Surging and LTFT questions on LS1

[abc]

Good morning fellas.
Can you guys help me with sharing some of your experience? I have a few questions and if you would rather me start separate threads for each question just say to, but for the mean time I'm gonna throw them all on here and see If I can get some clarity.

LS1 with a 4" crank, TEA 220 heads, 236 I, 248 E, @ .05", .620" and .610" lift and 115 LSA cam, centrifugal blower, stock intake and throttle body.

1. On a 2 bar operating system, does the LTFT function above 105 KPA? Depending on the answer I will probably have more questions. I was seeing LTFT affecting my first few low boost tests until I discovered a couple things. I learned how to clear the LTFT (could't find it at first) and Getting the right PID's and charts setup for monitoring the MAP correctly after moving to a 2 bar MAP sensor and latest version of HPT software. I'm thinking the ECM was still using the 105 KPA cell when I had the PID's setup wrong and was populating LTFT's even in the low boost areas.

2. Can you guys help me with idle surge/hunt and lite load/low rpm surge? I'm just now starting to dive into the drivability side as I finally feel somewhat comfortable that the combo does not have to come back apart. I took what I think is a big swing at the ignition timing tables in an attempt to stabilize the timing in the surging area, although it did help significantly there is still enough there I would like to see if I/we can do better.

3. What channels do you guys log in attempt to solve the surge/hunt type situations?

24-05-06 06-43-59.hpl

24-05-06 06-52-55.hpl

2000 fueling and timing 35.hpt

4. One of the logs is of only steady state surge/hunt. One on the logs is of a low boost short pull after I learned how to clear the LTFT's. Can you guys look over both along with the tune file and offer you insight please and thank you.

[edcmat-l1]

No mention of injectors......

[abc]

whoops... Injector Dynamics ID1050X

[Alvin]

Good morning fellas.
Can you guys help me with sharing some of your experience? I have a few questions and if you would rather me start separate threads for each question just say to, but for the mean time I'm gonna throw them all on here and see If I can get some clarity.

LS1 with a 4" crank, TEA 220 heads, 236 I, 248 E, @ .05", .620" and .610" lift and 115 LSA cam, centrifugal blower, stock intake and throttle body.

1. On a 2 bar operating system, does the LTFT function above 105 KPA? Depending on the answer I will probably have more questions. I was seeing LTFT affecting my first few low boost tests until I discovered a couple things. I learned how to clear the LTFT (could't find it at first) and Getting the right PID's and charts setup for monitoring the MAP correctly after moving to a 2 bar MAP sensor and latest version of HPT software. I'm thinking the ECM was still using the 105 KPA cell when I had the PID's setup wrong and was populating LTFT's even in the low boost areas.

2. Can you guys help me with idle surge/hunt and lite load/low rpm surge? I'm just now starting to dive into the drivability side as I finally feel somewhat comfortable that the combo does not have to come back apart. I took what I think is a big swing at the ignition timing tables in an attempt to stabilize the timing in the surging area, although it did help significantly there is still enough there I would like to see if I/we can do better.

3. What channels do you guys log in attempt to solve the surge/hunt type situations?

24-05-06 06-43-59.hpl

24-05-06 06-52-55.hpl

2000 fueling and timing 35.hpt

4. One of the logs is of only steady state surge/hunt. One on the logs is of a low boost short pull after I learned how to clear the LTFT's. Can you guys look over both along with the tune file and offer you insight please and thank you.

LTFT's can "stick" positive and run the WOT fueling rich. STFT's or LTFT's are not updated during PE, and you should be in PE at 100kPa or slightly less than that. So if the engine has been trending lean it will stick a positive LTFT when you go into boost. A extra 5% fuel on a engine already running on the richer side for boost can really kill performance.

for that reason I'm not a fan of leaving LTFTs ON with supercharged stuff. Only positive LTFT's in PE have effect. It's possible to have a negative LTFT but it will not actually change fueling.

As far as the surge goes.. This setup is on the edge of what will get along with stock o2 sensors. There is simply too much overlap for them to operate like they should. They will drive the fueling off which is often a case of a surge. There are things you can do as you approach this limit (has to do with overlap) but at a certain point it just doesn't work.

