Adding an intake makes the temperature increase??

Enclosures and acoustic damping to help quiet them.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Devonavar

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Adding an intake makes the temperature increase??

Post by hostile » Mon Oct 18, 2004 8:19 pm

Hi Everyone,

I just bought an Antec SLK3700AMB for myself, and have decided to stick with it for now, despite my reservations in this thread. As an aside, that was my first post in these forums, and I didn't get the SPCR customary welcome... I'm cut.. cut to the quick... *sniff*...

:D

Anyhow, I went out and bought a Vantec Stealth 120mm for an extra case fan.

After much reading about how to ventilate the case, I decided to place the stock Antec up front with it's 79cfm, and the Vantec at the back with it's 53 cfm, giving me a positive case pressure and thereby ensuring all my air is filtered. I even swapped out all my labouriously placed duct tape with HVAC tape.

My case temp shot up 7 degrees at idle with this setup! So, I swapped the fans so the Antec was at the back and the Vantec up front, and it was 5 degrees higher. After much fiddling, twiddling, installing a fan bus controller, swapping the fans with the bus controller at various settings, I find myself right back at the beginning, with the Vantec removed and my case as cool as it can get without the side off.

Can anyone possibly think of why this would be?

My guess is maybe the hard drives are being cooled more efficiently with the front fan in, so almost all of their excess heat is being kicked into the case, as opposed to being "trickled" off. Keeping in mind, I have 5 hard drives (3 x Ultra 160 SCSI drives and 2 x SATAs), which run a lot warmer than most run of the mill IDE drives.

I would still think that pushing more air through the case would result in lower temperatures, would it not?

Thanks!

sthayashi
*Lifetime Patron*
Posts: 3214
Joined: Wed Nov 12, 2003 10:06 am
Location: Pittsburgh, PA

Re: Adding an intake makes the temperature increase??

Post by sthayashi » Mon Oct 18, 2004 8:40 pm

hostile wrote:Can anyone possibly think of why this would be?
If you were riding a tandem bicycle with Lance Armstrong, how much faster would you go than if Lance were just biking by himself?

I suspect the principle involved here is the same, though I can't think of a better way to describe it now (stupid brain malfuctioning..... :? )

Edward Ng
SPCR Reviewer
Posts: 2696
Joined: Thu Dec 11, 2003 9:53 pm
Location: Scarsdale, NY
Contact:

Post by Edward Ng » Mon Oct 18, 2004 9:01 pm

It's a Vantec Stealth! It sucks so much it makes temps worse where other fans wouldn't! :lol:

Okay, okay--in all seriousness, any chance you could take pics of your rig and post them up for us to see? It will help us quite a bit in diagnosing your air flow problem.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Mon Oct 18, 2004 10:34 pm

Edward Ng wrote:It's a Vantec Stealth! It sucks so much it makes temps worse where other fans wouldn't! :lol:

Okay, okay--in all seriousness, any chance you could take pics of your rig and post them up for us to see? It will help us quite a bit in diagnosing your air flow problem.
The Vantec was the cheapest in my area by about $10. By researching the site, it would seem MikeC lives about 15 minutes away from me... perhaps he knows of a good place to buy all of the HQ fans... I digress...

Anyhow, I posted a pic of my rig in this thread, but here are some fresh ones:

Image
Image
Image
Image

I tidied up things a bit from the last picture. Anyhow, I hope this helps.

Trip
*Lifetime Patron*
Posts: 2928
Joined: Wed Sep 24, 2003 7:18 pm
Location: SC

Post by Trip » Mon Oct 18, 2004 11:11 pm

I guess the front fan just impedes airflow. Your front intake is so open and the pressure in the case so low, that the front fan doesn't help. Even the strong Antec is pulling freely at the back of your case.

That's my guess.

You say you're using a filter?
Last edited by Trip on Mon Oct 18, 2004 11:15 pm, edited 1 time in total.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Mon Oct 18, 2004 11:14 pm

Trip wrote:I guess the front fan just impedes airflow. Your front intake is so open and the pressure in the case so low, that the front fan doesn't help.

That's my guess.

