Scaled Jets (F-16)

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by Acrojet, Jun 6, 2016.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Jun 8, 2016 #101

    DangerZone

    DangerZone

    DangerZone

    Well-Known Member HBA Supporter

    Joined:
    Sep 5, 2011
    Messages:
    2,107
    Likes Received:
    370
    Location:
    Zagreb HR
    Jay, please notice that it is not my project but Peter's (user Acrojet), I was merely commenting an information provided on this thread.

    In terms of scaling, the KAI FA-50 is basically a lighter weight 85% scale F-16. Flies fine, has solid supersonic performance, weighs tons more than Peter's plane. At 900 lbs empty mass and 2100 lbs MTOW, the wing loading of such an aircraft is still very low even if he'd further reduce the wing area. At 60% the aircraft is still pretty huge and at the planned MTOW he could do fine even in a 50% scale.

    You are right about the giant inlet below the fuselage, it would ingest debris, gravel, dust and dirt from airfield runways. Your suggestion about a split inlet like on the FA-50 might indeed be a better solution, it's a good idea and up to Peter to see. I heard a rumor that the side inlets on the FA-50 was a suggestion from Lockheed Martin or some other American experts working on the project, to reduce the water/dust/dirt ingestion on wet runways that used to be a problem on the F-16 original design. This would also ease the design of the front retracting landing gear, so the FA-50 is indeed an F-16 with most design flaws resolved.
     
  2. Jun 8, 2016 #102

    Acrojet

    Acrojet

    Acrojet

    Well-Known Member

    Joined:
    Jun 6, 2016
    Messages:
    53
    Likes Received:
    51
    Location:
    Gardiner, Maine USA
    Well actually it was an "aft-fan" originally, but it was determined to be simple-yes, but the loss in efficiency, especially for such a small motor was too great. So it's been decided to move the fan to the front. It will be a little more complicated, but in the end still pretty simple and the efficiency is way better due to the fan "pre-pressurizing" the front of the compressor section thereby increasing compression ratio and giving better fuel specifics..
    Very similar to the Williams EJ22
     
  3. Jun 8, 2016 #103

    DangerZone

    DangerZone

    DangerZone

    Well-Known Member HBA Supporter

    Joined:
    Sep 5, 2011
    Messages:
    2,107
    Likes Received:
    370
    Location:
    Zagreb HR
    User Lucrum mentioned using a T-58 engine and removing the aft shaft part, would this be the same thing you had on your mind?

    2-general-electric-t58.jpg

    https://en.wikipedia.org/wiki/General_Electric_T58

    That's one powerful engine, installing a well designed composite fan (like some of those on modern turbofan engines) might show good performance for the desired subsonic speed range.
     
  4. Jun 8, 2016 #104

    Acrojet

    Acrojet

    Acrojet

    Well-Known Member

    Joined:
    Jun 6, 2016
    Messages:
    53
    Likes Received:
    51
    Location:
    Gardiner, Maine USA

    Man I looked half scaling to keep the weight down for the same motor, and it was just too darn small, especially if I was going to have a "jump-seat" to carry another person. So even 60% was a compromise. Once I looked at 2/3 or even 3/4 scale it became evident that was too big for my little efficient 650-700lb thrust motor to work. That meant going to a Modified GE-T-58, Marbore ll/ V or heaven forbid, the GE J-85/CJ-610. That takes it out of my "wheel house" and puts this into another catagory of jet..... Too heavy, too expensive, too much fuel burn, increased complexity, etc.....
     
  5. Jun 8, 2016 #105

    Acrojet

    Acrojet

    Acrojet

    Well-Known Member

    Joined:
    Jun 6, 2016
    Messages:
    53
    Likes Received:
    51
    Location:
    Gardiner, Maine USA

    Too big, too heavy and too much work to convert. I don't have the numbers in front of me, but I think that motor weighs roughly 240lbs and burns 125 gal/hr for 800-1000lbs of thrust and my little Tfan weighs roughly 90lbs and burns roughly 45gal/hr....for 650-700lbs of thrust.... No comparison. And if this motor doesn't pan out my model doesn't work..... ;-(
    So let's keep our fingers crossed and hope it all works out. The good thing is, the core is already built and running well now, so now it's the fan added and we're in business!
     
    Battler Britton likes this.
  6. Jun 8, 2016 #106

    Doggzilla

    Doggzilla

    Doggzilla

    Well-Known Member HBA Supporter

    Joined:
    Jun 2, 2013
    Messages:
    1,888
    Likes Received:
    396
    Location:
    Everywhere USA
    The problem with engines is that they are charging prices for hand made parts, yet everything is CNC now. There isnt anything required that cannot be bought off the shelf.

    The controls should probably come from turboshafts, as those are the most reliable systems from this power range, but otherwise the parts can be manufactured fairly easily to the standards we are looking for.

