A lot of heat air comes out from the top of my 3080 FTW3, directly hit my side case cover
I am a little bit worry the cover cannot hold a long time direct heat air 

irakandjii, I do not have the answers you want, obviously, but I would love to get your opinions on my case setup.  I have an MSI Sekira 500x (recently bought it, just for a change).  I have some BASIC airflow knowledge, mostly from reading about it online.  Anyway.  The case has 2, 200 MM fans in front as intake, 2 200MM Fans on top as that are now exhaust (they ship the case with them as intake which I thought was REALLY stupid so I flipped them. 1, 120 MM fan as exhaust on the rear.
 
I also have an NZXT Kraken x73 AIO liquid CPU cooler mounted horizontally at the top, under the 200MM exhaust fans in the radiator bracket.  I set the radiator fans as exhaust also so basically the radiator is sandwiched between the two sets of fans both exhausting air (2, top 200mm fans and the 3 120s that the Kraken came with.  
 
I originally, by mistake, had the Kraken fans as intake, and quickly realized my mistake and fixed it when the CPU started going over 45 at idle.  :)  Anyway, if I stress test for an hour, my 9900K maxes out at about 72C, and my 2080 Ti maxes out about 75C both respectable temps for a 100% load.  Under normal conditions while gaming, I get average CPU temps of 45C and average GPU temps of 57C.  I really have NO ISSUES right now, but I am planning on getting whatever 3080 I can get my grubby little paws on first.  I was wondering if the way I have the airflow setup now is as optimal as I can get it?  I don't want to mod the case, at all, or add anything that would take away from the aesthetics, it's a pretty setup.  Any thoughts?  I figures since heat rises top exhaust is always the best, the front intake seems optimal to me because they are large fans and there is a clear airflow path to the mother board (memory and VRMs) and the GPU as well.  It's a Strix 2080 Ti OC model, HUGE heatsink and 3 fans pointing down that rarely speed up much.
 
I was thinking about the rear exhaust fan, from what I have read a SMALLER fan is better back there for air pressure.  It's the only 120MM, but there is room for a 140 I believe.  I can post some pics of the system if that would help.  The case has hinged glass panels on both sides, and I can sort of hear and feel the change in pressure with I close them if that makes sense?  I tried running the system with the sides and top removed and it actually goes UP a few degrees on average (3-4C maybe). I tried with JUST the top, and just the sides removed and there's really no difference from running it full assembled (maybe 1C variation which isn't really anything to analyze). That would indicate a decent airflow patter/setup, right?

jilljill2002
A lot of heat air comes out from the top of my 3080 FTW3, directly hit my side case cover
I am a little bit worry the cover cannot hold a long time direct heat air 

Unless the side of your case is made of extremely cheap plastic or acrylic, you have nothing to worry about.  No metal or glass typically used in cases is going to "melt" or get damaged from whatever heat a GPU can put out, keep in mind the 3080 will thermally throttle in the 85-90C range at worst, and the 3080 FTW3 Ultra won't hit anywhere near that unless your case's airflow is atrocious (for reference I got my card today, my case has insanity-level airflow, and I only saw 63C peak load under 3DMark testing so far, with both vRAM & VRM temps on the iCX readouts being even lower than that, and that's with the fans maxing out at 75% so not even near full speed).
 
In case you're wondering what "insanity level" (in my opinion) airflow looks like:
 

Dabadger84

jilljill2002
A lot of heat air comes out from the top of my 3080 FTW3, directly hit my side case cover
I am a little bit worry the cover cannot hold a long time direct heat air 

Unless the side of your case is made of extremely cheap plastic or acrylic, you have nothing to worry about.  No metal or glass typically used in cases is going to "melt" or get damaged from whatever heat a GPU can put out, keep in mind the 3080 will thermally throttle in the 85-90C range at worst, and the 3080 FTW3 Ultra won't hit anywhere near that unless your case's airflow is atrocious (for reference I got my card today, my case has insanity-level airflow, and I only saw 63C peak load under 3DMark testing so far, with both vRAM & VRM temps on the iCX readouts being even lower than that, and that's with the fans maxing out at 75% so not even near full speed).
 
