Raptor Composite Aircraft

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by Dexacare, Mar 28, 2016.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Oct 17, 2019 #2201

    Scheny

    Scheny

    Scheny

    Well-Known Member

    Joined:
    Feb 26, 2019
    Messages:
    141
    Likes Received:
    64
    Location:
    Vienna, Austria
    He has a racing ECU, which do not care about pollution requirements.

    Diesel always inject what is demanded by torque and do not care about fuel to air mixture. This is only limited in cars through the injection mapping. The mixture is done by the injector which creates a swirling spray pattern, maximizing the surface area of the injected fuel. This mixture starts to burn then from the outside, aka flamefront. This is the reason that mixture ratio is nor of concern.
     
  2. Oct 17, 2019 #2202

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    6,859
    Likes Received:
    1,844
    Location:
    North Carolina
    I believe that Peter is using a single shot. I don't know if the stock engine uses multiple shots. Some modern diesels do, though.

    I'm going to measure cylinder pressure in my Briggs, I think that Peter should be too, though he would have no baseline now.
     
  3. Oct 17, 2019 #2203

    Rik-

    Rik-

    Rik-

    Well-Known Member

    Joined:
    Sep 14, 2019
    Messages:
    250
    Likes Received:
    49
    Location:
    California
    Quieter too.. Cummins shoot 3 shots of fuel per cycle as they claim it reduces smoke, vibration and noise as well as better emissions and power.
     
  4. Oct 17, 2019 #2204

    Bert

    Bert

    Bert

    Well-Known Member

    Joined:
    Mar 17, 2016
    Messages:
    45
    Likes Received:
    19
    Location:
    Stony Plain
    I seem to remember reading that the DuraMax uses 7 shots of fuel at a time. I would be very surprised to find that the Audi Peter is using only uses a single shot.

    Bert
     
  5. Oct 17, 2019 #2205

    thjakits

    thjakits

    thjakits

    Member

    Joined:
    Jul 4, 2014
    Messages:
    10
    Likes Received:
    2
    Location:
    Panama, Rep. of Panama
    Today's EFI Diesels are WAY more complex than the old "pump up the pump"-engines....

    AFAIK - the Audi/VW Diesels have very specific injetion timings - multiple times per compression stroke - varying according to load and rpm. Injection pressures are at levels where material elasticity becomes a serious factor (e.g. nozzle-expansion under pressure).
    Factories are running many hours on a test stand to find the application-correct settings.
    Even if the Aftermarket system used on the Raptor is able to accomodate the micro-second adjustments
    (at 3800 rpm you got 32 injection-sequence events per second...) - I doubt that our prolific builder has the neccessary background to make educated adjustments...
    With today's engines you have to watch a little more than EGTs and Cylinder pressures....

    I just come back to the same old - WHY has no one yet built a 3-liter Aerodiesel with 300-350 hp???

    I assume - "Because it isn't easy...." - tuning a Audi or VW to go like stink for a few seconds or even minutes is one thing - to get reliable 300-350 CONSTANTLY is another...

    thjakits
     
    akwrencher likes this.
  6. Oct 18, 2019 #2206

    Scheny

    Scheny

    Scheny

    Well-Known Member

    Joined:
    Feb 26, 2019
    Messages:
    141
    Likes Received:
    64
    Location:
    Vienna, Austria
    I was one of the developers at Bosch, who developed the ECUs.

    Even ten years ago, 3 injections where minimum and premium cars all used 7, with standard diesel ranging at around 5. With rising computation power were should now be with 9 injections in the current and next generation.

    There is pre dead point injection which is smoothing the temperature rise gradient during end of compression which eliminates the characteristic knocking.

    The main injection is used for the power stroke. It can be split to help the flame front propagation. The post-injections are then used for quality improvement. At high power out is used to prevent smoke by checking how much air is left which can be burnt. At low power it is used for pollution reducing.

