When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Working on smoothing out the idle in $8D. Comparing my settings to OEM. I am not sure that I have the correct OEM settings for Idle Spark Advance and Idle Spark Retard vs RPM error.
It makes sense to me that as the RPM moves higher or lower than Commanded RPM, you can advance or retard accordingly to bring the actual idle back to Commanded. What it appears is that the OEM settings (see attached PDF) have the Retard Correction at "zero" for a zero error in RPM. However, the Advance Correction is set at 4.92 degrees at zero error.
That doesn't make sense to me. Unless I am missing something, this table setting would advance the spark by 4.92 over the SA table value when you are exactly at the commanded Idle RPM.
Why wouldn't both values for zero rpm error be "zero" and the correction values begin in the 50 rpm entry?
Good question. Could never find an answer. I put 0* correction at 0 rpm error on my tune instead of OEM 4.9*
Maybe someone will chime in and explain the logic.
I would have to look, but 4.92 may be canceling a bias value. Remember an ECM cannot have a negative value without having a bias value. If the bias was not included in the XDF calculation from hex for that table it will look weird.
So if you have a range of say -10 to 10. You need 10 bias value in the code. So if you are looking at the table ignoring a -10 bias the value would show 0-20, but with the 10 bias its -10 to 10.
In the past I have re-written some XDFs that were floating around to have the correct bias values. For example the TBI ECMs did not have the bias values included for the main spark table and it looked as if they had alot of timing when in reality it was -9.81* less timing than the timing maps were indicating.
I've done something similar w VE tables so they show actual VE rather than NET VE after Bias. Part of the reason I did that was so when calcing a new VE frpm a datalog, it would take the BIAS into account, otherwise the changes were calc'd just on the net and would therefore be understated.
I've done something similar w VE tables so they show actual VE rather than NET VE after Bias. Part of the reason I did that was so when calcing a new VE frpm a datalog, it would take the BIAS into account, otherwise the changes were calc'd just on the net and would therefore be understated.
Yep I datalog VE tables on LS PCM stuff. MAF calIbration is alot easier to adjust quickly and I can make it run MAF only to where it runs and drives well. Then I can datalog the VE tables to make an easier starting point for initial speed density tuning.
Are you saying that the zero rpm error advance should be set at 4.92 (the equivalent of no change at zero RPM error.) If that's the case, doesn't all the 4.92 degree retards equal zero change.
I was looking at some really ancient post from Rbob and he posted that the OEM for retards was 9.84 degrees for 50 and 100 rpm. That would jive with the concept of a "bias" amount
Are you saying that the zero rpm error advance should be set at 4.92 (the equivalent of no change at zero RPM error.) If that's the case, doesn't all the 4.92 degree retards equal zero change.
I was looking at some really ancient post from Rbob and he posted that the OEM for retards was 9.84 degrees for 50 and 100 rpm. That would jive with the concept of a "bias" amount
Don
Williamsburg
Just to add to the confusion.
I was looking at some data logs and the commanded SA at idle is right on with the tables. So what ever the error is +/- rpm, there is KR or SA added according to the tables. Like I said my tables have be retuned for 0* if 0rpm error and it's what I get. If it's +/-25rpm, I get +/-2.5*, if +/-50rpm I get +/- 4.9*, etc.. Just like the tables are set for. I don't have any data log with OEM 4.9* for 0rpm error.. I could do some testing but the car is on the hoist for winter project!
What are the values in your tables for Errors advance and errors retard?
I am curious about the changes you say are at 25 rpm (table is in 50 rpm increments). I wonder about the increments. If a 5 degree change plus or minus at a 50 rpm error doesn't smooth out the RPM, why would the same change (5 degrees) be correct for 100, 150 or 200 rpm.
What are the values in your tables for Errors advance and errors retard?
SR
0_____0.0
50.___4.9
100___6.3
150___7.7
200___7.7
SA
0_____0.0
50.___4.9
100___4.9
150___4.9
200___4.9
Remember, the ECM extrapolates. So an error of 25RPM = around 2.5* using my tables. Also, maybe GM did not want to add too much SA for under rpm idle error to avoid surges.
