Sign in to follow this  
Followers 0
señorJumper

Pid Loops On Slc 5/04

3 posts in this topic

Hi guys, I have a problem with a wrapping machine that has 4 pid loops to control sealing heaters, and when the machine starts the temperature goes low and when the machine stops the temperature overshuts high, is there any good example on how to control temperature with an analog input to measure and digital output to control the heater? I have attached the file in case somebody wanted to take a look at it. Thanks for your help. Señor Jumper Cpr55101.rss

Share this post


Link to post
Share on other sites
It sounds to me like a case of integral response (or lack thereof). I don't know enough about your process to make that statement with alot of conviction but based on the description it fits. I'm at home and don't have a copy of Logix500 with me. But I assume you are using 'time proportioned output' or 'PWM output' or whatever they call it in Logix. It is the output style that drives a discrete output for some percentage of a time period to get a type of analog control. If you are using just plain old on/off control around a window that the stuff I'm about to spout is meaningless. With the machine sitting idle the PID output needs to send out pulses of some period proportion just to overcome the heat loss to the atmosphere. Just as a number let's say it's 20% time. Now when you start the machine the heat load increases due to the sealing. So the time duration needs to increase to match that load. Here's the downfall of pure feedback controllers; you need an error in order to react. In your case heat loss causes the temperature to drop. It sounds like you are set up with relatively low P and D term gains. So the correction is left up to the integral term. But that gain appears too low to keep you on track by itself, so the temperature drops. After some time the integral value increases to the point that the time proportion matches up to the heat load and the temperature stabilizes at the setpoint. But then when you stop the reverse happens. Less heat load > too much output on time > temperature rises. The way I see it you have two options: system modeling or higher gains. More P and D term gain may help if you are not running into temperature oscillation during normal operation. The other idea would be to get a handle on the integral value needed to maintain temperature during standstill and normal run and preload these values when you stop and start. The stop one is easy. The start one is a bit more complicated unless you run only one product at one sealing temperature and one speed. It might be pretty tough to hit the start case spot on. That said, something is better than nothing. If you can get a good median value for the integral sum diring normal run you could use that to get you close. The PID will pick up any slack from there. That would reduce your variations at start and stop. Keith

Share this post


Link to post
Share on other sites
I had a quick look at your program and my suggestion is to reduce the reset time. At its current setting of 20 minutes/repeat it is most likely way too slow. At the same time you will probably need to reduce the proportional gain from its current setting of 80.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0