In case you're wondering what "insanity level" (in my opinion) airflow looks like:
 

Holy #%$@# !!! lol, that is insane, what case it that?  Too much of a closeup picture for me to tell.  It's cool looking though.

njbongo
irakandjii, I do not have the answers you want, obviously, but I would love to get your opinions on my case setup.  I have an MSI Sekira 500x (recently bought it, just for a change).  I have some BASIC airflow knowledge, mostly from reading about it online.  Anyway.  The case has 2, 200 MM fans in front as intake, 2 200MM Fans on top as that are now exhaust (they ship the case with them as intake which I thought was REALLY stupid so I flipped them. 1, 120 MM fan as exhaust on the rear.
 
You did the correct thing here, using 4 200 mm fans to push air into the case with 120 mm fan pulling it out is just really, really BAD engineering.
The biggest advantage of 200mm fans is reduced noise through reduced rpm, that being said, the best rule of thumb I could find for a 200mm fan was 110 CFM @ 700 rpm. So you have about 220 cfm of airflow through the case.  (less case specific losses which I cannot determine in detail).  The 120 mm fan will have some impact, by forcing air out of the case it will reduce static pressure inside the case marginally and offset some airflow resistance to your main airflow.
 
I also have an NZXT Kraken x73 AIO liquid CPU cooler mounted horizontally at the top, under the 200MM exhaust fans in the radiator bracket.  I set the radiator fans as exhaust also so basically the radiator is sandwiched between the two sets of fans both exhausting air (2, top 200mm fans and the 3 120s that the Kraken came with.  
 
Usually, and a in a perfect world, I would try to set this up as a  "push / pull".  Use the Kraken mounted fans to push the air upward through the radiator and the 200 mm to draw the airflow away from the radiators and out of the case.  BUT!!!  it is unlikely to make a big difference.  The Kraken should be properly engineered with the appropriate fans (high flow or high static pressure) so changing the direction might actually reduce their effectiveness.
Bottom line, I suspect what you have done is very close to optimal here and any modifications would be very close to diminishing returns.
 
I originally, by mistake, had the Kraken fans as intake, and quickly realized my mistake and fixed it when the CPU started going over 45 at idle.  :)  Anyway, if I stress test for an hour, my 9900K maxes out at about 72C, and my 2080 Ti maxes out about 75C both respectable temps for a 100% load.  Under normal conditions while gaming, I get average CPU temps of 45C and average GPU temps of 67C.  I really have NO ISSUES right now, but I am planning on getting whatever 3080 I can get my grubby little paws on first.  I was wondering if the way I have the airflow setup now is as optimal as I can get it?  I don't want to mod the case, at all, or add anything that would take away from the aesthetics, it's a pretty setup.  Any thoughts?  I figures since heat rises top exhaust is always the best, the front intake seems optimal to me because they are large fans and there is a clear airflow path to the mother board (memory and VRMs) and the GPU as well.  It's a Strix 2080 Ti OC model, HUGE heatsink and 3 fans pointing down that rarely speed up much.
 
I suspect you will be fine, the Strix is a well engineered card.  The 3080 may demand some more air flow, but with the "guesstimated data" I have now, the air flow through your case sounds good.  If you have a problem the first thing I would do would be to add a 120 or 140 mm fan to push more air into the case and support the 120 mm fan you have in push pull.
 
I was thinking about the rear exhaust fan, from what I have read a SMALLER fan is better back there for air pressure.  It's the only 120MM, but there is room for a 140 I believe.  I can post some pics of the system if that would help.  The case has hinged glass panels on both sides, and I can sort of hear and feel the change in pressure with I close them if that makes sense?  I tried running the system with the sides and top removed and it actually goes UP a few degrees on average (3-4C maybe). I tried with JUST the top, and just the sides removed and there's really no difference from running it full assembled (maybe 1C variation which isn't really anything to analyze). That would indicate a decent airflow patter/setup, right?
 
There are large percentage increases in flow rate as you initially move up the "fan size" curve.  This is because of the hub in the center.  It is the blades that do the work, and in a shrouded fan, the outer edges that do the most work.  So as a simplfied example:
Hub Area (40 mm hub) =  pi * r squared  = ~1250 mm2
We need to subtract this from the total area of the fans since air cannot flow through the hub :)
So, by analysis:

• Working Area 120 mm fan = 11,300 mm2 - 1250 mm2 = ~ 10,000 mm2
• Working Area 140 mm fan = 15,400 mm2 - 1250 mm2 = ~ 14,150 mm2
• The 140 mm fan will move approximately 141% more air per turn.  
• The 200 mm fan will move 300% or 3 times the amount of air per turn ( I leave it you you to do the math and prove it)

What this means :

• You can stay at the same noise level -> rpm and move more air
• or,
• You can reduce the noise level (reduce rpm) and stay at the same amount of air.

