Help! Frustrated with inconsistent fueling in my CTS-V

This is my first post, and it will be long, so bear with me. This is also my first time tuning a gen 4 platform. I really thought I would be able to figure this out on my own, but after working on this tune for 6 months I've realized that I'm not as smart as I thought I was. I decided that its finally time to ask for help. Please help!

The Car- 2009 CTS-V M6
Mods- BTR stage 2 cam (Intake- .615 lift, 227 duration, Exhaust- .593 lift, 242 duration, LSA- 122), Airaid intake, ZL1 lid, 2.38 pulley, FIC 1000 injectors, TSP longtubes, catless x pipe, DSX aux fuel pump kit, AEM CAN bus wideband, track attack heat exchanger

The Problems- The main issue I'm having is a very lean tip-in. When coasting or decelerating and then getting back on the throttle, it goes very lean for a second or two, then it either comes down to commanded or overshoots rich depending on how quickly I stab the throttle. If I'm really abrupt with the throttle, it is not uncommon to see the afr jump up to 17 then overshoot to 10 before it settles somewhere in the ballpark of the 11.4 I'm commanding in PE. (I know I should be talking in Lambda, but I just can't get used to it.) I only have this issue when running in VVE only or blended mode- not when running MAF only.

MAF mode- When I run the car of of MAF only, I don't have lean tip in problems, but I have other issues. If I really spend a lot of time logging MAF error and putting the corrections in my tune, I end up with this super nasty lean spike around 2500 rpm when everything else is in line. As a result, my MAF curve is a little on the rich side to mask this and make the car driveable- not the correct way to do things! The car drives somewhat okay like this, but I don't like it being rich (or my ltfts being through the roof in closed loop), and I don't feel like it runs nearly as smooth as it does in VVE mode.

VVE mode- In VVE mode the car seems to be much smoother and just feels better in general. Except for the horrible lean tip in. Once again, I have to keep this table on the rich side in order to help mask this issue, which pisses me off. It is also pig rich (9 afr) coming out of pe, but I feel like this is a result of the table being too rich everywhere. The car will also occasionally set a P0068- MAP/MAF Throttle Position Correlation code and go into Reduced Power Mode. This makes me wonder if I'm failing the MAF correctly.

Blended mode- The car drives best in this mode, but it is not good by any means- just okay and driveable. This is what I use when I actually need to use the car as a car. The fueling still looks really wacky if I watch the wideband while I'm driving, but if I ignore it and just get where I'm going it can seem okay. The main issues in this mode are the same- lean tip-in, unacceptable ltfts, rich overshoots going into pe, rich coming out of pe. I feel like I'm going to have to ultimately have a blended tune to ever be happy with this car. I'm a mechanic by trade and good driveability means a lot to me.

Tuning process- I've been tuning in open loop only. I do MAF and VVE separately, logging afr error with the wideband and doing the copy/paste multiply by % - half just like everybody else. I know there are more advanced ways to do this, and I'm open to suggestions. I'm aware of the rich after flash issue and I only log if my injector tip temp is under 110 degrees. The only filter I use is to filter out deceleration [50091.156]>2. I don't know if the filter helps or not, but I live in a very mountainous area and I have trouble getting really good steady state data. The elevation where I live is 4100 ft. and this is probably adding to my challenges, but I'm too ignorant to know why. Maybe baro updates? I haven't done any spark tuning at all- I feel like I need to get the fueling in line first, correct? Any changes in the spark table are from the idiot previous owner- I've tried to put it back to stock for the most part

What I've done to try to fix it- If you're an experienced tuner, It should be pretty clear to you by now that I have no idea what I'm doing. Fixing the lean tip-in is my main concentration at this point. The first thing I tried was adjusting my injection timing- this didn't fix the tip-in, but it did make the car less smelly and make it run better in general. Enrich desoot mode is disabled, catalyst overtemp protection is disabled. DFCO on or off doesn't make any impact on the tip-in. I have double checked my injector data, and I think it is right. I have tried raising the impact factor and the evaporation factor in the transient fuel table. I have tried stock prediction coefficients and zeroed prediction coefficients with no change. I'm completely out of ideas at this point.

If you made it this far, thank you for taking the time to read all of this. I would appreciate any input and constructive criticism- even if you want to tell me that I should just throw in the towel. I'm attaching a tune file and a log file- I have many more if they could be potentially useful.

I forgot- I also tried increasing the manifold volume under the speed density tab.

I'll take a look at it when I get home later tonight. I struggled with similar issues at one point.

Do you have a ported Throttle Body? I have the same issue and when I ported my stock TB it helped.

Youre Zone MAP boundaries only go to 100 kpa, you can get more resolution in the top end by changing that. This doesn't match a supposedly stock CTS-V cal in the repository, but who knows for sure if that one has been modified.

Try logging DynAir, VEair, and MAFair. Dynair should be what it is running from. If your VVE is off you will see VEair deviating from Dynair.

