Can someone help me make sense of this? I am doing my linearization runs to begin tuning my PID and am noticing that with my WGDC fixed I am getting inconsistencies based on what looks like IAT or maybe something else temperature related.

I pulled over and flashed this file. Of course while flashing, the intake manifold will become heatsoaked and the first run after the flash recorded the highest IAT and lowest boost pressure. Within a few mins I recorded three other runs, each of which the IAT dropped (more airflow through the intercooler) and boost increased.

I am using KFLDRAPP and CWMDAPP to fix my WGDC and in the logs below was set to 33%.

So this poses my questions.

1: Does KFTARX/LDIATA have any influence when the duty is fixed?
2: Could it just be the difference in air density based on temp (which is why LDIATA/KFTARX exist)?
3: Could something else be at play here? I have heard that wastegate temperature can produce inconsistent boost levels? I figure since the IAT is dropping there's more airflow through the engine bay and to my external wastegates, cooling them down which could change the spring tension, changing the amount of boost with the same duty.
4: Other?

Car info:
2000 A4 1.8t, 018CH wideband ECU with 5120, BW S252SXE, twin scroll JDL manifold, two TiAl 38mm gates with .7bar springs (runs ~.9bar @ 0%DC), MAC solenoid running dual port, stock block, stock intercooler. Running no MAF currently while I'm waiting for more intake components.

Charge system has been checked to have no boost leaks.

Having these inconsistencies will make tuning my PID maps quite tricky I'm thinking.

Anyway, here are the logs. Any input would be greatly appreciated.

Your setup is bad ass man.  Your iat will affect boost with a fixed WGDC. The stock intercooler is restrictive so with higher iat the boost will be lower.


Sent from my iPhone using Tapatalk

I didn't read your post fully by "inconsistent with fixed duty". The OEM included PID so that it could reliably produce the required pressure from the turbocharger regardless of altitude, temperature etc. The pedal requests a torque figure that the ECU translates to load. That load request is made up with throttle angle and (usually) a boost pressure request. The pressure regulation module uses various factors to vary the duty cycle to meet the requested pressure, and then uses the PID routines to correct that duty request based on feedback.

50% duty at sea level and 10deg C will yeidla different boost pressure over 50% duty at 800m with 20deg C. Fixed duty is a good hack to get stuff going, but will result in massive variances in performance as the temperatures and pressures change externally.

IAT effect is insignificant.
What matters most is EGT. Higher EGT means more energy stored in the exhaust and more energy transferred to the turbine.
If you make two runs back to back, the 2nd run will always have more boost due to this.

I gather that you are setting fixed duty cycle in attempt to calibrate your pid with the "feed forward" method.   Between runs and flashes do some light driving to keep all your runs at consistent temps, and then do you your wot pull to get the most consistent readings.  I've found that after I generate a kfldrl based on these pulls, I've needed to modify the kfldimx axis a hair to account for over or under boost.

The plan is to use the feed forward and your tool to tune the PID. This was more a test for me to confirm my suspicion and not logs that I plan to use to read with your tool.

I'll definitely do a bit of driving prior to doing my fixed duty pulls to try and keep things as consistent as possible.

Quick question. Should my IMX pressure axis start at 0 or the lowest my system can possibly produce? Also, where should it max out? The most boost I ever plan to run, or the boost I want this file to produce?

For example, if I only want to run 18psi max right now, should I make the furthest right value ~18psi or can I/should I make the furthest right column the most boost I'd ever plan to run (maybe 25-26psi)?