web width control

S

Thread Starter

Sandeepa

I am looking into techniques for controlling the width of the film in blown film manufacturing process. I intend to measure the width of the film after collapsing and then carry out the control action.

Identifying the edges of the film and then calculating the width seems to be the way to go for the width measurement purpose. Since the web would be transparent, I was thinking of either using diffuse type of photoelectric sensors, or a would a photoelectric light grid make more sense?

Is this the right way to go? What more do I need to look into in this application?

Thanks in advance for any ideas, guidelines, help.
 
One method that I have used in the past is to mount a set of ultrasonic sensors on top of a calibration basket. You can calculate the layflat based on the input from the sensor and by knowing the distance from the front of the sensor to the center of the bubble. Use at least 2 or 3 to average out any error based on bubble movement. Then you can use a setpoint from an operator to compare the existing layflat. Send the error to the IBC system to change the bubble size.

Another method is to do what you suggest and use an edge sensor. BST Promark makes a web guide system that has layflat feedback.
 
D

Davis Gentry

Assuming that the film is transparent to visible light, and if other sensors such as laser edge detectors can't find it, it may be that a different wavelength will work. Perhaps the film is NOT transparent in UV, or IR, or at some other wavelength outside of visible. If so, then you can pick up the edge using a line scan camera with appropriate CCD and lensing. Note that if UV works then standard lenses will not - you will need to go to a sapphire lense which does not block the UV - Nikon makes good ones.

Davis Gentry
Delta Tau Data Systems
 
I am not too keen on using ultrasonic sensors for cost purposes.What I have in mind is something like this :

Have 2 diffuse type of photoelectric sensors mounted on a bar.And move them outward till the edge is detected.Then, based on the number of steps teh stepper motor has taken, calculate the width.

Not a very refined picture but hopefully gives the basic idea.
Does this sound like a feasible way of doing the required ?Would a diffuse type sensor work efficiently with a transparent plastic film ?
Thanks a lot, once again.
 
Top