Haven't seen anything here in months.
I decided to use the "Big Mike Procedure" (BMP) to calibrate the idle of my LS7 after some mods. I found this procedure quite helpful.
Thanks to BigMike42 and the others who've contributed to this thread.
One question I have is about the Idle Airflow Multi table...when performing the BMP, should it be left alone or have specific values entered?
Hib Halverson
Just Another Tuner in Training.
Not for sure what i missed but when i raise the RPM's to 1500 it stays there and doesn't come back down even though desired shows 900
disregard, i had added too much air in those RPMS and wouldnt let it come back down
Last edited by Jay@HAP; 01-31-2018 at 04:06 PM.
Good write up which has guided me on taming a BTR stg 4 cam on an LS7 with Nick Williams 102mm throttle body and lightweight flywheel all installed on 06 C6 ZO6.
I am sharing some of my findings to help those who don't know and to be critiqued by those who know.
A customer came to me with the same setup stated earlier that had an over oscillating idle between 500 to 1800 rpm. The idle would stabilize for 5 seconds and surge for another. It was quite embarrassing as stated by customer in parking lots or front a red light especially because of the straight exhaust system installed. Car went to several tuners but no
one put the effort to do it right. I wanted for a while to try the guide and see the outcome on a big cam light flywheel setup.
With that type of idle, I had to start from the base and set it up right which was the minimum air flow (RAF) table. I zeroed out all the Proportional and integral tables, set delays to maximum just like stated on the guide and set all timing tables to 20. Zeroed out the over and under spark table and set all idle target tables to 1000 rpm. For the RAF table, I have set all values to 20 g/s. Re-adjusted the ETC Scalar Area to 7400. Started the car up. The idle had a hard time to stabilize without me intervening and catching it by the pedal. With that 20 g/s, it was idling around 1200. I used the VCM scanner negative spark function to reduce idle to 1000 rpm but as soon as idle dropped below 1100 rpm (-5 ignition trim was needed) it would start to oscillate/surge between 600 and 1800. So, I reduced the airflow to 19 g/s and re-started the vette and the idle now is showing the same oscillation/surging when timing was retarded 5 degrees from base. The engine did not want to have a stable idle. It would surge endlessly between 600 and 1800. To my surprise, the logger showed throttle movement that followed the nature of the surge, when rpm dropped below the 1000 rpm mark throttle would increase until it reaches a maximum at 500ish rpm and from there it would reduce until a minimum at 1800 rpm. This was happing even though Proportional, integral and followers are zeroed out. At first I thought there is stall management option that was on found under the engine torque tab but none could be found. Since I have set my scalar minimum idle rpm to 1000, I thought this triggered the ecu to counteract and stop the engine from dying but reducing it to 300 still showed the same oscillation phenomenon. Changed the tune to mafless through tweaking the VVE to rule out any vacuum leak but still oscillation persisted. Even though throttle source was showing idle, I changed the virtual torque values and smoothed them out thinking it might be the cause to the throttle to oscillate with rpm but that also did not help.
The only thing left was the RAF table. All these trials to stop oscillation was due to a misconception I had of the RAF table especially since most of my experience is in universal wired in ECU's (MoTeC/Haltech/Holley...etc)
The concept I had regarding the RAF and E38 idle strategy is, the ecu would first determine the idle rpm target based on the coolant and gear table. Then ecu would move to the RAF table to lookup the base flow while cross referencing the Y-axis Rpm and the X-axis gear. The base flow will be translated by the ECU to a certain throttle opening ratio and be fixed. The adaptives then will take it from there to try to maintain the target idle RPM.
So basically the misconception was, the RAF table is for lookup only and used by the ecu to determine that starting point for the throttle position, while in reality the ECU interaction with RAF was irrespective of the set target idle RPM.
This was an eye opener to me, the whole RAF table was referenced depending on the engine rpm and gear. That's why a single value of 20 g/s inputted in the whole RAF was causing the idle oscillation. A 20 g/s commanded airflow at 500 rpm will translate to a bigger TPS percentage than at 1600 hence causing an unsteady high low throttle.