Increasing manifold volume may add to your delay, pre and post throttle changes, though I don't know the correct number to put in there for your setup.

You could increase your PE enrichment ramp in rate to help with the lean tip-in.

Do you have real gasoline where you are at? Or is it E10 or worse like most places? You don't have flex fuel enabled but it is set for a sensor.

The VVE seems lumpy, it can help a lot to do some major hand smoothing after pasting the EQ error, before calculating coefficients.

You have the ECT set so that it will never set a P0101, and it is currently in blended mode, MAF enabled, is that correct?.

Another note, you appear to be in closed loop, are you adjusting your AFR/EQ error based on the wideband and fuel trims as well? It will be easier to force it in open loop and just go by the wideband.

ROBZL1- thank you, I would appreciate your input

PGA2B- The throttle body is stock, unported

NotSure- 1. I copied the zone MAP boundaries from a 2012 ZL1. For some reason the my stock VVE table was missing a zone (maybe it was zone 11?) and it would cause a warning to come up when I would press calculate coefficients. I think this is a weirdo 2009 only thing. It wouldn't make calculations that made sense until I copied the ZL1 table. I changed my VVE table to a 210kpa axis, but I didn't change the zone MAP boundaries to reflect this change- I completely overlooked this. Thank you for pointing this out! I'm not sure about where to start creating my own zone MAP boundaries. Do you know any procedures for this?

2. I will try logging those 3 pids- I haven't tried logging these before

3. I've can't seem to make any perceptible changes playing with manifold volume

4. I haven't messed with my PE enrichment ramp-in rate since I initially set up PE. Maybe I should. As far a I can tell, I don't seem to be having any lean issues transitioning into PE. My lean issues happen when I initially touch the throttle after coasting or decelerating. The only time it is lean going into PE would be if I smash the throttle and go directly into PE from coasting or decelerating. I may be using the phrase "tip-in" incorrectly. I'm referring to "tip-in" as the transition from off throttle to on throttle, not as the transition from normal acceleration into PE.

5. I have mostly E10 where I'm at. Real gasoline is an option at a local station. I haven't ever tried it- maybe I should. Should I change my stoich to 14.13 to reflect E10? This vehicle already uses an extraordinary amount of fuel, and I don't really want to make it worse. The car does have the DSX flex sensor kit on it, but I don't have E85 in my area. The previous owner ran the car on E85 only- it made plenty of power, but it drove like poop with his awful tune in it (I realize that it still has an awful tune, but at least it is MY awful tune, and it drives WAAAYYY better now than it did when I bought it) I disabled the flex sensor because I thought I would be eliminating one more variable in my inconsistent fueling. I haven't tried turning the sensor on and off to see if it changes the lean tip-in issue.

6. My VVE table is very lumpy. I have had it smoother in the past, but the problem is, as I dial it in logging afr error and smoothing it out, the lean tip-in gets much worse as the whole table gets leaner and closer to stoich. Also, the lean tip-in contaminates my logs and causes lumps because weird areas end up logging as lean if I happen to let off the throttle and get back on while I'm logging. If I can log somewhat steady-state, then this is less of a problem (not really an option in the mountainous area where I live). I wish I had access to a dyno for dialing VVE, although it still wouldn't fix my tip-in issue.

7. It is currently in blended mode, MAF enabled, closed loop. I use this tune for driving, not for tuning.

8. Yes I'm in closed loop in the tune I use for driving the car (not tuning). The tunes I use for tuning are open loop and I adjust AFR/EQ error based on the wideband. I have not tried closed loop tuning with fuel trims. I will attach the tune I use for dialing VVE and the tune I use for dialing MAF.

Big part of your problem is injection timing - way too retarded. Next issue is transient fueling. VE and MAF both need to be dialed in without transient influences and then use transient adjustments to get what your after, but injection timing needs to be fixed before everything else... Also you might want to pull a lot of timing out while your tuning all of this in ;)

NotSure makes some good points. Things I would do:

1) Try changing PE ramp-in from 1.0 to 2.0. That should help with lean tip-in.

2) I think you just need more work on your VVE and MAF to get them to match and to smooth them out. As noted, log DynAir, VEAir and MAFAir to see where they deviate. You might find you want to drop the dynamic RPM limits down to 3000/2800 or 2200/2000 or something to start with, so you don't have to get as much of the VVE built out. With that cam/mods, you should be fine with just the MAF, actually. That said, I run blended up to 3000 in my car with only a slightly larger cam, so I understand. :)

3) Filtering your data- I'd add a steady-state filter to what you mentioned. I use: [50090.156.slope(3000)]<1. This only uses data when the throttle has changed less than 1% over 3000ms (change 3000 as necessary, but it works well for me). Also at some point, once you get the fueling in a happy place under 110 injector temp, you can start playing with the tip temp offset to keep your fueling good across the board, too.

4) O2 sensors- At some point, I'd add O2 mV to your log/graphs and work on dialing in the Proportional and Integral terms to get your O2 swings a little tighter, just based on seeing the wideband data. Also try dropping Proportional Limit to 1.00 or 1.25.

