[Seek]

I am not able to find anything relating exactly to my situation. Occasionally when driving around, my in gear idle, stock 624, is so smooth that I can't really "feel" the engine. Often times, the idle is rough (when newer/smooth running cars act up, people say it's "stumbling" - on this engine it's basically normal and runs great) and seems to jump between 590 and 670rpm's. Looking at the monitoring (Moates QH in BE), the spark advance is jumping between 16 and 27 degrees during this episode. I took the calibration and modified it a bit to try 600, 575, and 550rpm idles - they all lower some and remain the same level of roughness. 550 seems to run in the 575-600 range. I then wrote the 624 tune back to the chip and the timing advance jumped to ~28 degrees with much less up and down movement and stayed around there, still in gear, and the engine was a bit smoother, idling between 600 and 630 rpm's.

This is on a basically stock A9P tune with only displacement (bored .030) and disabled thermactor changes. I saw mention of monitoring where idle spark is coming from, but I couldn't find a source. Does anyone have any ideas where to start? Everything works right, I'm just trying to fix up the "little things" before going all out with new injectors, mass air sensor, etc. The basics are a 306 with gt40 intake, gt40p heads, ho cam, 2.5" exhaust, still stock 60mm TB and 55mm mass air sensor with panel air filter, 19lb injectors, A9P ecu, distributor at 10 degrees.

[Tipo874]

Gross idle spark comes from FN111 (Idle - Spark vs RPM) multiplied with FN841N/D (Idle - Spark Mult vs Neutral/Drive Error), and and a9p appears to have an idle advance target of 22.5° in drive and 18° in neutral.

There are some modifiers for ECT and dashpot, but they don't seem to have any effect on a standard tune a9p at normal operating temp.

One thing I recently found that was important to adjust for a raised idle is FN824 (Idle - ISC Gain Multiplier vs RPM). Adjusting this to match my target idle made the idle much more stable.

[Seek]

I should add that I only started messing with idle as with the stock idle setting, I was in traffic the other night while 2 of the 3 lanes on the interstate were closed for road work. Just letting off the gas on flat terrain, the car was able to idle up to speed to 15mph...it's making way too much torque for idle. Lowering the idle did not drop BE's estimated torque output though so I just set it back since that's where the tables begin at. The heavy idle power output seems extreme though, and this was with it still switching and shooting for stoich. The torque converter also stalls to about 2500 so I'm a bit shocked at it's idle behavior. I don't drive the car all that much so I don't notice it often.

[Seek]

I will be looking at this soon - I got some new datalogs and it appears to be jumping between 13 to 30 degrees of timing at idle. It shakes a bit much at idle when it's doing this but has perfect throttle response and doesn't stall. I don't see why the eec would ever want to cause such big swings in spark timing while it's keeping the rpm's between 600-640. It is difficult to know where to start as jumping in hasn't helped me this far.

[Seek]

I don't see what is affecting this. All I see for idle is the Idle Spark Multiplier vs Drive RPM Error and what it would cause but I don't see how it can affect it so much. Below I have an idle datalog when the rpm's are fairly consistent but the spark is all over the place. Any pointers?

[Tipo874]

Your log looks perfectly normal to me for a stock A9P in drive.

The calculations are:

From FN111 (Idle - Spark vs RPM) 30°
From FN841D (Spark Mult vs Drive RPM Error):
10 .883
0 .750
-20 .547

So at:
614 rpm (10 rpm under) spark will be 30 x .883 = 26.5°
624 rpm (on target) 30 x .750 = 22.5°
644 rpm (20 rpm over) 30 x .547 = 16.4°

[Seek]

I see what you're explaining but it doesn't explain my rough idle, yet sometimes it's perfectly smooth. Closed versus open loop isn't the difference but it "may" be warm ambient temperature vs cold to some degree. I noticed there are some extra rows in the scalar - should I start here and put some more fine tuning in for idle? You would think that it'd be possible to get a smooth 624 in there were enough rows - not that we can add anymore. There are no error codes and the cylinder balance test comes back fine so I'm not sure what else I could check to find if a specific cylinder is what is causing all the ruckus.

