Idle's good but fluctuates while coasting

[ForcedC5]

I've got a 6spd and when I approach out of gear the idle fluctuates from 500 to 1500.. As soon as I get to a stop she's steady..


I'll review the posts but if you info off the top I'd appreciate it..

[bink]

Bump up your Running Airflow - aka Idle Airflow.

Idle Airflow should be near your scanned MAF value (g/sec) at each temp range. jfPilla has an excellent writeup on this in the Tuning section ->http://www.efilive.com/forum/viewtopic.php?t=1169.

Engine Calibration-> Idle-> Idle Airflow.

Bump it at your operating temps to begin with ( +0.5 g/sec ?). Increase it until it "Cruise controls" a little....then back it down to stable. You will need to drive it for 30 - 45 minutes to allow for the Long term and Short Term Idle airflow trims to stabilize - between each value change.

Hope this helps some.

Cheers,
joel

[ForcedC5]

Just wanted to say that finally got my car back from the body shop and did what you said and it idles perfectly...


Thanks!!!!

[Aloicious]

okay, so I'm sorry for digging up this ancient thread, and I know both these guys haven't been around here for a while....

but this is the exact same problem I have going on. is he actually referring to the desired airflow table here? i.e. raising it outside of the RAFIG process?

here's whats going on with mine, at any speed, while coasting out of gear (M6), my RPMs will constantly fluctuate between 400 and 1200 RPMs. otherwise, when stationary it idles perfectly. I've done multiple RAFIGs and have the desired airflow table really good from about -4 through 92 ('good' as per the rafig process), I've logged spark, and done some corrections to my base spark table (to make sure a stable and constant advance is commanded in the problem areas, currently 28*), I've looked at TC and TF, made many many changes to those with little to no improvement, I've altered the under/overspeed correction tables for both airflow and spark several times, again no improvement. I'm working on the VE, its not perfect, but its not so far off that it would cause so much rpm fluctuation. sometimes it seems like something does help it, but then after a little while driving it just gets as bad as it was before. I'm going to do a nice log on my way to work today in a couple hours, so hopfully later I can post a log and tune for y'all to look at. I've read and re-read Sspdemon's idle turoial and tried all of that, nothing really helped my issue much, I also read Jfpilla's idle transition tutorial, and tried most of that, but changing the IAC effective area like is mentioned so that the IACDES_B matches DYNAIR, made the truck run horribly, stalling at every stop, etc. I'm about at my wits end with this.

here's some info on the truck in question:
1996 C1500 running a GenIe L31 vortec 5.7L, using LS2 coils with Mike's (eficonnection) coil near plug 24x reluctor setup, custom MPFI setup with ford (bosch) 39# blue giant injectors (flow'd and balanced) at 45psi, 12200411 PCM running a COS3 OS 02020003 (IIRC thats the OS #, its the cos3 OS for a '02 camaro M6 base tune), currently running OLSD (full OL, not semi-OL like the COS can do), ZZ4 cam (.474/.510 lift 208/221 dur, 112LSA), RHS torker heads (close to stock specs on the heads 1.94/1.5 valves, 170cc intake ports, 64cc chambers), intercooled L31 style whipple (side mount ones) with an air to water intercooler, it sees ~5ish psi peaks post intercooler (counting from baseline atmospheric pressure of ~85kPa where I live at ~4500 ft elevation), T56 M6....I think thats about it....

one issue I know I will have, is with the side mounted whipple, the TB (and IAC) is moved quite a bit earlier in the airflow path, so with my setup, both the whipple, and the intercooler is after the TB (and IAC) which makes the already slow IAC corrections even slower. I've tried lowering the slower IAC RPM corrections, and raising the faster spark RPM corrections to compensate for this, but it really didn't make much of a difference with the issue I am having.

it wouldn't bother me too much as it rarely actually stalls due to the fluctuations, but when it drops to 400rpm, my voltage drops and lights dim until it recovers, not good, and very annoying.

any help is appreciated, I'll try to get a tune and log up later tonight too.

