Optimized Induction Airbox for Reciprocating Engines

Discussion in 'Tests and Improvements' started by Toobuilder, Feb 26, 2019.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Feb 26, 2019 #1

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    PART I – GENERAL INFORMATION

    System under Test

    2005 Harmon Rocket II / IO-540-D4A5 (260 HP Nominal). The engine is a parallel valve model in compliance with the Lycoming TCDS with the exception of the addition of piston cooling oil nozzles and a SDS EFI and ignition system.

    Problem Statement/Opportunity for Improvement

    Existing induction air inlet and airbox configuration is sub optimal for pressure recovery (Manifold Pressure (MP)) as well as filtration of airborne particulates. Airbox provides poor aerodynamic shape for effective pressure recovery and is of minimal volume to feed a 540 CI engine. Additionally, induction air inlet is close to boundary layer of spinner and within the arc of the propeller root, compromising airflow into inlet. Finally, the plenum housing is sealed to the servo via a spring loaded flat plate seal which will break seal with engine articulation. Airbox system uses a cone filter, presenting media in a concentrated footprint that increases the risk of a single contaminant event (such as snow, ice or FOD ingestion) fouling the media.

    Description of Change

    The existing airbox, servo, and 90 degree servo adapter will be removed, and a large capacity, aerodynamically optimized filter airbox housing will be fabricated and installed. The servo/throttle body will be reoriented to vertical and rigidly mounted to the sump inlet. The airbox will be attached to the servo inlet throat directly, sealed with an O ring. The O ring will allow the airbox to articulate in pitch, yaw and roll as well as translate vertically in relation to the throttle body – all while maintaining an airtight seal. Support for the airbox is provided by a semi-rigid tension element between the airbox and sump as well as a foam doughnut seal between the inlet of the cowl and end of the airbox. The design of this system allows articulation of the engine in all expected maneuvers and conditions while providing seal integrity from cowl inlet to combustion chamber.

    Verification Methodology

    Pressure Recovery – In situ flight test using direct comparison with similar make/model aircraft with original induction configuration, data comparison with historical flight test data direct measurement of pressure.

    Filtration – Analysis and observation.

    Expected Behavior

    Primary – The optimized configuration will produce higher observed MP than the original configuration in all measured flight conditions.
    The additional MP will enable a higher TAS, overcoming the additional frontal and wetted area of the cowling changes.

    Secondary – The effective inlet pressure of the new configuration will significantly exceed that found within the cooling plenum. Verification or disproval will be used to determine if the cooling plenum is a viable source of air for the induction system.

    PART II – FABRICATION AND INTEGRATION

    Plan view of original 90 degree sump adapter, servo and filter housing inside “new” filter housing mold for size comparison.

    [​IMG]

    Plug, mold and airbox half shell: Classic shallow angle divergent duct shape results in minimal internal turbulence and pressure recovery with the reduction of velocity.

    [​IMG]

    Internal configuration shown. Bellmouth entry helps the airflow “turn the corner” Not shown is a deflector wall which has been installed to prevent high velocity debris from entering inlet and impacting the filter media directly.

    [​IMG]

    Controlled articulation airbox hanger. Rod end, bolt and spring.
    Airbox neck "floats" on the O ring seal on the throttle body, so needs something to keep the airbox from falling off and dropping into lower cowl.

    [​IMG]

    Tension element supports weight of airbox

    [​IMG]

    Installed airbox shows large capacity and divergent shape to advantage

    [​IMG]
     
    Last edited: Feb 27, 2019
  2. Feb 27, 2019 #2

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    Just wanted to remind everyone that this section (unlike the member project logs) is open for comment by anyone. Got a question, idea or critical review? Bring it!

    To be continued...
     
  3. Feb 27, 2019 #3

    12notes

    12notes

    12notes

    Well-Known Member Lifetime Supporter

    Joined:
    Aug 27, 2014
    Messages:
    916
    Likes Received:
    565
    Location:
    Louisville, KY
    You mention disadvantages to the cone filter. What are you using for the new filter?
     
