Tweet
Hello,
so we have a PID control algorithm to regulate the boost for 20VT turbos (Yes, I know it's maybe not a classical PID because it has adaptive part, switchable D-Term, anti windup limits...)
The N75 valve is the actuator of this control loop. In classic therory the actuator curve has to be linear so that changing P, I or D would always have the same effect for all operating points of the actuator.
Somehow the WGDC (duty cycle) pre control map might be responsible for that.
My thought is, if the WGDC in that map is already set the way it will generate the desired target boost (which is defined in the target boost map) the PID has very few things to regulate und the actual boost pressure should follow the requested boost very well. Of course this is only true in case of the turbo can physically deliver the target boost...
My question is how we can practically adjust this duty cylce to get the best result. Please see two traces attached. It contains the N75 Dutycycle according to the WGDC pre control map (N75_Req) and the actual duty cylce (N75, that what the PID adjusts to reach the target boost)
Is the strategy to adjust the WGDC precontrol map a way, that these curves are lying exactly over each other, a good one?
Classical strategy would be mabye to disable the PID loop und run a fixed DC to exactly hit the desired boost and than turn the PID back on.
As you can see in my traces, even if the WGDC is set too low in the WGDC map the PID will fix it with applying a higher one. But I wonder if the quality of the pressure regulating might be better if I adjust the DC to fit from the bginning.
Who has experience?
Best Regards,
Daniel
so we have a PID control algorithm to regulate the boost for 20VT turbos (Yes, I know it's maybe not a classical PID because it has adaptive part, switchable D-Term, anti windup limits...)
The N75 valve is the actuator of this control loop. In classic therory the actuator curve has to be linear so that changing P, I or D would always have the same effect for all operating points of the actuator.
Somehow the WGDC (duty cycle) pre control map might be responsible for that.
My thought is, if the WGDC in that map is already set the way it will generate the desired target boost (which is defined in the target boost map) the PID has very few things to regulate und the actual boost pressure should follow the requested boost very well. Of course this is only true in case of the turbo can physically deliver the target boost...
My question is how we can practically adjust this duty cylce to get the best result. Please see two traces attached. It contains the N75 Dutycycle according to the WGDC pre control map (N75_Req) and the actual duty cylce (N75, that what the PID adjusts to reach the target boost)
Is the strategy to adjust the WGDC precontrol map a way, that these curves are lying exactly over each other, a good one?
Classical strategy would be mabye to disable the PID loop und run a fixed DC to exactly hit the desired boost and than turn the PID back on.
As you can see in my traces, even if the WGDC is set too low in the WGDC map the PID will fix it with applying a higher one. But I wonder if the quality of the pressure regulating might be better if I adjust the DC to fit from the bginning.
Who has experience?
Best Regards,
Daniel
Comment