This is the real purpose of larger fans.
(Note: noise is strictly driven by the rpm, the higher the rpm the higher the noise)

I trust this helps.
 
Give a man a fish and feed him for a day, teach a man to fish and feed him for a lifetime.

njbongo
 I have an MSI Sekira 500x (recently bought it, just for a change). 

Garbage airflow, why did you buy it?

njbongo

kevinc313

njbongo
 I have an MSI Sekira 500x (recently bought it, just for a change). 

Garbage airflow, why did you buy it?

Because, that's not true, and I felt like it.
   

It's literally a glass box with tiny mesh side vents.

Airflow is one of the issues in any case
 
You need to remove the heated air from around the cooler, before it gets recycled into the cooler - this causes a heat loop

irakandjii

njbongo
irakandjii, I do not have the answers you want, obviously, but I would love to get your opinions on my case setup.  I have an MSI Sekira 500x (recently bought it, just for a change).  I have some BASIC airflow knowledge, mostly from reading about it online.  Anyway.  The case has 2, 200 MM fans in front as intake, 2 200MM Fans on top as that are now exhaust (they ship the case with them as intake which I thought was REALLY stupid so I flipped them. 1, 120 MM fan as exhaust on the rear.
 
You did the correct thing here, using 4 200 mm fans to push air into the case with 120 mm fan pulling it out is just really, really BAD engineering.
The biggest advantage of 200mm fans is reduced noise through reduced rpm, that being said, the best rule of thumb I could find for a 200mm fan was 110 CFM @ 700 rpm. So you have about 220 cfm of airflow through the case.  (less case specific losses which I cannot determine in detail).  The 120 mm fan will have some impact, by forcing air out of the case it will reduce static pressure inside the case marginally and offset some airflow resistance to your main airflow.
 
I also have an NZXT Kraken x73 AIO liquid CPU cooler mounted horizontally at the top, under the 200MM exhaust fans in the radiator bracket.  I set the radiator fans as exhaust also so basically the radiator is sandwiched between the two sets of fans both exhausting air (2, top 200mm fans and the 3 120s that the Kraken came with.  
 
Usually, and a in a perfect world, I would try to set this up as a  "push / pull".  Use the Kraken mounted fans to push the air upward through the radiator and the 200 mm to draw the airflow away from the radiators and out of the case.  BUT!!!  it is unlikely to make a big difference.  The Kraken should be properly engineered with the appropriate fans (high flow or high static pressure) so changing the direction might actually reduce their effectiveness.
Bottom line, I suspect what you have done is very close to optimal here and any modifications would be very close to diminishing returns.
 
I originally, by mistake, had the Kraken fans as intake, and quickly realized my mistake and fixed it when the CPU started going over 45 at idle.  :)  Anyway, if I stress test for an hour, my 9900K maxes out at about 72C, and my 2080 Ti maxes out about 75C both respectable temps for a 100% load.  Under normal conditions while gaming, I get average CPU temps of 45C and average GPU temps of 67C.  I really have NO ISSUES right now, but I am planning on getting whatever 3080 I can get my grubby little paws on first.  I was wondering if the way I have the airflow setup now is as optimal as I can get it?  I don't want to mod the case, at all, or add anything that would take away from the aesthetics, it's a pretty setup.  Any thoughts?  I figures since heat rises top exhaust is always the best, the front intake seems optimal to me because they are large fans and there is a clear airflow path to the mother board (memory and VRMs) and the GPU as well.  It's a Strix 2080 Ti OC model, HUGE heatsink and 3 fans pointing down that rarely speed up much.
 
I suspect you will be fine, the Strix is a well engineered card.  The 3080 may demand some more air flow, but with the "guesstimated data" I have now, the air flow through your case sounds good.  If you have a problem the first thing I would do would be to add a 120 or 140 mm fan to push more air into the case and support the 120 mm fan you have in push pull.
 
