Disks running hot regularly


Recommended Posts

4 minutes ago, JonathanM said:

All those openings not sealed are allowing air to go where it's not needed. Tape over all the random holes in the bottom, back, slot covers, etc, then put the side back on and see how it goes.

I've literally never seen anything like that (okay maybe except on servers) Are you really saying the air might flow in from the pcie slots, or from the bottom and go out on through the rear fan, so the air pushed into the front wont make it out?  (orange: bad, unwanted, unexpected airflow)

Because other than the bottom there's really no big holes anywhere that would make sense in terms of unwanted airflow.
I could try to build a "funnel" from the pull fan after the disks up to the cpu fan. (grey) but I dont really think that would work properly (and I'm not sure how to build it...

Green is the wanted airflow, where dark green is flow that might not be there right now.

Blue is the top fan, I really dont see a way to redirect it to the bottom, so the only thing I could do is lower that top front one even more. (to make sure green reaches the cpu/rear fan, and not blue)

InkedIMG_20220526_163450_airflow.thumb.jpg.3175afa1d676b1b20e478979d12281c1.jpg

 

Link to comment
3 minutes ago, wambo said:

Are you really saying the air might flow in from the pcie slots, or from the bottom and go out on through the rear fan, so the air pushed into the front wont make it out? 

Yep. I'm really saying that. Air moves from high pressure to low pressure. period. end of statement.

 

If you are doing a really good job of evacuating air from the exhaust fans, and the intake fans have restrictions and obstructions (the drives) the extra air being pushed out the back will come in from all the extra openings instead of helping out the intake fans.

 

To do a really thorough job, you really need a clear cover to temporarily seal the side but allow visibility, and a smoke pen or some other source of non-noxious or settling vapour, and observe airflow in realtime.

Link to comment
2 minutes ago, Frank1940 said:

Fan noise is not uncommon in servers.   There are fans which are quieter.  But be careful and do some research.  'Moving' air makes noise in itself.  (Just stand outside when a breeze is blowing.)  Carefully notice the relationship between these three factors:  (1) CFM, (2) Head-- often measured in 'mm of water' and (3) Noise.  You will quickly find that you don't get something for nothing!

 

Here is a one fan that I have used when I needed a lower noise level.  This company has several lines of fans designed to emphasis one of the three factors in the trade-off situation.

    https://noctua.at/en/products/fan

This is not the only company to make quiet fans so have a look around before buying...

Well the bequiet silentwings 3 are also among those... "silent but still a lot of throughput"
The problem is that at max speed they're all still audible (or they're artifically throttled)

I'll test with this setup (and some slight changes) for now, and when I open the case next time, I'll tape the bottom and pcie closed (shouldnt be too complicated).
And then I have a manual fan control  card coming (but its got some longer delivery time), with that I can probably fine tune things to a nice 80% rpm level which isnt as loud.

Link to comment
15 minutes ago, JonathanM said:

Yep. I'm really saying that. Air moves from high pressure to low pressure. period. end of statement.

What I'm questioning is why I would have lower pressure at the end of my push-disks-pull tunnel than on the openings with "atmosphere" pressure.  
Even if my fans did almost nothing, they'd still create higher pressure - the same as the rear fan would create lower pressure.  
and the air coming in from the bottom, would at least "sweep" along some of the air from the disk bundle (if that works only a bit), so its not too detrimental

the only one "bad" would be  "in through pcie" - which really needs a lot of pressure difference, to force the air around that bend instead

I cannot monitor the airflow with smoke, but I can still test whether there's relevant airflow through openings... with light tissue... and I can be a bit sure that there's no air going into the back.


But either way,  I can keep the disks cool enough, I only have to dial in to do it with a bit less rpm so it's not too audible. ( I regret not having ordered pwm fans...)

Edited by wambo
Link to comment
44 minutes ago, JonathanM said:

Yep. I'm really saying that. Air moves from high pressure to low pressure. period. end of statement.

 

Not quite when you are talking about fan CFM.  The amount of air that a fan can move is determined by the pressure differential across that fan.   Nothing else!  

 

In the case of a computer case,  all of the input ports and output ports of the computer case are virtually at the same pressure. The pressure differential between the back side (low pressure) and the front side (high pressure) of each fan is what determines the fan's CFM.  The higher this pressure differential is, the lower the CFM is.  Hard drives installation for servers are tough nut to achieve a high volume of air flow across the hard drives.  Most of the 'tunnel' through them is blocked by the drives and creates a pressure differential situation depending on which side of the 'tunnel' you are at.  At the input end, the pressure rises as you try to force more air through.  At the output side, the pressures drops as you try to pull more air out of the tunnel.  To achieve maximum cooling what is usually done is to provide a large area of case openings at the output side of the tunnel and equip each one with a fan.  If more cooling is required beyond this, openings at input side of the tunnel are also equipped with a fan.   This will usually result in a good cooling but at the price of noise. 

 

What really complicates analysis of cooling air flow through any computer case is the number of tunnels are present.  Every little leak into the case is a tunnel.  Every object (PS, cables, MB, CPU cooler) inside the case is a part of a tunnel.   In servers, we are usually only concerned with the hard drive tunnel.  If we can keep that at a reasonable temperature, all of the other electronics should be well within their limits. 

 

Many long years ago, an HVAC engineer told me some that initially blew my mind-- A fan with zero CFM of air flow does no work and uses the least amount of electricity while running!  (Mounting a fan in one wall of a otherwise sealed box would provide this condition! )

 

 

Link to comment
2 hours ago, Frank1940 said:

Not quite when you are talking about fan CFM. 

Not sure what you are saying here, but yes, a fan works the same way. The fan blades push the air, creating a high pressure zone in front of the pushing blade, that compressed air tries every path it can to get to a low pressure zone, it will flow through the path of least resistance. A blocked fan still makes a high pressure zone in front of the blade, but since it can't move that air very far, just inside the housing, it quickly circulates to the low pressure zone directly behind the pushing blade. The fact that it only moves a tiny amount, around the edges of the blades, means it's not creating much resistance for the motor, and it either speeds up or draws less power, depending on the type of motor. That's why when you clog the nozzle of a vacuum cleaner, the motor revs up, it's not moving much air, just thrashing around what's directly beside the blades.

 

Static pressure is the rating given to a fan to tell you how hard it can push air that's actively resisting, vs. free air that is effectively always the same pressure on either side of the fan regardless of flow. The tighter the shroud is to the blade tips, and wider blades with shallower angle of attack tend to be able to push harder against a flow restriction. However, that typically means less CFM in free air for a given power level.

Link to comment

@JonathanM, Basically, we are saying the same thing but looking it different ways.  For those who are just starting out trying to figure how to set things up to maximize cooling while minimizing noise, a graph may help.  I swiped this one from the Noctua website:

 

image.thumb.png.573185a2e0a835077b5fd951bb5b2ad9.png

 

Noise levels for each fan model are as follows:

NF-A12x25    22.6 dB/A
NF-12        22.4 dB/A
NF-S12A      17.8 dB/A

 

The dashed lines (labeled PC case, aircooling heatsink and watercooling heatsink) show the typical static pressure vs airflow through/across the device.   (I am surmising that PC case line is for a virtually empty case...)

 

To give you a feel of how much a mm of H2O static pressure is, the atmospheric pressure at sea level using water is 10.3M (meter).   Now further understand that well-designed noisier fans have much better pressure differential performance and will deliver much higher volumes of at any specific static pressure.   With these fans, you don't have to be as careful with the cooling setup because these noisy fans static pressure and airflow capability will usually overcome a marginal airflow design.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.