You say you're using a filter?
Ya, I always put filtering material in my intakes... I just put in those heat register filters you buy at Home Depot - maybe 1/8" thick, and very breatheable. I cut a piece to fit in behind the stock filter.

Trip
*Lifetime Patron*
Posts: 2928
Joined: Wed Sep 24, 2003 7:18 pm
Location: SC

Post by Trip » Mon Oct 18, 2004 11:21 pm

I'll bet the filter is impeding the front fan's air.

Air sucks through easily w/o a fan, but air blown against the filter is partly pushed back, creating turbulence.

So the impedance of the fan combined with its turbulence is worse than no fan.

The fan is blowing against the filter, right?

Could you rearrange things so that it is sucking through the filter?

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Mon Oct 18, 2004 11:24 pm

Trip wrote:I'll bet the filter is impeding the front fan's air.

Air sucks through easily w/o a fan, but air blown against the filter is partly pushed back, creating turbulence.

So the impedance of the fan combined with its turbulence is worse than no fan.

The fan is blowing against the filter, right?

Could you rearrange things so that it is sucking through the filter?
If I remove the filtering material, the temperature only drops by 1 degree.

:(

No, the filtering material is in between the front of the case, and the stock mesh filter, which are both in front of the lower intake.

Air is being sucked through the filter.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Mon Oct 18, 2004 11:38 pm

The fan is blowing against the filter, right?

Could you rearrange things so that it is sucking through the filter?
Here's a couple of pics to show you the filtering setup.

Image
Image

Trip
*Lifetime Patron*
Posts: 2928
Joined: Wed Sep 24, 2003 7:18 pm
Location: SC

Post by Trip » Tue Oct 19, 2004 12:23 am

I guess then the airflow does better following the path of least resistance through your case and the fan creates turbulence inside the case.

lenny
Patron of SPCR
Posts: 1642
Joined: Wed May 28, 2003 10:50 am
Location: Somewhere out there

Post by lenny » Tue Oct 19, 2004 12:46 am

It's late, and I'm sleepy, so I'm not going to search, but I seem to recall a thread on the Sonata where someone observed something similar. Essentially, the front fan appeared to be cooling the hard disks at the expense of case temps.

However, since the hard disks generate the same amount of heat regardless of airflow, I would think that having more airflow should result in lower temps. Perhaps without a front fan, the heat rises and is evacuated by the PSU fan. With a front fan, it might be blown over the case temp. sensor.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Tue Oct 19, 2004 12:58 am

...since the hard disks generate the same amount of heat regardless of airflow, I would think that having more airflow should result in lower temps. Perhaps without a front fan, the heat rises and is evacuated by the PSU fan. With a front fan, it might be blown over the case temp. sensor.

That's what I've been thinking too, but I noticed has been my CPU temp which has increased the most, but the system temp does rise too.

kesv
Posts: 300
Joined: Fri Nov 07, 2003 8:44 am

Post by kesv » Tue Oct 19, 2004 1:26 am

hostile wrote: If I remove the filtering material, the temperature only drops by 1 degree.
I'll suggest that it's the five drives that are in effect blocking the airflow from the front fan, creating turbulence and flow resistance. However, you should still check what the temperatures of your drives are when running the different fan configurations. The added cooling of the drives is likely to also contribute to the temperature.

It would also be useful to see what your cpu temp is, when running the front fan. Unless you are in danger of overcooking the cpu the added case temperature might be a good thing if it means your drives run cooler. I'd say a case temperature of
40C and a cpu temp around 60C are still acceptable for your setup.

With five hdds you are not bound to compete for the coolest running setup anyway ;)

Get one of the various drive monitoring tools. I'm assuming that since you have those highend drives, you do some important work on them. You really should monitor the health of your drives.

Edward Ng
SPCR Reviewer
Posts: 2696
Joined: Thu Dec 11, 2003 9:53 pm
Location: Scarsdale, NY
Contact:

Post by Edward Ng » Tue Oct 19, 2004 5:59 am

lenny wrote:It's late, and I'm sleepy, so I'm not going to search, but I seem to recall a thread on the Sonata where someone observed something similar. Essentially, the front fan appeared to be cooling the hard disks at the expense of case temps.