I was thinking about the rear exhaust fan, from what I have read a SMALLER fan is better back there for air pressure.  It's the only 120MM, but there is room for a 140 I believe.  I can post some pics of the system if that would help.  The case has hinged glass panels on both sides, and I can sort of hear and feel the change in pressure with I close them if that makes sense?  I tried running the system with the sides and top removed and it actually goes UP a few degrees on average (3-4C maybe). I tried with JUST the top, and just the sides removed and there's really no difference from running it full assembled (maybe 1C variation which isn't really anything to analyze). That would indicate a decent airflow patter/setup, right?
 
There are large percentage increases in flow rate as you initially move up the "fan size" curve.  This is because of the hub in the center.  It is the blades that do the work, and in a shrouded fan, the outer edges that do the most work.  So as a simplfied example:
Hub Area (40 mm hub) =  pi * r squared  = ~1250 mm2
We need to subtract this from the total area of the fans since air cannot flow through the hub :)
So, by analysis:

• Working Area 120 mm fan = 11,300 mm2 - 1250 mm2 = ~ 10,000 mm2
• Working Area 140 mm fan = 15,400 mm2 - 1250 mm2 = ~ 14,150 mm2
• The 140 mm fan will move approximately 141% more air per turn.  
• The 200 mm fan will move 300% or 3 times the amount of air per turn ( I leave it you you to do the math and prove it)

What this means :

• You can stay at the same noise level -> rpm and move more air
• or,
• You can reduce the noise level (reduce rpm) and stay at the same amount of air.

This is the real purpose of larger fans.
(Note: noise is strictly driven by the rpm, the higher the rpm the higher the noise)

I trust this helps.
 
Give a man a fish and feed him for a day, teach a man to fish and feed him for a lifetime.

irakandjii, THANKS !!!!  There's some confusion.  First, the rear exhaust is 140 not 120, my bad.  Also, the radiator does have a push/pull config already.  I flipped the top case fans to exhaust. mounted the radiator under them, and the radiator fans at the bottom also se to exhaust to the air is all moving in the same direction.
 
You are right about the ASUS 2080 ti, it IS designed really well.  But, with the RTX 3080 shortages, and EVGA's que system my chances of getting an EVGA card are obviously WAY better.  I read today that out of all the stores that took preorders, all the ASUS models had the highest numbers, it does not surprise me.  This is the one I had https://www.asus.com/Graphics-Cards/ROG-STRIX-RTX2080TI-O11G-GAMING/HelpDesk_Download/ I just sold it a few days ago for a ridiculous price because I MIGHT upgrade to the new AMD 5900x after I see third party benchmarks.  I'm currently using my old ZOTAC 1070 AMP EXTREME until I get a 3080.
 
Anyway, thanks for the advice again!  I guess at this point I will wait and see how the 3080 performs, but like you, I am interested to see if I get crazy high temps like all the people on the forum are posting.  I hope not, because I don't want to have to RMA or sell the card to get a different brand......who knows when I would get another in either option.

njbongo

kevinc313



You are not getting good results.
 
72C on a 9900K with a 360mm AIO with push pull fans is bad.
 
75C on a 2080 Ti is bad.

In what universe?  Those temps you took from my post are MAX temps during a 1 hour stress test at 100% load on BOTH, not average.  I also did not play with any fan curves for the case fans, NXZT Kraken or the GPU fans, and left everything at stock settings to see how everything performed.  The actual temps while gaming are much lower.
 
I didn't post my average temps for "normal use", which right now are CPU 31C and GPU 35C which is now my old 1070 since I sold the 2080 ti.  But that card was within -/+ 2 or 3 degrees at idle.
Not everyone wants a jet engine turbine on their desk, and compared to reviews of other cases, those temps are more than acceptable.  
 
I'm not looking for the lowest temps possible, I'm making sure NOTHING is getting throttled when playing games, and it isn't.
 

You didn't cite ANY sort of app or load or power level or OC.  So yeah, those temps are bad.

I would think that the computer would overheat long before any quality tempered glass case would have any trouble.

 
Sigh.
 
100% *utilization* is meaningless when attempting to describe cooling capability. On a 9900K, 100% utilization can be anywhere from under 100w to over 200w, on stock settings. More with an OC. On a 2080 Ti, 100% utilization can be under 260w with a light load that's hitting voltage/frequency limit with stock settings, or up to ~375w-380w with a power limited heavy load with a Galax or FTW3 bios OC'd.
 
Watts -----> Temps.
 
