Marine Intake

[BeXtreme]

FWIW I just found the GM equivalent to the $$$$ marine sensor. Looks like it was used on some of the 3.8L and 2003+ 4.3L engines.

12677837

More Information for GM GENUINE 12677837

Looks like the same upper sensor and connector as the corvette one, but with the NPT thread instead of the M16. The corvette sensor, pigtail, and ICT Billet M16x1.5 to 1/8npt adapter all together was cheaper than that sensor though.

I'll probably just replace the one I already ordered with that one if I have any issues with it in the future. I have plenty of room in that area currently, so it shouldn't be an issue to have the adapter there.

 

[L31MaxExpress]

I spent some time this afternoon getting the Proportional and Integral settings dialed in closer. The air/fuel no longer overshoots rich/lean as much. It has tightened the air/fuel ratio swings from close to a full point to about 0.3:1. It also drastically reduced the integrator/short term fuel trim swings. The engine itself is much smoother now. Biggest shock, adjusting these almost completely eliminated the tip-in hesitation in closed loop. Definitely now on my list of things to tune initially on a modified engine. Prior to this setup I have always treated this adjustment as something to do after the tune was mostly done. Now I am rethinking the process and these adjustments should make dialing in the fueling easier and more accurate. If you look at the 02 mV graphs it is easy to see how much smoother the air/fuel fluctuates around stoichiometric now.

Before

You must be registered for see images attach

After

You must be registered for see images attach

 

[BeXtreme]

I spent some time this afternoon getting the Proportional and Integral settings dialed in closer. The air/fuel no longer overshoots rich/lean as much. It has tightened the air/fuel ratio swings from close to a full point to about 0.3:1. It also drastically reduced the integrator/short term fuel trim swings. The engine itself is much smoother now. Biggest shock, adjusting these almost completely eliminated the tip-in hesitation in closed loop. Definitely now on my list of things to tune initially on a modified engine. Prior to this setup I have always treated this adjustment as something to do after the tune was mostly done. Now I am rethinking the process and these adjustments should make dialing in the fueling easier and more accurate. If you look at the 02 mV graphs it is easy to see how much smoother the air/fuel fluctuates around stoichiometric now.

Before

You must be registered for see images attach

After

You must be registered for see images attach

What do you mean by proportional and integral setting adjustments?

 

[L31MaxExpress]

What do you mean by proportional and integral setting adjustments?

It is part of the oxygen sensor calibration for how the PCM uses the o2 voltage data.

Integral Delay is a delay term that helps slow or delay the fuel change in response to the amount of time it takes for the fuel that is injected to burn and the resulting exhaust gases to reach the 02 sensors. Many different things contribute to this.

Proportional Base is the amount of fuel that is added or removed to help make the mixture richer or leaner. Too little and the o2 swing becomes slow and the PCM has trouble adjusting the fueling in closed loop. Too much and the air/fuel ratio swings become excessive with the resulting engine surging.

I first used 2004 Z06 settings they were very different than the ones in a lower powered truck. I probably made 10 different changes to it just to see how it responded. I left the stock Vette Integral Delay table. Then I greatly reduced the Proportional Base multipliers. For the idle table I ended up around 60% of what the Vette had stock. Off-idle I kept reducting the Proportional Base modifier until I had it down to 30% of what was there on the stock Vette.

What I discovered playing with say 50% changes is that for the most part, the length of the 02-voltage oscillation is mostly controlled by the Integral Delay. The height of the swing is mostly controlled by the Proportional Base Multiplier. Once I made a few aggressive changes to see what each were doing it made dialing it in much easier.

 

[L31MaxExpress]

Too much fuel added or removed too quickly result in oxygen sensor voltage and short term fuel trims that looks like this.
/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\

Get it closer to right and the sweeps are wider and more rolling with less height, the short terms should be fairly stable.
~~~~~~~~~~~~~~~~~~~~~

Edit--Forgot to also mention. Too much Integral Delay results in the 02 sensor hanging lean or rich for a long period of time, before suddenly jumping rich or lean. Looks something like this when you are looking at the 02 mV graph.

-----\_____/--------\_______/------------\________/----------\____

 

[PlayingWithTBI]

In Process Control, PID (Proportional Integral Derivative) is used extensively. In our case Derivative isn't.

www.en.wikipedia.org/wiki/PID_controller

PID controller - Wikipedia

en.wikipedia.org

 

[L31MaxExpress]

Also resolved another issue with some more work on the tune. A start and stall issue around 100-140*F coolant temp, where the engine would start, idle down over a few seconds and stall. It was starting and transitioning great to idle either cold or hot but struggling in the mid temps. The Startup and Friction airflow tables had some very low values in that area. Bumped the values up in the low spot and smoothed them out in line with the rest of the table. I added a delay of 50 camshaft or 100 engine revolutions to the Startup Airflow. At 1,000 rpm that this fires up at the delay should be about 6 seconds of added airflow to help the engine transition from startup to the PCM controlling the idle.

I will revisit the Transient Fuel if needed once I get the rest of the fueling the way I like it, but it seems to have decent enough throttle response to have good drivability now.

Start Up in the prior problem area. Gave it a couple of light revs in open loop with good response.

xc_hide_links_from_guests_guests_error_hide_media

Throttle Response with a fully warmed up engine in closed loop where throttle response has been sluggish practically forever on this van.

xc_hide_links_from_guests_guests_error_hide_media

 

[L31MaxExpress]

Tomorrow mornings forcasted at 27F. Going to make a Running Airflow datalog from cold start to get the Base Idle Airflow Table dialed after an overnight cold soak.

 

[Supercharged111]

I did a few desired airflow cold start cycles when the weather turned cold earlier this winter. I doubled with logging/changing my commanded open loop fueling, a 2D table that I have on EFILive. I found desired airflow to be a bit high even for my motor and left it largely untouched. I did observe some inconsistencies based on the starting temp. Turns out IAT is to blame there, something the GM engineers half assed out the gate and bent the rest of the tune around.

 

[L31MaxExpress]

I did a few desired airflow cold start cycles when the weather turned cold earlier this winter. I doubled with logging/changing my commanded open loop fueling, a 2D table that I have on EFILive. I found desired airflow to be a bit high even for my motor and left it largely untouched. I did observe some inconsistencies based on the starting temp. Turns out IAT is to blame there, something the GM engineers half assed out the gate and bent the rest of the tune around.

My open loop AFR has commanded A/F multiplier vs MAP vs CTS, same as my 0411 was.

If it is the table you are referring to, I already raised the IAC Park Position Airflow vs IAT table, it was set pretty low from the factory.