[joecar]

The link in post #2 got broken... does anyone have a link to that thread...?

[joecar]

...
Idle Airflow should be near your scanned MAF value (g/sec) at each temp range. jfPilla has an excellent writeup on this in the Tuning section ->http://www.efilive.com/forum/viewtopic.php?t=1169.
...

The link in post #2 got broken... does anyone have a link to that thread...?

I found it, this is the fixed link: showthread.php?t=149 (http://forum.efilive.com/showthread.php?t=149)

[Aloicious]

I found it, this is the fixed link: showthread.php?t=149 (http://forum.efilive.com/showthread.php?t=149)

yeah, thats the one by Jfpilla that I was referring to about the IAC effective area screwing stuff up. (I also don't run a MAF at the time being, so I couldn't do ALL of what was mentioned in that thread)

and just my luck I did a nice log on the way here to work, and it really didn't have the problem much, it would still fluctuate, but before, it would perpetually do it 400-1200 over and over while coasting, on this log it actually settles down to commanded idle after a couple swings. so something is going better.

let me get my patients settled here at work and I'll post up the current log, tune, and an older log that shows the problem a bit better.

[joecar]

If you don't have a MAF, then you can use GM.DYNAIR instead...

DYNAIR is the equivalent "airflow" computed from the VE table (as I understand it).

Edit: grammar.

[Aloicious]

If you don't have a MAF, then you can use GM.DYNAIR instead...

DYNAIR is the equivalent "airflow" computed the VE table (as I understand it).

good to know, I did log that in my log tonight, so you can see the difference between the desired and calculated. I've really been battling this lately, but tonights log shows the most improvement on it in the last 2 weeks or so. I've been working with the spark over the last several days, so that may be what has made the difference.

[Aloicious]

okay here we go, these are fairly long logs here, sorry for the size.

attached is my most current log (12-11-09_to_work_idle.efi)
this is the tune where I would still have fluctuations but they would eventually settle down to commanded idle. you can see how different my desired IAC airflow (GM.IACDES_B) is compared to my calculated air flow rate (GM.DYNAIR) which is kinda erratic at times and I think something within the DYNAIR which is causing the issue. some areas which show what is going on are frames (parenthesis used to separate frame segments for easier readability):

(17040 to 17295) worst area

(15523 to 15733), (10818 to 11102), (10232 to 10481), (9608 to 9848) these segments start off bad, but end up working their way to commanded idle

(16314 to 16460), (14464 to 14676), (13700 to 13820), (8905 to 8985), (4800 to 4940), (3480 to 3560) these all show the problem, but are fairly minor examples

that log was driven using the attached tune (12-11-09_coasting_idle_tuning.tun)....don't worry about the VE table, I know its not pretty, I've done alot of engine work recently and am still getting it in line, I'm also working on the KR too, but the KR isn't affecting the coasting problem at all.

this is an older tune, (12-10-09_to_work_VE.efi) it shows the worst periods of the issue, though I don't know what tune this was driven with. you can see the good amount of spark variation which I have worked on between this and the recent log. this log was just for VE so alot of the idle PID's aren't logged in this one. the most pronounced areas that show the perpetual fluctuations are at TIMEs:
18:42:10 to 18:42:38
18:44:03 to 18:44:93
18:51:57 to 18:52:40
(ignore the CALC.BOOSTNVAC, the PID is written incorrectly for this log and its not calculating anything right)

any help is much appreciated! though I think the spark alterations have at least started me on the right path.