Once I learned that I had the ability to fullfill my objective of a constant TPS percentage irrepictive to RPM just like a cable throttle through having positive slope values for airflow in the RAF table starting from RPM 0. I knew earlier that the TPS percentage needed to have a stable idle at 1000 RPM was 5.6 and used an error of actual TPS vs desired and created a histogram similar to the RAF table that was populated with percentage error and then special pasted that to the RAF table. Now the car will always hold 5.6 TPS percentage irrespective to RPM and acted like a cable throttle. Doing that is very important before letting adaptives take over. Cold start should also be re-checked to determine Engine Tempwrature Multiplier of the RAF table.
I also have found that if the throttle was changed and the ETC Scalar was not set up right for the bigger throttle, air flow values in the RAF table will not represent airflow indicated by the PIDs Dynamic or MAF sensor in the scanner. In other words, if the idle was stable at 1000 and the dynamic flow PID was showing 20g/s, a value of 15 or 25 might be needed in the RAF table depending on whether the ETC Scalar was under or over estimated.
This is akin to VE table and injector size, if injector was over estimated, VE values will have to increase and visa versa.
This got seriously long
Good write up which has guided me on taming a BTR stg 4 cam on an LS7 with Nick Williams 102mm throttle body and lightweight flywheel all installed on 06 C6 ZO6.
I am sharing some of my findings to help those who don't know and to be critiqued by those who know.
A customer came to me with the same setup stated earlier that had an over oscillating idle between 500 to 1800 rpm. The idle would stabilize for 5 seconds and surge for another. It was quite embarrassing as stated by customer in parking lots or front a red light especially because of the straight exhaust system installed. Car went to several tuners but no
one put the effort to do it right. I wanted for a while to try the guide and see the outcome on a big cam light flywheel setup.
With that type of idle, I had to start from the base and set it up right which was the minimum air flow (RAF) table. I zeroed out all the Proportional and integral tables, set delays to maximum just like stated on the guide and set all timing tables to 20. Zeroed out the over and under spark table and set all idle target tables to 1000 rpm. For the RAF table, I have set all values to 20 g/s. Re-adjusted the ETC Scalar Area to 7400. Started the car up. The idle had a hard time to stabilize without me intervening and catching it by the pedal. With that 20 g/s, it was idling around 1200. I used the VCM scanner negative spark function to reduce idle to 1000 rpm but as soon as idle dropped below 1100 rpm (-5 ignition trim was needed) it would start to oscillate/surge between 600 and 1800. So, I reduced the airflow to 19 g/s and re-started the vette and the idle now is showing the same oscillation/surging when timing was retarded 5 degrees from base. The engine did not want to have a stable idle. It would surge endlessly between 600 and 1800. To my surprise, the logger showed throttle movement that followed the nature of the surge, when rpm dropped below the 1000 rpm mark throttle would increase until it reaches a maximum at 500ish rpm and from there it would reduce until a minimum at 1800 rpm. This was happing even though Proportional, integral and followers are zeroed out. At first I thought there is stall management option that was on found under the engine torque tab but none could be found. Since I have set my scalar minimum idle rpm to 1000, I thought this triggered the ecu to counteract and stop the engine from dying but reducing it to 300 still showed the same oscillation phenomenon. Changed the tune to mafless through tweaking the VVE to rule out any vacuum leak but still oscillation persisted. Even though throttle source was showing idle, I changed the virtual torque values and smoothed them out thinking it might be the cause to the throttle to oscillate with rpm but that also did not help.
The only thing left was the RAF table. All these trials to stop oscillation was due to a misconception I had of the RAF table especially since most of my experience is in universal wired in ECU's (MoTeC/Haltech/Holley...etc)
The concept I had regarding the RAF and E38 idle strategy is, the ecu would first determine the idle rpm target based on the coolant and gear table. Then ecu would move to the RAF table to lookup the base flow while cross referencing the Y-axis Rpm and the X-axis gear. The base flow will be translated by the ECU to a certain throttle opening ratio and be fixed. The adaptives then will take it from there to try to maintain the target idle RPM.
