+ Reply to Thread
Page 1 of 2 12 LastLast
Results 1 to 15 of 20

Thread: Exhaust Augmenter / Cooling system

  1. #1
    Site Developer Jman's Avatar
    Join Date
    Oct 2002
    Location
    Pacific NW, USA!
    Posts
    2,387

    Exhaust Augmenter / Cooling system

    Anyone using one in there aircraft? I'm doing some research on them and I was hoping someone has some information based on some real-world experience. For those who arn't sure what exhaust augmenters are, here is a description:
    http://bd-4.org/index.html?/cooling_nicoson.html

    Basically, it is a device postioned at the cowling outlet that, when used in tandem with the engine exhuaust, serves to suction cooling air through the cowling and out the outlet. I'm interested in them because I plan on building a canard pusher with a liquid cooled engine. If the augmenters are as effective as they are said to be, this may be an ideal way to help cooling while in groud operations and slow climb-outs.

    If anyone has any pictures or information I would really like to hear about it. Thanks.

    Jake

  2. #2
    Super Moderator orion's Avatar
    Join Date
    Mar 2003
    Location
    Western Washington
    Posts
    5,945
    Hi Jake;

    Interesting link however it should be read with caution. While his discussion on rules of thumb and the augmentor are in the ballpark, his initial discussion is off base.

    The factor that is primary in determining your cooling efficiency is that of the exhaust flow, not the inlet. The rule of thumb I was taught many years ago is that you should always consider an inlet as just that: It "lets" in air, you cannot actually ram it through.

    The amount of air that actually moves through the engine is a function of the cooling fins and the exhaust geometry. Furthermore, the idea that on the ground the prop directs air in is also not quite correct. For constant speed props, the root of the blade is not a very good airfoil and therefore it does not contribute all that much to "pushing" air through.

    For fixed pitch props, while their root geometry is somewhat better, at idle they move very slowly at the base of the blade and thus again are not all that good at directing a substantial amount of air through the engine.

    For comparison, look at the new designs such as the Glasairs - the inlets are very small. Look specifically at the geometry of the exhaust portion of the cowl.

    One other point he forgot to mention - augmentor tubes are really efficient at cooling but usually at only a narrow band of operation. In order to do a good design, it is important to select which part of the flight you want to emphasize. If for instance you want good cooling at climb, you will not have as good an effect at idle (cruise rarely designs the cooling characteristics). When designing an augmentor, design specifically for the condition that you determine most critical for your application.

    Also, don't necessarily stick to standard components. He used a 2.75" radiator core. While that is a good selection and should work for a wide varitey of applications, if you're space limited, you may want to go to a deeper heat exchanger instead. The cooling effectiveness is not linear but using a deeper core geometry can save you valuable space. Personally, I like looking at cores about four inches deep.

    His comments on the inlet flow diffuser geometry however are right on - you do not want to diverge more than about six degrees as that may cause an undue amount of boundary layer separation and subsequent turbulence. Most general aviation configurations have a very poor diffuser at the inlet and thus lose a significnt amount of potential effectiveness due to the separation.

    Also, I too would not recommend diffusing the exhaust flow. One, it will create geometry drag and two, when the slower air mixes with the high speed air on the outside of the aircraft, it can set up a very turbulent condition and subsequent drag. Ideally, one wants to accelerate the air coming out so the "delta V" of the two air streams is as close as possible. This however is very difficult to achieve although with an augmentor it could be optimized more so than without one.

    And one last point - the biggest mistake people make is keeping the bottom of the firewall square with the exhaust air duct. Air flowing out will try to round the corner and will often seperate, dramatically reducing the effectiveness of the cooling geometry. To fix this, it is important to remember that this is a duct and thus, to control the flow, it too must have some sort of smooth inlet.

