AIO "incorrect" positioning according to GN

[tabascosauz]

I took a look around TPU, and didn't see any existing thread talking about GN's new video, so here it is:

The principles are simple enough to understand, though admittedly I am no watercooling aficionado. Seems easy enough for the vast majority of normal people with normal ATX mid-towers to find a suitable configuration that won't cause premature failure, poor performance, or excessive noise according to GN.

But this leaves no "acceptable solution" for a number of popular small form factor cases:

• SM570 and SM580 are okay.
• The Ghost S1 needs the largest top-hat size available to fit a top rad.
• SG05 and SG13 are probably okay due to the laid-flat design.
• FormD T1 can't, but luckily the entire case can be flipped on its head to render the bottom-mounted radiator "on top".
• M1 can't, and radiator-on-side-bracket mounting might be dubious as well, pump-block and rad are about level with each other, might depend on board's socket placement. Also, bottom rad placement is obviously "bad" according to GN. The only remaining acceptable placement would be a 92mm 645LT in the rear, but that's super weak.
• Ditto for the CM NR200, since it uses the M1 layout.
• Ditto for the Streacom DA2, since it uses the M1 layout.
• Ditto for the Cougar QBX, since it uses the M1 layout.
• Cerberus and Cerberus X can fit large front rads with the two rear-mounted PSU configs, but according to GN, those rads should never be mounted with the tubes/rad tanks at the top. Which makes for a potential headache with trying to snake the AIO tubes all the way around a very long/tall graphics card to the bottom front of the case. As for the side bracket, it might work better than in the M1 because sockets are generally located considerably lower on mATX and ATX boards.
• SM560 is a strong no-go according to GN, only fits rads in the bottom. Air cooling is not really viable for most full-TDP hardware either, so RIP.
• DAN-A4 won't work either, only fits rads in the bottom. Ditto on air cooling.

Anyways, just something to think about. Glad I don't have to deal with this in either my Cerberus or M1, but I know that a lot of SFF builders out there rely on AIOs.

I'm guessing custom loop builders in any of these cases would be just fine, since properly constructed loops would be bled of air anyways.

 

[Caring1]

It's been mentioned by a reviewer here that most AIOs don't have long enough hoses to wind down to the bottom front of a case, after I queried why they mounted them upside down with hoses at the top.

 

[Vya Domus]

I had a AIO mounted with the pump up for a short while because it wouldn't fit any other way in my old case and interestingly enough it's performance and noise was the same as it is now when it's mounted correctly.

 

[slightofhand]

Yep, it all seems so easy when watching Steve. Then reality sets in... are the hoses on my system long enough? For mine, no. So I mounted my radiator "upside down", and it's been working just fine the past 19 months... no problems, no strange noises.

 

[oxrufiioxo]

My corsair h150 was front mounted for 2 years with no issues with performance or pump noise at least above fans running at 600rpm.... I have it top mounted now and performance is actually slightly worse due to the GPU exhausting hot air into it..... although probably margin of error difference.

 

[Mussels]

steves orientation information is correct if you get any air bubbles in the loop - you may get lucky having it other ways, or you may end up like me with half a dozen RMA's cause your case made the bubbles go to the pump

 

[xtreemchaos]

very interesting, what i take from this is if you like watercooling build a custom loop. AIOs are dodgy .

 

[Vya Domus]

what i take from this is if you like watercooling build a custom loop.

You can position a pump in a custom loop incorrectly as well.

 

[xtreemchaos]

yes we can but not if we have morethan a short think about it, but with AIOs it looks as if we dont have super long pipes we have no choice.
i think its common knol that the pump gos at the lowest point in a custom loop.

 

[Waldorf]

problem is that no AIO is absolutely air free.
as soon as the rad is mounted top of case/with the hoses on top (rear/front mounted) , air will get trapped there,
especially for gpus, since they normally are below the rad.

can happen with any loop if you dont have a res or dont properly remove air bubbles form it.
part of the reason i like to run any loop (aio/custom or mix) outside the case and can have the rad below the res,
but also shake it like stupid until all air was flushed from the rad.
THE it doesnt matter if the hoses/pump etc are mounted facing up/higher than other things.

 

[silkstone]

I've had one of my AIOs for about 8 years, (CM Seidon). I initially had it the wrong way, rear mounted with hoses at the top, but lower than the CPU. After a year, I'd switched it around with the hoses at the bottom as it just made more sense. Though the pump speed has gone down over 8 years, I'm quite impressed by the longevity of it!

 

[de.das.dude]

steves orientation information is correct if you get any air bubbles in the loop - you may get lucky having it other ways, or you may end up like me with half a dozen RMA's cause your case made the bubbles go to the pump

from what i saw on the video.

