Linear motion

[laura]

Yes I have. A lot about cylinders, but not about controlling them.

laura

 

[laura]

Thank You very much.


I do not know, if You have understood me correctly.

When I say, that I want to control the traveldistance, I meen that I want to be able to tell the cylinder, how far it shall move. The traveldistance will NOT be the same from time to time. For example, the first time I want the cylinder to travel 4 inches, come back again, and then travel 5,5 inches. Then travel 2 inches backwards e.t.c..

Therefor, I need a precise controlling of the traveldistance. The plc needs to know exactly how far the cylinders have moved, and must be able to move them on further to a ´precise position.

Thank You again.

laura

 

[tom_stalcup]

Ahh.. Okay... That makes a little more sense then. It seems like the "robocylinder" link posted earlier would do the job, but it may be out of your price range... I'm not sure how much they cost, since I've never used one before.

The air cylinders being talked about before were more for a fixed distance than for a variable distance, so probably wouldn't help you very much.

And we're still going to need to know things like how often you need things to happen, how long they can take to happen, etc.

You could put a stepper motor that you ripped out of a printer, and hook it up to a small gearbox of some kind... It might work perfectly, but would take 5 minutes to get there... You could mount a small motor to a screw drive and move things that way... You could come up with a fixed length air cylinder for the move, and a variable position stop clamped in place by a brake.

We need more information if you want us to help you pick out the best(and cheapest for you) options.

Information is key! And trying to help you design a machine without knowing EVERYTHING that you need it to do won't work very well.

 

[n521554]

Laura,

If this machine's cycle is slow, a cheap method may be to use a small stepper motor and ball screw controlled by a contactor. Use a prox. to count turns of the screw as feedback to the PLC. You can use the pitch of the screw to determine linear distance (1 pulse = Xmm). This will not work at too high of a speed because the PLC will start missing pulses from the prox. We use this system in applications where speed and accuracy (+/- 1mm) are not an issue.

Wally

 

[laura]

Thank You.

I actually need several of the linear movements.

Ok, Here it goes:

I am trying to design a crane, that can pick up plastic soda bottles from a beercrate (I believe it is called, I am not that good at english), and can move them to another beercrate, one by one, without moving the beercrates.

I have made a very simple drawing, to explain what I mean:

The meaning is, that this crane can put the bottles in order, so that the green bottles are in one row, the red in another etc..

If I steer the crane to position X:A1, and save the position in the plc, then the plc is able to move to this position later on, by itself. This I will do for all the positions, and the the plc knows all 60 available positions.

Then I can tell the plc, to move a bottle from for example position X:A3 to position Y:C1.

This way the crane can sort all of the bottles, if I just tell the plc, where the bottles are.

Tell me if I have to explain more.


laura

 

[laura]

Hi again.


Here is another simple drawing of the crane, as I imagined it:

laura

 

[Steve Bailey]

I think a review of your mechanism design is in order. You are showing five axes of motion where only three should be sufficient to do the task you've oulined. Calculating a set of position targets for each of these axes to pick up a bottle at one position and move it to another will be a nightmare. There will probably be more than one set of target positions for any given position.

I assume axis b is attached to axis a similar to the way axis c is. Where is the second attachment point for axis d?

 

[Peter Nachtwey]

Wow, did anyone see that coming?

I enjoy watching threads like this.

 

[laura]

Sorry, the letters should not be on the last drawing, and have nothing to do with the axis.


Yes, I might only use 3 cylinders, but I don't know how to control them yet.

I have not been able to find a price for the Robocylinders.

laura

 

[geniusintraining]

In reference to your drawing...look here you may get a better idea
http://www.intelligentactuator.com/application_videos.php

This is one application the K.I.S.S concept is valid...speaking from experience

 

[laura]

Thank You for that.

This is one application the K.I.S.S concept is valid...speaking from experience

Sorry, but I don't understand.

laura

 

[geniusintraining]

Sorry, but I don't understand.

Here in the US (maybe elsewhere), KISS stands for Keep It Simple Stupid...not implying that you are stupid

I made a turner once that I over engineered it tried to make it spin (100 mph) stop (on a dime) up, down...everything...it was going to do everything, after I put it into production...it fell apart in a week, it was not good, I was embarrassed that my name was on it.

The KISS part came in on the second version...just a simple lift and transfer conveyor...thats all that was needed, nothing more just lift and move

When I saw your first drawing...it looked like my version 1

 

[laura]

Uuups, that does not sound good.

If my crane consists of 3 cylinders, I would estimate, that it is possible getting it to function good.

But maybe it is more difficult than that.

laura

 

[Lancie1]

Laura,

This is a very complex project. You would need precise motion control in several different directions. The PLC or master controller has to know the exact coordinates of 72 different locations. It has to then be able to move to each of those 72 different positions. Why do that when the final destination only requires 6 seperate positions (colors), not 36 as you are planning. Also, the bottles do not have to be removed from 36 individual positions, unless that is a fixed requirement of the project.

Instead of your "crane", I suggest looking at ways to simplify the task. Let us perform som "brainstorming", also known as the "What If" game, where negative thoughts are not allowed, but each idea or thought builds on and uses the previous one:

What if the bottles were not in the same type of beer crates (grids)?

What if "X" was a gravity-feed rack (chute or inclined ramp), where the "crane" (or whatever mechanical arm is needed) pushes out the bottom one and the rest of the bottles roll down, so that the next one is available (located in the bottom feed position)? Therefore we use the natural law of gravity to select each bottle, one at a time, with only 1 known position needed, instead of 36.

What if the "Y" receiving crate is not placed (located) horizontally? What if it were to be turned vertically, sitting on one edge?

What if "Y" was sitting on edge, with only 5 inner partitions to divide the 6 colors, so that the bottles could be pushed out of the "X" rack and into the "Y" receiver rack with a simple pusher rod?

What if "Y' was tilted slightly, so that if a bottle is pushed into the highest end, it rolls down to the lowest position in the crate (again using gravity), but still is located on the correct color level?

What if "Y" was mounted on a simple up-and-down elevator, with 6 different positions (levels). The "crane" looks at the next bottle, determines the color, then moves the elevator to the correct level. When the color level is in position, the push rod pushes the bottle out of the gravity-feed "X" rack and onto the correct level of "Y".

What if the cycle then repeats for the next bottle until all 36 are moved into the correct color levels?

What if the above could be done with only 1 vertical up-and-down "Y" crate using a rotatry screw, and a push rod actuated by an air cylinder?

To summarize, sometimes it is wise to redefine the problem so that it can be done with a simpler system.

 

[Peter Nachtwey]

Pneumatics will NOT work. No way. This project is not simple, and not linear. You can try hydraulic cylinders. The project would be similar to this one:
http://www.ironderrickman.com/simulation.htm
This was a 'fun' one because of the arcs.
BTW, this is where some, a very small fraction, of all of that extra money you have paid at the pump goes.

I agree with those that have said that you should use a X Y table. This would be relatively easy. I think this project requires more help than what can be provided by forum.