However, since the hard disks generate the same amount of heat regardless of airflow, I would think that having more airflow should result in lower temps. Perhaps without a front fan, the heat rises and is evacuated by the PSU fan. With a front fan, it might be blown over the case temp. sensor.
Same idea here as well.

markjia
Posts: 171
Joined: Sat Apr 24, 2004 4:31 pm
Location: Vancouver BC

Post by markjia » Tue Oct 19, 2004 10:40 am

Why not try negative case pressure? If you are concerned about dust, maybe you can filter the other opennings in the case.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Tue Oct 19, 2004 10:57 am

kesv wrote:I'll suggest that it's the five drives that are in effect blocking the airflow from the front fan, creating turbulence and flow resistance. However, you should still check what the temperatures of your drives are when running the different fan configurations. The added cooling of the drives is likely to also contribute to the temperature.

It would also be useful to see what your cpu temp is, when running the front fan. Unless you are in danger of overcooking the cpu the added case temperature might be a good thing if it means your drives run cooler. I'd say a case temperature of
40C and a cpu temp around 60C are still acceptable for your setup.

With five hdds you are not bound to compete for the coolest running setup anyway ;)

Get one of the various drive monitoring tools. I'm assuming that since you have those highend drives, you do some important work on them. You really should monitor the health of your drives.
The CPU temp rises when I put in the front fan. Without the front fan in, the CPU and System idle at 45C/25C. With a front fan, the temperatures rise to a very undesireable 51/30C (Undesireable for me, that is).

Unfortunately, I cannot tell what the drive temperatures are. None of the SCSI drives report back, and the SATA drives are in a RAID 1, and appear as a "RAID Device" and don't report back either. I have tried both DTemp and Hard Drive Temperature Monitor.

With that said, the drives are not cool, but only slightly warm to the touch, which doesn't bother me very much - I have felt these drives in a system which wasn't getting good cooling to them, and you could use them to fry an egg.

Overall, despite my growing fondness for this case, it may not be the one for me. I'm trying though. :)

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Tue Oct 19, 2004 10:59 am

markjia wrote:Why not try negative case pressure? If you are concerned about dust, maybe you can filter the other opennings in the case.
Currently it is running on a negative pressure setup. Just the stock fan is in place, and all the front is sealed up with HVAC tape, as you can see in the pics above.

So far, this setup seems to give the overall best cooling, which I find to be very counter-intuitive.

Rusty075
SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ
Contact:

Re: Adding an intake makes the temperature increase??

Post by Rusty075 » Tue Oct 19, 2004 11:21 am

sthayashi wrote:If you were riding a tandem bicycle with Lance Armstrong, how much faster would you go than if Lance were just biking by himself?

I suspect the principle involved here is the same, though I can't think of a better way to describe it now (stupid brain malfuctioning..... :? )
Fans in Series

I suspect that's what sthayashi is thinking of.

Here's Cliff Note version, quoted from an earlier thread, because I'm lazy.
Most (some? a few? insert whatever adjective you think applies) people think that in this situation having 2 fans means that you have twice as much air flowing through the case. Afterall if you have 2 fans, one pushing 79CFM and the other 53, then you must have 132CFM going through there, right? Wrong. You've got 53. Fans in series will only move as much air as the slowest of the fans involved will.

Now running fans in series does increase the static air pressure that they can build up, but for the airflow rates and fans involved in computers the pressure increase doesn't add up to anything significant. The second fan is literally doing nothing but making noise.
In your case, the intake fan may actually just be creating turbulance that is disrupting the previous airflow pattern.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Tue Oct 19, 2004 11:43 am

Well there you have it.

I wasnt under the impression I would be moving 132 cfm through there, but I was thinking that a mild draw up front would help bring in more cool air.

Now that I really think about it in the context of the information provided, one would be decreasing the pressure differential by adding a front fan, thereby shrinking the amount of air going through the case.

... now I gotta see if I can return this Vantec fan...

