PID control temperature of oven

[vitamina]

Hi! I should check the temperature of an oven through a PID control.
I wish that, after setting the reference temperature, the system acts on the power supply and change the current on the resistances of the oven, so that the temperature of the oven reaches the reference temperature.
Not knowing the programming language labview, I wanted to know if any of you have already saved the program in the PC and can post it, please.
Thank You!

[RavensFan]

Search the forums for PID control.  This question has been asked numerous times.  I bet you'll find some examples in the first few pages of the hundreds of hits you'll get.

[aputman]

I think you will find that using typical PID controls on an oven will prove unsuccessful because the temperature typically reacts too slowly after changes in the power supply.  You may end up having to write your own routine that checks the temperature at longer intervals and adjusts the power supply.

[nathand]

---

@aputman wrote:

I think you will find that using typical PID controls on an oven will prove unsuccessful because the temperature typically reacts too slowly after changes in the power supply.  You may end up having to write your own routine that checks the temperature at longer intervals and adjusts the power supply.

---

With the correct tuning parameters, PID works very well in slow systems such as large heaters with substantial lag between a change in the output and a change in the response. I've used the LabVIEW PID block in several such applications with good results.

[aputman]

That's why I used words like "think" and "may".  I haven't been successful at it but I am definitely no expert. 

[nathand]

---

@aputman wrote:

That's why I used words like "think" and "may".  I haven't been successful at it but I am definitely no expert. 

---

What sort of incorrect response do you get from the heater system? If you're actively working on such a project I'd be happy to help.

For tuning a slow system, I highly recommend starting with the Cohen-Coon method (or the similar Ziegler-Nichols Open Loop approach). Take the PID out of the loop. Set the output at a fixed level and allow the system to stabilize. Change the output, and allow the system to stabilize again (make sure you wait until the system is actually stable). Record the output and process variable the entire time. Then you use that data to make a couple of quick measurements, plug the numbers into the tuning formula, and you get your gains. Unless the system is horribly non-linear, these gains will work well. Make sure you get all the units correct (for example, the LabVIEW PID block expects Integral Time in minutes, and not an integral gain).

[aputman]

No this project is long gone but thanks for the info.