SDS EFI 540 Installation (Rocket)

Discussion in 'Member Project Logs' started by Toobuilder, Sep 20, 2016.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Sep 20, 2016 #1

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    I've begun the installation of Ross' EFI system on the Rocket and with all the mods I plan to make I expect it to be a several month long process. I figured this would be a good place to document the installation so those interested can follow along.

    [​IMG]

    Here is a picture illustrating the old and the new. Noteworthy is the throat area of the new throttle body on the left and the Bendix RSA-5 servo on the right. Also of note is the dual entry injector adapters that I had Ross custom make for me. The dual entry tops are not catalog items and I paid extra to have them made. More on the reason later.

    [​IMG]

    This is a mock up servo adapter I fabbed out of MDF to see if the servo will fit in the vertical orientation like I want. It will, and I'll explain why I'm going against convention with a vertical mounting later. Very obvious is the significant discontinuity in size between the bores of the sump and the TB.


    [​IMG]

    Here's the monster mounted up. Yes, it's mounted at an angle. The reason is so the throttle cable only has to make a single bend from it's approach off the left side of the firewall.
     
    Last edited: Sep 20, 2016
    Vector, blane.c, Pops and 2 others like this.
  2. Sep 20, 2016 #2

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    I have been interested in EFI for aircraft for a long time, but only as a curious onlooker. The mechanical injection systems used on aircraft have proven themselves to be reliable, and considering the limited demands, pretty good performers. Additionally, they have been shown to be very efficient, boasting impressive BSFC figures when tuned and operated properly. So why EFI? Well, for me it was largely the experience I had with the vendor of my CPI electronic ignition. His ignition delivered what he promised, and the more I looked at his other products, the more I learned. I came to realize that his EFI solution will enable some key performance characteristics that would be difficult to achieve with mechanical fuel injection.
    So as stated above, I have some design goals in mind beyond just the “gee whiz” factor of EFI. These are in no particular order:

    1. The ability to safely run mogas without restriction. This is a tall order considering my operating location sees OAT’s ranging from the low teens to over 115. The challenges imposed by the vapor point of auto gas is going to drive careful attention to the suction side of the fuel system as well as how to handle a quick turn refueling stop in the heat of a summer day. The engine in the Rocket features the standard 8.5 CR, so in theory should be able to deal with the lower octane of the auto fuel, but the ability to manipulate timing and mixture should provide even more detonation margin.

    2. The ability to optimize fuel trim for each cylinder as well as an overall ignition timing curve to suit the engine. I plan on tweaking induction and exhaust tuning quite a bit, and the ability to keep the fuel in line with a few keystrokes is a huge advantage over replacing the individual injector restrictors of the mechanical system. I do not believe there will be a large jump in TAS with an optimized engine, but I expect that I should be able to fly my current speeds a lot more efficiently.

    3. Improved induction plenum design/pressure recovery. Ross’ throttle body is significantly shorter than the either the RSA-5 or -10, and this allows a vertical mounting within the tight Rocket cowl. This will enable a very long induction air box for maximum pressure recovery and also allows a very large air filter element for low restriction. The typical forward facing servo typically uses a short plenum and a restrictive cone filter element. Both are sub optimal for a NA engine.

    4. Reduced workload with regards to engine management. I have no issue with managing the three knobs, hot starts, etc, of the legacy systems, but EFI should bring this down to a more modern level. I don’t consider this a primary reason to go with EFI, but it will be a nice benefit on top of the other considerations.
     
  3. Sep 25, 2016 #3

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    The SDS injection system is automotive style in that the pump(s) are electric and pressure is maintained by a bypass regulator. This makes the standard engine driven pump useless so it has been removed from my airplane. This brings up a need for a fuel pump block off plate. Since I also need a secure place to mount the fuel feed flex hose and also now a return line, I fabbed up this combination block off plate and bracket which takes care of the whole thing.

