ARGH!!! Surge is Driving Me Nuts! Help plz!

[Motorhead-48]

I tried an experiment. I set my idle coast table up so the values in the cells where the spikes occured were now 1 degree higher than the High Octane table values. What I now see is 1 degree upward spikes when the surge occurs. So I'm confident that the changes in timing seen in the logs are a result of the car transitioning from the High Octane Spark table to the Idle Coast Down Table. Not a eureka finding and it doesn't explain the surge but another data point.

Why it does that....dunno

For DSTECK: How low did you go with your timing in the areas affected by surge?

[DSteck]

I tried an experiment. I set my idle coast table up so the values in the cells where the spikes occured were now 1 degree higher than the High Octane table values. What I now see is 1 degree upward spikes when the surge occurs. So I'm confident that the changes in timing seen in the logs are a result of the car transitioning from the High Octane Spark table to the Idle Coast Down Table. Not a eureka finding and it doesn't explain the surge but another data point.

Why it does that....dunno

For DSTECK: How low did you go with your timing in the areas affected by surge?

22-24º

[Motorhead-48]

I spenty yesterday afternoon trying different combinations of timing changes while logging and reviewing data. To sum it up what I found is that timing mods don't stop the surge in my car. Reducing the timing tends to mask the surge somewhat but it is still there.

Looking elsewhere (in the tune) for a solution.

[wstaab]

LS2, LS3, & LS7 E38 ECU engines with high performance cams are very sensitive to A/F ratios. Stock cams have no overlap (negative overlap) for idle quality and emissions. After market cams have overlap (intake and exhaust valve open at the same time) to improve cylinder scavenging at mid to high rpm depending on how much overlap and the intake center line. Cam overlap makes an engine less efficient at low rpm. Most high octane fuel in the US is E10 today. 14.68 A/F is for 100% gasoline. E10 needs to run around 14.2 A/F. Cam overlap introduces some unburned fuel into the exhaust port at idle and low rpm. The more overlap the more unburned fuel is introduced into the exhaust system. If your Stoich is set to 14.68 and you are operating in closed loop mode and use a 4 gas exhaust analyzer to measure exhaust gas content along with a EGT probe located 4 inches from the exhaust port you will discover the actual A/F ratio in the combustion chamber is as lean as 16 to 1. in closed loop. The extra fuel in the exhaust manifold causes the ecu to adjust fueling to maintain 14.68 in the manifold which is too lean in the combustion chamber. Also the closed loop lean/rich switching for catalytic convertor efficiency aggravates the surging.
The fix.
This is assuming you have calibrated your MAF transfer using a wideband and either disabled VE or have calibrated it too. I prefer to use MAF only until everything is calibrated and then tune VE with a wideband with MAF disabled.
Also you need to make sure your airflow tables are calibrated for any mods such as ported TB, ported intake manifold, and minimum idle speed.
Most tuners try to idle a cammed engine too low for the sound, not performance.
Any cam with more than 10 degrees of overlap should have a minimum of 900 to 950 rpm idle speed on a V8 engine. Some pure race cams with 30+ degrees of overlap idle at 1500 to 2000 rpm.
If your fuel source is E10 set Stoich to 14.20. For pure gasoline 14.68(stock). Disable closed loop and LTFT in your tune. In your Open Loop Airflow Gain Table set your A/F ratio to 13.8 in the cells from 0 to 1600 rpm and 20 to 60 kPa. This will force the commanded A/F ratio to 13.8 at idle and low throttle settings and eliminate the rich/lean closed loop swing.
Drive the car fine tune the timing tables in the 0 to 1600 rpm and 0 to 60 kPa cells. This will eliminate 99% of the surging. The 1% you cannot eliminate is caused by reversion into the intake manifold at low speeds due to the overlap in the cam which causes port to port imbalance in the intake manifold. Pure race engines have individual throttle bodies for each cylinder to overcome reversion problems with high overlap cams.
For closed loop operation for fuel mileage at cruise and the ability to fine tune the A/F ratio with LTFT's because of changes in air temp, humidity, and barometric pressure there is only one way I have found to modify the A/F ratio in the 800 to 1600 rpm range in closed loop from Stoich. You can modify the PE tables to 13.8 A/F in the 0 to 1500 rpm cells and set the PE TPS threshold at 0 in the 500, 1000, & 1500 rpm cells. This will result in an open loop commanded 13.8 A/F below 1600 rpm and revert to normal closed loop above 1500 rpm.
If there is another way to modify closed loop Stoich based on rpm I have not discovered it in the E38 and E67 ecu's.
Any other tuning ideas are welcome. The learning process never ends in "electronic" tuning.