  4. Feb 27, 2019 #4

    Mad MAC

    Mad MAC

    Mad MAC

    Well-Known Member

    Joined:
    Dec 9, 2004
    Messages:
    500
    Likes Received:
    230
    Location:
    Hamilton New Zealand
    Can't quite tell from the last pic if you have a drain hole to let any accumulated moisture exit.

    I am not sure ether way, are you tempted to have a suck in valve for inlet / filter blockage case (the Apex Robin R21XX with a birdstrike in the air intake comes to mind)
     
  5. Feb 27, 2019 #5

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    The new filter is an oval K&N measuring 14x7x2. As a big oval, it should be less susceptible to a localized FOD event clogging the whole thing, and it also has more filter area overall than the cone that was in it before.
     
  6. Feb 27, 2019 #6

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    There will be a drain hole in a lower corner before it is complete, yes.

    There are provisions in the mold for an alternate air door. I left it sealed up for the test period so that I can evaluate how this thing works with a "perfect" airtight seal. If implemented it will open (via manual cable) in front of the filter and block the main inlet, injesting lower cowl air. Wont help if the filter is already fouled, but I think that is a fairly low risk considering this is a VFR airplane for now.
     
  7. Feb 27, 2019 #7

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    7,350
    Likes Received:
    2,117
    Location:
    North Carolina
    What about a sprung bird door? Just enough spring to resist vibration and it will open automatically should something plug the regular intake. If you need to know it has opened, a microswitch and lamp.
     
  8. Feb 27, 2019 #8

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    Considered the spring loaded door too. The original had that, in fact. But if I decide to add the door, the addition of a cable allows the door to act as a blocker in inclement weather and act as a positive lock to keep it sealed when stowed. In other words, the addition of a cable adds a bunch of utility (good ROI).
     
  9. Feb 28, 2019 #9

    TFF

    TFF

    TFF

    Well-Known Member

    Joined:
    Apr 28, 2010
    Messages:
    12,059
    Likes Received:
    3,477
    Location:
    Memphis, TN
    I think it is nicely done. The only thing that would have been nice would be more room around the filter from the walls on the back side. Probably not an issue in action. I like the shape and volume available.
     
  10. Feb 28, 2019 #10

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    Thanks.

    I completely get what you are saying but I dont think there will be any significant flow restriction through any particular area of the filter. It does not look right at first glance, but due to the large circumference (relative to the throttle body), the air should not even "feel" the media because it's moving so slowly. So no sharp corners to turn through the filter. If I had more room to go wider on the airbox, I would have gone with an even bigger filter.

    I do intend to run a test with the filter element removed to see if I get a MP change. I suspect (hope) I don't, but we will all know soon enough.
     
  11. Feb 28, 2019 #11

    TFF

    TFF

    TFF

    Well-Known Member

    Joined:
    Apr 28, 2010
    Messages:
    12,059
    Likes Received:
    3,477
    Location:
    Memphis, TN
    The only reason for more room would be so more air would be scrubbed over more of the filter all the time. The volume as is has got to be great already. It’s still going to be hard to get that filter dirty.
     
  12. Feb 28, 2019 #12

    fly2kads

    fly2kads

    fly2kads

    Well-Known Member HBA Supporter

    Joined:
    Jan 2, 2010
    Messages:
    1,498
    Likes Received:
    554
    Location:
    Justin, TX
    This is a great project! Aircraft intake systems seem to be designed more for operational and packaging considerations than performance, especially on carburetor-equipped engines. Your project is fairly similar in concept to the intake system I have been sketching for my little Supervee concept, so I am glad I'm not off in left field. (Or if I am, at least I have company!) I'm looking forward to your updates and conclusions.

    Question: did you calculate a specific target volume for the plenum, or was this driven by the space available?
     
  13. Feb 28, 2019 #13

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    Big filter means low loss. I understand that there becomes a point of diminishing returns, but the goal is to have a full time filter that the engine cant even see.

