Flow in a duct - question

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by BJC, Nov 7, 2018.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Nov 7, 2018 #1

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    8,899
    Likes Received:
    5,748
    Location:
    97FL, Florida, USA
    A question about flow in a duct; in this case, a duct under the fuselage, just aft of the firewall, exhausting engine cooling air.

    The crossection of the duct fairs into the belly of the fuselage as shown in Design A. Design B has slightly less area, due to squaring the inside corner where the duct attaches to the fuselage. The external surface is not drawn in Design B, but it, too would be faired into the fuselage belly.

    Does the acute angle between the duct and the fuselage in Design A cause more drag than the 90 degree angle in Design B?

    If so, is the drag difference enough to allow Design B to flow more air than Desigh A, given the same pressure drop in the duct?

    Thanks,


    BJC
     

    Attached Files:

  2. Nov 7, 2018 #2

    Lucrum

    Lucrum

    Lucrum

    Well-Known Member

    Joined:
    Jun 10, 2008
    Messages:
    955
    Likes Received:
    188
    Location:
    Canton, GA
    My sense tells me that B would be less drag
     
  3. Nov 7, 2018 #3

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    8,899
    Likes Received:
    5,748
    Location:
    97FL, Florida, USA
    Thanks, mine too; but that is why I asked the question. I am looking for verification.


    BJC
     
  4. Nov 7, 2018 #4

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    5,998
    Likes Received:
    2,000
    Location:
    Rocky Mountains
    Ditto - but the overall dimensions/ratios, if significantly different than shown, may trump our sense. Unless it'a a race plane, or the cooling has shown to be marginal - what ever gives the best ratio between looks and ease of construction is how I'd roll. = B
     
  5. Nov 7, 2018 #5

    Dana

    Dana

    Dana

    Moderator

    Joined:
    Apr 4, 2007
    Messages:
    8,457
    Likes Received:
    2,855
    Location:
    CT, USA
    Unless the corner radii are huge, I don't think you'll see a measurable difference between the two. Any flow in the corners will be slowed anyway. Ideally you'd put a radius on the inside, too.
     
  6. Nov 7, 2018 #6

    Sockmonkey

    Sockmonkey

    Sockmonkey

    Well-Known Member

    Joined:
    Apr 24, 2014
    Messages:
    1,568
    Likes Received:
    403
    Location:
    Flint, Mi, USA
    The taper of the duct is more critical anyhow.
     
  7. Nov 7, 2018 #7

    Lucrum

    Lucrum

    Lucrum

    Well-Known Member

    Joined:
    Jun 10, 2008
    Messages:
    955
    Likes Received:
    188
    Location:
    Canton, GA
    Norman likes this.
  8. Nov 7, 2018 #8

    Jay Kempf

    Jay Kempf

    Jay Kempf

    Well-Known Member

    Joined:
    Apr 13, 2009
    Messages:
    3,531
    Likes Received:
    865
    Location:
    Warren, VT USA
    Tricky question. Probably you have to ascertain first whether the entire flow at the wall is turbulent meaning above the Re threshold for that particular duct. If so then it won't matter. If below you might find some flow anomalies in the corner which would drive the corner turbulent and that would reduce the effective area. Most likely the corner will just stagnate so no difference that you could measure.
     
  9. Nov 8, 2018 #9

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    5,911
    Likes Received:
    1,504
    Location:
    North Carolina
    Radius both inside and outside for lowest drag.
    I have a feeling that A may produce the lowest total drag, though higher duct flow drag. This will depend on the velocities. You want to radius the lip to counter imperfect flow conditions. There are some NACA and RAE papers on intakes I haven't read in a few years. Kuchemann and Webers 'aerodynamics of propulsion' has a good section intakes, but is a $100 used book.
     
    mcrae0104 likes this.
  10. Nov 8, 2018 #10

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    11,420
    Likes Received:
    2,096
    Location:
    Port Townsend WA
  11. Nov 8, 2018 #11

    TFF

    TFF

    TFF

    Well-Known Member

    Joined:
    Apr 28, 2010
    Messages:
    11,139
    Likes Received:
    3,017
    Location:
    Memphis, TN
    If it's an actual duct B but The transition inside the cowl would have to take advantage of it. Area if exit is going to be prime. If it's just the exit at the firewall with no real length to the duct, I might say A is better.
     
  12. Nov 8, 2018 #12

    gtae07

    gtae07

    gtae07

    Well-Known Member

    Joined:
    Dec 13, 2012
    Messages:
    1,731
    Likes Received:
    1,031
    Location:
    Georgia
    BJC likes this.
  13. Nov 8, 2018 #13

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    8,899
    Likes Received:
    5,748
    Location:
    97FL, Florida, USA
    Thanks ‘07. Lots of interesting stuff there.

    I’ve been planning on many of those featured. What is hard to find (specifically related to the acute angle verses right angle duct wall juncture) is reliable data on actual drag reduction or cooling improvement for each individual change.


    BJC
     
  14. Nov 8, 2018 #14

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,550
    Likes Received:
    2,577
    Location:
    Calgary, Alberta, Canada
    I am assuming this is an exit duct for cooling air. The flow is almost certainly turbulent with all the junk forward of this and from my in-flight tuft testing. My guess is that there would be no measurable difference in drag between the 2 designs if they have the same exit area. As Dan Horton proved with his variable exit design, lowest drag is achieved by restricting the exit area and recovering some of the cooling air momentum.
     
    BJC likes this.
  15. Nov 8, 2018 #15

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    8,899
    Likes Received:
    5,748
    Location:
    97FL, Florida, USA
  16. Nov 8, 2018 #16

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    8,899
    Likes Received:
    5,748
    Location:
    97FL, Florida, USA
    Yes,
    I followed the links that gtae07 provided and found Dan’s postings. He has done lots of interesting experimenting. If he were to document it in a book, with measured results for each change, I would buy a copy.


    BJC
     
  17. Nov 8, 2018 #17

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,287
    Likes Received:
    3,034
    Location:
    Mojave, Ca
    I'm doing a similar thing with the exit on the Rocket. Mine is trapezoidal in area and spreads across the width of the fuselage. The reason is so that the OML will form a neat 90 degrees with the gear legs, significantly reducing my need for upper leg cuff fairings. Anyway, this does make things very tight in the corners of the IML just like your example. My initial thought is to make the IML a 90, just as you are thinking. Seems to me this is covered in Hoerner's Fluid Dynamic Drag book? I have a copy, but have not looked this up yet. Will let you know if I find anything.
     
    BJC likes this.
  18. Nov 10, 2018 #18

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,287
    Likes Received:
    3,034
    Location:
    Mojave, Ca
    Well, I dug through my copy of Hoerner and I can't find anything that addresses this issue directly. Plenty of info on ducts, protuberances, intersection drag, etc... Stuff we already know, but not this exact scenario. All the available info leads me to believe this severe angle is "bad", but I can't point directly to an experiment that compares multiple options.
     
  19. Nov 10, 2018 #19

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    8,899
    Likes Received:
    5,748
    Location:
    97FL, Florida, USA
    Too:

    Thanks for checking.


    BJC
     
  20. Nov 10, 2018 #20

    Swampyankee

    Swampyankee

    Swampyankee

    Well-Known Member

    Joined:
    Dec 25, 2015
    Messages:
    1,389
    Likes Received:
    355
    Location:
    Earth USA East Coast
    Inside the duct, lower internal losses would be the one with the least internal perimeter for the required area, as long as the internal expansion angles don’t cause separation. Those acute internal angles in A are not optimal.
     
    BJC likes this.

Share This Page

arrow_white