Pressure control pid help

[cbt]

metering valve info

http://www.autoclave.com/products/valve_actuators/electric_flow_control/index.html

I've taken a break on the code to work on other stuff will get back to it shortly

[zacharia01]

Thank you so much. They seem to be big for our applications. Zacharia

[cbt]

OK started to work on this control again and still need some help with the pid.

My control valve has a slow response time it has a motor for control and 4 ma is close and 20 ma is full open thats at 11 turns.

With my pid when it adjust it just goes from fully close to full open the pid doesn't seem to put out any other output signal than the max and min no in between

[cbt]

I do get good pid response if I change the out pput max to a lower ma output but at lower pressures the valve bleeds slower because it needs to open more curent pis setting

-1    p

.001 i

0     d

[nathand]

---

@cbt wrote:

I do get good pid response if I change the out pput max to a lower ma output but at lower pressures the valve bleeds slower because it needs to open more curent pis setting

-1    p

.001 i

0     d

---

Are you sure you want the I value at 0.001?  How did you determine that value?  As I wrote above, a small I time is actually a LARGE integral output, and if your valve is slow, it's very possible that the integral will windup quickly and saturate the output.  Try it with no I value and see how the control is.  Then go through a real tuning process to find a good value for I, or at least, start with a larger I value and then see if performance improves as you decrease it slowly (a large value depends entirely on your process - might be 60 minutes, might be 10 minutes, might be 30 seconds).  The I time is in minutes.

[cbt]

ok that helped alot on the response of the valve ok now Im going to try your code again with those changes

[Dskp]

Hi Cbt,

I too working on similar kind of Bleed Down Project using crio and NI 9263.

So,your are now able to achive bleed rate as expected using PID's

It will be better if u can guide me on coding to get better result.

Thanks in Advance.

[sassyaspy]

Hello selvan.skp,

Are you having trouble with your hardware, or just choosing the correct PID values for your system? If it is the latter, please explain more about what you already have and where it needs to change. In the end you might just have to play with the values until you get the response you are looking for. There are some recommendations given by this site,  however, in the last section:

http://www.online-courses.vissim.us/Strathclyde/tuning_pid_controllers.htm

Hope this helps!

[natasftw]

selvan, it's unlikely you'll get much of a response from cbt.  The last post was: 01-19-2012 02:41 PM  His last login was: 12-23-2014 05:03 PM

It's generally best to create a new thread to discuss your application than bump one that's 3 years old.  It's unlikely you're sharing things exactly.  Your post has already required questions to probe and find out where exactly you need help.  It helps us help you if you provide as much information as you can about where you're stuck.

[nathand]

There already is a separate thread started by selvan.skp: http://forums.ni.com/t5/LabVIEW/PID-Control-Design-For-Flow-control-Valve/m-p/3090596