[Motorhead-48]

wstaab,

Lots to digest here...thanks for taking the time to post this up. I gotta go read it a few more times to get my head around all of it.

Question: For the Open Loop Airflow Gain Table am I building a value off the new Stoich 14.2 for E10? i.e. am I doing a divide....i.e. 14.2 / 1.025 = 13.8 as I do with the PE values?

Second Question: Should all of my PE values now be based off of my new Stoich of 14.2 ???

[DSteck]

Second Question: Should all of my PE values now be based off of my new Stoich of 14.2 ???

No. PE is the inverse of lambda. Shoot for ~.85 lambda in PE.

[wstaab]

wstaab,

Lots to digest here...thanks for taking the time to post this up. I gotta go read it a few more times to get my head around all of it.

Question: For the Open Loop Airflow Gain Table am I building a value off the new Stoich 14.2 for E10? i.e. am I doing a divide....i.e. 14.2 / 1.025 = 13.8 as I do with the PE values?

Second Question: Should all of my PE values now be based off of my new Stoich of 14.2 ???

If you are tuning a E38 or E67 ecu and setting your Stoich in the in the fuel control, general, air fuel ratio, to 14.2 in the 0.00 % cell then yes.

Open Loop Airflow Gain Table would be 1.03 (14.2 / 1.03 = 13.78)
Remember once your RPM and MAP values are outside the low speed areas in the Open Loop Airflow Gain Table your normal open loop gas tables are controlling the A/F. Make sure you have the values set to 1.00 in the normal operating areas in those tables.

PE EQ ratio table would be 1.03 from 0 to 1500 rpm.
I use 1.14 (14.2 / 1.14 = 12.45) from 1750 to 8000 rpm on a 12.1 compression ratio, 12 degree overlap LS7 with 25 degrees of timing at WOT from 4400 to 7200 rpm. No knock retard with 93 octane BP pump gas with this configuration. PE table use the Stoich value in the in the fuel control, general, air fuel ratio, 0.00 % cell for calculations.

[wstaab]

No. PE is the inverse of lambda. Shoot for ~.85 lambda in PE.

Confused as to what your are saying.
Lambda A/F is always 1.0 for what ever fuel you are tuning.
Lambda 1.0 for gasoline would be 14.7.
Lambda 1.0 for E10 would be 14.2.
Lambda 1.0 for 100% ethanol would be 9.0.
85% of E10 (14.2) would be 12.07
88% of E10 (14.2) would be 12.50
97% of E10 (14.2) would be 13.77

[Motorhead-48]

If you are tuning a E38 or E67 ecu and setting your Stoich in the in the fuel control, general, air fuel ratio, to 14.2 in the 0.00 % cell then yes.

Open Loop Airflow Gain Table would be 1.03 (14.2 / 1.03 = 13.78)
Remember once your RPM and MAP values are outside the low speed areas in the Open Loop Airflow Gain Table your normal open loop gas tables are controlling the A/F. Make sure you have the values set to 1.00 in the normal operating areas in those tables.

PE EQ ratio table would be 1.03 from 0 to 1500 rpm.
I use 1.14 (14.2 / 1.14 = 12.45) from 1750 to 8000 rpm on a 12.1 compression ratio, 12 degree overlap LS7 with 25 degrees of timing at WOT from 4400 to 7200 rpm. No knock retard with 93 octane BP pump gas with this configuration. PE table use the Stoich value in the in the fuel control, general, air fuel ratio, 0.00 % cell for calculations.

Got it and thanks for the doing the math to back it up with...helps me confirm my head is pointed in the right direction.

My initial run with those values made a huge difference in surging even before I've had a chance to recalibrate the MAF and VE.

Any reason I couldn't leave the 500rpm value alone in the PE table to allow the car to idle in normal closed loop?...trying to minimize the "stink". I'm assuming those cells/values represent 0rpm to 500rpm, 500rpm to 1000rpm, etc.

[wstaab]

Any reason I couldn't leave the 500rpm value alone in the PE table to allow the car to idle in normal closed loop?...trying to minimize the "stink". I'm assuming those values are 0rpm to 500rpm, 500rpm to 1000rpm, etc.

If your idle is ok in closed loop at 14.2 A/F then set the PE enable TPS back to normal in the 0, 500, & 1000 rpm cells. The TPS % has to be less than the value in these cells to enable PE. I use 45 in the 0, 500, 1000 cells and 25 in all the rest if I do not want PE enrichment at low, rpm low throttle settings. I use 0 in 0, 500, 1000, & 1500 cells when I want open loop 13.8 A/F below 1600 rpm.