    To do this, install a rounded lip so that the air can attach itself smoothly and flow out with a minimal amount of turbulence. The diameter of the lip will depend on the available space but the larger the better. Grummans (the general aviation models) for instance have incorporated a semi-circular tube lip that wraps around the nose gear torque tube (mounted transverse across the firewall bottom). It is about two inches in diameter and attaches onto the firewal above the torque tube and at the fuselage bottom. Personally, I would normally like to see a larger diameter but the mod was very effective here in that it solved virtually all the cooling problems the airplane had early on.
    Last edited by orion; June 29th, 2003 at 10:17 AM.

  3. #3
    Site Developer Jman's Avatar
    Join Date
    Oct 2002
    Location
    Pacific NW, USA!
    Posts
    2,387
    Orion,

    Thank you for the thorough reply. This is a fascinating subject and I am attempting to get a hold of as many technical papers and articles on the subject as I can. There are some interesting NACA papers as well as some articles in Contact! That I am trying to track down for study. It would seem that I have a LOT to learn on the subject.

    Here are a couple of questions you brought to mind.

    His comments on the inlet flow diffuser geometry however are right on - you do not want to diverge more than about six degrees as that may cause an undue amount of boundary layer separation and subsequent turbulence. Most general aviation configurations have a very poor diffuser at the inlet and thus lose a significant amount of potential effectiveness due to the separation.
    Could you comment on the use of NACA recessed inlets and their use with heat exchangers. The general consensus seems to be that they really do not work for trying to pass air through a core. Iíve read that the pressure build up in front of the core causes problems with the effectiveness of the NACA inlet. Do you think that the use of a well designed augmenter could help alleviate this problem?

    In order to do a good design, it is important to select which part of the flight you want to emphasize.
    I assume that ground operations and climb-out would be the area of emphasis for me. I plan on building the Cozy MKIV with a Mazda Rotary. I would love to be able to use the stock NACA scoop and Iím hoping that the augmenter would help suck air through the core during low airflow operations. What do you think?

    The cooling effectiveness is not linear but using a deeper core geometry can save you valuable space.
    Good point. I will have to find data that will tell me just how much effectiveness will be lost the thicker the core gets. Would that number be a linier number that would have a rule of thumb associated with it?

    Thanks for your response and your continued support of this board!!

    Jake

  4. #4
    Super Moderator orion's Avatar
    Join Date
    Mar 2003
    Location
    Western Washington
    Posts
    5,945
    Hi Jake;

    NACA and Contact! are probably the best sources of information available to the general public. This is especially true for issues such as cooling as there are not too many good technical publications that deal specifically with this subject otherwise.

    When using recessed (or flush) inlets, keep in mind that these have no capacity to develop any "ram" pressure. Also, if they are installed on an unfavorable surface gradient, they may actually have just plain lousy performance. This also applies when these inlets are mounted far aft on a body and thus are submerged in a pretty substantial boundary layer.

    Yes, creating suction with an augmentor will help but you will still have to pay attention to the particular requirements of the inlet. What some folks have done to make them work better in less than ideal circumstances is to incorporate a raised lip. This lip is created so that it sticks up a bit from the OML of the shape (that's Outside Mold Line to those who are not familiar with the term) so that the inlet has the benefits of the flush shape yet it can still generate a bit of ram pressure.

    Suction aft of the core is still the key item in determining the performance of your cooling installation.

    As far as the operational issue is concerned, I'd recommend designing the system for climbout. For ground operation, install an electrical fan just like you would in an automotive application. Yes, you now have to account for the airflow loss due to the fan when it's not operating, but it is the simplest solution to ground overheating when the cooling system is designed for a flight condition.

    As far as the radiator core is concerned, a rule of thumb is that by doubling the thickness, you can only increas the cooling ability by about 1.5 or so. This however varies depending on what the original thickness is that you are comparing against. The number works only if you are making a comparison of an original core of about 2 inches deep, maybe a hair more. Beyond that point, the air going through gets saturated (it can absorb only so much heat) and so extra thickness only creates duct losses and results in only a small benefit, if any.

    In my experience, four inches deep is pretty much a practical limit however for higher speed applications, I've looked at using as much as six inches.

