Hello Eric , My system Is as following ,I have a YIG which is a curent controlled oscilator.
The YIG output frequency is controlled by the current flowing threw the coil.
the error is the output of a DSB mixer which acts as a phase detector.
My YIG has two coils :
1.TUNE which is the larger coil and its a driven by steady voltage source ,however this coil is drifting.
2.FM coil which is smaller ,but this is the coil of the YIG which input is connected to the output of the PID controller.
(servo amplifier) shown below.
The feedback PID mechanism needs to compensate fot the DRIFT of the tune.
What kind of system is that?
So I'll know what PID to design?
Thanks.

 

Hello , I posted the system I need to controll in the previos post.
Could you please say what PID is mostly suited for this task?
Thanks.

 

hi yef,
On the circuit you have posted where is the Set point Input?
Also, the PID path requires a circuit element to control, I understand on your project it will be a YIG oscillator, so add that circuit to the simulation, so that we can observe the response.

Also, what the coefficients for each the PID paths, are you going to determine them empirically?
E

 

Hello Eric, Thank you for the responce.
regarding things you adreesed
1. " On the circuit you have posted where is the Set point Input?"
-Its a feedback system so its circular. We have YIG signal and cavity reference signal.
We alway compare the two using mixer and try to make the YIG be as close as possible to the REF signal.
The mixer acts as phase detector.
2. " so add that circuit to the simulation, so that we can observe the response. "
-Is there a way to represent the YIG as behavioral?
Its problematic to recreate the circuit behavior of the YIG because the oscilatio happens from inducing magnetic field on a sphere.
My input is DC current and some AC(very little frequency)
Is there a way you can reccomend me that I will put certain inputs and record the output (frequency ,amp, phase)
So i could use it in the feedback loop?

3. " what the coefficients for each the PID paths, are you going to determine them empirically "
Yes I want to simulate a closed feedback system and tune the coefficients.
Thanks.

 

Hello Eric, I have answer as best as I could to your questions.Could you please respond so I'll know what type of PID to use?Thanks.