Case fan grills are evil...

jimmyfergus · Post by **jimmyfergus** » Tue Sep 21, 2004 9:33 am

Hi All, just a quick thowaway post to encourage everyone to open up their fan grills.

This site frequently recommends unrestrictive fan grills for airflow and silence, and I've taken this in, but looking at the grills I have (on Dells), they look quite free-flowing, so I wasn't too worried.

However, in an excellent show of customer-responsiveness, Antec, as a thankyou for sending them product suggestions, sent me a free FanPal - a USB-powered desk fan for cooling the person rather than the computer. This is basically a novelty 80mm case fan with LEDs. See here for a review and good photos.

Well, at risk of being ungrateful, out of the box it just doesn't work very well, which is probably partly why they're giving them away. You can barely feel the airflow 1ft away. So within 30 seconds I had my screwdriver out and was taking it apart.

Freed from its grills, a much more significant amount of air flows. Even just holding only the relatively open looking back grill next to the fan cuts airflow, by my very unscientific estimate, by about 75%! Don't ask me for units

. The increase in noise was very significant too, even over this relatively noisy fan.

Interestingly, noise and airflow reduction is worst with a given grill behind (upstream) of the fan. Also, the more restrictive looking front (thin, metal) grill is actually a little less restrictive than the back (thicker, plastic), though just as noisy if placed behind the fan.

I grabbed a wire fan grill, and happily I can confirm that they have a relatively minimal effect on airflow and noise - up or downstream. Good news for those with pets and ankle-biters.

In the end, with its faults, the FanPal is a fun thing to have, especially for free. I expect I'll be cutting out its grills, and probably replacing the fan with something quieter, for next summer. At the least, it's given me a USB pass-through power-tap plug, and a cool little red-green-blue LED strip to play with.

I know all my case fans will shortly have nothing but wire grills protecting them. If this inspires anyone to cut up their case, don't forget this great forum article on metal cutting. It doesn't sound too difficult.

frosty · Post by **frosty** » Thu Sep 23, 2004 11:48 am

My pets are smart enough to stay away from my pc, I cut out all the grills even on my psu, let the air flow in and out.

Post by **MikeC** » Thu Sep 23, 2004 12:39 pm

Wire grills also have an effect: ~20%. This was checked with an anemometer on several different fans of 80, 92 and 120mm size, at full and lower speeds. The airflow reduction went as high as 24% but it was never below 19%.

At 12V, the reduction in airflow with most fans will be relatively harmless, though there is an increase in turbulence noise (audible to me). But at the low airflow rates SPCR members run fans, I think the airflow reduction is significant.

sthayashi · Post by **sthayashi** » Thu Sep 23, 2004 1:08 pm

Whoa.... Time to start removing wire grills....

jimmyfergus · Post by **jimmyfergus** » Thu Sep 23, 2004 2:30 pm

MikeC wrote:Wire grills also have an effect: ~20%. This was checked with an anemometer on several different fans of 80, 92 and 120mm size, at full and lower speeds. The airflow reduction went as high as 24% but it was never below 19%.

At 12V, the reduction in airflow with most fans will be relatively harmless, though there is an increase in turbulence noise (audible to me). But at the low airflow rates SPCR members run fans, I think the airflow reduction is significant.

Interesting... Yeah - I should have been more careful with my statement about noise with the wire grills - after all, I was talking about noise over the definitely non-silient fan provided with the FanPal. Also, my - ahem - "measurements" consisted of putting my face in front of it, so no surprise that the reduction is actually less than "minimal".

... I now have a grill-less FanPal with, and some nice Wiss tin snips to hack at my cases with. Even without grills, that FanPal fan has no place on a PC-silencer's desk though.

Jan Kivar · Post by **Jan Kivar** » Sat Sep 25, 2004 8:39 am

MikeC wrote:Wire grills also have an effect: ~20%. This was checked with an anemometer on several different fans of 80, 92 and 120mm size, at full and lower speeds. The airflow reduction went as high as 24% but it was never below 19%.

At 12V, the reduction in airflow with most fans will be relatively harmless, though there is an increase in turbulence noise (audible to me). But at the low airflow rates SPCR members run fans, I think the airflow reduction is significant.

Is this info/test part of the long-awaited fan test? I've never seen it posted anywhere eise.

Cheers,

Jan

pony-tail · Post by **pony-tail** » Sat Oct 02, 2004 11:32 pm

I had a wire grill on a 92mm panaflo l on a swiftek heatsink until i read this post . the CPU that this is cooling is a 3.06ghz (rig 1 in my sig )
I removed the grill to see what would happen .
I do not see how it is possible but it dropped my load temps by 2 deg C.
with no other changes
The fan is running at about 9v (diode in series ) any lower and the swiftek
gets too hot under load in summer . during winter I just hook it to the Antec fan only header and it gets what it gets
Back to topic
idle temps have not changed by any measurable ammount just the temps whilst in use- down by 1deg while surfing the web and 2 deg while running prime 95 .
Although this is onlly a small ammount taking a wire grill off is not a major mod either and costs nothing.

