TuneECU for 990 - got tunes?

Discussion in 'Crazy-Awesome almost Dakar racers (950/990cc)' started by jetpoweredmonkey, Oct 31, 2010.

  1. Alleycatdad

    Alleycatdad Unbunch yer panties!

    Joined:
    Feb 6, 2008
    Oddometer:
    1,451
    Location:
    True Norcal, not that Bay Area/Sacto Crap...
    I think I'm about done with my "R-Map-w/slipons" experiment; did a little over 800 miles last weekend (http://www.advrider.com/forums/showthread.php?t=702919) and got some interesting results:

    Slab, steady throttle, loaded, 60-65 mph, 200miles, right at 45 mpg. Riding with a buddy on an XRR, did no hard accel or high speeds.

    MIxed use, mostly dirt, riding it like it was meant to be ridden once I quit pussyfootin' around: 27 mpg; same for both days of the dualsport ride. Was really riding it hard, though, both in the dirt and on the pavement, lots of throttle changes and high-speed pavement. Also had to fill with 87 octane, so that confuses the issue further.

    The ride home, 384 miles of mixed dirt and pavement, slow on the pavement as above: 30 mpg on the mixed segments, right at 45 on the slab again.

    That said, the bike worked really well. Still feels fat to me, though, and I'm ready to execute step 2 of my evil, seat-of-the-pants plan...

    I've transplanted the final numbers from the map I built based on the R-with-slipons map in the areas that I adjusted to improve rideability to the stock/stock exhaust R map, smoothed the resulting transitions, matched the F-L switch, and also transplanted the 2nd throttle numbers from the region I changed on the map that I've been running.

    Clear as mud? Basically I took the parts of the slipon R map that I changed and dropped them into the stock R-stock exhaust maps. This resulted in some interesting transitions (the graphic display is good for this) around the changed regions that required some creative smoothing; this requires lots of "save" commands....

    Loading it up tonite and will test once all the freaking tourists leave town after the holiday.

    Steve
  2. Power-Tripp

    Power-Tripp Speed Savant

    Joined:
    Feb 27, 2011
    Oddometer:
    39
    Location:
    Deep South, USA
    Having used TuneBoy/TuneEdit since 2004, and TuneECU since its release, while dyno mapping KTM, Aprillia, Triumph, Duc, etc., I may be able to help.

    A few details that may clear things up a bit:
    •The F_L switch points listed in the table are NOT in TPS percent.
    •Both F and L tables are used at all times. It is just that one has priority over the other based on the switch points, and the other is far less effective.
    •The stock narrow band sensor has a very narrow range for closed loop operation Lambda .98-1.02
    •Watch your Wideband Commander's accuracy. The sensor/meter is not the most accurate unit, and the sensor does degrade over time.
    •Ignition advance alters the readings of the wideband sensor - even when no fueling changes have been made. This is a result of changes in residual O2 in the exhaust gases due to differences in combustion efficiency and heat with changes in ignition advance.
    •The fuel used alters the stoichiometric ratio, as well as the best output ratios. Oxygenated fuels (such as with ethanol), produce more residual O2 in the exhaust. This alters the Lambda/wideband sensor reading.
    •There is no perfect AFR/Lambda reading that gives best output and response at all engine speeds and loads. Give the engine what it wants, not what you think it wants.

    I hope this helps.
  3. DirtJack

    DirtJack Adventurer

    Joined:
    Aug 12, 2007
    Oddometer:
    2,843
    Location:
    San Antonio, Texas
    Power-Tripp, Thanks for your wisdom. What are the values, then?
  4. Power-Tripp

    Power-Tripp Speed Savant

    Joined:
    Feb 27, 2011
    Oddometer:
    39
    Location:
    Deep South, USA
    The values are closer to 4 per 1% TPS, but they seem to differ a bit as you get higher or lower. There is a bit of interpolation going on.

    26 is roughly 6 - 6.5% TPS, 3-4 = ~1% TPS.

    Look at the engine speed where you make peak power. At roughly 1/3 of this rpm, taper the F_L Switch off to 3-4... not 0.
    This allows good low rpm resolution for smooth response, still allows for reducing decel enrichment/enleanment, and moves the change-over point out of the way so that there is not a sudden change from one table to another right where you ride most.

    The L (MAP) tables are more important than many realize. While difficult to work with without an eddy-current brake on a dyno, they influence fueling and ignition advance throughout the map, giving good transient response.