I'm currently running CPU-Z stress test which is a light 117W load (100% utilization) on my stock 9900K, while simultaneously running Unigine Heaven at stock 260W TDP on 2080 Ti (power limited, 85-100% utilization). 9900K package at 68C after a half hour.........with my TINY 120MM Evga CLC120 AIO. My GPU is at a luke warm 49C with a hybrid kit. I could run this for a day and my temps wouldn't budge. Everyday fan settings.
 
Oh.....and my case is a 4U rack case with a 1814 in3 volume, less than half your case's ~4000 in3 volume. With four 120mm fans and two 80mm fans.
 
So yeah, your temps are bad and you should feel bad.
 
Please go run Prime95 with an AVX load and report back with power and temp readings. Good luck.


 
 

Actually, they aren't, I probably made a typo in my original post, and did not even catch it until now. My max temp under load is 62, not 72. Really wasn't relevant to my question for irakandjii since it was about case fan placement and direction, which he answered thoroughly. The screen shot is from Prime95, w/AVX and you can see the timestamp. And before you bother commenting, the 97% CPU load in the screen shot is because I clicked on the screen grab icon on the task bar to tale take the picture. It was running at 100% up until then for about 30 mins.  

I can't wait for the next issue I "should feel bad" about.
 
https://drive.google.com/...NPfvoJzTcVXFGZyst/view
 
 

kevinc313

njbongo
Actually, they aren't, I probably made a typo in my original post, and did not even catch it until now. My max temp under load is 62, not 72. Really wasn't relevant to my question for irakandjii since it was about case fan placement and direction, which he answered thoroughly. The screen shot is from Prime95, w/AVX and you can see the timestamp. And before you bother commenting, the 97% CPU load in the screen shot is because I clicked on the screen grab icon on the task bar to tale take the picture. It was running at 100% up until then for about 30 mins.  

I can't wait for the next issue I "should feel bad" about.
 
https://drive.google.com/...NPfvoJzTcVXFGZyst/view
 
 

OK, almost there.  Run Prime95 AVX FFT for 5 minutes and post a screen shot of HWiNFO64 or HWMonitor showing CPU POWER along with temps. 
 
You can't make a claim about temps and having effective cooling without citing power draw.

Really I can, because my question was to the OP about fan placement and mounting direction, asked an answered.  My goal in life is not sitting around running stress tests all day, the only reason I even bothered to post this picture was because of my typo in my first post.  We're done, none of this is going to simulate using my PC to play games, or help.

kevinc313

njbongo

kevinc313

njbongo
Actually, they aren't, I probably made a typo in my original post, and did not even catch it until now. My max temp under load is 62, not 72. Really wasn't relevant to my question for irakandjii since it was about case fan placement and direction, which he answered thoroughly. The screen shot is from Prime95, w/AVX and you can see the timestamp. And before you bother commenting, the 97% CPU load in the screen shot is because I clicked on the screen grab icon on the task bar to tale take the picture. It was running at 100% up until then for about 30 mins.  

I can't wait for the next issue I "should feel bad" about.
 
https://drive.google.com/...NPfvoJzTcVXFGZyst/view
 
 

OK, almost there.  Run Prime95 AVX FFT for 5 minutes and post a screen shot of HWiNFO64 or HWMonitor showing CPU POWER along with temps. 
 
You can't make a claim about temps and having effective cooling without citing power draw.

Really I can, because my question was to the OP about fan placement and mounting direction, asked an answered.  My goal in life is not sitting around running stress tests all day, the only reason I even bothered to post this picture was because of my typo in my first post.  We're done, none of this is going to simulate using my PC to play games, or help.

Your temp claims are completely pointless if you're not citing power figures.
 
I'm not the one who hijacked this thread.

And your comments on this thread are completely pointless in regards to my question to him.  If criticizing the products (with baseless information) other users on the forums have, or like, does something for you, maybe find some new hobbies.  I'm also not the first person to comment on the thread after he posted it, but was genuinely trying to help the person that "hijacked the thread" asking about his side panel.  The only reason I even asked irakandjii a question was because I saw his background, that he posted, and thought he would have a decent opinion on fan placement and it was sort of related to what he was talking about.
 
And by the way, that comment you posted to me that you decided to delete or edit, I actually saw because I got the Subscription Notification email before you made the change, and it proves my point.  You're not trying to help anyone. "sigh" to you too.

Use this link, you need it
 
https://www.nimh.nih.gov/.../find-help/index.shtml