    If you have a lot of computation power like in the standard ECU, it can even distribute the injections asynchronous to create sound engineering. The new Mazda does active noise cancelling of the first vibration orders. Like I said, 7 injections where standard for us, but you don't need more than three to run it good and safe. I bet his generic ECU has 3.

    As for fuel to air mixture, all car diesel nowadays have exhaust gas return valves. If you demand 20%, they will fill the cylinder with 80% exhaust. This way you always have best mixture ratio.
     
  7. Oct 18, 2019 #2207

    Hephaestus

    Hephaestus

    Hephaestus

    Well-Known Member

    Joined:
    Jun 25, 2014
    Messages:
    1,178
    Likes Received:
    256
    Location:
    YMM
    What i dont quite understand is why he's stuck on diy'ing it... Unless he lives in buttfrick nowhere - its not terribly expensive to have a pro do it. I've had more than one ecu programmed by a pro and shipped to my door.

    Its one of those rabbit holes I dont understand why he's gone down.

    This is the engine, these are the mods, these are the requirements/safeties & a week later you plug it in knowing its pretty dang near perfect.
     
  8. Oct 18, 2019 #2208

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    6,859
    Likes Received:
    1,844
    Location:
    North Carolina
    Maybe he asked and didn't like the price he was quoted. To do it properly requires a lot of dyno time. Unless someone has mapped your exact engine configuration, they will need to put the engine on a dyno to do it. He doesn't seem to understand the intricacies of tuning the injection system, so he thinks that it is an easy DIY job...
     
    wsimpso1 likes this.
  9. Oct 18, 2019 #2209

    christos

    christos

    christos

    Well-Known Member

    Joined:
    Nov 9, 2018
    Messages:
    50
    Likes Received:
    12
    Location:
    Greece
    Could you share with us your calculations? you told that it is a 200hp engine if i remember right.
    It is a heavy aeroplane, it will not meet Peter's pre-construction performance goals, but it doesn't mean that it will not fly. In any case it is a new canard design, so it needs at least an experienced pilot.
     
  10. Oct 18, 2019 #2210

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    5,893
    Likes Received:
    3,120
    Location:
    Saline Michigan
    Much of what we know about making gasoline engines (Otto Cycle engines) work is based upon us putting in fuel well ahead of intended combustion, keeping it from igniting prior to the spark tripping ignition, and letting the fuel burn progress smoothly through the combustion chamber. Early fuel intro (carbs, port fuel injection, even much of the operating range with direct injection) disperses fuel in the chamber. Yeah, there is some attempt at stratified charge in direct injection, but bear with me. To get a spark plug to initiate combustion, the fuel air mixture in the vicinity of the plug has to be between the upper and lower explosive limits of the fuel in air. Too lean or rich and it will not fire. Stoich is somewhere between the upper and lower explosive limits with gasoline, alcohols, propane, etc.

    Diesel cycle engines run much higher compression ratios than Otto cycle engines, resulting in the air at TDC being hot enough (from adiabatic compression) to cause auto-ignition of fuel introduced then. So, a Diesel with no throttle plate and plenty of air can still run at very small fuel charges. The rest of the picture is that fuels for diesels also have fairly low resistance to ignition and that the mixture local to the region just beyond the injector is rather rich, and the mixture becomes leaner as fuel spreads away from the injector. At some point the temperature and fuel-oxygen ratios supports combustion, and the fuel burns. Once the fire is lit and fuel keeps on coming in, the fire continues. Whether the overall charge is to lean or not does not matter... What matters is enough air and enough heat and enough fuel out in front of the injector.