The only course of action is to run the pump separately with the radiator high up, pump low. And shake the living hell out of the pump. He did say antec said start stops help bubble flow

 

[ThrashZone]

Hi,
Manufactures problem 5 year warranty nonissue.

 

[Bartonion]

Yep, it all seems so easy when watching Steve. Then reality sets in... are the hoses on my system long enough? For mine, no. So I mounted my radiator "upside down", and it's been working just fine the past 19 months... no problems, no strange noises.

 

[mushmx]

interesting video, i always use the AIOS on the top, just because thermodynamic, hot air tends go up.
but with this video i understand im wrong thinking that, i mean its ok that systems go up, but not forthe reason ive think.
Steve make things easy

 

[elghinnarisa]

Funny enough, as I was reading this I figured I would see if it would even work in my case with my H115i. I had mine front mounted with the hoses at the top for years.

I removed the screws, dusted it off a bit. Propped it up inside the case as a temporary test to see if it fits and works etc. with the hoses at the bottom.
Turn the PC on, roughly 50 seconds later I hear a... sizzle? Like water in a hot pan. And all of the sudden my pump RPM went from 3000 RPM (max) to 6800RPM, and CPU temp went up to 99C in less than a minute.
Now it sounds like a rusty chainsaw.

Sooo there went that H115i.

 

[Waldorf]

@mushmx
contrary to you and certain others in this forum:
that hot air goes "up" is completely irrelevant in a "forced" airflow system (read: pc case with at least one fan blowing air out the case).

even if i take my ultra low rpm silent fans doing 800rpm max, and turn them down to around 200, where they barely move air.
any smoke going thru the case is not "rising", and airflow follows in/out of case with fan setup...

try it yourself:
colored smoke

@elghinnarisa
most likely because the trapped air went straight to the pump, making it run dry..
time for a proper aio like the eisbaer 280

 

[CrAsHnBuRnXp]

Yep, it all seems so easy when watching Steve. Then reality sets in... are the hoses on my system long enough? For mine, no. So I mounted my radiator "upside down", and it's been working just fine the past 19 months... no problems, no strange noises.

My OCD wont allow me to have the logo's upside down from how they are printed on the RAD let alone the pump on the CPU.

 

[Caring1]

Funny enough, as I was reading this I figured I would see if it would even work in my case with my H115i. I had mine front mounted with the hoses at the top for years.

I removed the screws, dusted it off a bit. Propped it up inside the case as a temporary test to see if it fits and works etc. with the hoses at the bottom.
Turn the PC on, roughly 50 seconds later I hear a... sizzle? Like water in a hot pan. And all of the sudden my pump RPM went from 3000 RPM (max) to 6800RPM, and CPU temp went up to 99C in less than a minute.
Now it sounds like a rusty chainsaw.

Sooo there went that H115i.

@elghinnarisa
most likely because the trapped air went straight to the pump, making it run dry..
time for a proper aio like the eisbaer 280

Yep, the system would have lost fluid content over the years you had it, if it was installed correctly from the beginning the air if there was any would have been trapped at the top of the radiator and not harmed the pump.

 

[elghinnarisa]

Yep, the system would have lost fluid content over the years you had it, if it was installed correctly from the beginning the air if there was any would have been trapped at the top of the radiator and not harmed the pump.

Indeed, though it was running (and cooling) just fine up until I unscrewed it and started to move it around.
I did also anwser the question that it was not possible to mount it the proper way, the hoses are too short. It would have to be top mounted in that case, which could work.

Edit: Corsair offered me a H115i pro RGB as a replacement, too bad I already got myself a replacement.

 

[Waldorf]

Thats where you want it anyway, dumping the heat outside the case,
lowering temps for gpu/chipset etc

 

[CrAsHnBuRnXp]

All you need to remember is that top mounting the rad is the BEST outcome. If you have to place it in the front of your case, just make sure the coldplate/pump is below the inlet and outlet holes of the rad and you will be fine. If they are above the inlet and outlet holes, youre going to get air in the pump.

 

[Caring1]

If you have to place it in the front of your case, just make sure the coldplate/pump is below the inlet and outlet holes of the rad and you will be fine. If they are above the inlet and outlet holes, youre going to get air in the pump.

Yeah nah, I'm going to have to disagree with you there.
You will only get air in the pump that way IF the radiator is mounted incorrectly with the hoses at the top.

 

[CrAsHnBuRnXp]

Yeah nah, I'm going to have to disagree with you there.
You will only get air in the pump that way IF the radiator is mounted incorrectly with the hoses at the top.

That is false and I can prove it to you.

 

[Mussels]

The whole issue only matters for loops with air in them, you can likely jiggle things around and get the most special of layouts working just fine