Thanks Rusty!

burcakb
Posts: 1443
Joined: Tue Mar 09, 2004 9:05 am
Location: Turkey

Post by burcakb » Tue Oct 19, 2004 12:04 pm

lenny also has a point. Most case temp sensors are on or very near the southbridge chip. And again on most boards, this chip is below the graphics card. This means at on a rear pull system, air doesn't circulate around it much and case temps are higher. Add a front fan that blows some air toward the southbridge and you get lower case temps. Now this is the normal case. You've got 5 drives there and I'm betting some of them run very hot. So your front fan is blowing "hot" air over the southbridge. Furhermore due to cable clutter and some air going below the graphics card, your CPU area is getting less air than with one fan. So CPU temps also rise.


Another possibility worth exploring is, seeing how you're still keeping your noisy fan brackets, your stamped fan grills are also in place. If you'll look closely, you'll see that the holes on the front are larger than those in the back. So even using the same fans, there'll be a difference due to different pressures the fans have to operate against.

All in all, Rusty's explanation is probably the best :)

Rusty075
SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ
Contact:

Post by Rusty075 » Tue Oct 19, 2004 12:18 pm

burcakb wrote:All in all, Rusty's explanation is probably the best :)
Now just keep repeating that over and over, and everything will be ok. :wink:

What I meant to add to my explanation, and then forgot, is that I think lenny is right about the temp sensors thing. "case" sensors on motherboards are just about useless; change the airflow direction and the temps can jump all over the place.

An aside: my NF7 has the case sensor located right between the CPU socket and the first RAM slot, so that the air wash from a downblowing HSF blows right across it. Flip the fan over into "suck" mode, and my "case" temps drop by >10°.

I'd use the CPU temp, and the HDD temps (along with your ears) as the benchmark for airflow effectiveness.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Tue Oct 19, 2004 12:28 pm

I think it's probably a combination of all things.

1) Adding intake fan = lower pressure differential = less total air flow through case

2) Intake fan = extra hard drive cooling = more warm air in case

3) 1) + 2) = greatly increased overall temps


This board here is an NF7-S, so I probably have the same thing going on as you, Rusty.

All in all, no matter which extra setup I run, the stock one still gives the best CPU temps, which is my primary concern. No matter which senario, the hard drives seem to be cooled adequately.

I'm going to buy another one of these cases and put the wife's machine into it. Hers though I will be able to turn into a "near silent" computer, as she doesn't have the heavy iron inside hers like I do. :D

Thanks guys!

kesv
Posts: 300
Joined: Fri Nov 07, 2003 8:44 am

Post by kesv » Wed Oct 20, 2004 12:02 am

hostile wrote: The CPU temp rises when I put in the front fan. Without the front fan in, the CPU and System idle at 45C/25C. With a front fan, the temperatures rise to a very undesireable 51/30C (Undesireable for me, that is).
Most of us silence enthusiast would be very happy with 51/30C temperatures. Should be well within safe limits.
hostile wrote: Unfortunately, I cannot tell what the drive temperatures are. None of the SCSI drives report back, and the SATA drives are in a RAID 1, and appear as a "RAID Device" and don't report back either. I have tried both DTemp and Hard Drive Temperature Monitor.
It can be tricky to find a util that is good enough to figure out how to access the SMART information on your non-average devices. It should still be possible though. Nothing in your setup prevents the SMART values from being read as such. Particularly the scsi should not be a problem. I wonder why those utils don't pick those up.

The maker of your scsi card should have a util for monitoring the drives or at the very least an idea wich utils work with that card.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Wed Oct 20, 2004 12:38 am

kesv wrote:Most of us silence enthusiast would be very happy with 51/30C temperatures. Should be well within safe limits.
True, those aren't *bad* temperatures, but still not what I'm looking for, as I know I can get significantly lower.
kesv wrote:It can be tricky to find a util that is good enough to figure out how to access the SMART information on your non-average devices. It should still be possible though. Nothing in your setup prevents the SMART values from being read as such. Particularly the scsi should not be a problem. I wonder why those utils don't pick those up.

The maker of your scsi card should have a util for monitoring the drives or at the very least an idea wich utils work with that card.
It's just a vanilla Adaptec 29160, nothing special or out of the ordinary. I even tried the SCSI version of HDDTemp, and nothing registered.