    [​IMG]


    [​IMG]


    [​IMG]
     
    BoeveP51, jmt1991, blane.c and 2 others like this.
  4. Oct 14, 2016 #4

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    [​IMG]

    [​IMG]

    Nothing too spectacular in the above photos. Today saw me fabricate the main fuel return line from the regulator down to the blockoff plate/ bracket. The 3/8x.035 stainless tube is major overkill for this application and a ***** to bend, but its far lighter than hose and will last until long after the final drop of gasoline is produced on this planet. The Borla regulator has two inlet ports, one each going to the 3 injector banks on opposite sides of the engine. Those are the next lines to fabricate, and boy are they going to be fun!

    [​IMG]

    Speaking of fun, getting the left tank off the airplane was far from it. I need to add a return line to this tank (both, eventually), which I could have done on the ship, I suppose, but future plans call for the long range (29 vs 21) gallon tanks so I needed to have a look see for planning purposes.

    I hope I remember how all this stuff goes back on.
     
    Last edited: Oct 14, 2016
    Pops, Jump4way, Joe Fisher and 2 others like this.
  5. Oct 17, 2016 #5

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    [​IMG]

    Spent the day bending and flaring tube - good thing I enjoy it.

    The left side of the engine is pretty much plumbed now, though I still have some adel clamps to install.

    It should be noted that I'm going on my own with the plumbing here... Ross has a much simpler system that uses a deadhead system. I want to emulate the typical automotive common fuel rail method. Because I want to use auto fuel and live in an inferno, I want the fuel to recirculate through the injector bosses to keep them cool. Experience will tell if this overly complex plumbing scheme has any benefits beyond the deadhead system. Its an experimental - time to experiment!

    Some of you may wonder why so many bends in the tube. The bends are there to try deal with the relative movement of the individual cylinders without adding lots of stress to the injector bosses. Nice straight tubing becomes a structural member, while a 90 degree bend cripples it and makes it fairly pliable.
     
    Last edited: Oct 17, 2016
  6. Nov 1, 2016 #6

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Not exactly required for the EFI install, but the mount is off to blast and powder coat.

    Makes it a bunch easier to climb into the cockpit at least.

    [​IMG]
     
    jmt1991 likes this.
  7. Nov 7, 2016 #7

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Here's why repair projects go off the rails so easily:

    While digging around behind the firewall, clearing our some dead wood structure I contemplated the butchered rudder pedal brackets...

    ...well, as long as I'm in here....

    [​IMG]

    So about a day lost remanufacturing new brackets produces something I can live with.

    [​IMG]

    What does this have to do with a fuel injection installation?

    ...nothing!
     
    Last edited: Nov 7, 2016
  8. Nov 14, 2016 #8

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Once again, not directly EFI related, but at least stuff is going back together for a change.

    [​IMG]

    Today saw me shoot in my new rudder pedal brackets, cleaned up the firewall, and bolted my freshly powder coated engine mount home.
     
    Joe Fisher likes this.
  9. Nov 26, 2016 #9

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    I had to address some fretting issues on the socket portion of the upper gear legs and now those are nice and snug again.

    The method of attachment of the stub axles has always bugged me however. My airplane had the bolts going through the aluminum axles and threading directly into the stub sockets. This loaded the threads in shear, which is almost always a bad idea.


    [​IMG]


    I reamed 3 of the holes and installed Hi-Lok pins from the inboard side, secured with nuts on the wheel side. Not only are these pins much stronger in shear than the AN hardware that was being used, it is now installed with the threads far removed from any shear plane. I'll sleep better now.

    [​IMG]


    And of course while I was in the area, I could not let the existing wheel fairing brackets go as they were.

    [​IMG]
     
    Last edited: Nov 26, 2016
  10. Nov 28, 2016 #10

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Ok, FINALLY back to the EFI stuff!

