threadripper vids when it came to the Thermaltake AIO and coverage of the IHS).