1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Effects of Grill Patterns on Fan Performance/Noise

Discussion in 'Overclocking & Cooling' started by a111087, Oct 8, 2012.

  1. a111087

    a111087

    Joined:
    Apr 2, 2007
    Messages:
    2,760 (1.00/day)
    Thanks Received:
    200
    Location:
    US
  2. newtekie1

    newtekie1 Semi-Retired Folder

    Joined:
    Nov 22, 2005
    Messages:
    20,033 (6.15/day)
    Thanks Received:
    6,095
    Surprising, I would have through the mesh would be the worst CFM wise. It would have been nice if they tested with a filter in place as well. I personally use filters on my main rig, but I have to, because the only place I can put it in the room is on the floor and I have 4 dogs. Without the filters the case fills with fur in a week...
     
    Crunching for Team TPU 50 Million points folded for TPU
  3. a111087

    a111087

    Joined:
    Apr 2, 2007
    Messages:
    2,760 (1.00/day)
    Thanks Received:
    200
    Location:
    US
    Yeah, the mesh is getting pretty close to being the worst as far as CFM goes.
    I actually asked the author to do another review and include air filters. So, will see...
     
  4. sneekypeet

    sneekypeet Unpaid Babysitter Staff Member

    Joined:
    Apr 12, 2006
    Messages:
    21,619 (6.94/day)
    Thanks Received:
    6,143
    Saw a bunch of this data from SilverStone at CES. It may interest you to look at the fan grills on GD07 and 08. That is the design they found to perform the best for them. Just very interesting as to what sorts of cuts and adjustments need to be made as to just not increase the dBA levels. Something most filters don't even take into consideration, like most of the ones tested in the OP.

    Only thing is, I don't think they sell the filters without the cases.:(
     
  5. newtekie1

    newtekie1 Semi-Retired Folder

    Joined:
    Nov 22, 2005
    Messages:
    20,033 (6.15/day)
    Thanks Received:
    6,095
    Yeah, but really, even with the mesh, you are only loosing 2-3CFM, I can live with that if filters give about the same drop.
     
    Crunching for Team TPU 50 Million points folded for TPU
  6. MT Alex

    MT Alex

    Joined:
    Aug 2, 2009
    Messages:
    3,793 (1.99/day)
    Thanks Received:
    1,627
    My 700D has the honeycomb top where my 360mm rad is mounted, and it's loud as hell, very annoying. I've always wanted to cut it out, but I don't think black wire grills would look as nice.
     
  7. entropy13

    entropy13

    Joined:
    Mar 2, 2009
    Messages:
    4,946 (2.40/day)
    Thanks Received:
    1,212
    :confused:

    Are the filters Silverstone sells (two kinds, magnetic and loose swirl, and screw-on and honeycomb; magnetic only comes in 120/140mm while the screw-on have 80/120/140mm sizes) different from the filters you are talking about?
     
  8. FordGT90Concept

    FordGT90Concept "I go fast!1!11!1!"

    Joined:
    Oct 13, 2008
    Messages:
    13,787 (6.26/day)
    Thanks Received:
    3,671
    Location:
    IA, USA
    Figured wired would be best and turbine would be worst. Turbine has the most surface area blocked and blocked creates drag and drag is bad.

    Wired is awesome because they are round which tends to reduce drag. If all styles were rounded, the one with the biggest holes would win.
     
    Crunching for Team TPU
  9. a111087

    a111087

    Joined:
    Apr 2, 2007
    Messages:
    2,760 (1.00/day)
    Thanks Received:
    200
    Location:
    US
    that's what she said
     
  10. a111087

    a111087

    Joined:
    Apr 2, 2007
    Messages:
    2,760 (1.00/day)
    Thanks Received:
    200
    Location:
    US
    Last edited: Oct 14, 2012
    _JP_ says thanks.
  11. LAN_deRf_HA

    LAN_deRf_HA

    Joined:
    Apr 4, 2008
    Messages:
    4,555 (1.90/day)
    Thanks Received:
    952
    I hate the noise when I do a cheap Rosewill build. They use circles sized somewhere between mesh and the honeycomb and it makes for bad noise, and it's built into the case.
     
  12. a111087

    a111087

    Joined:
    Apr 2, 2007
    Messages:
    2,760 (1.00/day)
    Thanks Received:
    200
    Location:
    US
    Yeah, with all the progress that we see in computer industry, you would think that companies will wake up and start cutting the grills out to save us some time and effort (and metal for them to recycle).
     
  13. Completely Bonkers New Member

    Joined:
    Feb 6, 2007
    Messages:
    2,580 (0.92/day)
    Thanks Received:
    516
    I thought all enthusiasts cut out stamped grills and replaced with wire!

    Note that all these results are fan and RPM dependent. It would be good to see a review of performance at different fan speeds.

    I'm also doubting the real-world accuracy of the airflow CFM tests. I'm pleased they made a best effort at measuring it... but I don't believe that a baffle covering 50% of the "open space" e.g. angled slats, turbine and mesh could deliver 85-93% of the CFM airflow. That just isnt right. I think there are some problems with the 120mm to 80mm funnel technique that they used... which has painted a misleading picture of how these grills are really affecting airflow. Anyone with even an armchair understanding of fluid dynamics would know that changing the "r" will SIGNIFICANTLY affect airflow. Isn't the relationship r to the power 4? So between 120mm and 80mm there is a factor of 5-ish difference in airflow. Pushing 120mm airflow through 80mm means that the resulting changes in forced air-pressure/resistance might totally screw up the observed measurements.

    Anyone interested in this might want to look here http://en.wikipedia.org/wiki/Airway_resistance just to get an idea of the issues.

    Far better if they DID NOT funnel from 120mm to 80mm but had a 120mm measuring instrument... and failing that... 2x 80mm measuring instruments... and a circle to a figure of 8 tube adapter/modifier.
     
    Last edited: Oct 14, 2012

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page