Wednesday, February 19, 2014

Tuning: Closed Loop Tuning with O2 Maps



Oh the bane of piggybacks.... Closed loop tuning using the O2 map. The most finicky of settings! I can't promise to solve all your problems, but hopefully give you some insight and some 'gambatte' with regards to this challenging issue.

The way I initially setup my O2 map was by following AEM's guide (1NZFE uses a narrow band O2 sensor). The guide has an excellent overview on O2 tuning, so have a look at that!

Remember, you should only be tuning the O2 map for Close Loop; there's no point in Open Loop as the factory computer does not use the O2 sensor to calculate fuel delivery; it usually just pumps out a constant output in accordance with its base fuel map.

O2 map tuning is essentially Fuel Trim "tuning": making sure the tune you apply doesn't cause factory fuel trims to fluctuate too much. This is the key goal for O2 tuning! Your OBDII tool that can read the factory computer's fuel trims and open/closed loop indicators is absolutely a requirement.

In partial throttle and Close Loop, you should only be tuning the transition and boost regions of the maps (fuel, O2 etc):


In the vacuum region, you are not making any boost so there is no need to add more fuel and hence no requirement to adjust the O2 map. In the picture above, I have also added another region to our map that I've dubbed the "dead zone".

The dead zone is a portion of the map where, under normal operating conditions, the motor should never be able to get to. For example, from revs 0-3k; it is not expected that the motor will get to MAX boost, not possible for my turbo anyway. Look at your own map and determine from your RPM and load ranges what is indeed your dead zone.

It's still a good idea to put values in there in the unlikely event the motor will go into that region, but don't be tinkering around that area as it's most likely not a region of concern.

Now, the motor uses the O2 sensor to deliver an optimum amount of fuel in accordance with the base fuel map, usually stoich. However, at wide open throttle or at a certain increased load and RPM conditions, the motor switches from Closed Loop to Open Loop as it doesn't care about accuracy and falls back to its own fuel map. This transition and its key drivers is what you need to look out for when tuning!

This is what I've generally found:


At partial throttle with lower loads and low RPM, most likely the car would be making boost but would be in Closed Loop. As the revs and load builds, there is a switch to Open Loop and suddenly the fuel signals becomes constant.

Keep in mind that at wide open throttle, even at low-medium RPM, the above depiction becomes irrelevant as the factory computer would be in Open Loop. To reiterate, wide-open throttle usually puts the factory computer in Open Loop; but check and confirm the behaviour of your application!

So now we've narrowed the problem down to only that redish region in depiction above. Not so hard now huh? LOL not really..

When it comes to build your map, usually you would start with Open Loop tuning and build your fuel map. Then when you have attained your target AFRs, fuel trims, power etc; you can then work on your Close Loop tune.

With Closed Loop tuning you want to start slightly rich, just in case. Extrapolate your Open Loop tuning all the way down to the lower loads and RPM regions where you would have partial throttle. Do your run, adjust the O2 map and check fuel trims using your OBDII tool. Your tuning sequence should generally be:
  1. Setup fuel values in your fuel map.
  2. Perform run.
  3. Check target values (AFR, power etc)
  4. Adjust O2 map to minimise fuel trim fluctuations.
  5. Repeat from Step 1 until you reach your target values

Some tips and tricks:
  • The O2 sensor needs time to warm up before it can provide reliable measurements to the factory computer. So always do your tuning on a warmed up car!
  • In essence, O2 tuning is about minimising fuel trim fluctuations, so always check your fuel trims!
  • If you are getting frustrated and trims seem to be all over the shop, reset them and start fresh by resetting the factory computer (usually power cycling it). Then, warm up the motor and re-tune.
  • If your fuel trims at idle is already fluctuating significantly, this will significantly affect your closed loop tuning. Try to determine why your idle is too rich or lean. E.g. vacuum leak? new injectors pumping too much fuel?
  • Make values between cells transition smoothly and gradually. Don't put one cell as, for example, 10(%) and then a nearby cell as 25(%). The erratic jump may confuse the factory computer.
  • Try to work with your factory computer, instead of against it. If it wants to lock down a long-term fuel trim at a certain point but you have your target values with little swing in short term fuel trims, then leave it be; don't need to try to tune out long-term fuel trim. Even if it resets, the short term fuel trim will bring back the long term fuel trim to that level anyway.

I hope this overview and the tips above helps out with O2 map and fuel trim tuning. This is the most challenging part of piggyback tuning as you not only need to make the motor happy, but also the piggyback computer and the factory computer.

Any comments, suggestions or your thoughts; feel free to leave a comment! The more pieces of info we all get, the more ideas we can come up with to help make piggyback and closed loop tuning a little less of a headache for all of us!

Gambatte!



18 comments:

  1. The new layout looks awesome! Would it be possible to see some screen shots of your maps? It's hard to visualize some of the steps when the maps are all 0's. Keep up the good work, this site has helped me tremendously :)

    ReplyDelete
  2. Aw thanks! Yeah I'm considering of making another post about maps as it's basically all you see when tuning lol. Stay tuned!

    ReplyDelete
  3. Hi am just starting DIY tuning with this FIC with obd2 Honda. At the moment am trying to attempt O2 control without changing to new injectors (IRT = 0 as stock injectors as what I was told). With respect to the period ms setting in the O2 setting, what would be the best tuning methodology? i.e. O2 offset then add fuel to stabilise st or the other way around? also did you do O2 offset table? care to share what to look out for in order to get fairly stable AFR? thanks

    ReplyDelete
    Replies
    1. Heya!

      Yup, this whole post is about the O2 map offset table! Have another read! Basically, you need to set your fuel first, then adjust the O2 map to hold your AFR and to keep your ST fuel trims!

      The nature of the beast will prevent anyone from getting AFRs perfect and stable.. So if you can get it to a manageable state, then you're all good!!

      Delete
  4. how can i contact u. want to ask something

    ReplyDelete
    Replies
    1. Hi there! Please ask via the comments so we can all learn! Thanks!

      Delete
  5. So I have been running the aem fic6 on a turbo 2.5 jeep wrangler for a while.

    I thought I had the fuel map dialed, would go rich under throttle and lean under vacuum but i'm still geting hiccups at 2 to 3k no load that i cant seem to shake. Then last night i removed the battery. Reloaded my last tune and the computer was giving me fits. Had to load an old stock tune to even drive it.

    ReplyDelete
    Replies
    1. Heya! Ahh! 2-3k is about the middle of the transition between vacuum and boost, so the factory ECU may be getting confused if you have a too greater jump in additional fuel when you go into boost.

      Now, if you remove the battery, what happens is; the factory ECU clears its Long Term Fuel Trims and brings the factory tune right back to baseline. If you have been tuning on the factory ECU before this, then you could be adding too much/little fuel at the baseline maps!

      For piggyback tuning, it's good practice to do a full factory ECU reset by removing power to the ECU and let it sit for awhile (I like to leave it for like an hour to make sure it is totally cleared). Then with the baseline, do your tune. Just note though, at baseline, the tune assumes your car is perfect from factory. After awhile of daily driving there will be inefficiencies. So the factory ECU will immediately begin to adjust for these and will be evident when you check the Long Term Fuel Trims after a couple of drives. But that's fine, just take this into account when you tune.

      To assist with the 2-3k transition, make a gradual increase from vacuum to boost. So start adding fuel a little early like at 1,900 and then reach your target fuel value by 2,500 or so. Don't forget to adjust your O2 map accordingly just in case you're not in Open Loop yet and the factory ECU is still relying on the O2.

      Let us know how you go!

      Good luck!!

      Delete
  6. Have you ever considered writing an ebook or guest authoring on other websites?
    I have a blog based upon on the same subjects you discuss and would love to have you share some stories/information. I know my visitors would
    value your work. If you are even remotely interested, feel free to shoot me an email.

    ReplyDelete
    Replies
    1. Ah hey there! Aww I haven't thought about it! Been super busy lately so I haven't been able to write any new posts.. but if there's a topic in particular you're interested in you could always lemme know and I'll see what I can do! Thanks for popping by!

      Delete
  7. Hey Adam! Yeah I haven’t posted much since I’ve let go of my little beast but been trying to reply to comments!

    With regard to your question, trust the wideband.

    The wideband should give you the true AFR. Adjusting the O2 map, adjusts the trims i.e. hacks what the factory ECU sees. Hacking will never be accurate so you’ll need to fudge it enough so it works.

    Remember, adjusting the O2 map, adjusts what the factory ECU sees, it does not directly adjust fuel. Meaning, if the motor is idle and sees stoich, but then you adjust the O2 and then the ECU suddenly sees lean, the ECU will compensate and adjust trims to bring it back to what it perceives as stoich through the factory O2 sensor. At the same time, the external wideband would “see” the true AFR which could be rich/lean. The factory perceives AFR through the factory O2 sensor which you are hacking it’s signals through the O2 map.

    So changes to the O2 map augments what the factory ECU sees which means trims are not true anymore so you need to rely on the external wideband.

    Have a play at adding/removing only tiny amounts on the O2 map and record what happens. If the factory O2 sensor is narrowband, tiny amounts create big-ass fluctuations. Plus as you experiment, trims will continue to be saved which will further throw out even more changes so I suggest before you start you do a factory ECU reset by disconnecting power to the factory ECU and leave it for 10 mins or so, then turn it on and idle for 5-10 mins to get a solid baseline. Then tinker, then reset again.