    After looking at a bunch of locations for the ECU and other electronics, I decided that the area fwd of the panel would work well if it was mounted on its side, paralell with the fuselage skin. This keeps it well clear of my feet, safe from inadvertent water intrusion, and it remains somewhat accessible.

    Here is the flat plate mockup for the general layout. You will notice that I am using shock mounts for the assembly. As the ECU is potted and very robust, this is probably overkill but I have a bunch of these mounts laying around so why not?

    [​IMG]

    Here is the finished plate with the hardware mounted, but LOTS of wiring to add. Also noteworthy is the fact that I used titanium NAS6403 bolts to secure the ECU and relay boxes... No particular reason exept that I have a whole bag of them I picked up from salvage sales at work and they are exactly the right size.

    Besides, Ti is "cool".

    [​IMG]
     
    Last edited: Nov 28, 2016
  11. Dec 5, 2016 #11

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    FINALLY got the ECU mounted. The bracketry was "fiddly", with some challenging riveting situations. Glad this is behind me now.


    [​IMG]
     
  12. Dec 12, 2016 #12

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Ok, time for some of the more sexy hardware for this EFI installation.

    Time to do the throttle body adapter for real. Shown below is the significant discontinuity in the size of the TB and sump.

    [​IMG]

    Once again my ancient but reliable South Bend lathe makes a lot of aluminum chips

    [​IMG]

    Here is the MDF template and the finished part.

    [​IMG]

    Its standing proud of the mount surface more than it needs to, but it is the same height as the tuned 300 hp sump I intend to use later. If it works out right, all the integration hardware I'm doing now will allow a direct swap with the new sump. And more importantly, there will be only one change (the sump), so I'll be able to tell if the change helps or hurts performance.

    [​IMG]

    Another graphic display of the size difference between the sump and the TB. Obviously not ideal, but it should work for a while.

    [​IMG]
     
    Last edited: Dec 12, 2016
    Joe Fisher, weasel, Jump4way and 2 others like this.
  13. Dec 19, 2016 #13

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Shown below is my new electrical hub. I am now dual battery and the former battery position behind the baggage compartment will now exclusively serve the needs of the ECU, fuel pumps and ignition. I will retain the large Odyssey 925 for this purpose, but it will only feed a #10AWG wire.

    [​IMG]

    The ship, on the other hand will be served by this added PC 680. Since its primary function is to crank the engine, I want the wiring to be as short and low loss as possible. To that end, I have placed the battery just forward of the spar and reduced the starter cable down to #6. The #1 cable I removed weighed 3.5 pounds, BTW. I hope that carefull attention to efficiency will allow acceptable cranking power with minimum weight. It is expected that Ross' EFI will enable minimal time on the starter, hot or cold.

    [​IMG]

    As an aside, I had to make a hard decision today and let one of my favorite cars go. At least it was to another Rocket buddy. These are the choices an airplane nut makes. "First world problem".

    What does this have to do with the EFI, you ask? Well, I met this guy through his decision to go with Ross' CPI ignition. I helped him through some technical issues, and we became friends. On a recent visit to my hangar, he spied the Cadillac and had to have it.

    [​IMG]
     
    Last edited: Dec 19, 2016
  14. Jan 1, 2017 #14

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Not sure if this is the light at the end of the tunnel, but at least the engine is going back together. The sump is cleaned up, painted and ready to go. The throttle body adapter is mounted permanently (well, its not coming off before flight I hope).

    [​IMG]

    [​IMG]

    Thought I would go with a more traditional aircraft engine color but decided at the last minute to go with hot rod orange - paying homage to my muscle car upbringing. A blending of two worlds, if you will. I even bought some polished valve covers for it.

    I was hoping to get the bottom end together this weekend and got the rods clearanced and hung on the crank. When I tried to do a dry assembly of the case to check the main bearing clearance I discovered that my engine guy ordered the wrong bearings and they don't fit this case. So now I'm dead in the water until I get the correct bearings. All I can do is fiddle around with assembly of minor accessory stuff like pipe plugs and fittings, etc.