    Bill

  5. #5
    Site Developer Jman's Avatar
    Join Date
    Oct 2002
    Location
    Pacific NW, USA!
    Posts
    2,387
    Thanks for that additional info. The Cozy has been using the recessed inlet with success in air-cooled applications but I havenít researched how many are flying with the stock inlet and a liquid cooled engine installation.

    I've seen one installation that was having cooling problems and the builder put two vortex generators in front of the inlet and was able to see improved cooling performance. Off hand I don't remember if this was in an air-cooled or liquid-cooled installation.

    In the case of a pusher, does it make more sense to try and duct all the way out the cowling or to dump the air coming from the radiator into the cowling and allow it to find its own way to the outlet / augmenter. It would seem to me the first option would be more efficient but more complicated to fabricate.

    Anyway, thanks for the comments, they are very helpful.

    Jake

  6. #6
    Registered User robert.lees's Avatar
    Join Date
    May 2003
    Location
    Leicester, UK (EGBG)
    Posts
    61
    Jake, in 1961 the one and only Auster Mk 11 was built with augmenters. It still flies (I saw it last weekend). 260hp, it is the final derivative of the humble Taylorcraft, which started out as 40hp.

    The attached photo is from the Farnborough Airshow, 1961.

    Rob
    Rob Lees

  7. #7
    Registered User robert.lees's Avatar
    Join Date
    May 2003
    Location
    Leicester, UK (EGBG)
    Posts
    61
    OK, picture too large...

    ...and I can only post every 120 seconds? Why?
    Attached Thumbnails Attached Thumbnails Exhaust Augmenter / Cooling system-austeraopmk11farn1961.jpg  
    Rob Lees

  8. #8
    Registered User robert.lees's Avatar
    Join Date
    May 2003
    Location
    Leicester, UK (EGBG)
    Posts
    61
    Rob Lees

  9. #9
    Site Developer Jman's Avatar
    Join Date
    Oct 2002
    Location
    Pacific NW, USA!
    Posts
    2,387
    Thanks for the info and links Robert. This is a very interesting subject to me and any info about an actual flying example is helpful.

    The reason there is a mandatory 120 second buffer between posts is because I had a web robot cover this site with porn spam advertisements not too long ago. These robots can post spam just as fast as the server will let them amd by putting a 120 second buffer between posts this limits the robots ability to post to 30 posts an hour. This will give me a much more manageable mess to clean up when I finally notice what is going on.

    Jake

  10. #10
    Site Developer Jman's Avatar
    Join Date
    Oct 2002
    Location
    Pacific NW, USA!
    Posts
    2,387
    One other issue I am looking at is that I am hoping to use a turbo charger in my installation (I know, I want it all ) but I am concerned that the turbo will slow the velocity of the exhaust gasses to the point that it will make the augmenter less effective. Does anyone understand how the exhaust gas velocities will change with the addition of a turbo?

    Jake

  11. #11
    Registered User StRaNgEdAyS's Avatar
    Join Date
    Oct 2003
    Location
    Northern NSW Australia
    Posts
    822
    The engines I m working on make use of augmenters, though I will not be testing their alleged efficiency until I get into some larger scale models, but here's what my information has to say on them....
    Augmenter
    An augmenter is simply a duct positioned behind the engine that allows air to be drawn into the duct where it is heated by the exhaust from the engine.
    Because the exhaust is very hot, the cool air that was drawn intothe augmenter is also heated and expands rapidly. This expansion causes an increased pressure and when properly harnessed additional thrust.
    By making the duct expand towards the back, the pressure causes air to be forced out of the back to increase thrust. An augmenter can be as simple as a single divergent cone, however, to draw the largest volume of air into the engine there should be a convergent intake cone positioned in the front of the second divergent cone.
    The intake angle of the first cone should be less than 45 degrees. The divergent section should be at least 3 times longer than the intake. To keep drag down, the intake and exhaust diameter should be the same, but for a static augmenter this has little impact.
    Aluminum should not be used for the tailpipe, but is acceptable for making an augmenter since it will be cooled by incoming air.
    Not sure if it helps, but there you have it.
    Life is short,
    Live it HARD
    Live it FAST

  12. #12
    Super Moderator orion's Avatar
    Join Date
    Mar 2003
    Location
    Western Washington
    Posts
    5,945
    The augmentor, as discussed here, is primarily to aid the extraction of air from the engine cowling in order to improve cooling with a minimal amount of drag. Thrust is not a consideration.

