ECMTuning - wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.

<h1>Quickie A/F Setup</h1> <div class="level1"> <p> I was asked to store this post away somewhere for future reference. For now, I'll just dump it in here as is and maybe come back later and edit/expand upon it. It's really just documentation of my personal process for running through a "quickie" setup with a new car using DSMLink. It's intended to help the user get a handle on whether fuel delivery and airflow metering all make sense. </p> </div> <h1>Process</h1> <div class="level1"> <p> First, if you don't already know what fuel trims are or how they're used by the ECU, then I would strongly recommend you start by reading through <a href="/wiki/doku.php?id=fueltrimupdatepoints" class="wikilink1" title="fueltrimupdatepoints">Fuel-trims: What They Are and How They Work</a>. Knowledge of this fundamental concept is key to understanding everything else below. </p> <p> I generally monitor <code>STFT</code> and <code>AirflowPerRev</code> at idle. I want to see <code>STFT</code> cycling around 0% with <code>LTFTLo</code> around +/- 5% or so. You'd expect to see 0.28 g/rev for <code>AirflowPerRev</code> at idle ^{<a href="#fn__1" id="fnt__1" class="fn_top">1)</a>}, but I don't get too caught up on that if the fuel trims are inline with global fuel settings that make sense for my injectors. </p> <p> If fuel trims at idle are not where I want them, I either use the lower airflow sliders (50 and <em>maybe</em> 150hz sliders) or I use deadtime to tweak things. Which I use usually depends on whether <code>AirflowPerRev</code> is inline or not. If <code>AirflowPerRev</code> is off from the expected value in the opposite direction from <code>LTFTLo</code>, then I adjust the lower airflow sliders more than deadtime. If <code>AirflowPerRev</code> is around what I'd expect, I might focus on deadtime instead for idle fuel trim adjustments. Just keep an ear out for little misfires at idle that can sometimes be introduced with deadtime adjustments, particularly if the injectors are not well matched at low pulsewidth. </p> <p> Next I do some cruising around, again watching <code>STFTs</code> to see where they like to hang out. I don't really have a good airflow figure to watch for at cruise because it depends much more on throttle position and road inclination than idle airflow does. So it's hard to pick a value there. I just watch <code>STFTs</code> and try to adjust airflow sliders or fuel settings (primarily deadtime and/or global fuel sliders) accordingly. I would consider airflow slider adjustments if I know I have reason to suspect airflow to be off a bit (maybe I'm dialing in a hacked MAF or something). I would focus on fuel settings more if <code>LTFT</code> were off for both idle and cruise. </p> <p> Then I do a few pulls and look at <code>BoostEst</code> around 5500-6000 RPM to make sure it matches with what the gauge was reading. Some deviation is expected, especially as revs increase. But more than a psi or two and I feel like I want to tweak the airflow sliders some more. </p> <p> At that point, I'm usually pretty happy with the setup and can start tuning. </p> <p> If you have a WBO2 wired into the ECU for datalogging, then you can do even more checks and verification. In that case, I set the estimated A/F Ratio properties to match my global fuel settings and see where the estimated A/F Ratio differs from the measured. </p> <p> In all cases, I constantly do cross checks. For example, if <code>LTFTLo</code> at idle is positive by a good bit (more than 5%) yet <code>AirflowPerRev</code> is also higher than 0.28 and my deadtime has already been increased a good bit from where I'd expect it, then I have to question what's going on because the numbers don't add up. Maybe fuel pressure at idle is lower than what I assumed it was, for example. </p> </div> <div class="footnotes"> <div class="fn">^{<a href="#fnt__1" id="fn__1" class="fn_bot">1)</a>} <div class="content">Automatics and 2.3L engines will generally have a slightly higher AirflowPerRev value at idle.</div></div> </div>