What to use to measure fine movements of press. Havent done this before

[JJ1234]

I have a mechanical press. We are having problems and we believe it may have something to do with the movement and speed of the press. The press is old and problems seem to go away with a fresh die capsule change but we can't pinpoint the problem.

I need to measure and record the movement profile of the press throughout its cycle. Fill to eject can range from about 50 - 70 thousands and I
would not consider the press high speed but it is not slow either, it moves about 15 to 20 cycles a minute.

While I've done several automation projects, I've not really done anything like this and could use help. Could someone point me in the right direction for what equipment I would need.

I assume you could do this with some type of DAQ and linear sensor though in the future I might want to do some programming with alarms and such so I would like to use a PLC.

So my questions would be:

I like automation direct plcs for little projects like this. I will need to record the data for the day, haven't looked to see if there PlCs can do this.

Can I just use an analog sensor with an analog card?

Any recomendation on with type of sensor would be best? something like an LVIT?

 

[I_Automation]

Automation Direct makes encoders that will be perfect for this.


This one is 1,000 pulses per revolution. Have it report to a high speed counter input. Add a TOP sensor to set the count at 0.000 at the top of the stroke and divide the count by 2.7777777 to get the degrees of the ram.


If 0.36 degree resolution isn't good enough they have higher PPR encoders in the same line going up to 5000PPR.


I have worked on many stamping presses (that's where I was volunteered to learn PLC's) and they had encoders like this while running up to 160 strokes per minute. (Also saw a display press making tops for soda cans that was running at 1,600 SPM, which basically was vibrating at that speed.

 

[JJ1234]

Thank you,

The way the press works I will actually need sometype of linear reading. Its a PTX press which has a lower punch and anvil. The lower punch is driven by a cam and treadle as well as a couple cylinders that control compensation and eject.

Therefore i need to mount the sensor basically on the tooling thats driving the lower punch that goes up and down so that i can get its exact position.

 

[JJ1234]

You can see here
https://www.youtube.com/watch?v=VpxK-0i7om0

 

[drbitboy]

You can see here
https://www.youtube.com/watch?v=VpxK-0i7om0

Wow, I could watch that all day. And it's in waltz time!

For which part of that are you trying to measure position vs. time? You said "lower punch" which I assume is the part underneath in the middle that seems to be changing direction more than once per cycle?


And this is about data acquisition to diagnose and/or characterize the process, not control the process, is that correct?

 

[JJ1234]

Yes, all of that is correct though i would like to incorporate a PLC so I could set off alarms or lights or such in the future to assist with set up/making sure things are in spec, etc but no controls at all.

 

[Peter Nachtwey]

I need to measure and record the movement profile of the press throughout its cycle.

Yes!

1000 counts/rev is not very good. If a cycle takes 4 seconds then that is only one count every 4 milliseconds.
BTW, Gasbarre uses our controllers. All their camming is done electronically. The controller can get record multiple pen trends at 500 micro second intervals.

 

[JJ1234]

Our models are about 20 years old. One of them is like the original prototype. Very low tech. Our units dont even have that control box on the right side.