    Changes in F or L switch points require work to the F and L tables, as well as ignition advance.

    I hope this helps,

    -Wayne
  5. Dusty

    Dusty Long timer

    Joined:
    Jun 15, 2007
    Oddometer:
    1,415
    Location:
    Bismarck
    Power-Tripp, thanks for stopping by and the valuable info.
  6. Power-Tripp

    Power-Tripp Speed Savant

    Joined:
    Feb 27, 2011
    Oddometer:
    39
    Location:
    Deep South, USA
    Happy to help Dusty.

    A KTM customer suggested that I read through this thread and help clear up some confusion.

    There is a good bit of information on many of the different model Triumph forums related to work with TuneECU and TuneBoy/TuneEdit. These are similar interface software packages, and learning how each works and differs between different makes and models can help KTM and Aprilia ADV owners.

    [​IMG]
  7. k1973

    k1973 Been here awhile

    Joined:
    Dec 28, 2010
    Oddometer:
    100
    Location:
    Austin, TX
    Hey Wayne,

    Thanks for stopping by. I've just started my work on the SMT map. Going to use the same method I did with Speedy to help improve fuel economy. I was hoping to have time to get to 'bama this summer, but doesn't look like I'll have time. Anyway, good to have you on the site. Your skills will definitely be appreciated here.

    -Keith
  8. RoundOz

    RoundOz Plenty of seasoning

    Joined:
    Jun 10, 2009
    Oddometer:
    1,540
    Location:
    West Nebraska
    Power-Tripp
    I am sure all the guys here welcome your input, especially considering your obvious expertise and experience tweaking FI systems with a dyno. I hope I speak for many of the contributors here in saying that your revelations that the relationships between the maps and ignition curve are too complex to completely master with seat-of-the-pants testing, leads to an obvious question:
    What is the chance that you, working with customers bikes, can come up with a tune that is maybe somewhere between the standard map and open pipe map, definitely better than both, with good economy, plenty of power and perhaps most importantly, smooth operation at low revs. Something that you feel safe enough distributing to the masses without fear of being too lean on anyone's bike, but eliminates the compromises that KTM applies to comply with emissions and noise regulations.

    Obviously all machines are different (and there are some hardware changes between the 2009+ bikes and the older models, so maybe 2 different tunes?) so this would still be an approximation, but should be useable for all bikes and form a better base map for further tweaking either on a dyno, by datalogger feedback or by test riding. My personal preference would be for something suitable for an open 2-1 or 2-2 rather than with the cats (and I would think that most users who are this in depth with making the big Katoom run right have already ditched those 40 pound tumors)

    You may be reducing the need for your services locally but, and I am just throwing this out there for comment, maybe you could be compensated by some kind of group-buy by forum members?
  9. Power-Tripp

    Power-Tripp Speed Savant

    Joined:
    Feb 27, 2011
    Oddometer:
    39
    Location:
    Deep South, USA
    Keith - Nice to see you on this forum as well.

    RoundOz,
    I do what you ask regularly... no payment required. A gift to forum members, if you will. I still haven't had enough 990 models on the dyno to make a good base map available to the masses, but I have another scheduled this week, and possibly one more later in the month. This (as well as the other's we have mapped this year) should give me a good baseline to put together something that will work for most 990 riders by the end of the month.

    Does this forum have the ability to upload/download .dat files? If not, can a moderator set it up to do so?

    The reason one map will not work perfectly in other bikes - The injector industry's dirty little secret.

    -Wayne
  10. Swedemarine

    Swedemarine n00b

    Joined:
    Jul 16, 2009
    Oddometer:
    9
    Location:
    Sweden
    That would be great!

    I have been seat of the pants fiddling with TunECU, and it hasn´t become noticeably worse anyway. So I guess you would be able to come up with something a lot better.

    I understand that there is an individual variation, and that custom tuning is the best, but a "canned" map should be better than the driveability of the lousy standard programming, right? And bikes with carburettors work pretty well. The way they work there are bound to be compromises in some areas of the fueling, but they still manage to give good driveability.

    I would also like to make a donation of some sort, if you make this happen for pre 09 Adventures.

    Greetings /Swedemarine
  11. DirtJack

    DirtJack Adventurer

    Joined:
    Aug 12, 2007
    Oddometer:
    2,843
    Location:
    San Antonio, Texas
    Wayne, the dirty secret is, indeed, an eye opener! I guess the injector industry is relying on closed loop operation, but will the narrow band sensors used by KTM handle the varibility of injectors?