    The most expensive single tool somebody would need is the balancing machine, which is like $50,000 used, and which needs programming to be able to do whatever specific balancing task is required. These machines are what makes it possible to make small engines and turbochargers. The same machines balance full size turbines. It is the most important tool. They spin the part and then tell you which part needs a little shaved off of it, which allows you to make up for manufacturing imperfections. This dramatically improves the reliability of components, and is basically required.

    The parts can be made on CNC or cast, then they are balanced using the balancing machine to make up for any imperfections. This greatly improves reliability and simplicity.

    Controls are the most important component and so Im going to say its best to simply buy existing controls that are proven, most likely from turboshafts. Having familiar controls means that mechanics will be more confident working on them and be much more trusting of the engines. If new controls are developed, mechanics may refuse to work on the engines because they are unfamiliar and seem untrustworthy. Using existing proven control equipment will give people much more confidence in the engine, since it will seem much more similar to them. Thats really the only components I really think needs to be bought off the shelf.

    Components like combustion chambers can be copied from large turboshafts if required. Since there arent many from pure jet engines.
     
  7. Jun 8, 2016 #107

    DangerZone

    DangerZone

    DangerZone

    Well-Known Member HBA Supporter

    Joined:
    Sep 5, 2011
    Messages:
    2,107
    Likes Received:
    370
    Location:
    Zagreb HR
    Yeah, weight is an issue too, a 50% scale would be better for a single seater but a tandem seater surely allows more comfort at 60%. Did you scale the wings exactly as in the original F-16 design or did you additionally adapt them for better efficiency of the chosen airfoil?

    A Czech company was offering some turbojet engine projects a couple of years ago at the Friedrichshafen Expo. An engine looked like a larger size TJ100 of PB, can't remember the name. Anyway, the fuel consumption was horrific, so your figures of 45gal/h at 650-700lbs are really looking good.

    Are the fan blades gonna be metal or composite?
     
  8. Jun 8, 2016 #108

    nerobro

    nerobro

    nerobro

    Well-Known Member

    Joined:
    Nov 6, 2011
    Messages:
    1,109
    Likes Received:
    312
    Location:
    Northern Illinois
    Scaling jet engines doesnt' work well. The pressure ratios are pretty low, which leads to really bad fuel economy.

    I wonder how much recouperators could affect that..
     
    BoKu likes this.
  9. Jun 8, 2016 #109

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,542
    Location:
    Rotterdam, Netherlands
    Any hands-on experience with CNC?

    A typical blisk for such a tiny engines will have dozens of machining hours. A 5-7 axis CNC machine that can do that doesn't run for less than a few hundred bucks an hour. Plus CAM patterns. By the same multiple blisks are machined, you're tens of thousands of dollars down, and that's just the shaft and compressor/turbine.

    Just try finding a machine that can do nickel steels is a major challenge.

    A double shaft and it gets way more complicated.
    Injectors, ignitors, flame holders, all specialized materials. Tubing for high temps. Bearing for those kind of speeds.
    Forget about casting. You have to machine the final part, since casting is about 2 orders of magnitude too rough.

    Seals? Biggie too for such an application.

    Tolerance stacking gets a major nightmare on such parts.
    You need a unique one. An Arduino or such would work, but just finding the sensors is going to be very $$$.


    I think doing a small jet engine is incredibly hard. The FJ22 is an amazing achievement, but it's telling that even the best experts couldn't get it to a satisfactory stage after cubic feet of dollars were spent on it.
     
    gtae07, DangerZone, casperbd and 3 others like this.
  10. Jun 8, 2016 #110

    BoKu

    BoKu

    BoKu

    Well-Known Member HBA Supporter

    Joined:
    Aug 16, 2013
    Messages:
    2,322
    Likes Received:
    2,904
    Location:
    Western US
    When I was interviewing with Eclipse, there was a display case showing a bunch of the EJ22 compressor and turbine parts. It was, without exaggeration, an absolute jewel, like something from the watchmaking world championships. I've worked in a Silicon Valley machine shop that specializes in tiny work, and spent a lot of time drilling and tapping holes for #0-80 screws. The EJ22 was like the best of what we'd ever done there, but in materials known for their toughness and intractability.

    Of course, what the display didn't show was what we now know to be its "secret sauce;" that it was in fact a three-shaft motor--a feature normally found only in the largest turbofans. So in addition to all the tiny compressor and stator and turbine blades, there are three tiny concentric shafts, and three sets of tiny bearings. And as we also know now, there was very little to it that could be serviced in the field, so every minor failure means a trip back to Merry Olde Englande. After a while, the shipping company just started leaving one of their jets on the ramp outside the Eclipse hangar because they knew there'd be another overnighter to Williams not many days hence. Given that the Eclipse 500 only flew once or twice with those engines, it is probably not too far off to guess that they accumulated 100x the miles being carried to the factory for service than they did actually propelling an airplane.

    Thanks, Bob K.
     
