Positive computer case pressure: myth? - Page 6 - Overclock.net

Forum Jump: 
Reply
 
Thread Tools
post #51 of 84 Old 09-02-2016, 11:59 PM
New to Overclock.net
 
doyll's Avatar
 
Join Date: Aug 2011
Location: UK
Posts: 22,798
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1629
The case airflow .. the paths the air flows through the case .. determine how well components cool by how cool the air is reaching the components. These airflow paths flow the components heated air out of case .. and the better this is done the better they can supply cool air to components. Case fan placement and speed have a huge impact on how and where these airflow paths are. thumb.gif

post-flame-small.gifPhanteks Owners Club post-flame-small.gifpost-flame-small.gifWays to Better Cooling post-flame-small.gif
post-flame-small.gif1. Controling PWM Case Fans post-flame-small.gif2. Optimizing Case Cooling post-flame-small.gif3. Cooler Size Comparison & Cooling Comparison
post-flame-small.gif
=i7 980 @ 3.55GHz =PH-TC14PE w/2x TY-143 fans =Crucial Ballistix 3x4GB =GA-X58A-UD5 =ASUS GTX580 DirectCU II =Enermax Modu84+ =Define R2 w/3x TY-140 case intake fans; all PWM controlled by CPU fan socket
Phanteks Enthoo Primo MoBo Rampage III Extreme CPU i7 980xCooler R1 Ultimate w/ TY-147A fans RAM Dominator GT 6x2GB GPU ENGTX580 DCUII PSU Corsair TX850 Case [/COLOR]Phanteks Enthoo PrimoBase custom caster base
=i7 980X @ 4.0GHz =R1 Ultimate w/2x TY-143 fans =Dominator GT 6x2GB =Rampage III Extreme =ASUS GTX580 DirectCU II =TX850 =Enthoo Primo w/ custom castor base
doyll is online now  
Sponsored Links
Advertisement
 
post #52 of 84 Old 09-03-2016, 01:50 AM
New to Overclock.net
 
Blameless's Avatar
 
Join Date: Feb 2008
Posts: 29,123
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 3130
I'd like to add that when delivering cool intake air isn't always practicable, warm air flow is better than no air flow, and more air flow is better than less. As long as the air flowing over components is cooler than the components involved, the components will be cooled.

This is why a lot of datacenters and other installations are moving from conditioned/chilled air to outside air. It's often cheaper and more effective to push two or three times as much ~40C intake air through hardware than it is to cool a lesser volume of air to just above the dew point.

Clearly, if you have the option to blow cool intake air directly at a component without disturbing airflow through the rest of the case, one should take it, but if not, don't be afraid of air that's exhaust from something else.

...rightful liberty is unobstructed action according to our will within limits drawn around us by the equal rights of others. I do not add 'within the limits of the law,' because law is often but the tyrant's will, and always so when it violates the right of an individual. -- Thomas Jefferson
Blameless is offline  
post #53 of 84 Old 09-03-2016, 05:13 AM
New to Overclock.net
 
doyll's Avatar
 
Join Date: Aug 2011
Location: UK
Posts: 22,798
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1629
Quote:
Originally Posted by Blameless View Post

I'd like to add that when delivering cool intake air isn't always practicable, warm air flow is better than no air flow, and more air flow is better than less. As long as the air flowing over components is cooler than the components involved, the components will be cooled.

This is why a lot of datacenters and other installations are moving from conditioned/chilled air to outside air. It's often cheaper and more effective to push two or three times as much ~40C intake air through hardware than it is to cool a lesser volume of air to just above the dew point.

Clearly, if you have the option to blow cool intake air directly at a component without disturbing airflow through the rest of the case, one should take it, but if not, don't be afraid of air that's exhaust from something else.
40c component intake is about 15c above room.
If that's the best you can get, okay.

I optimize and help others optimize case airflow almost every day, and if we start with a system that is supplying 15-20c air to component we end up being able to low that intake air temp by 5-10c without much effort. It is not hard to get component to <10c .. and attaining 5c is usually not a problem. This is with CPU and GPU at full load.

One of the biggest problem I see is systems with fixed speed case fans .. meaning they supply more air then wanted 68% of the time, and not enough 30% of the time .. with only the right amount 2% of the time. The case's flow paths are changing almost every time the component fans change speed.

Few of us live in environments where our computer rooms are above 28c .. most of us have 20-25c rooms. There is no logical reason to have a system that cannot supply it's components with air less than 10-15c above room ambient. All of mine have less than 8c above ambient component intakes and that 8c one is HTPC that operates at 60c at 100% load.

Warm airlfow may not be better depending how warm 'warm' is.
As often as not three times as much 10c warmer airflow will not cool as well as 66% as as much 5c cooler air does.