    The Diesel cycle as historically modeled had energy put in at constant pressure starting at TDC and running some time after the piston started down. This is primarily because of several phenomena:
    • Injection of fuel significantly ahead of TDC will give difficulty igniting - Air temperatures are not high to reliably light the fuel coming in much before TDC;
    • Injection of fuel significantly ahead of TDC may not burn at all - a misfire - due to the fuel not lighting but adding considerable cold mass to the charge air, reducing temps and pressures and suppressing ignition of the rest of the fuel charge;
    • Injection of fuel ahead of TDC that makes any significant pressure will cause net torques to drop, not rise, due to the pressures being applied in the reverse of normal direction - negative power from that part of the cycle;
    • Diesel cycle engines with mechanical fuel injection commence fuel flow very near to TDC in order to light the fire on each firing stroke. They then ran fuel until the intended amount was injected. The pumps were turning with the engine and the amount of fuel injected in each shot was set by fuel lever. Oh, and mechanical injection diesel injectors will only open when pressure is above certain level, they shut when pressure drops - so they are regulated by the duration of the high enough fuel pressure pulse;
    • Mechanical fuel injection generally could not put in all of the fuel at TDC, but took some time after TDC to get it all in, at least at higher rpm and power levels. The faster the engine is running, the more spread along the power stroke is the injection, reducing both torque and efficiencies as rpm goes up. This is in addition to breathing ability of the engine as rotation speed goes up that hinders all engines.
    Common Rail systems used in current Diesel cycle engines run much higher pressures and electronically controlled injectors can be used to tailor the fuel flows for conditions. The result of all of this is that fuel flow can be input starting at TDC to get the fire lit, and then the balance of the programmed flow put in more quickly. Add turbocharging to make the oxygen density of the charge air higher, and peak temperatures go higher.

    In total, we can now get the whole fuel charge done earlier which gets the burning done earlier, so we extend the rpm range with good torques and run cleaner with regards to smoke. This has a downside in road vehicles - the high peak temps contribute to oxidizing nitrogen in the air. To keep NOx down at mid level operation, programming is to inject just enough fuel at TDC to get the fire lit, continue with a pilot stream that keeps the fire lit until some amount of expansion and cooling has occurred, then inject the rest of the fuel charge. This somewhat lowers torques and power, and raises fuel use, but allows you meet the NOx limits on the emissions/fuel economy tests. When a manufacturer cheats (Hello VW) they program to get low emissions for the test and let it run higher efficiency and power the rest of the time. This was supposed to be a "bet the business" level crime, but VW appears to have survived it, even though everyone else in the industry knew they had to be cheating.

    Anyway, the best results in diesels still occur when you have more air than you need to burn the fuel. For a given amount of fuel, if you let the air charge be close to stoich, combustion is slower and air mass per cycle is smaller, raising the temperatures through the end of the power stroke, where it is not well converted to crankshaft torque, and it spews much hotter gases over the exhaust valve, the turbocharger, waste gate, EGR system, etc. Hotter gases on these parts equates to shorter device lives. In general you should run more air than you need so the engine is more efficient at making power and so the exhaust valves and turbocharger will hold together. And a system that does not move enough air is hotter later in each power stroke and has the symptom of high EGT's.

    Sorry for the long answer, but the topic seems to require the more sophisticated set of thinking involved.

    Billski
     
    Topaz, akwrencher, BoKu and 2 others like this.
  11. Oct 18, 2019 #2211

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    5,893
    Likes Received:
    3,120
    Location:
    Saline Michigan
    Ideally, yeah, dispense with the pilot flow, start the fuel in a skosh ahead of TDC and get it in as fast as you can. Power and economy are both better this way, while NOx will be higher but is not regulated in airplanes.

    But who says some of the emissions based fuel programming had been eliminated?

    Billski
     
  12. Oct 18, 2019 #2212

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    12,043
    Likes Received:
    2,347
    Location:
    Port Townsend WA
    Sometimes a long answer is helpful.
    My impression is that a modern diesel might make sense for long range aircraft. But diesels are heavy, so maybe a diesel/electric hybrid. The smaller diesel could provide the long range at 50% power in cruise, the electrics 50% added for takeoff.
     