I never really gave a second thought on this because idle is stable and Idle SA never really gets corrected more tan 5* most of the time
Idle stability has been a big challenge for this car. Since changing from the original 502 block/cam to the ZZ502 block/cam, idle has been the one noticeable rough spot. It's tolerable now, but still seeking to smooth it out, if possible. Candidly, it is idling faster than I would like as it pulls at the brakes. I have tried lower that 900 9850-800) using the APU, but smoothness goes away quickly. At 900, the TC is the issue, but I am not about to change that. The csv log would show that the movement is about 75 rpm from high to low. Commanded idle is 900 rpm. The range seems to be 925 to 850. Idle SA is 22.15 degrees. I stole your idle error values from a post from you some time back where you posted your "final" BIN.
Am I chasing a ghost or is there something I could tweak to narrow the variation to plus/minus 25 degrees. Attached are my BIN settings and a screenshot of idling in gear.
My 383 set up likes 900 rpm at idle, that’s how I solved the rough idle due to O2 sensor voltage swings. The swings are normal at idle for catalytic converter action. Fortunately I don’t have a TC. Also, with under-drive pulley, voltage was getting too low for my likings while idling at red light.
It runs way better at idle in open loop with no swings. Very very smooth, but I do have a Cat Converter, so I run closed loop at idle.
I played with the O2 sensors windows for idle and it helped a bit.
Since my biggest swing is down (900 down to 850), I am thinking I should bump up the Idle Error advance at 50 rpm from 4.92 to something higher. Assuming the ECU interpolates between values, perhaps a higher advance at 50 rpm in the table would keep the drop in idle to the 25 rpm range.
I have tried OL idle and OL full time. I think the idle rpm error tables still apply in OL. I will try OL idle again.
You have to find out what causes the roughness and make rpm go down.
Try data logging warm idle for a few minutes and look at the O2 voltage swings. Make an average ans see if you are on the lean side. If so, up the idle O2 voltage parameters.
You should hear the idle roughness changing as it follows the O2 voltage sin wave (that’s what it looks like on a scanner). It’s going from lean to rich to lean on and on. In my case that’s what was causing the roughness. Hell, it even does it on a stock engine LB9.
I fought my idle for years with a mild cam. I finally went to open loop idle. My combination likes 28 degrees advance (it only fluctuates 27-28) at 750 rpms and afr of 13.8 at idle and 50 KPa. Now it idles great with only a slight lope that most people would not even notice.
I fought my idle for years with a mild cam. I finally went to open loop idle. My combination likes 28 degrees advance (it only fluctuates 27-28) at 750 rpms and afr of 13.8 at idle and 50 KPa. Now it idles great with only a slight lope that most people would not even notice.[/QUOTE]
I don't remember why I set them that way! According to my notes it was supposed to make idle smoother. I must have read something on TGO lol. I haven't played with that in a while and kinda left the idle that way since the same thing was happening with my 2 other LB9s but a little less noticeable. Maybe I should increase them all to get richer. I'll tinker with that one day maybe.
I did that on my stock LB9 '89 Camaro which was recording lean at idle on AFR gauge. It would swing from low 15s to mid 16s AFR and O2 average voltage was in mid 0.300. It did smooth out the idle and got richer. The engine hesitates less when the swing goes leaner. It is a '7165 with $6E so there is no closed loop parameters.
Last edited by SbFormula; Jan 19, 2023 at 12:07 PM.
I do not have any Cat converters. I am going to try two things on the idle first.
1. On the RPM errors, my Commanded idle is 900 rpm at 22.15 degrees. The variation on the log is from 850 to 925, meaning more under rpms than over. I am going to bump up the Idle RPM advance from 4.92 to 7.73 degrees. That should give a quicker advance change when the rpm drops below commanded.
2. I am going to set idle in open loop (CL over 1,000 rpm.
Will see what happens. Revised Error tables attached
[QUOTE=Sharp38;6490770]1. The difference between OL idle and CL idle is noticeable.