Sure, we can operate components at 50c with 40c airflow, or we can operate them at 38c with 28c airflow from a 22c room. It is not really all that hard to select case and components that the case cannot supply components with airflow 4-8c above room ambient .. rather than the 20-34c warmer hotter component intake air temp resulting in 20-25c hotter components with fans screaming at full speed we often see and hear in poorly setup hard working gaming systems inexperienced people build.

Generally all it takes is 30 minutes of reading the right guides, an inexpensive $5 digital indoor / outdoor wired remote thermometer, and maybe an hour or two of playing around changing fan placement & profiles, and maybe getting a couple fans better suited to the task.

post-flame-small.gifPhanteks Owners Club post-flame-small.gifpost-flame-small.gifWays to Better Cooling post-flame-small.gif
post-flame-small.gif1. Controling PWM Case Fans post-flame-small.gif2. Optimizing Case Cooling post-flame-small.gif3. Cooler Size Comparison & Cooling Comparison
post-flame-small.gif
=i7 980 @ 3.55GHz =PH-TC14PE w/2x TY-143 fans =Crucial Ballistix 3x4GB =GA-X58A-UD5 =ASUS GTX580 DirectCU II =Enermax Modu84+ =Define R2 w/3x TY-140 case intake fans; all PWM controlled by CPU fan socket
Phanteks Enthoo Primo MoBo Rampage III Extreme CPU i7 980xCooler R1 Ultimate w/ TY-147A fans RAM Dominator GT 6x2GB GPU ENGTX580 DCUII PSU Corsair TX850 Case [/COLOR]Phanteks Enthoo PrimoBase custom caster base
=i7 980X @ 4.0GHz =R1 Ultimate w/2x TY-143 fans =Dominator GT 6x2GB =Rampage III Extreme =ASUS GTX580 DirectCU II =TX850 =Enthoo Primo w/ custom castor base
doyll is online now  
Sponsored Links
Advertisement
 
post #54 of 84 Old 09-03-2016, 06:12 AM
New to Overclock.net
 
Blameless's Avatar
 
Join Date: Feb 2008
Posts: 29,123
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 3130
Quote:
Originally Posted by doyll View Post

40c component intake is about 15c above room.

In the datacenter example I was using, the only air conditioning in the entire building may amount to 20k BTU, while the computers may put out 150-250k BTU of heat. AC is only used to keep the staff comfortable while they monitor things, or to cool specific problem racks so they can be worked on without the technicians risking heatstroke. Normal cooling for the computers would be outside air, which during the summer in a warm area could easily approach 40C. It's still way more efficient to use this air than to run 300k BTU of AC, because removing that amount of heat from the air costs and order of magnitude more than raw air flow using outdoor air, which is plenty sufficient to keep everything within safe operating temperature.

It was just an example of how air temperature doesn't always trump air flow.
Quote:
Originally Posted by doyll View Post

As often as not three times as much 10c warmer airflow will not cool as well as 66% as as much 5c cooler air does.

Depends on thermal density and the resulting temperature deltas between the cooler (or coolant) and ambient air. If the delta is low, then cooler ambient will help more than extra flow, but if the delta is high, the opposite is usually the case.

Many components that are traditionally passively cooled (chipsets with an MCH, VRMs, etc) or GPUs near the limits of reference air cooling, often run with fairly extreme temperature deltas, as can some whole systems focused on silence.

...rightful liberty is unobstructed action according to our will within limits drawn around us by the equal rights of others. I do not add 'within the limits of the law,' because law is often but the tyrant's will, and always so when it violates the right of an individual. -- Thomas Jefferson
Blameless is offline  
post #55 of 84 Old 12-23-2017, 12:29 AM
Graphics Card Aficionado
 
atomhard's Avatar
 
Join Date: Apr 2011
Posts: 52
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 0
Awesome, awesome discussion here.
Just wanted to thank everyone involved biggrin.gif
atomhard is offline  
post #56 of 84 Old 01-02-2018, 07:35 PM
 
honegod's Avatar
 
Join Date: Jul 2017
Posts: 61
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 3
Quote:
Originally Posted by 8051 View Post

Considering computer cases are hardly air tight it's hard to believe any measurable pressure could be created in them. Would it be possible to measure the pressure w/a barometer? Has anyone ever measured the pressure created in a positive pressure case?

As mentioned, a manometer is the tool.
A clear tube bent into a U
add water to halfway up the vertical sides
Seal a hose to one side of the U and run it to the inside of the case

Presto smile.gif case differential pressure gauge.

Positive pressure pushes the water in the tube side down,
and the water in the other side rises an equal distance.

Negative pressure would allow ambient pressure to push the open side water down
And case side water to rise.