So basically the misconception was, the RAF table is for lookup only and used by the ecu to determine that starting point for the throttle position, while in reality the ECU interaction with RAF was irrespective of the set target idle RPM.
This was an eye opener to me, the whole RAF table was referenced depending on the engine rpm and gear. That's why a single value of 20 g/s inputted in the whole RAF was causing the idle oscillation. A 20 g/s commanded airflow at 500 rpm will translate to a bigger TPS percentage than at 1600 hence causing an unsteady high low throttle.
Once I learned that I had the ability to fullfill my objective of a constant TPS percentage irrepictive to RPM just like a cable throttle through having positive slope values for airflow in the RAF table starting from RPM 0. I knew earlier that the TPS percentage needed to have a stable idle at 1000 RPM was 5.6 and used an error of actual TPS vs desired and created a histogram similar to the RAF table that was populated with percentage error and then special pasted that to the RAF table. Now the car will always hold 5.6 TPS percentage irrespective to RPM and acted like a cable throttle. Doing that is very important before letting adaptives take over. Cold start should also be re-checked to determine Engine Tempwrature Multiplier of the RAF table.
I also have found that if the throttle was changed and the ETC Scalar was not set up right for the bigger throttle, air flow values in the RAF table will not represent airflow indicated by the PIDs Dynamic or MAF sensor in the scanner. In other words, if the idle was stable at 1000 and the dynamic flow PID was showing 20g/s, a value of 15 or 25 might be needed in the RAF table depending on whether the ETC Scalar was under or over estimated.
This is akin to VE table and injector size, if injector was over estimated, VE values will have to increase and visa versa.
Originally Posted by BigMike42
Smoothing out the 2-1 in automatic cars
Cammed autos sometimes may have an issue going 2-1 coming to a stop. If torque management is engaging on the closed throttle downshift, what happens is your idle will drop way too fast, and then overcompensate back the other direction. Idk about you, but this pisses me off and is extremely annoying.
The first thing you need to do is eliminate the timing pull. I can't speak for all ECMs, but removing limit torque management gets rid of the timing pull for my car and I still get to keep my rev matching. If that doesn't work, you need to turn off CT downshift, which will unfortunately get rid of rev matching. Try the limit tq management ect enable first to see if the timing pull goes away.
Ok now that you have the timing pulled, engage a 2-1 stop. One of two things will happen:
- The car is either going to stop with minimal fluctuation
- The car is not going to dip down below idle, but the RPMS will annoyingly rise 500+ rpm from the lack of timing being pulled.
If you're the first one, congratulations. Normally this will happen if you only had to run minimal timing and airflow for your cam. But if you have a cam with some overlap, you probably have pretty decent spark and airflow values, so you will get a spike.
My "trick" to fix this, and it is a bit of a trick...almost a hack, but it works pretty well, is to use the coastdown spark table as my pseudo torque management downshift table.
What I do is put around 8 degrees in the entire coastdown table. If your car has over/under speed for coast, make sure they are zero'd out completely. That's part 1 of the trick. This is going to guarantee that the car will only see 8 degrees of timing during any coasting downshifts. 8 degrees is about the highest you can go before the RPMs start to flare.
The problem now is that the RPMs are going to drop faster because we have such little spark in the coast table. You can solve this by doing either of these methods or both:
- Base airflow table. Add some air to the cells above your idle. Go 1g at a time, log and retest.
- If that is not enough, subtract from the integral airflow table between the -96 and the -512 cells. Take away 10% at a time and retest.
Once these are set properly, there should be no dip below idle RPM as you let off the throttle.
Another problem with such little coastdown spark is low speed coasting is going to sound like your car is dying. What we need to do is bump the idle speed further out.
Set the adaptive idle MPH to the same value as your 2-1 closed throttle shift MPH. In my case, it is 7.2 mph.
Ok you may be thinking now, if we set the idle mph to the same mph as the 2-1 shift, we won't get the 8 degrees of timing.
That is where the idle adaptive delay comes in. Increase the adaptive over/under delay to hold the idle spark out until the shift is complete. The bigger your idle spark/coast spark disparity, the larger the number will be. My example below is going from 8 to 22 degrees spark and while there is a slight RPM jump, it is vastly improved over before when it was just shifting at the full 22 degrees spark.
