Bad Kr 3800 No Power / L26 Top Swap

[AlexTheRoach]

Hi guys, this post is being made out of desperation as I'm really not sure what to check next. Also I'm new to the forum so I'll do my best to link everything correctly.

As a breif backstory, back in august I did a "top swap" on my 3800 SII in favor of a ported aluminum upper manifold and throttle body. Also the throttle body is an L67 with the gen V adapter to connect to the manifold and a custom CAI system was made.

Other than this mod all others are bolt ons such as the front power log and DP which shouldn't affect the engines core function.

The real issue here is knock. Ever since the swap the engine is mostly down on power feeling sluggish as if lean and knock is all over the place. I have tried MAF tuning and adjusting about every variable possible to no avail.

Here is the link to my most recent log. Note that at one point I commanded open loop to see what would happen so some data may get weird. 2004 Pontiac Bonnevile SE (16th UMass Drive).hpl

Also heres an image of the mods showing the intake and the catch can which I forgot to mention.04 Pont Top Swap.jpg

Finally my suspicion is a bad maf tune or failing front O2 causing it to be lean, I thought LTFTs were supposted to iron out fueling errors so my only last suspicion is the O2.

[5FDP]

Your fuel trims are almost always negative, that means it's a little rich and not lean at all. I honestly see no problems with the fueling at part throttle right now. You need a wideband oxygen for wide open throttle tuning though because fuel trims do no function in open loop and you need a way to watch what the actual air fuel ratio is doing in the exhaust.

I've seen part throttle knock like that if something is rubbing or if the motor mounts are trashed. It could just be picking up a lot of vibrations.

[AlexTheRoach]

Your fuel trims are almost always negative, that means it's a little rich and not lean at all. I honestly see no problems with the fueling at part throttle right now. You need a wideband oxygen for wide open throttle tuning though because fuel trims do no function in open loop and you need a way to watch what the actual air fuel ratio is doing in the exhaust.

I've seen part throttle knock like that if something is rubbing or if the motor mounts are trashed. It could just be picking up a lot of vibrations.

I appreciate this perspective, I think for some of the Kr like when it spikes up to 10 out of nowhere it may very well be false knock, but I am very sure that most of the knock is real... at least the car is very down on power compared to how it used to be before the swap.

I have log files of this drive going very far back, and comparing them is quite revealing.

As an example, driving up one of the very steep hills yesterday, the car didnt have enough power to stay in third to get up the hill at any resonable throttle compared to all previous drives where it could chug up the hill at 14 ish % no problem and hold its speed. I had to drop into second at 23% throttle to keep it moving up the hill.

[GHuggins]

Are you actually hearing any of this knock? Only reason I ask is because that CAI is huge which means it's going to throw off all of your airmass and associated calcs. Torque models won't agree. Won't be running in the same spark rows and on and on. As at test throw the stock airbox back on with the stock MAF curve and give it a drive... Also as pointed out you need to be using a good wideband not only for wot tuning, but it's good to know if your O2's are working in general.

[AlexTheRoach]

Are you actually hearing any of this knock? Only reason I ask is because that CAI is huge which means it's going to throw off all of your airmass and associated calcs. Torque models won't agree. Won't be running in the same spark rows and on and on. As at test throw the stock airbox back on with the stock MAF curve and give it a drive... Also as pointed out you need to be using a good wideband not only for wot tuning, but it's good to know if your O2's are working in general.

What would I hear, I know they say pinging but it would have to be pretty loud to hear in the cabin.

The only real sign assuming its from knock is on the highway sometimes it would randomly buck almost where the power would seem to cut for a split second (rpms and everything seemed normal however).

Also I chucked the old intake but I got 10 or so scans with it on the new setup before I swapped to the CAI, the issues were far less apparent then. I'll link those later.

So my question then is can I even tune it from just the maf curve or will I need to mess with VE spark and others? Thanks

[GHuggins]

Yes it can be calibrated. You'll have to adjust VE and MAF both for the added airflow. I wasn't thinking earlier. I forgot the MAF on the 3800's was in the TB inlet and that should have stayed the same. Just make sure it's clean. Those were halfway bad for MAF's reading wrong. You'll need to pull some timing regardless for the higher cylinder pressures. You can use the logs with it in it's stock form to see how much things have shifted if need be to make adjustments.

[AlexTheRoach]

Messed around today a bit with VE and MAF. Feel like VE might be the answer to getting it dialed but I still don't understand why when I change the MAf freq table the pcm just pulls any fuel I add via fuel trims.

After changing the ve it seemed like it was actually changing the fueling but still weird. These are the logs from today for the diff ve and maf values... didnt seem to change much but ve at x1.15 seemed to help.2004 Pontiac Bonnevile SE (VE x1.1 MAF Stock Quick Drive49).hpl 2004 Pontiac Bonnevile SE (VE x1.05 MAF x1.05 Quick Drive50).hpl
2004 Pontiac Bonnevile SE (VE x1.15(+Smooth) MAF x1.1 Quick Drive51).hpl

[blindsquirrel]

You cannot make it run richer or leaner than Stoich while it's in closed loop. That's what closed loop is for. Trims go away and it runs solely on whatever is in the MAF/VE when it goes open loop under heavy throttle (or during warmup before the temp threshold is reached).

If you are tuning using fuel trims, all you can do is adjust to get the trims as close to zero as you can. This is always a moving target though because of weather/temperature/whatever changing all the time. You will never get it to where trims are always zero all the time everywhere.

I have no idea if you are setting the file up properly to do MAF and then to do VE, because I don't see a tune file posted here or see a description of what you are changing for each.