    Most of us replace the stock mufflers and start with the Akra maps which are open loop. This setup has low speed drivability and MPG issues. Solving these two issues would make most of us happy. Most would replace the O2 sensors if necessary to get better mileage.

    As you probably know, the TuneECU site supports uploading of tunes.

    Thanks for trying to help the orange crowd. We can probably figure out a way for financial support of your efforts.
  12. bouboule

    bouboule Adventurer

    Joined:
    May 1, 2011
    Oddometer:
    52
    Location:
    France - EST - ALSACE
    Someone can say me when the L maps are used and when F maps are used?
    If it's not a trottle values in f/l switch, whare are them?

    How is a correct value for off trottle between 4000tr and 2000tr??
  13. Power-Tripp

    Power-Tripp Speed Savant

    Joined:
    Feb 27, 2011
    Oddometer:
    39
    Location:
    Deep South, USA
    Motorcycle ECUs are fairly simplistic in their design and operation compared to modern ECMs used in automobiles. This is a cost-cutting situation, and makes these ECUs very similar to how cars operated 15-20 years ago.

    Mapping engines is not rocket science, but it does take a good bit of knowledge of engine operation, ECU/sensor operation, and combustion chemistry.

    The stock narrow band Lambda (O2) sensor only has a range of .98 - 1.02 Lambda (roughly 14.28:1 - 14.86:1 AFR), and only works in an on-off-on-off manner at low engine speeds and loads - idle, off-idle, and cruise up to around 6% TPS.

    [Many do not realize that Lambda (O2) sensors are calibrated to Lambda 1.0 = 14.57:1 AFR (not 14.7:1), are not particularly accurate, and degrade over time. The problem is that pump fuels with ethanol tend to have a stochiometric ratio of 14.1-14.2 AFR, and this also alters their power producing AFR, but the oxygen in the ethanol results in higher residual O2 levels in the exhaust that makes the Lambda sensor read inaccurately. Also, changes in ignition advance alter residual O2 levels, and make the Lambda sensor read differently - even when there has not been a change in fueling. Become knowledgeable on how Lambda/UEGO sensors operate, their limitations, accuracy, and you can see where mapping to a set AFR/Lambda ratio, or "auto tuning" is a marketing ploy.]



    Yes, the stock calibrations (maps) are not good for smooth operation, or best output and response. KTM has to account for variations in injectors, variations in fuel quality (which can be very large), different charging system voltages, operation at different altitudes and atmospheric conditions, as well as noise and gas emissions.

    A custom map, is not a full re-calibration because the tuning solutions like TuneECU do not allow access to all parts of the calibration - sensor tables, coil dwell, cold and hot starting offsets, and much, much more - as is done using aftermarket race engine management systems with data logging. However, we test and modify all tables as part of "giving the engine what it wants" for best output and response. This tends to require 4-5 hours on the dyno, burning a full tank of fuel. The result is smooth, crisp operation and best output, with good fuel consumption, idle, and off-idle characteristics.

    Bouboule - The L tables are manifold air pressure (MAP) tables, and are displayed in hectopascals based on the MAP sensor's input from engine vacuum. At low rpm, the ECU uses the L tables to give more resolution at low engine speeds and loads than can be had from throttle position percentage alone. At higher loads and speeds, the ECU reverts to using the TPS tables for main fueling, and uses MAP input for sudden increases in throttle/vacuum under acceleration, which require fueling and ignition advance alterations.

    Use of just TPS input is termed Alpha-n. Adding MAP improves resolution, and is termed speed density.

    This differs from most automobile applications that do not use individual throttle bodies, but instead use a plenum intake and mass air flow meters (MAF) for good resolution at low engine speeds and loads.

    With as much bad information and marketing-speak floating around on the 'net today, I consider a smart and informed customer a valuable commodity.
    I hope this helps.

    -Wayne
  14. RoundOz

    RoundOz Plenty of seasoning

    Joined:
    Jun 10, 2009
    Oddometer:
    1,540
    Location:
    West Nebraska
    :clap
    and you have given me another reason to avoid ethanol blend fuel (besides the possibility of expanding gas tanks). It is my understanding that here in Nebraska, at least, gas companies are not required to show which fuels contain ethanol unless the amount is over 10%, and it is regularly used to boost octane levels, or simulate boosting octane levels. All this is heresay, but I usually try to hunt down the rare ethanol-free blends when I can. I resorted to the lowest grade at midnight last night in the middle of nowhere and found that the engine ran fine, with no indications of pinging.