Play with the idle MPH and idle adaptive delay until you like how the car is stopping. If you can get it to look like the below chart, it will feel pretty smooth.
One more trick that can help the little under rpm blip....
Give the rolling idle 10-15 mph of lead time. So if your target idle is 700 rpm, set 710 rpm to all the cells over 0mph. Some cars come like this already from the factory.
Maybe I missed in the thread.
On the 2-1 downshift. Cant you just use your fast torque exit to calm down the blip?
This was a super helpful post. I like the idea of creating a histogram logging RPM error. I gave it a try, but I found that the logged RPM errors were way more significant than the change you need to make to the airflow table so I'm not sure how you simply copy and pasted the errors into the airflow table.
Also, you mention that you strived to obtain a constant TPS irrespective of RPM. If in closed loop with locked timing isn't the throttle blade the only other factor to raise and lower RPM? In other words don't the higher RPM cells correspond to higher TPS? If not, how would the RPM increase or decrease??
I've been trying to figure this idle tuning out for three years on and off and I still feel lost. I like the tips at the beginning of this thread to keep timing and things constant which is helpful, but I'd like a way to log error to better develop most of the airflow table. Simply using the same numbers for like 1000 and lower doesn't make sense to me. Shouldn't the lower RPMs need less airflow? I know the answer is yes. I mean that is how it is stock and it seems logical.
The method where you zero the airflow table and then command different RPMs while logging dynamic airflow makes sense, but I still don't understand what's going on very well, especially the part where it's recommended to multiply the logged airflow by 0.8 or 0.9.
All that I can think in terms of your statement in regards to multiplying the logged airflow by .8 or .9 would be to correlate the logged airflow with the TPS % as opposed to the TPS actual.
07 Silverado 5.3 full exhaust, CAI, BTR Stage 2 turbo cam and springs 226/231 .605"/.598" 113+4
Thank you.
Being playing with that devil on 2 years.
Somehow it was much worse on 2 bar CoS, but now not sure how much my lower rpm RAF values were related to higher rpm's, but I surely played with them as well.
Now I'll need to try with stock GM OS when got a chance.
Thank you again.
Tried Big Mikes approach and having issues holding idle when out of adaptive after the throttle blip (wants to stall right away). Dont think its strictly related to Base airflow as I went high and low and still happening. When I could get it to idle, I found that 9.7 g/s was as close as I could get to steady idle at commanded RPM. The spark control through the scanner doesn't work for me...tried on and off multiple times. Whats the trick there?
I still need to sort out my idle fueling (seeing between -7 and -20 negative stft, trying to figure out the spread there...doesnt make sense.) I figure the base airflow is going to have to change when I get the fuel right.
Now that I think about it, maybe I should just focus and turning VE back on and getting those two right, since the MAF is more finnicky at low air mass.
For now it would be nice not to have a wild swing at idle, so I think I will make some subtle changes to the Prop and Int tables to slow the movement and hopefully reduce my overshoot on the way down. Then focus on fueling and come back to idle when I like what I have there? Can anyone to affirm this approach?
So I started messing with adaptive idle last night and made some progress with P and I tables, with more tweaking to do. TPS is not doing much at all to control idle (IMO) with shifts back and forth between 10.6 and 11 percent for the most part.
It looks like spark correction is doing most of the work. Wondering if I can pull back the spark over/underspeed tables without having to add back in more airflow idle control?
disable spark control and see what the idle airflow is like alone, if its bad without spark then dial it in as much as u can and then re enable spark, if its still ok then u can lower spark correction but also check for errors by taping the throttle various amounts to see if is still reacts enough dosnt stall or take too long getting back to idle
So I cut spark over and under speed correction values in half and it wouldnt even run...
Put them back in. Messed with P and I more last night and now I'm idling at slightly higher TPS for the same target idle speed and spark is down just a touch.
I blip the throttle to 1500 RPM and it falls too fast and far (2-300RPM) but recovers. Small blips with small error and it recovers pretty nice. Im thinking I need more correction for low error values so that as it overshoots on the way down it starts pulling back faster, rather than dropping so low and then jumping back up.