    Regarding the use of an augmentor along with a turbo, you are correct in your assumption in that the tubo will extract energy from the flow, thus slowing and cooling the exhaust, making the augmentor less effective. However, only testing the installation will most likely give you the level of the reduction of the effect.

  13. #13
    Registered User
    Join Date
    Jan 2003
    Location
    West Palm Beach, FL
    Posts
    167
    I am planning to adapt this technique in my turbo rotary Cozy IV installation. The exhaust from the turbo will be a 20 inch straight 2.5 inch SS pipe directly out of the turbo to just short of a 4 * 6 inch exit oval in the back of the cowl. I plan a duct from the NACA to the "front" of the turbo, expanded around the turbo heat shroud, then surrounding the exhaust as it goes to the cowl. This SS shroud will mate with the cowl protruding an inch or so beyond the end of the exhaust pipe.

    I'm hoping that the suction and air movement generated in the outer pipe will provide two benefits - increased airflow and better cooling during taxi, and better protection in the cowl from the heat of the turbo.
    I'm also hoping that the mixing of exhaust and "clean" air at the cowl exit will help protect the wood 3 blade prop from heat.

    I'll be building the above during the next few weeks, so I'd be interested in any suggestions to help make this scheme work better.
    Flying Cozy IV Turbo Rotary

  14. #14
    Super Moderator orion's Avatar
    Join Date
    Mar 2003
    Location
    Western Washington
    Posts
    5,945
    Your reasoning sound good - actually, the turbo will take a bit of heat out of the exhaust so, coupling this with the cooling air, the overall configuration should go a long way in protecting your prop.

    The only thing I'd be curious about would be the behavior of the prop itself. The air coming out of the cowl will be of sufficiently lower density than the surrounding atmosphere. As the prop swings around, its perfomrance will vary as it passes through the different flow densities. On some installations in the past, this variation has caused some problems and, as you may know, this is one of the main reasons wooden props are recommended as they can better deal with the possible resonsnce and vibration that can set us as a result.

  15. #15
    Registered User
    Join Date
    Jan 2003
    Location
    West Palm Beach, FL
    Posts
    167
    Good point, Orion. The issue of mixing different airflow densities is something I was expecting. We get that anyway with the pusher configuration. I made the cowl curved so that the outlet is at an angle to the airflow and about 8 inches forward of the prop. Maybe I'll get some mixing before it all hits the prop. We'll find out soon enough.

    You can see pictures at the end of http://www.kgarden.com/cozy/cowl.htm

    Thanks for the confirmation on the general principles. The design is mainly somewhat educated guesswork compiled from what I've learned from other lists while building.
    Flying Cozy IV Turbo Rotary

+ Reply to Thread
Page 1 of 2 12 LastLast

Similar Threads

  1. Automated Assisted Flight System
    By jlong in forum For Sale / Wanted
    Replies: 1
    Last Post: April 3rd, 2006, 06:42 AM
  2. Automted Assisted Flight System
    By jlong in forum Instruments / Avionics / Electrical System
    Replies: 0
    Last Post: March 4th, 2006, 05:24 PM
  3. emergency blowdown system
    By mreno in forum Firewall Forward / Props / Fuel system
    Replies: 8
    Last Post: February 5th, 2005, 09:24 PM
  4. Exhaust system alloy
    By Dust in forum Firewall Forward / Props / Fuel system
    Replies: 17
    Last Post: December 1st, 2004, 04:14 AM
  5. Fabricating an Exhaust System
    By Frang_TX in forum Firewall Forward / Props / Fuel system
    Replies: 0
    Last Post: November 10th, 2004, 08:34 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts