Announcement

Collapse
No announcement yet.

Tuning K constant using different gears.

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Tuning K constant using different gears.

    I'm bored so I thought I'd share this little tidbit of information. This is more for those of us that are anal retentive about having our K being spot on. Some of you may already have thought of it.

    I've been doing a lot of reading about dyno tuning ecus because we're getting a dyno soon. Long story short, I've read from a number of people that its common for AFRs to change based on which gear you are using during a run. In other words if you tune for a particular AFR using 3rd gear pulls when you do a pull in 2nd gear it might run leaner or in 4th it might run richer. It's likely this happens for a number of reasons including acceleration enrichments, the way the load and rpm transition through the maps, and lag in the efi system.

    So I got to thinking. Most of us are probably tuning our K by logging AFRs and trying to get them to match what's on the fuel map. Obviously it works but could it be better? There has got to be some variation due to gear changes and different driving conditions that are encountered during logging runs.

    What I'm wondering is what the afrs would look like in steady state mode on the dyno. If we were able to do steady state it should allow the ecu to stabalize and get rid of any error due to acceleration enrichment, lag, or transitioning affects. We could set the entire fuel map to a constant afr and then pick a couple load/rpm points to steady state in and log the afrs. I think it would be interesting to see how close the ecu could keep the afrs relative to the target fuel map with a good K. How close would a stock ecu with stock injectors and afm get to the target afrs?

    It seems to me that this would be the ultimate way of tuning the K constant. The only remaining question would be, does a K tuned in this manner result in accurate AFRs on the road? This is something thats going on my list of things to test when we get the dyno.

    Questions, comments, did you fall asleep?

  • #2
    `

    Comment


    • #3
      this would explain why I can do a third gear pull and be a little rich and then a fourth gear pull and be close to my target afr. I wonder if the z32 guys run into this issue. Be nice if the maps could be tuned for each gear and individual cylinder trims but that’s a little excessive.

      by the way what gear does everyone tune in?

      tbs
      Blue & Silver 81 zxt, 88 ECCS
      Z32 mas,450cc injectors
      ported exhaust manifold, cable throttle linkage
      t3/t4e 12psi , 3inch exhaust, K&N,
      ACT clutch, 240sx t/b, 1G DSM BOV
      Custom Intake FINISHED
      Ported and Polished N42
      Tan 88-89 conquest seats
      Sperco IC 18.5x6x3.5
      2.5 IC pipes
      Suspension work!!
      Nistune
      Zeitronix Wideband
      240sx transmission
      Soon:
      95 Maxima Alternator

      Comment


      • #4
        I have noticed both the gear and transitional effects you are talking about, but only to a very small degree. I believe bernard's statement about hitting lower load TP in lower gears seems like it could be part of it. I'm thinking what I see is mostly from a small amount of lag in the system due to having so much intake piping volume. I have done all of my tuning in 3rd gear on the street and 4th gear on the dyno. I really only notice it being far richer in first and second gear, but as my car is supposedly in those gears for less than 2 seconds each (0-60 was 3.7 on g-tech), I can't imagine hitting a very high load number. For that reason I started tuning it to come in richer slightly sooner instead of having it out near 14.7 until hitting boost. You may see what I'm hinting at already; it may be an issue of TP "lag" or even something as simple as AFM voltage response time?

        Comment


        • #5
          the afm has to take time, the element cools as the air passes over it changing the resistence, likewise the o2 takes time to guage a/f to further adjust it

          I would imagine the throttle position logging is the only real way to increase accuracy and efficiency relative to our ecu/sensors... if you'll note most new cars come with mass-air and absolute-pressure sensors, and they all take throttle into account to be 100% sure

          we're working with an inferior system so there's only so much to expect, but you are all correct the ratios and sensor lag have to be tuned for but after a point it won't get any better, you'll always be off when driving agressively or lazily, you can't have both perfect

          Comment


          • #6
            For a number of reasons, I got into the habbit of plugging in a Consult and checking the Alpha (o2 correction) table, then adjusting K untill Alpha = 100%. You need a Consult-happy box to do this, but it seems to work pretty well. Since the Alpha table is only updated when the ecu is in closed loop, it is a steady state condition, eliminating transient enrichments, etc. Checking against a wide-band, the fuel table comes in pretty dead-on.

            Any of you with a NIStune box, can you see the Alpha table in there?

            Feel free to chime in on the following:

            MAF's are pretty fast at responding to flow changes, as long as they aren't huge. Probably faster than MAP's, as pressure "follows" flow. TPS doesn't really have anything directly to do with absolute flow through the engine, just differential. o2 sensor lag seems to be very dependent on temp and position. (I'd like to see real testing on this though.)

            Therefore, a correctly sized MAF should give you the best (quickest) air-consumption data. Adding TPS enrichments for short transients covers the bases.

            The idea of "lag" in the fuel system would imply the engine should be running lean and advanced in low gears with high dN/dt. Shouldn't it? Since what's being said here is that the engine is moving across the map faster than the ecu can keep up. I don't buy it. Engines are slow in CPU time.

            I'm thinking everyone is using injectors that are much larger than stock. K, offset, and maps are all corrected. But maybe the accel enrichment tables are now too fat, and taking too long to time-out. Thus transients are fatter than they should be...

            Has anyone played with these tables. Do you know / can you share the units / scales?

            Thanks!
            It's RWD in reverse.

            Comment


            • #7
              Hmmm, thinking about this more while not working...

              Has anyone tried doing a full power 3rd/4th gear pull, logging (calculating) Tp vs. RPM, then tweeking the TTPMax table to limit at this level? It seems you'd still get enough fuel to make your power, but pinch off any potential "plenum surge" type excess enrichment.
              It's RWD in reverse.

              Comment


              • #8
                j30_vg33et wrote: Hmmm, thinking about this more while not working...

                Has anyone tried doing a full power 3rd/4th gear pull, logging (calculating) Tp vs. RPM, then tweeking the TTPMax table to limit at this level? It seems you'd still get enough fuel to make your power, but pinch off any potential "plenum surge" type excess enrichment.
                Then what happens when the ambient air temp drops 40+ degrees?

                Comment


                • #9
                  You run leaner and make better power!
                  It's RWD in reverse.

                  Comment


                  • #10
                    or have a structural oil pan failure....

                    Comment

                    Working...
                    X