If you bent the U into an V the angled sides would make the water move farther along
The tube for a given delta P

Flattening the tube almost completely, with only a 1 degree bend would
Crank up the sensitivity where Case Pressure could be measured with a ruler,

Since it would be a unitless measure the same tool would be useful for direct comparisons
Between different fan/vent setups or even different cases
honegod is offline  
post #57 of 84 Old 01-02-2018, 09:01 PM
 
honegod's Avatar
 
Join Date: Jul 2017
Posts: 61
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 3
[doyll quote ] "I often call it positive airflow instead of positive pressure because as has been repeatedly said here the pressure difference is minimal. The rate of airflow we need in our cases is quite different from the 'positive pressure' of clean rooms or inside of military vehicles. We need to move a large volume of air through our case while clean rooms are only keeping dust out.

What this means is we want our intake vents to have fans with a little more intake airflow potential into case then the exhaust airflow potential of our exhaust vents (with/without fans). This creates a case environment that is pushing just enough to keep air (and dust) from leaking into our cases" [/doyll quote]

The entire point of fans is to create a pressure differential.
Key to that is that air is a compressable fluid, where water is incompressable.

Put a intake fan on a sealed box and the fan will pressurize the air in the box to whatever
The stall pressure the fan is rated at.

If you put a variable Opening in the box and make a exhaust hole the air in the box will lose pressure
Until the rate of the mass of air escaping perfectly matches the fan flow/pressure curve.

Note that the speed of the air flowing out of the case can be much faster than the flow speed through the fan

Making the hole bigger lowers the pressure in the box and the air speed through the hole.

Eventually the hole size reaches the point where it no longer restricts the fan and the fan uses all
Of its pressure creation potential to move air.
Increasing the hole size beyond that point does nothing.

Stick another fan into the hole as exhaust and it reduces the area of the hole,
While at the same time making the hole think that the ambient pressure on that side of the case
Is lower. So for an exhaust fan to work it needs to create enough of a pressure drop for the intake fan to
Move more air.

I would like to see pressure/flow curves for fans to see how they would interact for intake/exaust.
honegod is offline  
post #58 of 84 Old 01-02-2018, 10:34 PM - Thread Starter
 
8051's Avatar
 
Join Date: Apr 2014
Posts: 1,541
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 18
Quote:
Originally Posted by honegod View Post

As mentioned, a manometer is the tool.
A clear tube bent into a U
add water to halfway up the vertical sides
Seal a hose to one side of the U and run it to the inside of the case

Presto smile.gif case differential pressure gauge.

Positive pressure pushes the water in the tube side down,
and the water in the other side rises an equal distance.

Negative pressure would allow ambient pressure to push the open side water down
And case side water to rise.

If you bent the U into an V the angled sides would make the water move farther along
The tube for a given delta P

Flattening the tube almost completely, with only a 1 degree bend would
Crank up the sensitivity where Case Pressure could be measured with a ruler,

Since it would be a unitless measure the same tool would be useful for direct comparisons
Between different fan/vent setups or even different cases

The only pressurized vessels I've ever dealt with (in chem labs, in motorcycle crankcases, cylinder heads) all featured gaskets, seals and many bolts to insure proper sealing, computer cases have none of that, none at all. It's hard to imagine any computer case being pressurized at all.
8051 is offline  
post #59 of 84 Old 01-02-2018, 11:11 PM
New to Overclock.net
 
Blameless's Avatar
 
Join Date: Feb 2008
Posts: 29,123
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 3130
Quote:
Originally Posted by honegod View Post

" I often call it positive airflow instead of positive pressure because as has been repeatedly said here the pressure difference is minimal.

Positive pressure is more accurate as the amount of air flow out must match the flow of air in. The difference is where the impedance, and thus pressure differential, is.
Quote:
Originally Posted by 8051 View Post

The only pressurized vessels I've ever dealt with (in chem labs, in motorcycle crankcases, cylinder heads) all featured gaskets, seals and many bolts to insure proper sealing, computer cases have none of that, none at all. It's hard to imagine any computer case being pressurized at all.

The pressure vessels you are talking about are for resisting significant pressure differentials. It shouldn't be hard to imagine smaller differentials as innumerable every day things rely on or are affected by them.

...rightful liberty is unobstructed action according to our will within limits drawn around us by the equal rights of others. I do not add 'within the limits of the law,' because law is often but the tyrant's will, and always so when it violates the right of an individual. -- Thomas Jefferson
Blameless is offline  
post #60 of 84 Old 01-02-2018, 11:41 PM
 
spddmn24's Avatar
 
Join Date: Oct 2015
Posts: 252
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 7
Quote:
Originally Posted by 8051 View Post

The only pressurized vessels I've ever dealt with (in chem labs, in motorcycle crankcases, cylinder heads) all featured gaskets, seals and many bolts to insure proper sealing, computer cases have none of that, none at all. It's hard to imagine any computer case being pressurized at all.

You need to think on a much smaller scale. Airflow is the result of pressure differentials. If air is flowing out of the vents in a compute case that means the pressure inside the case is higher than the pressure outside of it.

spddmn24 is offline  
Reply

Quick Reply

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off