IF you stick the car in OL and tune it with the wideband do you still have the surge?

[edcmat-l1]

The one log where it does have a good amount of idle roll, it's in open loop (so it's not trims causing it in that log), the spark is barely wavy, the throttle blade is very active and the RPMs just roll.

Doing a compare with your file, it looks like a lot of the idle RPM adaptives have been changed. I'd put them all back to stock and make your spark over/under tables more aggressive.

And do like Alvin said, turn off the long terms and leave them off.

[abc]

Thanks again for the response. I have not researched your comments yet so these questions may be premature but I have to ask:

Would it be OK to just lock the LTFT down to say 5% or should I just do as you say and disable them all together?
As Alvin said, I now notice every time I clear the LTFT and then go make a small pull, each time it will continue to add fuel in the LTFT so I did change the targets in the boosted area and locked the LTFT down to 5%, that was going to be my next series of tests unless of course I understand you guys correctly and should completely disable them.

I haven't attempted to tune it in OL yet at any rpm or load point, I imagine I should test it soon?

When you mention the throttle plate is very active, can I assume you know this by watching the TPS %? Yes, the log you referenced was a cold start from this morning.

I will check the original as found tune file as a compare, but I don't remember changing a lot of the adaptives. Is there anything in there that really stands out to you or are you just saying start over from the beginning so I know I did or did not tune myself into a corner?

As far as being more aggressive with the over/under tables, do I understand you correctly with this statement: instead of adding 1? of advance when the idle is 25 rpm low, add 2? and when the idle is 25 rpm high subtract 2? instead of 1?

Here is the stock, as found file and I believe it was never worked on prior to me.
2000 vette original.hpt

I don't see much changed in the idle adaptives other than the stall saver and I haven't figured out just yet how big of a change I need to make before it's actually a help.

[Alvin]

On the LTFT question I would disable them. I would try to keep STFT active but your cam size is at a line where the fuel trims will drive it into bad fueling. There is a point that overlap is just too much for stock O2 sensors. It's not that it won't work period. It just might not drive as good as it could. It would be worth experimenting to see what drives better. OL or CL.

If you want to be more complicated about it. Some OS's allow for max and min. You could set LTFT up like min = 0.75; max = 1 which would allow for negative long term fuel trims for normal driving conditions but no positive.. Remember negative LTFT's do not effect WOT fueling..

[abc]

In order to put it in OL, I assume I just set the temp. limits to something it won't reach, such as: "Closed Loop Enable Coolant Temp. vs IAT" table set at 275?f?
The cyclic pattern in the STFT does seem to match the surge/hunt I'm feeling in the lite load steady state situations. I will put it in OL for an easy test.

I don't see it right now, but I'm assuming there is no way to turn CL on after a certain RPM like the stand alone ECM's?

Can you guys also give me any insight on how to fix if possible the difference in Kpa readings in my PID list vs the chart. I assume it's some kind of scaling issue but now I don't know which one is correct?
The one in the chart is correct with key on engine off and I know this due to having another stand alone Holley equipped vehicle in the garage I can compare it to.

One more, I've been digging around and haven't yet been able to figure out how to get an actual Lambda reading in any of the displays even though in the units display for my AEM WBO2 sensor clearly states Lambda in the options for the PID list.

[abc]

One more quick question: Assuming it's not octane limited in lite load situations, could I add more ignition advance in attempt to settle down the STFT's, or do you know that has little to no benefit and is not worth the time it takes to test it?

[Alvin]

There is a table, O2 readiness vs ECT. set it to something high to keep it in OL. The closed loop vs ECT or IAT doesn't work on all OS's.

There are ways to get a OL idle doing some PE tricks. This is an old strategy. I don't see many talking about it anymore. I wouldn't think its worth the time on this.

IF you want to try to keep CL you can make it better by adjusting the proportional and integral tables. This is really more for when the o2 sensor has been moved further away from the engine but you can use them damp the over control fuel trim swings.. I can get by on quite a few like this.. but like I said before.. there will be a point where there is just too much overlap for the stock o2 sensors to work properly. The output of a NBo2 isn't proportional. Its really just a rich/lean switch type of a sensor.