[Tipo874]

What I'm saying is that your problem is not that spark jumps all over. That is a symptom of some other problem that makes the engine stumble, and the EEC tries to compensate by varying spark advance.

It can be for lots of reasons, and I'm no expert at that.

My 351C was fouling #3 and #7 plugs when I bought the car, and it would miss on low rpms. Turned out to be glazed bores and I rebuilt the engine, and the problem went away. That was easy to spot though as the plugs were oily.

[cgrey8]

I used to get idle hunt that was caused by spark, not AFRs. I confirmed that it was caused by spark by pulling the SPOUT and having the engine run perfect. Connect the SPOUT and it would gradually start hunting again. I solved that by limiting how much the EEC could adjust spark when the RPM moved up and down at idle. It removes the ability of the EEC to stablize RPM at idle, but in my case, it was my fix. My idle is pretty dern stable even without the EEC fluctuating spark for minuscule RPM changes. So I'm happy with basically neutering the EEC's ability to do large spark adjustments based on RPM fluctuations.

[Tipo874]

I agree that spark can cause idle hunt, but shouldn't on a healthy stock motor.

One could try to modify FN841D so that the error term becomes less agressive.

[Seek]

The only things I can think of trying to fix it if it is indeed an issue somewhere is to replace the sparkplugs with another set, replace the wires, rotor and cap, tfi module, and pull all the connectors apart and reconnect again with generous amounts of dielectric grease. These were all replaced a year ago when the new motor went in, minus tfi, so it is possible I had gotten a bad part. All spark plugs look the same though, engine goes WOT smoothly, AF never leans out, no error codes, EGR functioning normally (new valve), etc. Runs better than when the motor had 220k miles but idle may be about the same as it was then with the standard output parts and eec.

[Tipo874]

Have you checked the base spark advance?

[Seek]

The base table is stock - load-30 being 25 degrees and load-20 being 28 degrees. I believe I have a load of ~22 at idle.

[Tipo874]

I meant the base 10° with the SPOUT plug removed.

[Seek]

I can't remember the spout behavior - I think it is rough-ish but I will double check tomorrow. I haven't datalogged it with the spout out.

[Tipo874]

Datalogging with the SPOUT removed is of no use. Just verify, with a timing light, that it is at 10° with the SPOUT removed.

I don't know if the 5.0 dampers suffers from this, but old 351C dampers fairly oftens has the outer ring slipping, which offsets the timing indicators, making it very hard to set timing properly.

[Seek]

I've never had timing jump all over when setting it - I use a 1-year old Pioneer SFI balancer. Timing is set to exactly 10 degrees base but I will verify.

[BobCat]

Datalog ISC Integrator as well as ISC Duty Cycle so you can see exactly what the IAC is doing.

[Seek]

IPSIBR shows a change about every second of .001, between .018 to .022. I did pull the spout and noticed some interesting behavior though - with the idle screw where it's at, the rpm's hunt. I turned the idle screw inward 1/2 turn and it didn't change this. I can also turn it out quite a bit more and it will hunt but not stumble/die. I'm not sure on the point or changes, but I've turned it in so it idles at 800rpm and it was fine (same rpm variations, not percentage but +/- 20 rpm's, but less effect since it's a smaller percentage of change). In park, the rpm's vary between 700 and 750 rpm's, both with the spout connected and not.

With the fluctuations in drive/idle, I see the computer wanting the timing to be anywhere between 16 degrees to 29.8 degrees to correct for the unstable idle. I didn't have any problems like this hunting idle when initially putting it together last fall.

I have taken my enjoyable rush hour datalog and put it on the net if anyone cares to take a look - I only got it working halfway through the trip as Binary Editor was not wanting to load the window again without reinstalling the application. I've narrowed it down to being either the Config or Database folder that is causing BE to randomly quit working when restarting the app. The datalog has me messing with the SPOUT at the end, with the car in park and drive. Link to datalog: http://home.comcast.net/~seekproj/2009_ ... -04-10.zip