_ _ ( ) ( ) | `\| | | , ` | | |`\ | (_) (_)
_ ( ) | | | | _ | |_( ) (____/'
|\ /| | ) ( | | | | | | | | | | | | | | (___) | (_______)
___ ( _`\ | (_) ) | _ <' | (_) ) (____/'
###### ## ## ## ######
IMPORTANT: enter the case-INsensitive alphabetic (no numbers) code AND WRITE SOME SHORT summary of changes (below) if you are saving changes. (not required for previewing changes). Wiki-spamming is not tolerated, will be removed, so it does NOT even show up in history. Spammers go away now. Visit Preferences to set your user name Summary of change: This '''audi 5cyl works''', but experiences "idle fluctuation problem" We've created this editable page to be able to help him, summarize the findings, suggestions in a thematically organized way. * does it rather happen with cold air/warm air ? ** cold engine / warm engine ? ---- The warmer the engine, the worse it gets. The air seems not to make any difference ---- We've experienced that '''similar idle problems are often caused by VE table'''. Can be very sensitive in some cases. Because of the wideband measurement and EGO delay it's hard to diagnose properly (the leaning shows up later in the log than the actual problematic loadsite). In '''many cases the idle fluctuation was cured when the VE was fixed at the right loadsite'''. * Temporarily '''disabling EGO correction''' (or even permanently for idle) also worths a try. ---- * injector size= 370 ** req_fuel= 7.8 ---- '''History''' - Mattias wrote on chat: * some guy with an Audi 5 cyl posted on [http://www.vemssupport.com/forum/index.php/topic,1503 this forum thread] * he was having trouble with the engine falling through when letting it down to idle http://www.savarturbo.se/~mattias/motor/vems_data/log-pics/overrun-1.png http://www.savarturbo.se/~mattias/motor/vems_data/log-pics/overrun-2.png At that point, fuel should be resumed after overrun, shouldn't it? Expression grid says ecu pulse width = 0, and so it is.. but the overrun fuelcut settings don't say so. ---- '''LOG FILE''' http://ranz.planet.ee/idle2.vemslog '''Important''': In the log he tries at first with no overrun fuelcut, later with the ones in my screenshots. Interesting thing after TPS pulse during engine breaking and RPM is falling, '''observe injector pulsewidth''' : * There are 3 tps pulses in the log. After the TPS pulse when the RPM is falling the pulsewidth is not 0, and match with the VT calculations. The two later pulses when RPM is falling the pulsewidth is 0 and does not match with VT calculations in expression grid. You can see the config change around 22 seconds into the log (orange marker in the multigraph, see picture above). Changes to '''overrun fuelcut''' goes from : *pressure based: 0 -> 18kPa *Fuelcut rpm: 8000 -> 2500 rpm *Fuel resume rpm: 500-> 1800 rpm In the log the fuel resumes at higher pressure than 18 kPa and at much lower RPM than 1800 rpm. So the settings are not interpreted correctly. More like 25 kPa and at 1174 RPM, which is not intended. '''Other observations on the configuration''' * MAT/TPS enrichment table is wrong, should be copied from a default firmware config to enable proper scaling with MAT. * MAT/TPS ignition retard is not too safe for turbo * Boost target is above fuel cut ---- I´m going to change the MAT/TPS tabel, never paied any attention to it :( What is wrong with MAT/TPS ignition retart? Boost target is above, but real boost is under. In real life it does not go over 260kPa (stock K24-7000 turbo) ---- '''The log/config has some config errors which WILL cause idle problems.''' (Reidar, some of these were suggested in the forum thread already. All the config changes are in the log file, look above how to see this) IAC PWM Frequency: is set to 250hz. 25hz works fine on the audi aan/aby/adu/3b Set the whole idle area in the Ignition table to the same ignition advance.. 