Since this has a 2 bar custom OS.. it is normal for the MAP sensor to not read correctly until the engine fires.. This is just a known issue with those operating systems. If your trying to check KOEO map reading.. fire the car.. then key it off gently while the scanner is running. But the 2 bar OS map will not work until the engine has fired.

[abc]

There seems to only be the closed loop enable ect vs iat table in my os or software. Maybe I'm over looking the readiness table, where would it be?

I like the idea of telling the ECM the O2's are farther away as a test. Can we talk more about that?

I'm used to the MAP not reading normal when koeo, but when I cross past 100kpa my two map readings don't match and progressively get farther apart the higher I get in manifold psi. Thoughts?

Also, any info. on the Lambda reading issue?

[edcmat-l1]

What 2 map readings? You should only be logging 1.

[abc]

Good question. I will look at my channels and chart and try to figure out what I may have done. I know I only have one MAPS but is it likely I have selected a different sensor for the chart than I have in the PID's list?

Is there any info. available that more directly describes what each PID is? There seems to be many options when it comes to MAPS and there is nothing that truly identifies which one is in the PID list after you select it so I can make sure I select the same one for the chart?

[edcmat-l1]

In the drop down menu for your channels you will only have 1 2 bar pid. This is a 2 bar COS correct? That should be the only MAP pid or channel to log.

[Alvin]

In the drop down menu for your channels you will only have 1 2 bar pid. This is a 2 bar COS correct? That should be the only MAP pid or channel to log.

be sure to delete the orignal map sensor pid when you do this.. Only have 1 in your list.

[edcmat-l1]

be sure to delete the orignal map sensor pid when you do this.. Only have 1 in your list.

I thought that was self-explanatory when I said "That should be the only MAP pid to log". But thanks for double clarifying in case I wasn't clear enough.

[abc]

10-4, thanks guys.

Is there any info. on how I can get my AEM 30-0310 wide band to read in Lambda in my channels?

I may have found it, I'm gonna go test OL and the new settings. It appears in the transform button for the AEM PID that I was not dropping down for the conversion.

I also found an Intake manifold absolute psi PID in my list. I removed it so we will see how things go now.

Thanks for the help.

[abc]

Well, switching to OL definitely smooths out the light load drivability, the idle not so much right now. I assume I will be able to dial it in with some fueling adjustments.
No that it has been confirmed that the NBO2's are causing the surging, can we talk about testing the PID's for the NBO2's in an attempt to keep some form of CL? I guess it just feels odd to me to have CL available and choose or be forced to not use it. Should I just get over the feeling of it being odd and embrace the OL?

I didn't have any luck getting the WBO2 to read in Lambda, any suggestions?

[Lakegoat]

Open Channels, Click on add channel, Click on + External Inputs, click on the + at AEM, +at UEGO and then select and double click the number of your sensor. Save IT. Then go to Graphs and set up an EQ Error graph.

[abc]

Alvin, your advice for me so far has made a massive change in the way the lite load drivability feels. It feels darn near stock now. I was willing to leave it the way it was as a compromise for having the big power on tap, but deep down I knew it should be better.
I put it in OL via the ECT vs IAT table and that immediately solved the lite load cruise surge/hunt and did what I expected it to do and exposed areas of the base map that needed more work. Thank you again.

I can't help but think I should try to come up with a good compromise with CL enabled, are you willing to help walk me through it or offer up some reading that will help? I guess my ignorance on the subject has me concerned I don't know what I don't know. The thought of permanently throwing out the crutch of CL comp. has me concerned I'm not good enough to keep it alive properly.

In your experience, what concerns, advice, and so on can you offer for those looking to operate a daily driver in OL long term?
Part of me thinks I will need to do a log frequently just as a sanity/safety check.

I did solve my two different MAP readings and you guys were correct, I had an additional MAP sensor listed in PID's.
I did figure out how to display and log in Lambda, it was as simple as using the transform function and adding the function: input divided by stoich plus zero equals output. (input/14.2)+0=output.

Thanks again guys for your help.