SDS EM-5 ECU (EM56)

Discussion in 'Mazda Rotary' started by rv6ejguy, Nov 18, 2015.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Jan 19, 2016 #21

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,449
    Likes Received:
    3,247
    Location:
    Mojave, Ca

    Prediction: BETTER GET CRACKING ON THE SIX!
     
  2. Jan 20, 2016 #22

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    Would love to but have a mountain of other things that need attention in the coming weeks. I know the Rocket and RV10 guys would love to have this capability.

    I came up with a new idea tonight which may allow us to fast track this a bit...
     
  3. Jan 20, 2016 #23

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,449
    Likes Received:
    3,247
    Location:
    Mojave, Ca
    Question about the injector trim: How stable is it from day to day, season to season? I suspect that with the ability to read EGT down to 1 degree on the current engine monitors, people will be tempted to fiddle - attempting to chase an exact EGT value on every flight. I suspect that in reality trim is a set and forget feature, but just the sight of 4 exact EGT values slightly off from day to day is going to bug some people. We see it today with people obsessing over CHT and EGT on their EMS display when 20 years ago most pilots didn't even know about either.
     
  4. Jan 20, 2016 #24

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    We really designed it to equalize EGTs once people set cruise power but I suppose if people see one cylinder always leaner than the rest, they'll leave the trim on that one upped a bit under all power settings.

    I think many VFR pilots already have their heads down in the cockpit way too much with all the new gadgetry, fussing about this and that. Absolutely right that the new glass displays give us unprecedented information about all things- important or not. We've just added a way to make it easier for pilots to be happy about four near equal EGTs where before they could only sit there and suffer the trauma of looking at those mismatched bars. Think of the stress levels we are lowering out there... ;)
     
    Midniteoyl likes this.
  5. Jan 21, 2016 #25

    rbarnes

    rbarnes

    rbarnes

    Well-Known Member

    Joined:
    Aug 28, 2015
    Messages:
    47
    Likes Received:
    11
    Location:
    Texas
    hmmm, neat feature....

    with the ability to individually tune each injector while in flight would it be possible to work up a true closed loop tune that could do this automatically ?
     
  6. Jan 21, 2016 #26

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    Without four widebands, you can't do this.

    It's possible to run in closed loop with different AVERAGE target AFRs at different load ranges and rpms and even re-write the values in the map, as long as the wideband does not take a dump. We'll be working on the first part of this in the coming months, time permitting. The latter part may eventually become an option too but we must test this extensively before release. There are many possible pitfalls and some of the other "self tuning" aftermarket ECUs don't work very well.

    For now, SDS is so easy to tune using a wideband, this is a medium priority project. We've got a lot of other stuff in the wings to get to market first.
     
  7. Jan 22, 2016 #27

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
  8. Jan 26, 2016 #28

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    V26 software is now flying on an RV in Texas. The initial test flight saw one cylinder trimmed +2% and another -9%, the other 2 were very close in EGT when running LOP. An 11% imbalance is more than I thought there would be, but there it is.
     
  9. Jan 26, 2016 #29

    Toobuilder

    Toobuilder

    Toobuilder

    Well-Known Member

    Joined:
    Jan 20, 2010
    Messages:
    4,449
    Likes Received:
    3,247
    Location:
    Mojave, Ca
    Ross, are your customers typically flying with dual ECU's? And if so, since there is only one "fuel system" (injectors and TB) what is the mechanism to bring the "backup" ECU on line to take over the failed brain?
     
  10. Jan 26, 2016 #30

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    Our high time Lycoming user has over 1600 hours with a single ECU but I'd say about 50% of people want dual ECUs. We typically use one board to run top plugs and other for bottom plugs and just switch the injector outputs with a small relay box. We also have a new programmer option which can access either board with a single programmer.

    I just got off the phone with Rusty in TX and he said this stuff was magic, all cylinders can be made to peak at the same time and the engine is uber smooth, even though all cylinders obviously are not producing the same hp each. He spent more time waiting for the thermocouples to stabilize than to trim. He says it's very repeatable and did several rich/ lean exercises to verify. Ran to 200F LOP and still smooth but lost big power there of course.
     