    Try to reset often so you’re always back to a baseline.

    If you can avoid O2 map tuning that would be fantastic. But if not, try following the advice in the post above and just focus on the transition portion just before you go into open loop. A dyno helps so you can hold your revs just before open loop and see what trims are doing and you can tune accordingly. You can road tune but it is challenging with plenty of logging and desktop analysis.. but doable! Just not significantly as accurate as on a dyno.

    Any questions lemme know!

    All the very best!!

    ReplyDelete
  8. I cannot believe that I missed your response. I only discovered it today. Thanks so much! I have tried to eliminate some wild AFR fluctuations and bucking that I get at around 3 - 4 psi boost. The fluctuations and bucking occur at steady throttle. If I give it more gas or let off on the gas, the bucking stops. I attempted to follow the tutorial linked below, but no matter what voltage I enter in tiny increments, my fuel trims (OBD2 on Torque Pro) go to -20. As such, I can't get accurate trim data for my breakpoint table.

    https://www.youtube.com/watch?v=GZNDCT9UjQQ

    The tutorial suggests that if I don't perform O2 skewing, my fueling enrichment will eventually be trimmed out the factory ecu.

    I will re-read your explanation above, but I can help but think that some of the radical changes in adjacent cells of my fuel map are causing the wild afr fluctuations.

    I did some experimentation by entering different values in my O2 map and can achieve good enrichment at different boost levels (for example, I'll get 12.0:1 at 3 psi and 11.5:1 at 5 psi, with voltages of 2.8 and 2.6 respectively), but I still get the bucking at steady throttle around 3 - 4 psi.

    I am wondering if you would be willing to look at my maps and see if anything jumps out at you. If you were willing, I could send you either pics of the maps or the tune file itself. I'd be very grateful as I'm getting to the point where I'm considering ditching the F/IC and going with a standalone computer and having a tuner tune it.

    ReplyDelete
  9. I don't know why it published my above comment as anonymous, but this is Adam, in the event that this post too is "anonymous. Thanks!

    ReplyDelete
  10. Adam, I am no expert and am still figuring this system out the same way you are, but from what you described in your post, your external wideband sensor and your OBD2 scanner are actually showing the same results. Whatever tweaking you made to the 02 map in that area, made the factory ECU want to add fuel. So the 02 voltage you plugged in was telling the computer that the mixture is lean. The fuel that the ECU is adding to combat this perceived "lean" mixture can be seen by your wideband reading rich and your positive fuel trims. So in order to prevent the ECU from making large trims and running out of adjustment on your tune, you would want to add fuel to the fuel map in these areas until the trims you are seeing on the scanner are back to near-zero. Hope that helps you out some! -Lake

    ReplyDelete
  11. Hey team! Apologies for the late response! But Luke above has a good point!

    I’ve had some sputtering during cruise, towards acceleration. This was due to that finicky transition point between NA and boost. At around that breakpoint you’d probably be beginning to add fuel, but because the factory ECU still in closed loop, it’ll trim out that extra fuel.. after awhile, the trim may be so severe that the big jumps in fuel signal can cause the sputtering.

    Try resetting the factory ECU by disconnecting the power from it and let it sit without power for 30 mins or so. Then reconnect and start it up. Monitor the trims using a OBD2 tool and see what happens when you accelerate.

    As Luke says, you’ll want to make sure your trims are near zero. You can do it by adjusting fuel and then adjusting your O2 map accordingly..

    And this tinkering is the biggest work you’ll be doing with piggy back tuning. Getting fuel on open loop, in boost is a piece of cake. The issue is keeping the factory ECU happy at all other times and in particular, the transition period in closed loop to open loop.

    See how you go! And don’t fret because you won’t ever get that part perfect. As long as it’s smooth enough and workable. Even if you have to go a bit too rich during transition just to be safe.

    Once you feel that you need more control and to optimise tuning, you’ll need to look into standalone.

    All the best!

    ReplyDelete
  12. Hi...maybe im already late for this post but im just trying my luck here...hehe
    May i know...do you have to add resistor to factory o2 sensor to be able to tune o2 map?tq in advance

    ReplyDelete
  13. Hey there wea! Thanks for dropping by! If I recall, no I did not use resistors for the O2 sensor wires. Only the CAM and Crank signal wires I used resistors.

    The O2 sensor wires are an intercept connection so you shouldn’t need resistors as far as I can tell. But maybe it’s something different to your application.

    Let us know how you go! All the very best!!

    ReplyDelete
  14. Thanks for the reply...may i know why do you put resistor on both signal? Sorry because im very new in this...tq btw

    ReplyDelete