[Aloicious]

okay, so I'm at work here, and I've been looking over logs, and reviewing the vast differences in the desired and calculated airflows I think something with my IAC effective area is probably incorrect (quite possibly since the B4403 in my tune is from a 2002 L31 xvan table, but the 2002 L31 xvan doesn't use the same IAC as the 1996-1998 L31 trucks, at least according to the part numbers I can find)...so I started looking at other B4403 curves and found lots of different stuff. so I tried an experiment. I took the B6667 from a 1998 C1500 (closest truck to my 96 that EFILive supports) which is the IAC effective area table, but its IAC %, not sqmm....so I extrapolated the 1998 B6667 into a possible B4403 in order to compare the IAC curves and what is going on with them for various vehicles, and perhaps find a curve closer to the 98 to use rather than the '02 xvan one.

basically I took 0% from the 1998 tune to mean 12 in sqmm (since at IAC fully closed the PCM expects a 12 mm^2 open area in the throttle blade on a stock TB), so starting at 12, and going up to 120 mm^2, I extrapolated the percentages and used the 1998 values, then used the linear fill tool in EFILive to estimate the intermediate cells. and plotted it against various other B4403 tables for comparison. so the physical curve between the known cells probably aren't correct, but we can get a general shape of the overall area under the curve to compare.

its kindof interesting to see them plotted against each other, though none of the IAC tables I compared it to matched too close, it is, if nothing else, rather interesting to check out. I may look into this area a little more and see what I can find, because I know stock L31 calibrations were different on a number of simmilar stuff through the 1996-2002 years.

if anyone is interested in looking at it I'll attach the spreadsheet.

Edit- spelling/wording

[Aloicious]

okay, so after some research I'm not going to mess with the IAC effective area table, i may have figured out why i cant get DYNAIR to match IACDES, I believe my idle trim limits are too low, so after doing a RAFIG, it gets it closer, but only a slight bit, so I'm going to try some of the advice given by Ross and Jesse in this thread first, posts 11 and 12:

http://forum.efilive.com/showthread.php?t=473&highlight=dynair&page=2

that should get my idle to hold itself without relying on the IAC position, that way any RPM drops will be minimal and easily corrected by the spark correction, then I can get more productive RAFIG sessions with a wider Idle trim learn limit.

I do have a question though. the problem seems much worse when my heater is on (which is alot nowadays, its cold here), does the heater count in the A/C tables?

[joecar]

I'm still reading along...

[Aloicious]

I'm still reading along...

heh, yeah I put alot in there. I'm at work again tonight, I did another log/tune I can post a little later. after adjusting the IAC, like Ross mentioned in that post I linked to, it dropped my IAC counts much lower (which was expected), like idling at 30, where it was at 80, and it still has the fluctuations, but doesn't seem to drop as far, as much. so thats good. but its still does it, and its still much worse with the heater blowing.

I'm also wondering about RAFIG, I've been monitoring the DYNAIR vs IACDES for several logs, and with doing the rafig processess it doesn't make them coinside any better. I thought that was the purpose of doing the RAFIG??

so then if the rafig process isn't making the dynair/iacdes correlation any better. I may have to manually alter the B4307 to make them better (I'm thinking that the dynair/iacdes mismatch is a big part of my problem). but what does increasing/decreasing the b4307 change? does it increase/decrease the IACDES_B? wouldn't altering it also alter the dynair too, making altering b4307 to correct them moot?

[mr.prick]

I've always heard that you want IAC counts to be 60-80 at idle.
So the IAC has enough room to move up or down. :nixweiss:
I increased {B0108} because TPS% fluctuates from 0%-.4% at idle :confused:

[Aloicious]

I've always heard that you want IAC counts to be 60-80 at idle.
So the IAC has enough room to move up or down. :nixweiss:
I increased {B0108} because TPS% fluctuates from 0%-.4% at idle :confused:

yeah I adjusted the throttle screw slightly so it doesn't have the .4 TPS anymore. I have also heard the 60-80, and once I get the swing figured out I will most likely move it back up some, but for now, its still hunting, but not droping to like 400 rpm and droping voltage. it only drops to maybe 600 now, but swings up higher.

[Aloicious]

hmmm...so I've been looking at this and it seems whatever dynair is logging is having the main effect on the problem, does anyone know whate xactally DYNAIR is? Jocar, you said its looked up from the VE, but after several VE sessions it doesn't change the fluctuations. I wonder if there's some other variable that is included in what is logging that I'm missing.