    The accessory case shows the results of going with EFI - nothing but block off plates for magnetos and fuel pumps.

    [​IMG]
     
    Last edited: Jan 1, 2017
  15. Jan 16, 2017 #15

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    I tried two different paint schemes before settling on the "aluminum" head. I sent the picture to a bunch of friends and it came back the unanimous choice.

    [​IMG]


    A weekend spent thrashing saw many small and expensive parts turn into a unified lump that resembles an aircraft engine. I had hoped to hang it on the mount today, but the intercylinder baffle sealing burned a bunch of time. I still need to mount the sump and induction tubes, but we're close now.

    [​IMG]
     
    Last edited: Jan 16, 2017
    BJC and rv6ejguy like this.
  16. Jan 28, 2017 #16

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Not much to report, but at least the sump is bolted down and sealed. The engine is now structurally ready to hang on the airplane, but I need to get the rear baffles installed first.

    [​IMG]
     
    cheapracer, Joe Fisher and rv6ejguy like this.
  17. Jan 29, 2017 #17

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Finally- the engine and airframe are together again

    [​IMG]


    Trying something new with my oil cooler. It is now mounted solidly to the accessory case and will be fed air with a yet to be fabricated duct. The duct will be semi structural, the remainder handled with a simple welded tube support.

    [​IMG]

    [​IMG]

    [​IMG]

    It was hung on the right side aft baffle before and did OK there. Oil temps were perfect, but I was battling some cracking issues as well as the #6 cylinder was running a touch hotter than it should have. Both issues could have been solved, but I think (hope) the new scheme will end up lighter and more effective, if somewhat of a packaging nightmare.

    And before anyone calls me out for using aluminum fittings, they are just for mock up. They wiil be replaced with steel before flight.
     
    Last edited: Jan 30, 2017
  18. Feb 20, 2017 #18

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Here is my attempt at a plug for the oil cooler duct. First step was to define the inlet and outlet position and shape, then build a simple skeleton to hold it together, then fill the space with tooling foam.

    [​IMG]

    [​IMG]

    [​IMG]

    Once it's shaped nicely, I'll glass over the waxed plug and then dig out the foam.
     
  19. Feb 20, 2017 #19

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    This is the start of a new cowl inlet ring. Simple MDF plug, will be glassed to make a mold, then crank out a pair for the cowl. At 6 in diameter, these have a bit more area than the stock Rocket, but will be moved fwd and outboard to capture the local acceleration effects of the prop.


    [​IMG]
     
    Last edited: Feb 21, 2017
  20. Mar 13, 2017 #20

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,441
    Likes Received:
    3,235
    Location:
    Mojave, Ca
    Here's the next step of the inlet ring: Making a mold with my trusty trash bag and repurposed air compressor vacuum bag system.

    [​IMG]

    ...And the next step for the oil cooler scoop:

    After getting the plug nice and straight, I covered it in packing tape as a mold release. I then covered the plug with 2 layers of 6 oz glass and vacuum bagged that down nice and tight. I followed that the next day with a light sanding and 1 layer of 6 oz carbon, which was once again vacuumed down.

    With that cured, I started digging out the plug. Messy work, but the packing tape really worked well and left a very nice finish inside. Only missed a couple of spots, but light sanding will take care of that.

    [​IMG]

    Here it is fitted to the cooler with nut plates on the flanges.

    [​IMG]

    And here is the position back on the airplane. It clears everything just as the plug did. I still need to trim the cooler side flanges and add a nice radiused flange to mate with the aft baffle wall, but this it pretty much it. Turned out quite nice and is incredibly light and stiff. When it's all said and done, I will weigh the components of the "old" vs. the new. I'm anticipating a net weight savings with this setup.

    [​IMG]

    [​IMG]
     
    Last edited: Mar 13, 2017

Share This Page

arrow_white