Also realize I might should just return to factory and only apply changes to the 'gain' essentially, as with PI controller tuning. I have been tweaking specfic ranges and may need to revert and just apply multipliers to the factory values.
u also have .25 sec delay for an over/under event can adjust that to suit if u need a faster response, log adaptive idle advance also it will show u what spark is doing then u can see if its over working because air is too high or low, it should still run without spark correction so maby the airflow is out, set all idle/idle coastdown spark and main spark to 15 deg and all spark correction to zero then start it up and it will be only airflow correcting and using so if it dosnt start or idle pretty good then u need to do more on the min airflow/ I and P to get it right also throttle follower torque plays a roll too, u can make a pid for desired idle rpm and rpm set them up from say 0-1500rpm and the desired rpm will be a line and rpm will be moving as close to that line as it can be for a correct idle if its more under or more over u can see it and make adjustments, can also use the follower steps to help if rpm falls too fast
I think the dying may have to do with rich after reflash and Open Loop at cranking?
When it is idling smoothly, the airflow is close to by BRAF values. Ive continued to have issues with the approach BigMike outlines...When I flash to work on BRAF and spark outside of adaptive, and blip the throttle, it dies or oscillates...Ive tried bumping air flow way up and still stalls. Then every 5th flash or so it would finally work for me. So since it was running decently with some adaptive PI changes and the same BRAF I had prior to the cam, I left BRAF alone.
RPM def falls too fast..the follower steps are a whole nother thing I dont understand...
I will log adaptive idle advance as well. Ive been logging RPM vs Map and DSRD Idle RPM, the error is minimal when steady (within 20 rpm of dsrd)
Attached is the latest file I used to try and follow BigMikes approach to dialing BRAF...did I miss something? I tried messing with the ECT multiplier like @Ghuggins mentions earlier in the thread, when I did it would hold at like double the airflow of the BRAF table (16 g/s idling at 1200 rpm vs the 8.5 g/s in BRAF for 800)
Thanks a lot for your input, and to all those who have contributed to this thread so far! Maybe @schpenxel will meet up with me for a beer and idle conversation seeing as we're in the same town! lol
This is all good but what about the E78 truck?
What I have found is. FIRST, find the idle rpm in which you have the best vacuum with your timing map at stock. SECOND, Fluctuations/oscillations in idle I?ve found to be due to fueling, if you dial in your vve and MAF the idle issues resolve.
Done this many times over with cam swaps and turbo builds with great success. E78 are fun to tune but they can kick your butt and get you chasing your tail if you try changing too many tables at once.
Last edited by RySmDPT14; 12-04-2018 at 11:48 AM.
OK So this looks way intimidating! I just ordered an MPVI2 for my 07' 5.3L LC9 in my jeep to resolve just two issues. When I get the module and software started, I don't know where to start. The background is the motor was new, I had the ecm programmed by sending it to a programmer a few years ago. The configuration is automatic 4L65E trans, no cats, not running flex fuel but still have stock injectors. If the outside temperature is above say, 50 deg.F it starts, idles and runs exceptionally well. The one issue is on deceleration and idle it smells quite rich. The other issue is when outside air is lower that 50 deg. F it barely idles and won't take any throttle without dying until it warms up a little. SO, I don't want to mess up how it performs (Tons of power and great throttle response) but need to find the cause of the cold start and rich smell on decel and idle. If I start the jeep and start taking scans within a few minutes it will be warm and the one issue won't be there to look at. Should I start it and warm it up then try to solve the rich smell issue first and then let it cool off overnight to see if the cold start issue is better? Also, What should I look for first? Download the tune and look at all the tables for something irregular over the ECT vs. spark, or fuel or air or whatever on the cold side of the table? Should I immediately try to start scanning from cold all the way until warm and just study the graph first? You can see I am clueless but I'm intent on finding the issue(s) causing these two minor problems. Any direction would be appreciated.
you will probably want to start a new thread.
It is specifically about Gen IV idle tuning right?
Yes, but I am just trying to help so that you get more views. Do you know if the engine has the stock cam?
Reply With Quote