5) Idle. I know you're not there yet, but here are some thoughts. That's a lot of Airflow Final Minimum for a BTR PDS 2 cam; I'd take about 3 g/s out of there across the board for starters and see how it does. Idle spark seems fine at 21. The start-up idle RPM tables are a little odd/high, too. Once you get your fueling squared away, I'd check those out. You could also reduce Integral Airflow to about 50% of its current value in the entire "normal" line for starters and it would probably smooth things out a bit.

The biggest thing is just spending more time on the MAF and the VVE. You'll probably find that you will have some spikes in the VVE and small humps in the MAF in certain ranges. I couldn't get rid of them on my car. It's the only thing I could do to keep fueling in line. I also left the prediction coefficients zeroed- otherwise crazy stuff started to happen.

Here are my current VVE and MAF just for reference. Notice the hump in the MAF at 7500-ish, and the obvious spikes in the VVE around 1600 and 2600. I had to do it to get fueling right. If there was another way, I couldn't find it. After several months of two hours worth of logging a day on my drive back and forth to work, with many combos of trying different things, I finally gave up and decided the hump and the spikes were just going to have to stay!

Best of luck! I'm sure you'll tame the beast.
Attachment 96040Attachment 96041

EDIT- Was working on the post while everyone else posted... I know Greg is correct on the boundary (he's kinda the boundary guru). I was going to note you might find 650 to be really late in the top end, and getting there way too soon in the RPM band. For starters, I might just put everything back to 520 boundary until you get your VVE/MAF tuning process worked out, then adjust from there.

Greg, do you think I should just go back to the factory 520 boundary from 2000 rpm on up and try to start from scratch with VVE? I came up with these injection timing numbers using the excel spreadsheet that you and a few others on here developed, but I'm probably misinterpreting something. If nothing else, the car sure smells a lot better at idle.

How can I dial in MAF and VVE without transient influences without a dyno? Should I be zeroing out something?

Does my timing look to aggressive? It is pretty seldom that I experience any knock retard, although the previous owner had disabled burst knock and I have left it that way.

As a side note- this is Russell from Highlands Automotive. We know a lot of the same people. I think you understand why I don't want to ask for advice from my local tuning shop...

:D Were you by chance driving down Buck Creek this Sunday? If not I met someone coming down in a gray V while I was headed up to the hospital and then that same car was in a pull off when I came back down the mountain about an hour later... Yeah, I recommend anyone and everyone stay away from there...

Rob, thank you for the detailed response

1. I'm going to try raising the PE ramp-in. I have not tried this yet.

2. I have completely overlooked the DYNAir, VEAir, and MAFAir pids altogether. I start logging those. I can lower my dynamic rpm limits, but I don't want to go below 3000 because I feel like it sacrifices too much driveability. In the past when I've had the MAF dialed in closer (leaner), I get this lean spike at 2500 that makes the car horrible to drive. The only way around this has been to make a lump in the curve by hand- but this makes PE way rich in that same area (6500-7500ish hz).

3. I had no idea that a steady state filter like this existed. Thank you for that! This will be very useful.

4. I haven't touched my O2 settings yet, I'll try dropping the proportional limit first.

5. The idle is definitely a little high and a little irregular. I think I may go ahead and try a few of these things even before I finish with my fueling, although I may be asking for trouble by changing more than one thing at a time...

I'm really relieved to see the humps and spikes in your tune, because I feel like mine is going to have to have these for decent driveability. Are you highlighting and multiplying the trouble areas? This is what I have done in the past, but I would imagine there is probably a more scientific way to do this.

I felt like I had hit a brick wall with this thing, and now I have plenty of new ideas to try. I can't wait to make some changes and do some logging this weekend. Thanks!

That was me Greg. I was making the same crappy implausible log over and over like a crazy person. Small world.

Quote:

---

Originally Posted by grayv

I'm not sure about where to start creating my own zone MAP boundaries. Do you know any procedures for this?

---

Just try to have MAP zones in the major areas that the engine sees most. One way to visualize this is to look at one of your graphs that has the row/column axis in MAP and RPM, doesn't matter which one, and look at the counts per cell instead of the data. You may have copied a ZL1 zone made by GM, but that doesn't make it right, LOL. Also don't waste resolution by having multiple zones in areas that your engine will never see, such as 0-1000 rpm and 100+ kpa. One big zone can cover the areas it won't, or will barely, see.

Quote:

---

Originally Posted by grayv

The car does have the DSX flex sensor kit on it

---

Is there a wire from the sensor to the ECM? Need to enable Flex Fuel in the tune to take advantage of it. You already had it set to Sensor, just need to change to Enable. You can log Alcohol % to see if it is working. If so and it is reading around 10% like the fuel you are using, it will automatically change Stoich appropriately. Yes it will go a bit richer but only around 4%, and at least you will be dialing it in with the correct fuel data.