Any other utils you can suggest?

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Wed Oct 20, 2004 1:48 am

Well, after much more fiddling, tweaking, and frustration, I have a setup which I think will work for me.

I re-installed the Vantec up front, and put it on the fan bus controller, turned it all the way down (to maintain a high pressure differential as previously discussed), turned down my Zalman HSF fan a wee bit just so the obvious noise of it is clipped, added more HVAC tape (I must have a whole roll on this thing now!) and here are the results:

CPU Idle: 43C, System Idle: 24C

CPU Load: 48C, System Load: 28C

I finally found a hard drive monitoring util which recognised the drives, but it's only a trial version (Active SMART SCSI v2.41), and my SCSI drives are reporting back at 29C, 30C, and 30C, going from newest model to oldest, at idle.

I think I'm going to take back the Vantec fan, and get the 36cfm Zalman 120mm, since this Vantec Stealth is ANYTHING BUT "stealthy".

I wanted to thank everyone for all of their time, help, and suggestions!

Overall, this case is damn quiet for me, but after having used wind tunnels for the last 5 years, and being just sick and tired of it, I'm starting to get hooked on this whole "quiet/silent" thing. It's amazing that I can actually hear hard drive seeks on my box now. With 3 SCSI drives and not being able to hear their seeks, you can imagine how loud my box was before the Antec!!

Again, thanks everyone! :D

-=HoSTiLe=-

teejay
Patron of SPCR
Posts: 749
Joined: Mon Aug 02, 2004 2:23 am
Location: The Netherlands

Post by teejay » Wed Oct 20, 2004 3:03 am

hostile wrote:I think I'm going to take back the Vantec fan, and get the 36cfm Zalman 120mm, since this Vantec Stealth is ANYTHING BUT "stealthy".
If you're going for a 36-ish cfm 120mm fan I'd strongly consider the Nexus fan. Okay, it's a rather ugly shade of orange, but it is also very very quiet, even at 12V.

The Zalman fans do not get very high marks from the SPCR experts. My personal experience is limited to one 92mm Zalman which is not too bad at low voltage/RPM... but it's no Nexus! Check out the new sticky, The Top SPCR Quiet 120mm Axial Fans Compared. It's basically a summary of what people have been saying about 120mm's around the boards.

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Wed Oct 20, 2004 10:07 am

teejay wrote:
hostile wrote:I think I'm going to take back the Vantec fan, and get the 36cfm Zalman 120mm, since this Vantec Stealth is ANYTHING BUT "stealthy".
If you're going for a 36-ish cfm 120mm fan I'd strongly consider the Nexus fan. Okay, it's a rather ugly shade of orange, but it is also very very quiet, even at 12V.

The Zalman fans do not get very high marks from the SPCR experts. My personal experience is limited to one 92mm Zalman which is not too bad at low voltage/RPM... but it's no Nexus! Check out the new sticky, The Top SPCR Quiet 120mm Axial Fans Compared. It's basically a summary of what people have been saying about 120mm's around the boards.
Thanks Teejay, I'll get the Nexus instead.

Trip
*Lifetime Patron*
Posts: 2928
Joined: Wed Sep 24, 2003 7:18 pm
Location: SC

Post by Trip » Thu Oct 21, 2004 7:27 am

When you put the intake fan on, did you seal the area around the intake fan?

hostile
Posts: 17
Joined: Sat Oct 16, 2004 12:38 am
Location: Vancouver, BC, Canada

Post by hostile » Thu Oct 21, 2004 10:38 am

Trip wrote:When you put the intake fan on, did you seal the area around the intake fan?
Only on the outside front of the case, behind the bezel, as shown in the photographs above.

I didnt seal around the fan mounting clip on the inside of the case.

Trip
*Lifetime Patron*
Posts: 2928
Joined: Wed Sep 24, 2003 7:18 pm
Location: SC

Post by Trip » Thu Oct 21, 2004 2:04 pm

So air could still go in around the intake fan?

It must have been either the turbulence or the heating of the case sensor, then. I never thought about the sensor.

Post Reply