Linear Ramp down in Micrologix 1400

Could somebody suggest to me how I would be able to ramp down an output linearly based on the state of an input?

The process is as follows: The PLC is driving the speed of a wrapper up to the user entered speed setpoint. When the roll length gets to be 95% of the target length, I'd like to start ramping the speed down. When the roll length reaches 99% of the target, I'd like the output to be zero.

Is the time from 95% to 99% constant?

Ken Moore said:

Is the time from 95% to 99% constant?

Click to expand...

That will make no difference to the solution.

I wonder if the Micrologix 1400 is fast enough. The question I have is how long does ABnewb expect the ramp down to take? If it is many seconds then OK. If it 200 milliseconds the micrologix may not respond fast enough. How much time is there between the time the distance or position is read to the time the control output is changed may be problem.

ABnewb needs a velocity as a function of distance or position equation.

Peter you are correct, the question should have been, "Can the time from 95 to 99% be a constant."

What's actually controlling the speed of the wrapper - a vfd?
Are you using the analog outputs or pulse outputs to control the speed?
Sounds like you should have a target speed and slow speed on the HMI either way.

If analog + VFD: When the input goes active, have the PLC output the slow speed. Let the decel of the VFD control the actual ramp duration.

If pulse output + VFD: There's an accel/decel duration of the pwmx function file (add in 10ms increments). You can alter that value to change how long it takes to reach the new target freq.

If no VFD, then let's re-visit the first question

Also, how do you know how the roll length in relation to target?

Could always scale 95->99 so that it is 0->1, and multiple your speed by that number. If its less than 95%, make your scalefactor 1. It would scale linearly.


You can also use a PID loop whose output is scaled between 0 and 1 to use as trim.

Come on guys.
This is simple.
v^2=2*distance*deceleration_rate
Assume v95 is the velocity at 95%.
At 95% roll length calculate the deceleration rate.
distance is 99%-95%=4%
deceleration_rate=v95^2/(2*dist))
The deceleration rate is assumed to be constant.
Then every scan do
Then v = sqrt(2*deceleration_rate*(99%-roll_length%))

My doubt is that the micrologix 1400 can do the initial calculation exactly at 95%. Of course the deceleration_rate could be calculated at 94% or 93%. It wouldn't make much difference.

Unfortunately, the deceleration rate is not constant. It will vary based on the operator entered speed set point.

The roll length is calculated based on the result of a high speed counter. That portion of the logic works well.

I currently have the speed control set up on a PID loop, but it isn't working as seamlessly as I would have expected it to. I'm not great with tuning, so that plays a part.

Currently, the process is as follows:
The process variable is the wrapped roll length
The setpoint is the desired wrapped roll length
The output is a 4-20mA speed signal to the vfd


When the roll length gets to be 95% of the desired roll length, the wrapper needs a linear ramp down. When the length gets to be 99% of the desired roll length, the speed reference to the VFD needs to be 4mA (Because the roll coasts long after the output goes to zero).

I've tried using the scale with parameters, but I can't get the output calculation to work as intended.

I could add a bunch of sloppy logic to get it done, but I know there's a better way.

By "wrapper" I assume you mean winder. A wrapper is a whole different beast that wouldn't necessarily base anything on material length.

Why do you want to start to decel unconditionally at 95% of full roll? Generally speaking we would approach this like Peter Nachtwey indicated, with a fixed decel rate and a variable stop initiation length based on current speed. Also, what is the coating thing about? Presumably you have control of this roll with the winding motor. I would think you could get it to stop wherever you want by issuing a zero speed command (4mA in your case).

Keith