    Keep up the great work of explaining FI from the point of view of a dyno-tech - will be great to try a modified tune of yours when you get it squared away!
  15. Alleycatdad

    Alleycatdad Unbunch yer panties!

    Joined:
    Feb 6, 2008
    Oddometer:
    1,451
    Location:
    True Norcal, not that Bay Area/Sacto Crap...
    Wayne, thanks for your input! There's a lot more to all of this than simply changing carb jets...

    What tuning method are you using? You mention .dat files; tuneEcu are .hex files. If you can't (or don't want to, or whatever) upload your map to the tuneEcu site, I've got some space you can use. Your call...

    As useful for understanding what's going on as an actual map would be, would it maybe be possible for you to document your process and/or changes to the stock map for us? I'd like to understand more about what's going on here. I understand that this would probably take longer than the mapping process itself, but as a learning tool it would be great to have. Again, your call; we're grateful to have your input and the offer of posting a "generic" map is definitely appreciated!

    Thanks again!

    Steve

    PS-thanks for the great read on injector consistency. Interesting stuff!
  16. Tom-Hamburg

    Tom-Hamburg Adventurer

    Joined:
    Nov 19, 2010
    Oddometer:
    84
    Location:
    Hamburg
    07-04-11 TuneECUv1.9.7 now released

    * Bug in the copy / paste of one or more cells (v1.9.6), corrected.

    BR, Tom
  17. Power-Tripp

    Power-Tripp Speed Savant

    Joined:
    Feb 27, 2011
    Oddometer:
    39
    Location:
    Deep South, USA
    I sent Alain an email Sunday about the copy/paste issues, and Monday morning the new version was released.

    GREAT service from Alain.
  18. gixxersteph

    gixxersteph Been here awhile

    Joined:
    Nov 7, 2003
    Oddometer:
    655
    Location:
    Northern NJ, land of the SUV
    Indeed Tune-ECU has been very responsive.

    Now with Wayne on board I am beginning to understand a little of what goes on in my KTMs brain. I knew that somehow the name Power-Tripp rang a bell. Just now I figured it out. I bought a Shorai battery from Power-Tripp and I must say the service from taking the order to delivery was among the best I've enjoyed.

    When I was speaking to them and asked about their tuning facility and range of possible tuning types from Teka to Power Commander and so on I was assured that it could be handled. Furthermore I was invited to stop by when in the area. Well it's not a hop and a jump for me, but if/when I get to Barber I WILL stop by.

    thanks for the support one and all.
  19. Spark01

    Spark01 V-Twin

    Joined:
    Aug 29, 2010
    Oddometer:
    309
    Location:
    The Netherlands
    I made a change in the Secondairy throttle map a couple of weeks ago(more open)
    And made a big trip in the italien and Swiss mountains past week.

    I have the following findings.

    Good:
    -smoother on the throtle in low rpm's/TPS
    -More intake noise and the exhaust pops a lot after braking hard on the engine, i like it :evil.
    -Fuel consumption went down from 1:12-1:15 l/km to 1:13-1:16 l/km?

    No good:
    -When opening the throttle fast full open in low rpm's/TPS, the engine doesn't pull as hard as before, it hasitated for a very short moment.

    This needs some improvement, i'm going to hook up the datalogger to the map sensors. And try to see what the pressure drop is when moving the throttle.
    And then try to fine tune the 2nd map to get the best results.
  20. Power-Tripp

    Power-Tripp Speed Savant

    Joined:
    Feb 27, 2011
    Oddometer:
    39
    Location:
    Deep South, USA
    Secondary throttle operation strategy is a controversial issue. In some cases, the OEM is attempting to simply reduce intake noise or limit rpm transient response. In other cases, it is used to keep operation smooth, or to limit output at higher rpm. In each case, the mapping to fuel and ignition advance tables are different.

    Changes to secondary throttle opening alter several things at once.
    If you open them earlier, or more at the same rpm point, the engine may be getting more air. But this can change the port velocity and pulse tuning at this particular point, as well. This can make this region of the map become either richer or leaner, and require different ignition advance.