Larry: what does your SA table look like coming off of a 28 degree idle advance (or did you control that through the Closed TPI idle This is with GM fastburn heads and 10.25 Compression ratio and 93 octane fuel. It likes a lot of advance down low but not too much on top. I run open loop all the time because idle to 1200 rpms is too lean in closed loop.
Last edited by Larry; Jan 19, 2023 at 12:42 PM.
Reason: spelling error
I do not have any Cat converters. I am going to try two things on the idle first.
1. On the RPM errors, my Commanded idle is 900 rpm at 22.15 degrees. The variation on the log is from 850 to 925, meaning more under rpms than over. I am going to bump up the Idle RPM advance from 4.92 to 7.73 degrees. That should give a quicker advance change when the rpm drops below commanded.
2. I am going to set idle in open loop (CL over 1,000 rpm.
Will see what happens. Revised Error tables attached
Remember, your rpm variations at idle are almost normal. I don't' know if you can ever get them super tight. OEM will do the same +/- 25 rpm both ways.
Changing the Idle error table or idling in OL made no difference in the rpm stability. I found this on the DIYAutotune website.
Engines don’t like to idle lean; a lean air/fuel ratio is one of the biggest causes of a surging idle. On an engine with sequential injection, this is simple enough – get the engine idling at 14.7:1 for gasoline, and you’re good to go. Batch fire engines, however, have issues at idle with intake pulse reversion sending some of the fuel into an adjacent cylinder. So, if your engine is batch fire, you’ll need to add a little more fuel so all the cylinders can stay at 14.7:1 or richer. Batch fire engines typically idle best in the mid 13’s, so if you’re leaner than that, put in some more fuel and see if that stabilizes the idle.
I am idling right at stoich or a tad lean (1.01 lambda). I am batch fired, so will richen it up a bit and see if there is any change.
Changing the Idle error table or idling in OL made no difference in the rpm stability. I found this on the DIYAutotune website.
Engines don’t like to idle lean; a lean air/fuel ratio is one of the biggest causes of a surging idle. On an engine with sequential injection, this is simple enough – get the engine idling at 14.7:1 for gasoline, and you’re good to go. Batch fire engines, however, have issues at idle with intake pulse reversion sending some of the fuel into an adjacent cylinder. So, if your engine is batch fire, you’ll need to add a little more fuel so all the cylinders can stay at 14.7:1 or richer. Batch fire engines typically idle best in the mid 13’s, so if you’re leaner than that, put in some more fuel and see if that stabilizes the idle.
I am idling right at stoich or a tad lean (1.01 lambda). I am batch fired, so will richen it up a bit and see if there is any change.
Sounds all true! Definitely doesn't like being lean at idle that's for sure.
What was your AFR in warm idle open loop? I have my 383 set-up high 11s to mid 12s (real AFR on gauge) for cold start-up and first minutes of warming-up. It loves it. Runs so smooth. Then it gradually goes up in the 13s. As soon as the closed loop kicks in, it starts being a tad rough. Again that rich-lean-rich-lean wave.
I actually changed my closed loop O2 windows on the 383 and will test it when I get a chance. It did help my '89 Camaro LB9. From factory it was idling quite lean.
0.633___0.677___0.721
Last edited by SbFormula; Jan 21, 2023 at 09:39 AM.
I actually changed my closed loop O2 windows on the 383 and will test it when I get a chance. It did help my '89 Camaro LB9. From factory it was idling quite lean.
0.633___0.677___0.721
Well, well, well!!!! I finally got to test this. Absolutely better. I settled at:
0.581___0.625____0.669
I was even capable of lowering idle to 850rpm without sacrificing smoothness. I might try 800rpm in hot summer, see what it does. However, if voltage goes down, my idle will increase. It's just the way I tuned it. The under drive pulley on alternator can be a pain.
So the ECM still swings the AFR (doing it's job for the cat converter) but it barely crosses 15.0:1 now. It corrected that hesitation it had every time the AFR would get too high in the 15s while swigging.
Last edited by SbFormula; Feb 2, 2023 at 04:42 PM.