    Last edited: Jan 26, 2016
  11. Jan 26, 2016 #31

    Midniteoyl

    Midniteoyl

    Midniteoyl

    Well-Known Member

    Joined:
    Sep 4, 2003
    Messages:
    2,406
    Likes Received:
    500
    Location:
    Indiana
    11%? I guess it could have been worse, but still that leaves a fair bit of power on the table.
     
  12. Jan 26, 2016 #32

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    I believe this engine had vertical induction and was a known bad case. We'll be waiting to see how Dave Ander's RV4 works with the Sky Dynamcs horizontal sump and Lycon flow matched heads. It should be much closer in mixture distribution.

    dave1.jpg
     
  13. Jan 26, 2016 #33

    Midniteoyl

    Midniteoyl

    Midniteoyl

    Well-Known Member

    Joined:
    Sep 4, 2003
    Messages:
    2,406
    Likes Received:
    500
    Location:
    Indiana
    My problem is I would be using this trim system to chase down all those distribution problems and end up spending a fortune trying to improve it :)
     
  14. Jan 27, 2016 #34

    gtae07

    gtae07

    gtae07

    Well-Known Member

    Joined:
    Dec 13, 2012
    Messages:
    1,812
    Likes Received:
    1,080
    Location:
    Georgia
    Ross,
    I think I've managed to confuse myself... when you input fuel trim per cylinder, and then change the power, do you then program another offset (such that you've built up a table of offsets for each 100RPM or manifold pressure increment), or does that trim setting get applied all the time at all power settings until it's canceled or changed?
    Or am I really just off in another zip code?
     
  15. Jan 27, 2016 #35

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    Trim just adds or subtracts at percentage to the injector open time in that cylinder and is applied throughout the entire map until you change the value. So if the ECU is outputting a 10ms injector pulse nominally and you trim one injector +5%, the resulting pulse width delivered is now 10.5ms.
     
  16. Feb 1, 2016 #36

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    Here is some data from actual flight testing in Texas on an IO-370 using the new trim software. This engine originally had a rather large GAMI spread. Now essentially zero with the trim feature. Able to run 200 LOP now, super smooth.

    1507.jpg

    rop.jpg
     
  17. Feb 1, 2016 #37

    Midniteoyl

    Midniteoyl

    Midniteoyl

    Well-Known Member

    Joined:
    Sep 4, 2003
    Messages:
    2,406
    Likes Received:
    500
    Location:
    Indiana
    Whats the AFR work out to be @ 200 LOP?
     
  18. Feb 1, 2016 #38

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    I didn't get that info. I'd estimate leaner than 18 to 1. Try to see if they looked at that during the flight and report back.

    A new method is emerging on injected Lycomings to run WOT as much as possible and lean like crazy any time you're below 75% power. In this way, mixture controls the power at altitude. Some guys run way LOP in the climb even. The CHT/ EGTs stay cool and the fuel flows are way down.
     
  19. Feb 1, 2016 #39

    Midniteoyl

    Midniteoyl

    Midniteoyl

    Well-Known Member

    Joined:
    Sep 4, 2003
    Messages:
    2,406
    Likes Received:
    500
    Location:
    Indiana
    Well thats leaner than what I thought my 'safe' would be which is ~16:1. you plan on running that lean regularly and bore scoping the cylinders to see if any wear on the valves? Or rings? Or detonation?
     
  20. Feb 1, 2016 #40

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,790
    Location:
    Calgary, Alberta, Canada
    There is less wear because everything runs cooler. Running this lean in cruise, it's actually hard to get CHTs in a good place in cooler weather. No detonation because cylinder pressures are low with the mixture this weak. Oil temps down as well as are carbon and lead deposits. You should not run LOP above 75% power, that part is important.

    Too many people equate lean to bad. Running at 50 ROP to peak is not good but at least 75 LOP and everything is happy.

    Anything above 4000 feet MSL @ 2500rpm and say 150 LOP, and you'll be below 75% and can safely run LOP. At that setting the lean mixture will drop power about 8-12% from best power mixture. This, combined with the reduced rpms and MAP, even at WOT will put you in a safe cylinder pressure zone- out of the "red box" so to speak.

    The injected R3350TC engines in airline service were regularly getting over 3000 hours time on wing running LOP back in the mid '50s. The idea is well proven but only recently re-discovered for opposed engines maybe 10-15 years ago. There's still reluctance to the idea from many locked in the old mindset though.
     
    Midniteoyl likes this.

Share This Page



arrow_white