[Motorhead-48]

If your idle is ok in closed loop at 14.2 A/F then set the PE enable TPS back to normal in the 0, 500, & 1000 rpm cells. The TPS % has to be less than the value in these cells to enable PE. I use 45 in the 0, 500, 1000 cells and 25 in all the rest if I do not PE at low, rpm low throttle settings. I use 0 in 0, 500, 1000, & 1500 cells when I want open loop 13.8 A/F below 1600 rpm.

Copy all..thanks...beer owed!

[DSteck]

Confused as to what your are saying.
Lambda A/F is always 1.0 for what ever fuel you are tuning.
Lambda 1.0 for gasoline would be 14.7.
Lambda 1.0 for E10 would be 14.2.
Lambda 1.0 for 100% ethanol would be 9.0.
85% of E10 (14.2) would be 12.07
88% of E10 (14.2) would be 12.50
97% of E10 (14.2) would be 13.77

A lambda of one is always stoich regardless of fuel, yes. The numbers in the Power Enrichment table are 1/lambda, also known as inverse lambda. With pure gasoline, a 12.5 AFR is ~0.85 lambda. The inverse of that is 1.176, which is what you'd put into the PE table for commanding that air/fuel ratio.

The stoich air/fuel ratio of your gas doesn't mean you have to change the values in the PE table based on it, because you should be thinking in terms of lambda to begin with. 0.85 lambda is a good target, even for 10% ethanol gasoline.

[E85 LSX]

Wataab,

are you saying that the cylinder is very lean but the oxygen is then mixing with unburnt fuel in the exhaust and burning so the wideband reads stoich?

[wstaab]

Wataab,

are you saying that the cylinder is very lean but the oxygen is then mixing with unburnt fuel in the exhaust and burning so the wideband reads stoich?

NO

At low rpm an engine with cam overlap will have unburned fuel (hydrocarbons) introduced into the exhaust system due to the scavenging effect of cam overlap.
This unburned fuel causes the exhaust gas to have less oxygen content which the narrow band o2's read as a rich mixture. The actual A/F ratio in the combustion chamber is leaner. In closed loop the ecu will modify the fuel flow (injector cycle time) to bring the exhaust mixture to Stoich as set in you table.
14.7 A/F in the exhaust manifold will be leaner in the combustion chamber due to cam overlap.

[MCScott]

NO

At low rpm an engine with cam overlap will have unburned fuel (hydrocarbons) introduced into the exhaust system due to the scavenging effect of cam overlap.
This unburned fuel causes the exhaust gas to have less oxygen content which the narrow band o2's read as a rich mixture. The actual A/F ratio in the combustion chamber is leaner. In closed loop the ecu will modify the fuel flow (injector cycle time) to bring the exhaust mixture to Stoich as set in you table.
14.7 A/F in the exhaust manifold will be leaner in the combustion chamber due to cam overlap.

Actually, not quite. It is true that a large cam will cause some unburned A/F to leak out the exhaust valve. Additional partly burned mixture is exhausted due to the partial misfire condition caused by big cams (that's what makes them "lope"). Both of those things actually cause more oxygen to appear in the exhaust, which in turn causes the O2 sensor to read lean, which causes the closed loop correction to richen past true stoich in an attempt to correct. That's why really big cams have a tendency to idle stinky rich in closed loop.

You can clearly see this effect using an exhaust gas analyzer on a car with bypassed cats - the CO level goes way up. Some folks use the "low-RPM PE" trick you describe to make the car idle OL, in order to correct this condition.

Your point overall about heavily cammed cars running better at slightly richer than stoich at low RPM I have also found to be true in my experience.

[E85 LSX]

Agreed ^^

[Motorhead-48]

Well although there is some difference of opinion in the above threads about the "science", the one thing that I have to say as the OP of this thread looking for assistance is that after chasing about a dozen different techniques over the past several months to eliminate this surge, the technique passed to me by "WStaab" just plain works! My surge problems are gone.

Thanks again!

[MrDrezzUp]

If the gas being used does not have ethanol is it acceptable to change only the "PE Enable TPS" and the "Power Enrichment Fuel Multiplier vs. RPM"?

[wstaab]

If the gas being used does not have ethanol is it acceptable to change only the "PE Enable TPS" and the "Power Enrichment Fuel Multiplier vs. RPM"?

Yes

[E85 LSX]

Does the timing still jump up and down when using this open loop method?