Originally Posted by
Haans249
Hello All,
Providing some updates here. At the end, I am in need of some additional assistance!
1) EXTREMELY LEAN Cold Starts
After some more investigation in the calibration, I found a table that effectively offsets the lean COLD start condition. The table of concern is located under Engine / Fuel / General / ECM 33380 Temperature Adder. This table provides an offset to the Pulsewidth of the injectors depending on Injector Tip Temp and Fuel Rail Pressure. You will need to log the Injector Tip Temp, Fuel Rail Pressure, Injector Pulsewidth, and the ERR correction from wideband when in SD mode or the STFT if in CL. Create your Graph in VCM scanner and log the correction factor. Since the table is initially not populated at all, this was the procedure I went through starting with just idle conditions:
a) You will need to initially disable STFTs/LTFTs or you can use the STFTs as the error correction - both methods should work fine.
b) Start from a cold start and log the aforementioned PIDs until you reach full engine temperature, and specifically until your Injector Tip Temp reaches at least 158 degrees. I found in my particular setup that the overall error correction when generally logging over time started out with needing +20% range when cold to basically nothing at above 158 degrees for ITT.
c) Once you have your table populated, you should see a decreasing amount of correction from cold to warm. I decided to put corrections until 158 degrees.
d) In order to populate the first iteration of the table, I suggest setting each column break to the same correction factor (hence starting with just Idle to get the curve). Note what the average injector pulse width is for your idling, and take the correction factor from each column and multiply that by the pulsewidth. For instance, if your pulse width at idle is 1.2ms and your correction factor is 20%, your adder will be .24 in the table and so on.
NOTE: Oddly enough, this temperature adder table doesn't seem to have any effect if you start an already warm vehicle, but not completely cold. Not sure why, but thats what I noticed. It might be some other setting in my tune from start-up that is affecting it, but thats what I noticed.
2) VVE Table Changes
I appreciate some of the tips here. I added transient filtering for throttle as suggested, but I also added additional filtering for both ECT above 190 degrees and Injector Tip Temp over 176 degrees. This helped with the VVE table tuning and getting things dialed in. I also setup an excel workbook to focus the adjustments to the areas that needed greater than 2% correction and to do a % compare of the before and after VVE table to ensure the adjustments are inline after Extrapolating Coefficients.
In addition, the HD columns and rows that I posted above also made a significant difference in getting the VVE table fine-tuned and dialed in properly. Unfortunately the stock columns/rows of the VVE table leaves a lot "hidden" in the details. Expanding these out to the ones I posted above (and creating a graph in VCM Scanner) is what I suggest if your laptop/pc can handle the extra calculations.
NOTE: You can actually input as many columns/rows as you'd like to have details as this actually does not affect the ECM. The Virtual VE Table is used to calculate the coefficients, so it is an "In-software" function and has no affect on the ECM itself. So there is no "danger" to the ECM as I was unsure about in my previous post.
3) MAF Re-scaling
I was able to rescale the MAF table. This one cannot have "infinite" columns, so I just created a custom excel spreadsheet to build out the new MAF table. In the lower rpm/idle areas I created it with smaller distance between each column and grew it to larger distance between frequencies towards the upper RPMs. Doing that seems to have helped quite a bit with smoothness of driveability.
4) Flashing CEL at WOT
This seems to have cleared up a little bit, but I did check my calibration and it still has the stock misfire detection.
NOTE: Does anyone have any suggested settings for misfire detection?
Those are the updates from my previous questions. I do have the VVE table and MAF dialed in and have returned the calibration to CL with O2's operation.
ADDITIONAL ITEMS NEEDING HELP:
Now that those items are dialed in (as far as I can tell in my inexperience), I have begun the process of adjusting my Virtual Torque Tables. I have made those adjustments, including adjusting the external load table, but I am still having a couple issues that I cannot seem to figure out.
1) When I go WOT, I am having an issue where I get Trq Mgmt kicking in and dropping spark to basically 0, while my throttle goes from like 84% to 50% through the remainder of the run. You can see that my MAP drops when this occurs showing that my throttle blade is closing. You can see this happen at 2min 51 seconds in the Part 1 log I provided. I believe this has to either do with Trans Torque Management or just the Torque model since the Final Throttle Torque Req shows 588ish, but Delivered Engine Torque is 640ish, so seems its trying to get delivered down to final throttle torque request so initially pulls timing and closes throttle blade.
QUESTION: Can someone go through the log and point me in the right direction? It really is pretty frustrating as I'm definitely leaving power on the table with this issue. Do I need to add MORE torque to the Virtual Torque Tables or do I need to REDUCE the virtual torque tables in those areas?
2) My original tune has the TCC locking at WOT for 3, 4, and 5th gears. After reading materials, you are supposed to get faster ETs with the converter unlocked, so I switch it to run WOT unlocked. Prior to this the tune would result in a lot of short shifting in the gears, however, after I switched it to unlocked WOT, I am having some issues with hitting the rev limiter (6,850rpm) only during a 4-5 shift. The lower shifts are just fine, but 4-5 seems to be a bit of a problem. I have reduced the shift vehicle speeds to get the shifts happening sooner, but doesn't seem to help much with the 4-5 shift. I was getting shifts taking 1.1 seconds for 4-5...
Could this be due to the virtual torque tables or trans torque management?
Thanks again for any additional help!!