15deg on idle or so. The PID settings is somewhat off in my opinion and experience with these cars. '''P''' = Increase this, try value: 70 '''I''' = Lower this value, try value: 3 '''D''' = Increase this value, from experience a really high D often helps on cars with AirCond where they got stall issues when AirCon active, try a value of 150. Use the ignition based idle control to get a steady idle. I usually tune it a bit rich on idle. 0.95 or so. - Reidar ---- IAC PWM Freq was @25Hz when there was no diode @ valve cables and the valve made disturbing click noise, after adding diode I raised it to 250Hz and that worked a lot better. As I recall, the problem was same before installing diode. Ignition is set to 15 all around IDLE. I´ll try changing PID settings next week, for now I am away from the car. * pleas share a new log file when you get a hold of the car. this way, we don`t have to filter out all the bad parts ---- I don´t have a log file from my own car, but just installed one ECU to audi S4 and it had same problem with the idle (same 1.1.81FW and almost same conf as my own car). Tried different things: Lowering IAC PWM Freq did not help, back to 250Hz. Higher P, lower I and higher D did not help also, it got worse... the idle jumped up and down over 5 seconds before stopped @850RPM. * A to high '''P''' will cause it to jump up`n down. Using Ignition based IDLE did not help also, no difference at all. *Using the Ignition based IDLE is to help get a steady idle after you got the correct IAC PWM. But what helped... using all other settings but incrising D from 1 to 100 :) now the car does not drop below wanted idle RPM more than 20-30RPMs :) * With a to low '''D''' the car will never settle at the given RPM. It will bounce up`n down all the time. I´ll now try it on my own car also and make a log from it. * '''A little IDLE tuning help:''' *Disable EGO correction. Disable Ignition control. Disable PID (P=0 I=0 D=0)Set your preferred IDLE RPM under -> Idle control general (hot) and have the car at operating temperature. *Now you can use the Ignition lock under -> tools and set it to 15deg or your wanted idle adv-ignition. Dont have the ignition pending in the idle aerea of the ignition table. *Start tuning the IAC refrence position until you get as close to your IDLE RPM target as possible. *Tune your VE table to keep it in the range of +- 0.03(lambda) in the idle aerea. *Now you can rev the engine to aprox 3000 rpm. let it fall, and see what happen. exit the ignition lock. *At this point your engine might stall, or bearly catch itself. Start tuning the '''P''' in the pid settings. Increase untill the engine revs up to a little (50-100rpm) above preferred RPM value. Now it will probably be pending up`n down and wont settle at the given RPM. Set the '''I'''' to: 1 or 2 *Start tuning the D untill the car settles at the given RPM value. On the audi`s it usualy quite high. 100-170, even above 200 for a car running AirCondition. *Now enable Ignition based idle control. I usually allow it to change the ignition aprox 8deg advance and 8deg retard. '''advanced change of rate=15''' and '''threshold for activating IAC=3-5deg''' *Keep an eye on Lambda. Is it the same as you just tuned it for? When the Inntake manifold heats up, the sensor might heatsoak some, and it will lean out on you. Some tuning in the MAT/TPS fuel enrich table might be needed at 0% Trottle position. Enable EGO if you want. '''PS!''' this might not be the correct way, but it sure works for me! -Reidar ---- Thaks for the help tutorial, but that did not work for me. I disabled EGO, locked ingnition and set PID to 0. Now changed the IAC ref when I got IDLE wat I wanted. Now revving up it catched the wanted RPM pretty good??? the RPM dropped somwhere 750RPM and catched 850 target pretty well (ignition lock disabled). Now I tried to increase P, 10, 20, 50, 100, 200, 255... nothing, the RPM stayed @ it´s position. So I put it 50 at first (aftewards tuning other fields lowered it to 30, so it worked better). I to 2 and D to 170... ignition based idle control enabeld as you sugested and now it seems to work pretty well:) http://ranz.planet.ee/idletest.vemslog * Looks good! :) Your PID settings are now verry simular to the ones i usually end up with. you can allso see in the log that the ignition control is active, and helping you get a steady idle. The '''P''' might be more sensitive at a lower (25hz) frequency without an diode. just a tought. glad to see it's working :) -Reidar ---- Optional: Add document to category: Wiki formatting: * is Bullet list ** Bullet list subentry ... '''Bold''', ---- is horizontal ruler, <code> preformatted text... </code> See wiki editing HELP for tables and other formatting tips and tricks.