R
Rob
Hello,
I need to create a PID controller in a Cyclone FPGA, and I am wondering
if someone could help me with a few questions?
I need to control a PWM driver which is generating a single phase AC
waveform under load.
Is a PID controller the way to go? Lots of articals on PID controllers
seem to work in slow systems like,
temperature, or motors which are set to run at a selected speed. More
static type systems.
The system I need to design for is a very dynamic system. Can a PID be
as dynamic as I need it?
I am currently working on a system which works off a high low trip
point then calculates what type of equation to use. If the ADC sample
breaks the trip point, use the waveform data. If the ADC is in the
trip points but is not on the waveform point then nug the output by 60%
up or down (depending on which side of the waveform point its on). The
ADC is 5 times faster then the waveform table clock. So I can correct
the output 4 times before the waveform table moves to the next point.
The above describes a P type controller. This is also faily straight
forward to put into an FPGA.
Why would this not be a good way of doing this?
I don't have a very good grasp on figuring out the equations for a full
PID controller. I also don't really know how to describe it so an FPGA
can use it (I would probability be able to if I understood the math).
What would be the best way to design a system like this?
Is there an example of a PID controller which describes the thinking
process?
(Something which describes the math and walks throught how the
constants are selected, shows a basic VHDL implementation, along with
how to go about tuning?)
I am hoping that someone could point me in the right direction.
Thanks
Rob
I need to create a PID controller in a Cyclone FPGA, and I am wondering
if someone could help me with a few questions?
I need to control a PWM driver which is generating a single phase AC
waveform under load.
Is a PID controller the way to go? Lots of articals on PID controllers
seem to work in slow systems like,
temperature, or motors which are set to run at a selected speed. More
static type systems.
The system I need to design for is a very dynamic system. Can a PID be
as dynamic as I need it?
I am currently working on a system which works off a high low trip
point then calculates what type of equation to use. If the ADC sample
breaks the trip point, use the waveform data. If the ADC is in the
trip points but is not on the waveform point then nug the output by 60%
up or down (depending on which side of the waveform point its on). The
ADC is 5 times faster then the waveform table clock. So I can correct
the output 4 times before the waveform table moves to the next point.
The above describes a P type controller. This is also faily straight
forward to put into an FPGA.
Why would this not be a good way of doing this?
I don't have a very good grasp on figuring out the equations for a full
PID controller. I also don't really know how to describe it so an FPGA
can use it (I would probability be able to if I understood the math).
What would be the best way to design a system like this?
Is there an example of a PID controller which describes the thinking
process?
(Something which describes the math and walks throught how the
constants are selected, shows a basic VHDL implementation, along with
how to go about tuning?)
I am hoping that someone could point me in the right direction.
Thanks
Rob