Hi all, I think this belongs here rather than the rate my logs section. Continued from the thread here:
The pid is adapting and I can feel the car driving very differently even by the time I finish my commute (approx 12 miles). I'll look into disabling boost pid adaptation.
regarding the tune, I am not there yet, still working on understanding some things. I saw the following post after doing testing today:
Step 1) log ldimx_w and extrapolate a general ldimx OFFSET from existing KFLDIMX
Step 2) with linear KFDLR log boost vs duty
Step 3) using spread sheet extrapolate new KFDRL AND new temp KFLDIMX
Step 4) extrapolate new final KFLDIMX by running new temp KFLDIMX thru new KFDRL then minus OFFSET
The above tells me I did almost everything wrong:
Step 1 - Didn't use an offset, and I do not understand the point of the offset.
Step 2- Did not make KFLDRL linear before taking logs. Blindly set CWMDAPP = 8 and used KFLDRAPP to modify wgdc. Logging ldtvm showed expected results (wgdc locked at specified %) however ldtvm and ldtvr_w still were moving around. I continued as ldtvm made sense.
step 3 - I did this correctly as best I can work out, thank you prj and elRay for sharing your excel sheets.
step 4 - I did not do this. I understand what to do here, but I don't understand why. (IMX looks clipped to 0 on some regions, I will post the file if anyone wants to comment on this)
I am working on calibrating the wastegate actuator following this thread
http://nefariousmotorsports.com/forum/index.php?topic=517.90Using the
- calculated the offset for KFLDIMX to be approximately 10% (step 1 in the quote above) using the graph on tab 3 of the spreadsheet I attached
- Did a series of pulls with CWMDAPP = 8 and KFLDRAPP set to 0-60% to obtain the response of the system at WGDC of 0,10,20,30,40,50,60% as seen in tab 1 of the spreadsheet (green highlight)
- I used prj's excel calculations to come up with a new KFLDRL and KFLDIMX.
- Added offset to the new KFLDIMX .
- modified prj's excel sheet to run the new KFLDIMX backwards through the new KFLDRL, documented in detail on tab 2 of the attached spreadsheet
- Subtracted the offset from the resulting new KFLDIMX after being run through KFLDRL
I'm about to flash this to the car and go take logs. Have I screwed up anything obvious here? Is my overall thought process correct?
Thanks in advance for your help.