    The plenum volume was not targeted - it was driven by space available and geometry and that's as big as I could get.
     
  14. Feb 28, 2019 #14

    Little Scrapper

    Little Scrapper

    Little Scrapper

    Well-Known Member HBA Supporter

    Joined:
    Jan 3, 2014
    Messages:
    5,397
    Likes Received:
    3,318
    Location:
    Wisconsin
    You completely knocked it out of the park on this first thread in the new section. Well done.
     
    Topaz likes this.
  15. Feb 28, 2019 #15

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    Thank you. I hope to continue the trend with Part III - Test and Evaluation, and Part IV - Final Report.

    I have plenty of other ideas to test, but I look forward to other's contributions as well.
     
    Little Scrapper, rv6ejguy and Topaz like this.
  16. Apr 30, 2019 #16

    dtnelson

    dtnelson

    dtnelson

    Member

    Joined:
    Apr 30, 2019
    Messages:
    22
    Likes Received:
    22
    Awesome! As a retired engineer, I particularly appreciate the use of the good 'ol scientific method. I'll be very interested in your specific measurement techniques! And, Timely! I'm in the process of putting in an updated induction air system in my Long-eze project (starting with a Vans induction box).

    One comment; and I know this is highly debatable; I've never had an air filter on my Velocity (IO-360, 1500+ hours). I'm not bragging about this... but in my installation I've had no issues (and yes, I know it only takes one...). My induction air comes in via the roof mounted NACAs which helps to limit the normal runway debris. Also, FWIW, I've always done regular oil analysis, and I've never had high silicas (i.e., not much dirt in the oil). Again, I'm not making a recommendation here, only an observation.

    Looking forward to your next installment!

    Dave
     
  17. Apr 30, 2019 #17

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    Thanks Dave. I know plenty of people get by with no filtration but considering my environment I just couldn't let myself go down that path. Part of this experiment will determine if I really have achieved "no loss" filtration.
     
  18. May 2, 2019 #18

    Victor Bravo

    Victor Bravo

    Victor Bravo

    Well-Known Member

    Joined:
    Jul 30, 2014
    Messages:
    6,594
    Likes Received:
    5,375
    Location:
    KWHP, Los Angeles CA, USA
    Really impressive looking to my amateur eyes. Questions:

    Is that the textbook 7 degree divergent angle from the inlet to the pressure recovery plenum? Would this duct benefit from flow straignteners or "splitter" baffles to reduce the angle of divergence that the air "feels" as it travels through the duct?

    Other than ease of mfg. the plug and mold, is there an aerodynamic reason you made the cross section of the duct rectangular instead of oval or matching the inside contour of the cowling? Don't the corners create a little disturbance or change in velocity?

    Did you "clock" the inlet and propeller to time the pressure wave off the back of the propeller into the intake port?
     
  19. May 2, 2019 #19

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,568
    Likes Received:
    3,407
    Location:
    Mojave, Ca
    The divergence is actually less than 7 degrees so no splitters are required. The square corners are for manufacturing ease as well as provisions for a possible alt air door. The square corners only exist at the rear portion of the airbox where the local velocity is low. The high velocity forward section is circular in cross section.

    There is no attempt to time the blade passage with an individual cylinder aspiration event. The large volume of the airbox (and the future installation of the 300 HP sump) should serve to average out any prop pulse.
     
  20. May 2, 2019 #20

    dtnelson

    dtnelson

    dtnelson

    Member

    Joined:
    Apr 30, 2019
    Messages:
    22
    Likes Received:
    22
    Just as a point of interest, in the pusher world, there have been some measurements looking for a pressure wave caused by the prop. None was found. Instead, surprisingly, the air is mostly static right behind the prop (especially inboard). This is one of the reasons that on the faster 'ez's, you'll usually see the exhaust kept completely inside the cowl, and cut off 3 or more inches inside the cowl edge. The exhaust obviously creates a strong flow, which augments the flow out of the back of the cowl, resulting in better cooling.

    Dave
     

Share This Page

Group Builder
arrow_white