[AlexTheRoach]

You cannot make it run richer or leaner than Stoich while it's in closed loop. That's what closed loop is for. Trims go away and it runs solely on whatever is in the MAF/VE when it goes open loop under heavy throttle (or during warmup before the temp threshold is reached).

If you are tuning using fuel trims, all you can do is adjust to get the trims as close to zero as you can. This is always a moving target though because of weather/temperature/whatever changing all the time. You will never get it to where trims are always zero all the time everywhere.

I have no idea if you are setting the file up properly to do MAF and then to do VE, because I don't see a tune file posted here or see a description of what you are changing for each.

I guess I kinda knew that before as I have tried doing open loop tuning but I didn't commit to it propely and just messed with ve and maf keeping CL on. Which now that your mentioning this makes no sense that I was trying to adjust these parameters in CL.

Tomorrow I'll go for a drive but disable maf (set hz to 0) cot is off and try to get data from that. I guess just refer target afr to what roughly the o2 and reported values are and adjust based on that? I want to focus on fueling first then dial in timing.

Here's my current file, maf and ve reverted to stock for tomorrow, didnt set the maf to fail yet. Appreciate all the help
04 Pontiac Bonnevile SE.hpt

[SiriusC1024]

If you fail the MAF it probably won't run. Are you able to log Volumetric Efficiency Airflow and Dynamic Airflow?

[AlexTheRoach]

If you fail the MAF it probably won't run. Are you able to log Volumetric Efficiency Airflow and Dynamic Airflow?

Just checked and it doesnt show me any channel values for either. The P04 pcms are pretty archaic it seems. Heres what the log was last time I tried to do no maf tuning. I left alot of other tables alone thinking it would auto into OL but did not. So far I have only forced OL by using vcm controls. I will try to force OL by setting O2 temp closed loop enable to max value with maf failed and see if that works.
2004 Pontiac Bonnevile SE (No MAF Calibration Quick Drive33).hpl

[SiriusC1024]

So yours ran without a MAF? Those 3800s normally don't. Did you have to do something special like figure out a good VE table in order for it to run?

[AlexTheRoach]

Ran totally fine. I have a couple logs from yesterday with MAF failed. It just pulls the values from the MAF table it seems so it takes out any errors from the maf not reading right. One of the big issues seems to be the sharp bend and no screen for the maf which I eventually will try to fix. Until then I'm going to tune without it.
2004 Pontiac Bonnevile SE (MAF Fail (x1.05) OL Force (PE Ex Fuel) Quick Drive54).hpl

[SiriusC1024]

It's not something dumb like the catch can being full then spilling oil into the vacuum side during acceleration or on an incline?

[SiriusC1024]

Yeah man what did you do to the front valve cover? There's not supposed to be a vent there. The PCV valve is over by the alternator. Catch can would go inline with PCV normally, but the PCV on these is self-contained within the intake. I hope you're not applying vacuum to that line. Even if you weren't crankcase pressure will push oil out of that unbaffled vent you made then eventually into the intake. If you don't believe me then run the car without the oil cap and watch the mess.

[AlexTheRoach]

Yeah man what did you do to the front valve cover? There's not supposed to be a vent there. The PCV valve is over by the alternator. Catch can would go inline with PCV normally, but the PCV on these is self-contained within the intake. I hope you're not applying vacuum to that line. Even if you weren't crankcase pressure will push oil out of that unbaffled vent you made then eventually into the intake. If you don't believe me then run the car without the oil cap and watch the mess.

I had created a mini baffle inside the valve cover behind the pcv because as you mentioned some oil would end up in the catch can if it was not there. Because of the manifold swap pcv gets a bit messed up so I researchd it and arrived at this solution. Stock pcv valve location is blocked off, the air for the crankcase enters as metered air from a tapped hole under the throttle body (between maf and blade), then a hole was drilled and tapped into the upper intake manifolds pcv line (because of porting I had to weld the intake tract which sealed off the pcv line going straight out the top of the manifold seen on stock SIII, so right below that weld where the line was in tact i drilled and tapped allowing for the pcv inlet routing to be used). Then the front valve cover is tapped for a PCV valve and run to a catch can. From there it runs to the upper intake manifold where the evap would be on Series III cars. This design retains stock function while still using some of the stock pcv routing.

Here are some images of the porting when I was almost done with the rough port. You can see on the manifold where the pcv was running, so I welded that and tapped below it where it wasnt filled.

L67 04 Pont Throttle Body Port.jpgL26 04 Pont Intake Port.jpg

[SiriusC1024]

Any oil in the catch can? Check down in the intake, too.

[AlexTheRoach]

After almost 2,000 miles the catch can only had filled about 10%. I've only gone about 300 since then and check it here and there. Havent looked down the intake but did check the blocked off pcv and no oil or air was getting by. I do eventually want to improve this setup with an improved valve cover baffle but for now I think its fine.

[GHuggins]

Last log didn't show any MAF dtc's. Need to log MAF state to know it's failed. Also don't tune in OL. You'll just waste time. Tune it how it's going to be driven. Then only make OL corrections for when it's in OL such as when it's in PE.

[AlexTheRoach]

Don't have the log to show it, but adjusted the maf curve based on a ve calculator using rough ve values for this engine setup. Result was the engine running the best I've seen it before. MAF curve was hugly adjusted and looks very different from stock, seemed to work and now just have to dial it in. Only got into it about 80% but approximate torque from the pcm was up to 275 where I've only seen it up to 230 and that was wot in the past. Definetly pulled nice with almost no Kr (I know this calc is not super accurate).