    BoKu and cheapracer like this.
  13. Oct 18, 2019 #2213

    Venom

    Venom

    Venom

    Well-Known Member

    Joined:
    Jun 26, 2016
    Messages:
    49
    Likes Received:
    20
    Location:
    Pacific NW
    Wouldn't it be ironic if the VooDoo flew before the Raptor.
     
    Mark Z likes this.
  14. Oct 18, 2019 #2214

    bmcj

    bmcj

    bmcj

    Well-Known Member HBA Supporter

    Joined:
    Apr 10, 2007
    Messages:
    12,916
    Likes Received:
    4,862
    Location:
    Fresno, California
    Maybe he asked and was told it can’t be done?
     
  15. Oct 18, 2019 #2215

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    6,859
    Likes Received:
    1,844
    Location:
    North Carolina
    Ignoring mechanical considerations, the optimum time powerwise for peak xylinder pressure is somewhere after TDC. Too soon, the charge has cooled down a bit before it has worthwhile leverage on the crank. Peter is carefully setting his injection pulse to be symmetric around TDC...

    What he should be doing is tweaking the timing for each injection pressure and duration, with rpm as a guide. He has a variable pitch prop as a brake, so this should be easy. This assumes that he doesn't exceed mechanical limits. But since he hasn't bothered with any instrumention there, he can't test that factor unless it breaks...
     
  16. Oct 18, 2019 #2216

    Hephaestus

    Hephaestus

    Hephaestus

    Well-Known Member

    Joined:
    Jun 25, 2014
    Messages:
    1,178
    Likes Received:
    256
    Location:
    YMM
    Perhaps, i can think of 2 people locally who'd have enough knowledge to get a very close baseline map based on configuration then do final tweaks even without dyno time.

    But this probably ends up back in the hardware doesnt work for what he wants, programmers told him that... So he decides his math is better?
     
  17. Oct 18, 2019 #2217

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    6,859
    Likes Received:
    1,844
    Location:
    North Carolina
    Eventually it will dawn on him that his maths is usually flawed.
     
  18. Oct 19, 2019 #2218

    Andy_RR

    Andy_RR

    Andy_RR

    Well-Known Member

    Joined:
    Sep 29, 2009
    Messages:
    181
    Likes Received:
    105
    Location:
    Melbourne, Australia
    Well at least the re-drive works better when you tighten the belt up! You could hear some serious flapping about in earlier videos but today's is a distinctive improvement. I like to think it was my prompting him to video it in action which concentrated his attention here.

    So anyway, he's got the fuel pressure up and hey-presto the EGTs drop! Well colour me surprised Sherlock! It's almost certainly advanced the end of injection angle and increased the expansion work, dropping the gas temp at EVO. Demonstrates to me that tge engine controller is completely uncalibrated and a wild surprise that it hasn't blown up yet!

    I'm guessing though that running it at high power levels with a sub optimal tune has cooked an injector with too high CHTs and now said injector solenoid (or piezo stack?) is beginning to break down once the CHTs rise. Electric thingies don't last long outside their temperature limits...
     
    Doggzilla, cheapracer and wsimpso1 like this.
  19. Oct 19, 2019 #2219

    flywheel1935

    flywheel1935

    flywheel1935

    Well-Known Member

    Joined:
    Nov 1, 2018
    Messages:
    101
    Likes Received:
    42
    Location:
    Downham Market, Norfolk, UK.
    Here in the UK I,m seeing developments at Raptor "Goundcraft" about 5/6 hours before you guys in the States, Latest news, One of the drive belts has jumped a tooth .Just
    had to post on the Raptor YT , also, its a Prototype, please take the bloody plastic of the seats, its not a customer plane !!!!
     
    bmcj and pictsidhe like this.
  20. Oct 19, 2019 #2220

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    6,859
    Likes Received:
    1,844
    Location:
    North Carolina
    A belt jumping is very bad news. But only one of the belts jumping is worse!
     
    BoKu likes this.

Share This Page

arrow_white