so umm, yeah.

Ever since this firmware update my car has been acting a little goofy. It's got a rythmic misfire (does not appear to be on 1 rotor or 1 face, but it's maybe once every 1/2 second, faster at higher RPM) until the car is TOTALLY warm.

Car has also been a PIG on gas lateley too (and wasn't before)... last tank of gas I got a whopping 8.4mpg. $65 worth of sonoco's finest netted me 200km. Yay fun. last year I was getting 18-20mpg with about 60% hwy/40% city. I think my city mpg was bad... but not that bad. But even my hwy MPG is awful now, 50km trip from windsor used 1/4 tank of gas.


Seems to work fine at WOT. - I did some dyno tuning yesterday and leaned out the top end a ton, runs much cleaner up there but still misfires a bit. AFR's are very smooth now WOT.

at least, they were until last night, now I'm afraid to even venture into boost.

Cruising around 27-2800RPM in 5th on my way home last night, with the little green ball staying in /exactly/ the same spot, (which I think was 2700rpm, 88kpa or something) I was watching my AFR's, they'd stay at a steady 14.2 for a while, then peg the wideband full rich and bounce between 10.0 and 11.5 for a while, then peg the wideband full lean, etc... (car would stumble/misfire as this happens).

Also, trying to adjust VE in the lower buckets, it seems like a 1% change is huge, cruising at say 13.2 AFR, lean it out 1% and it is now a 16.5 AFR... any way to fix that?

I've attached my latest MSQ hoping it's a settings screwup or something but I really don't know what to make of this.

 

Can you datalog what's going on? That would probably be more helpful at this point.

Your cranking advance angle is 10 degrees when it should be 0, but that's probably not hurting anything.....

Your closed-loop settings basically disable closed loop. Also probably not the problem. Your closed loop settings are set to only come on above 7000 rpms, with a wideband at 0 volts... but only with 1% authority. Did you do this on purpose? You also have the AFR target table disabled.


Your fuel cut settings are probably wrong... with these settings, below 63 kPa, above 2000 rpms, and at any throttle position, you're going to enter fuel cut... This could be causing some of your cruise problems.

Your staging off delta is set to 31, so that means once you get into staging, it's going to stay on until the engine goes down to 200 rpms... you probably want to turn staging on at something like 4000 rpms (40 in the setting window) with a delta of 3 (300 rpms less than 4000, staging will turn off). You made the same mistake for the staging second parameter... you put an on point of 120 kPa, and a delta of 110 kPa... that means it won't shut off the secondaries until you get to 10kPa... The delta should probably be 10 or 20 kPa max... This problem is probably the cause of at least some of your odd behavior.

 

Car definatly starts easiest with 10 degrees advance.

yes I have closed loop disabled, it was making the car run funny last year. I am not running a TPS.

63kpa is rarely seen unless I'm closed throttle. I have seen as low as 18kpa (closed throttle at 8000RPM) - but with the throttle open even a tiny amount I'm above 63kpa unless the rpm's are very high, somewhere I'm rarely at.

I would like to have a TPS but I haven't figured out a solution for that just yet.
I have AE disabled (I just set it to TPS mode instead of mapdot) as it seems to be unnecessary at this point, and was being triggered while cruising sometimes.


I JUST changed those staging settings the other day, I misread the window and thought both numbers were absolute! I'll definatly fix that :P I guess my secondaries were staying on always then?

well anyways, I'll try these tomorrow and report back with my results. I'll try and log as well.

 

You're not actually changing the advance since you have it set up for trigger-return... you're just changing when dwell starts... so if it starts better with it set up for 10 degrees of advance, change it to 0, and increase the cranking dwell a bit.

That seems a bit high to me... even on an ITB-equipped 20v 4age I often see below 63kPa during normal operating conditions (cruise, etc..)

If it's unnecessary, you might have the off-boost parts of your table tuned too rich... I've needed some AE for every 7 I've tuned.


They were probably not on until the first time you staged... then they were staying on until you shut off the engine. You should datalog, and take a look at your logs, things like that are a lot easier to find in the logs sometimes than they are when you're looking at gauges or settings.


Cool!

Ken

 

well... only drove it 10km today.

I changed the deltas on the secondaries... don't know if that will help my fuel economy ornot as I didn't drive it far.

it is still running like *** 90% of the time, the other 10% of the time it runs fine. Seems to tend to want to run the best (but does not always) cruising between 2800 and 3k in 5th.

I did get 3 datalogs, I think the 2nd and 3rd one show it the best, MAP stays the same, RPM's stay the same, IDC's go nutty.
Back and forth between 5% and 12% or so. in one of them the car is still cold, the other one it's warm and the other it's hot. Sometimes conditions can be identical and the car runs smoothly. Othertimes, it's like this. The car is nearly undrivable in this condition, it sure isn't "fun" to drive. :P I did not attempt anywhere near WOT at any time, but when it was starting to do this (did it just much much less severe) it was fine at WOT.

I also attached my newest .msq though it is basically identical to the old one, just different deltas.

 

I'll take a look when I get home from work.

 

Your ###24.xls datalog shows you are repeatedly triggering accel and decel mode. This is what is causing your erratic fueling. Opening your .msq shows you have tps accel enabled, but you have no tps. Perhaps noise on the disconnected tps circuit is causing you grief, but your datalog shows a throttle position of zero throughout. I suspect there's a mismatch between your .msq and datalog.

In any case, go back and set accel to map-based (move the slider to the left for 100% map-based accel), and run another log. If you still have false-triggering, then you either need to dampen the map line (small fuel filter in-line, along with a restrictor), grab your map signal from somewhere else on the intake, or increase the threshold setting.

Have you installed the MegaLogViewer yet? You really need to do that, as these issues become obvious when you see them graphed. Also, put a tps on that throttle body asap, as it's a very useful input for more than just accel.

 

I set it to TPS mode after this started happening as a quick way to disable accell enrichments. I will try putting it to map mode and setting all of the settings to zero.


the MSQ and logs are matched, from the same day.

I don't have the log viewer installed, I could never get it to work. I will try again, it does look handy.

thanks alot for your input.

 

To disable, don't switch to TPS mode, just set the threshold to a large number, and/or set the accel enrichments to zero. For MegaLogViewer, follow the installation instructions on the download site regarding java runtime environment, and you'll be fine.

 

thanks renns, I got home from work fairly late and have only just gotten time to actually look at this.

 

something I noticed... with the misfire.


as it misfires, that seems to sync with "fan/output 4" going on and off, how can I disable fan/ouput4? I'm not using it.

 

Did you ever take the suggestions that renns made with Accel enrichement?

Ken

 

yep, I sure did.


I switched to an old .msq from september, then modified it with my new cold start tables.

I have the accell enrichment set to mapdot, and all the values set to 0. so it's never adding or removing any fuel.

I haven't set up a fuel filter buffer yet though.


I got the MS log viewer working

 

The log viewer is going to be VERY helpful in debugging problems...

 

the car still has a misfire.. it's FAR FAR less severe or frequent, but you can still hear it.

 

So the misfire that's still occurring... can you give a detailed description of the conditions it happens under? Send a datalog and point out which time periods it happens in, I can use that to look for any code problems (although the code is well-tested at this point, so I doubt that's the problem).

Ken

 

sure, I'll take a log today.

it's most severe while the car is cold, can't feel it while driving but you can really hear it... instead of a smooth "hmmm" it sounds like it's missing a face or something, constantly.

Once it's warm it's mostly gone away but still present.

Completely gone under WOT.

anyways, yeah... log tonight.