E67 weird ignition cut while transitioning into boost?

[SlvrSky07]

Hello. I am trying to get the SD tune completed on my Saturn Sky. I have 80lb injectors with a fuel return system with 1:1 AFPR. I am running into an issue when I am going from VAC to Boost ONLY while tuning for SD. When the MAF is enabled and is the primary sensor for fueling it does not do this issue. If you look at the attached log, one example is at 6:42:15 time stamp. I will be trying to spool the turbo up and accelerate but it feels like the engine just quits and you hear a fast fuel cut noise until you let off and try it again. During that time my wideband shows its running really rich (about .70 lambda). I have been constantly taking fuel out of those effected cells on the VVE table but when logging its constantly saying to remove 25% or so which doesn't seem right to me. Is there some parameter causing this issue?

Thanks!4.hplsd tune e-85 new zone 16.hpt

[SlvrSky07]

Anyone??

[jaydubb71]

I will take a peek when Im off duty

[smokeshow]

Looks like you are trying to calibrate the VVE but forgetting that the cams are phasing as well. That is my presumption as you did not log the phase angles yet they are still enabled and moving. You can't just make edits based on MAP and RPM...you have to incorporate cam phase angle. This is the whole reason VVE exists - it can represent more degrees of model freedom than a simple x vs y table can. Unfortunately this is a task I would only recommend or explain for people already expert level in calibration.

[SlvrSky07]

I was always told to log that way, but what you say makes sense. I guess I figured it it was kind of like a base map for them at 0 and it adjusted from there. I will have to look into that more.

[SlvrSky07]

smokeshow, I came across your post here: https://www.performancetrucks.net/fo...ration-551391/

You mentioned '#3 If a MAF is installed and calibrated*, it is possible to plot VVE airflow against MAF airflow to rough in the majority of the VVE surface. Again, I can detail this later if there is interest."

Can you elaborate more please?

[smokeshow]

smokeshow, I came across your post here: https://www.performancetrucks.net/fo...ration-551391/

You mentioned '#3 If a MAF is installed and calibrated*, it is possible to plot VVE airflow against MAF airflow to rough in the majority of the VVE surface. Again, I can detail this later if there is interest."

Can you elaborate more please?

I see google search is workin for ya lol. I sort of detail this procedure in my thread here https://forum.hptuners.com/showthrea...ation-for-gen4

While that procedure doesn't focus on VVT applications, the math is almost exactly what you'd need to simplify the VVT cal process. However the data collection process is far more lengthy and susceptible to mistakes. I don't know anyone in the aftermarket who bothers with it other than me.

[SlvrSky07]

Thanks for pointing me in the right direction. Hopefully I can get this understood and figured out before the car goes away for the winter lol.

[SlvrSky07]

Would you recommend changing the desired angle for the intake and exhaust cam (low-med-high baro) to 0, doing the "classic" SD tune to make sure the logged data is not skewed due to the cam angles changing or would you set up a histogram for each cam angle (the 40 intake and exhaust tables) and log error and make the coefficient changes that route?

[smokeshow]

Would you recommend changing the desired angle for the intake and exhaust cam (low-med-high baro) to 0, doing the "classic" SD tune to make sure the logged data is not skewed due to the cam angles changing or would you set up a histogram for each cam angle (the 40 intake and exhaust tables) and log error and make the coefficient changes that route?

Yes. lol

No matter which way you spin it, its gonna take some legwork. This is realistically a task for an intelligent data processor - I use MATLAB to handle that sort of work. Histograms will be a major pain in the ass unless you simplify the cam phasing schedule. GM does this on stock stuff by discretely calibrating out as far as possible and then using a regression model in hopes of covering all the edge and corner cases. Not a layman task - so simplification is your friend here.

[SlvrSky07]

Yes. lol

No matter which way you spin it, its gonna take some legwork. This is realistically a task for an intelligent data processor - I use MATLAB to handle that sort of work. Histograms will be a major pain in the ass unless you simplify the cam phasing schedule. GM does this on stock stuff by discretely calibrating out as far as possible and then using a regression model in hopes of covering all the edge and corner cases. Not a layman task - so simplification is your friend here.

Would you recommend a certain cam angle to lock in and use that as base? Or use 0 since that the lowest number ?

Also thanks for taking time to explain this to me. On all the VVE stuff I have read, I didn’t come across much information on what to do with VVT when collecting % error data. All I ever see is the disable maf, open loop settings, dfco, cot etc and log the error and try to keep the table as smooth as possible.

[smokeshow]

Would you recommend a certain cam angle to lock in and use that as base? Or use 0 since that the lowest number ?

Also thanks for taking time to explain this to me. On all the VVE stuff I have read, I didn?t come across much information on what to do with VVT when collecting % error data. All I ever see is the disable maf, open loop settings, dfco, cot etc and log the error and try to keep the table as smooth as possible.

Since you have two independent camshafts, things are far more complicated. You could lock out the exhaust for now and just capture data as the intake cam moves, see how close that gets you. But if you do intake first and exhaust later or vice versa, don't forget that this guy exists...

Capture.PNG

[SlvrSky07]

I cant seem to find any information on what you would log or put in that ExhaustCam.IntakeCam table. What would be the difference between that table and the Constants table? I imagine that the constants from the numbers calculated from the VVE table. The other question I have about logging the error for each cam at each angle, when you select any angle other than 0 intake, 0 exhaust, it does not allow you to do any other table changes besides the paste, no smoothing or multiplication. Am I going about that wrong?

I parked the intake cam by setting it to always be 0. Next, I set up 20 histograms to log the ve error at each exhaust cam angle. Then paste by % into the matching VVE table, keeping intake at 0 on the vve table and going through 1-20 of the exhaust vve tables.

After locking the intake cam and letting the exhaust move as the original tune intended, the car is super sluggish and has a hard time even building boost.
Is this the correct way to do this as you mentioned before or am I not following correctly?