    Glider, bmcj and rv6ejguy like this.
  11. Jun 8, 2016 #111

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,788
    Location:
    Calgary, Alberta, Canada
    It seems to me a low cost turbofan would be better off using twin centrifugal compressors (staged) rather than multiple axial stages. Way cheaper to machine and assemble and would get you to the 9-10 pressure ratio range. A geared fan would massively increase cost and complexity of design/manufacturing.

    No problem to find CNC machines capable of machining nickel alloys. Ceramic tools can wail on Inconel or hard 17-4 at amazing rates, but that machine time will be expensive for sure. Luckily we're talking pretty small parts here for this thrust range so material costs and machining time is lots less than 5000 lb. class engines.

    IMO, getting a design that works well and reliably with decent SFC and surge margins will be a lot harder than machining the parts for it.

    The key to this whole project is the engine and that's going to be the hardest part too I think.
     
    BoKu likes this.
  12. Jun 8, 2016 #112

    BoKu

    BoKu

    BoKu

    Well-Known Member HBA Supporter

    Joined:
    Aug 16, 2013
    Messages:
    2,322
    Likes Received:
    2,904
    Location:
    Western US
    I've wondered about that myself. I think that the big challenge of serial centrifugal compressors is the topology of delivering air from the output of one stage to the eye of the next. I'd bet that there would be a lot of friction and inertia losses from the convoluted ducting.
     
  13. Jun 8, 2016 #113

    nerobro

    nerobro

    nerobro

    Well-Known Member

    Joined:
    Nov 6, 2011
    Messages:
    1,109
    Likes Received:
    312
    Location:
    Northern Illinois
    There are several designs on the market that use a few stages of axial fans leading to a centrifugal compressor. Having the off axis burner can makes it more likely you can make a recouperator work.
     
  14. Jun 8, 2016 #114

    Battler Britton

    Battler Britton

    Battler Britton

    Well-Known Member

    Joined:
    Nov 14, 2015
    Messages:
    504
    Likes Received:
    339
    Location:
    Montpellier,LFNG (Candillargues)
    Some pictures of , maybe a smaller ,but quite similar and 2 seats aircraft. JUST with a 200 hp Lyco

    2014-10-19_1.jpg

    DSC_0145.jpg

    commande_de_vol_m.jpg

    I saw quite often this " very" small plane, and the canopy could be like a 60% F16, but is still big inside


    http://cc02aircraft.free.fr/fr/index.php?option=com_content&view=article&id=52&Itemid=57


     
    Last edited: Jun 8, 2016
    Glider and DangerZone like this.
  15. Jun 8, 2016 #115

    flyvulcan

    flyvulcan

    flyvulcan

    Well-Known Member

    Joined:
    Aug 6, 2007
    Messages:
    525
    Likes Received:
    311
    Location:
    Adelaide, Australia
    WR24-7-2

    For those that are interested in small gas turbine technology, the above link is an interesting one to see how a small gas turbine engine is constructed, albeit an older design.
     
  16. Jun 8, 2016 #116

    Lucrum

    Lucrum

    Lucrum

    Well-Known Member

    Joined:
    Jun 10, 2008
    Messages:
    956
    Likes Received:
    189
    Location:
    Canton, GA
    Interesting
     
  17. Jun 8, 2016 #117

    Doggzilla

    Doggzilla

    Doggzilla

    Well-Known Member HBA Supporter

    Joined:
    Jun 2, 2013
    Messages:
    1,888
    Likes Received:
    396
    Location:
    Everywhere USA
    Williams has been doing this since the 1970s with the FJ-33, well before CNC or laser balancing equipment existed. It should be relatives straightforward to produce an engine with the same efficiency using off the shelf controls and CNC parts. The average machine shop actually exceeds most aerospace companies back when this stuff came out.
     
  18. Jun 8, 2016 #118

    Doggzilla

    Doggzilla

    Doggzilla

    Well-Known Member HBA Supporter

    Joined:
    Jun 2, 2013
    Messages:
    1,888
    Likes Received:
    396
    Location:
    Everywhere USA
    This is actually the generation before the FJ engines, which are 24-40 years old. The only reason its so simple is because its a one use engine. Its for a cruise missile. Williams also makes some for Tomahawk cruise missiles, but most of that stuff is pretty hush hush.
     
  19. Jun 9, 2016 #119

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,788
    Location:
    Calgary, Alberta, Canada
    CNC was actually around in the early 1960s. The aerospace industry was using CNC machines extensively by the mid to late 1960s... Giddings and Lewis, Sunstrand and Cincinnati were the main suppliers at that time.

    McDonnell Douglas had 80 CNC machines in 1969.
     
    Last edited: Jun 9, 2016
    BoKu and bmcj like this.
  20. Jun 9, 2016 #120

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,788
    Location:
    Calgary, Alberta, Canada
    Yes, you need a lot of funky tubes to do that but 2 centrifugal stages gives you a decently high PR which would require 4-5 times as many axial stages. 3D printing of the "funky tubes" might be a big time saver. Axial is elegant and small diameter, centrifugal is cheap.
     
    flyvulcan and DangerZone like this.

Share This Page

arrow_white