PID for Pipe Burst Testing (Finite Time)

[ulloap]

Hi Guys,

I'm a beginner in the PID World, so right now I'm trying to program one controller to do the following:

I have a Burst Machine to test pipe samples -big sizes, over 12" - 36" Outer Diameter by 40"long-.

So the test logic is as follow:

I need to reach a specific pressure (psi) in specific time (65 sec) ASTM Standard Test, the pressure rate depends on the thickness wall of each sample, but the time to reach the pressure keeps the same, no matter of the size. That means I need more or less flow-rate -depends on the sample I'll be testing-.

So I have a centrifugal pump to fill the pipe sample, when its filled ,I switch to a pressure pump, where the PID comes into play, it needs to sense the current pressure inside the sample -tru a Pressure Transducer-, send the signal tru the PLC "makes the magic" and adjust to frequency of the pressure pump VDF needed to achive the pressure set point in to the specified test time....

I'm stuck in how to setup it, due the finite time -65 sec-, how I need to tune that PID...

Hope you guys can help me out!

Best Regards
Ulloap

 

[duckman]

To save one of the guys time who may be able to help, please supply all info as well as PLC and the program you have written so far. No ones is likely to write it from scratch for you. People are willing to assist you in getting YOUR code to work correctly.

 

[psgama]

Just tune the PID for a damped 5 sec response without overshoot and step your setpoint in 5 second increments in your PLC program over 60 seconds.

 

[Peter Nachtwey]

There is a formula for the rate of change of pressure with respect to fluid being added to the volume under pressure

dP/dt=RateOfChangeInPressure=BulkModulusOfWater*Flow/Volume

You should be able to predict the control output signal to the pump pretty closely and use a bias that ramps up as the pressure SP ramps up.
My company gets involved in a lot of pressure testing.
The bulk modulus of water is about 300000 psi. That means adding 1% more fluid to the volume under compression will increase the pressure by 3000 psi after the pipe is full of water. That is a lot. You will need fine control of the pressure pump. The size of the pressure pump is critical. You don't want it too big or small.

Read up on the bulk modulus of fluids. What pressure are you testing too?

 

[shooter]

A PID is only good in one situation, so you will need a lot of settings for all the different pipes,
so a matrix will be best choice here, containing the PID settings.
Use a gliding setpoint so the pressuresetpoint is slowly pushing up,
so the setpoint should move every second for 1/60 of the actual setpoint.
The pump will slowly reach its point, and controlled by a good setting of the P and I and D factor.
I doubt you can get a good control for all types with only one pump, but you can try.