[macca_779]

hmmm...so I've been looking at this and it seems whatever dynair is logging is having the main effect on the problem, does anyone know whate xactally DYNAIR is? Jocar, you said its looked up from the VE, but after several VE sessions it doesn't change the fluctuations. I wonder if there's some other variable that is included in what is logging that I'm missing.

What AFR is your Wideband seeing during these conditions. Usually the most common cause of hunt I see is being to rich. As a test use the DVT controls and command a progresively leaner mix and see what happens. You have to rule that out before you continue chasing your own arse looking at air flow.

[Aloicious]

What AFR is your Wideband seeing during these conditions. Usually the most common cause of hunt I see is being to rich. As a test use the DVT controls and command a progresively leaner mix and see what happens. You have to rule that out before you continue chasing your own arse looking at air flow.

commanded is 14.68, the WB is actually runs a shade on the lean side, average BENs are around 1.01 to 1.02ish in the affected cells. there is small fluctuations with AFR but the wide RPM swings dont match any AFR swings. I'm going to go do some more logging tonight so I'll keep a closer eye on the AFR/RPM correlation.

[Aloicious]

okay I think I've got it figured out for the most part. it ended up being a lot of little things with subtle changes needed. the main issue was the spark, after smoothing it out at the affected cells, (ended up using 28*) this helped some but didn't cure it completely, then I found that I could recreate the oscillations while commanding zero IAC (with bidi) which showed it was not IAC fluctuations causing the RPM changes, so while logging this, I found that the AFR's were a little out of whack like was mentioned earier. turns out it was slightly lean at low RPM cells, a little rich, at higher RPM and kPa areas of the fluctions (800-1000, 45kPa), and about stoich at correct commanded idle (800-850RPM, 35kPa), so changing my VE to correct this, also helped out, but again, it didn't cure it, and put a larger ugly downward spike in my VE. so currently, I,ve been trying to cure that spike and the minor fluctuations still left by changing the cracker tables to supplement the airflow in the 1000 especially, which should help that VE spike by supplementing the off throttle airflow, as well as raising the 1000+ RPM TC table to also assist in the off throttle airflow and hopefully slow down the RPM drop.

after all that I adjusted the TB screw again, to bring my IAC counts back up to ~60 steps at warm Idle A/C and heater off. since at the lower IAC counts I was using earlier, I would see the steps drop to 0 at times. and afterwards did a couple RAFIG's to bring the desired airflow table in line, also of note is that I did the RAFIG with the heater ON, which gave slightly higher trims than with the heater off, and I'd rather have the desired flow slightly higher, and trimmed down a little than a little low and having the slow IAC trying to trim it up.

and on top of all that I raised the commanded RPM at idle to 850 (from 800) which helped out as well. and retured the IAC correction tables closer to stock (they were lowered alot when I thought the IAC was causing the problem) to give it the ability to make changes as needed, as they're needed.

after all that it seems to be much better, there are still very minor fluctuations but NOTHING like it used to be. and with some fine tuning on the TC I bet it will pretty much be eliminated

[Aloicious]

thanks for the help and suggestions that were given. they did help out and are appreciated.

[joecar]

I'm reading along to learn (idle is not my best subject)... looks like you're getting there, keep us posted.

[Aloicious]

I'm reading along to learn (idle is not my best subject)... looks like you're getting there, keep us posted.

yeah, it still does it a little at times, which I'm still working on, but its so much better, in fact I may try dropping the commanded idle back down a little, it just seems kinda high to me. I'll see if I can find one of my zero'd IAC logs to show you what it was doing in those. maybe someone will have an idea on more possible issues as well. my setup is pretty one-of-a-kind, so it makes issues like this a little more difficult.

Idle isn't my forte either but I've learned alot from this little escapade.

also worth noting is that I've also been using the stock IAC effective area table from the '02 corvette/'00 silverado also, since although its not perfect, I think the area under the curve better matches my extrapolated '98 effective area table better than the '02 xvan especially at the low extreme which is most important IMO. I don't know if that made a difference or not, of course had to do a couple more RAFIG's but it seems to be working fine.