Katana Man · Post by **Katana Man** » Thu Oct 07, 2004 7:45 pm

Finally, some people are seeing the light! Thank you guys

I've been firm believer in the low resistance, low fan speed, low noise technique.

- A stock LX-6A19 case setup has 8 restriction pieces.
These include 1 front bezel, 1 filter, 2 case grills, 2 fan holders, 1 finger guard, and 1 PSU grill.
- My modified LX-6A19 case has only 2 resitriction pieces.
1 Modder's mesh grill, and 1 PSU grill.

It seems that the fans are more affected by these restrictions if they are running slower around 5V. I don't have any equipment to test, but I'd bet that more than 50% of the airflow is blocked when you have so many restrictions.

Read this post for more info.

Post by **MikeC** » Thu Oct 07, 2004 8:09 pm

Katana Man, seems like it's been a while since we were on the same thread! Good to run into you again.

Any new projects worth turning into a SPCR article? We haven't had a good DIY in ages.

It was your posts -- the ones you linked above -- that made me first look at the role of wire grills. Until you mentioned it, I had presumed they were harmless, but after your posts, I examined them manually and felt you were right, saw the effect they had by checking CPU temp with & with one the HSF. Now, with that anemometer, I confirmed it for sure: It is around a 20% drop in flow. Not much noise difference with a low speed fan, but you really want/need every single cfm at those low speeds.

Katana Man · Post by **Katana Man** » Thu Oct 07, 2004 8:25 pm

Not much new here, except for the fact, that I'm getting a bigger solid state hard drive being made for me as we speak.

They are combigning my two 1.2GB drives into a single 2.2GB drive. For a test, last week, I put Win98 on the SSD drive and it boots in 4 seconds! hehe

2.2GB will be plenty for XP especially after I clean up all the junk that comes with it.

As for DIY projects, I've really been happy with the low resistance projects. I don't see much room for improvement. The noise and temps are way down. However, I do plan on updating my D8000 so all fans are 120mm. Thanks to Thermalright for the XP120 heatsink.

With your anemometer, did you ever test more than just the fan grill? I'd be curious to see how much air flow is restricted for a front 120mm fan at 5V-7V, when there is a bezel, filter, case grill, and fan holder (like the fan LX-6A19 fan holder) in the way.

Post by **MikeC** » Thu Oct 07, 2004 8:38 pm

With your anemometer, did you ever test more than just the fan grill? I'd be curious to see how much air flow is restricted for a front 120mm fan at 5V-7V, when there is a bezel, filter, case grill, and fan holder (like the fan LX-6A19 fan holder) in the way.

In time... all these new tools take study and practise to really understand the results thoroughly.

4 secs, eh?

Sizzle · Post by **Sizzle** » Fri Oct 08, 2004 5:41 am

What about PSU's? Do you guys remove the grilles on the back of your PSU's?

Dave404 · Post by **Dave404** » Fri Oct 08, 2004 6:20 am

Fortunately, my Nexus NX4090 never seems to spin up worth talking about - if it did, I'd think about it. It would be a act of horrible butchery, though.

Post by **MikeC** » Fri Oct 08, 2004 6:31 am

Not sure what is meant by grilles on the back of your PSU's. I always remove the fan grill on the PSU (thought I have not done so with any 120mm fan PSUs, mostly because I hardly use them and I wprrt about the risk a wire or tool sticking in the PSU through the fan. On the intake side, with an 80mm PSU that is fresh-air ducted, cutting or minimizing a restrictive grill can definitely be worthwhile.

Sizzle · Post by **Sizzle** » Fri Oct 08, 2004 8:21 am

MikeC wrote:Not sure what is meant by grilles on the back of your PSU's. I always remove the fan grill on the PSU (thought I have not done so with any 120mm fan PSUs, mostly because I hardly use them and I wprrt about the risk a wire or tool sticking in the PSU through the fan. On the intake side, with an 80mm PSU that is fresh-air ducted, cutting or minimizing a restrictive grill can definitely be worthwhile.

The plural meant for "you guys" as in people in this thread. But thanks for the info Mike, you answered my question.

MagnusT · Post by **MagnusT** » Fri Oct 08, 2004 2:37 pm

Turbulence is evil....

I got an Antec Sonata about a year ago. You can mount a front intake fan either in front of the HD-rack (on the case front) or at the back of the HD-rack (inside the case). Both do the same thing. But both are also too close to parts that generate turbulence. I'm still using my home made 12 cm-fan foam-tunnel-intake. Turbulence is evil....