Announcement

Collapse
No announcement yet.

Semi OT, automating a conveyor

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Semi OT, automating a conveyor

    Ok I need a little help with steppers or servos on this one.

    I currently have a friction conveyor belt mounted 90 degrees at the end of the sheeting conveyor on our flexo printing press. The conveyor is operated via a clutch and constant running motor, the clutch is triggered by a batch counter/process controller to engage the clutch every 50 cards that pass under a photo-electric sensor. There are two problems with this design. First I've never had much luck getting the belt to track straight and second I can't use "paddles" with a friction drive. My plan is to move to a link belt conveyor with paddles appropriately spaced, let's say 8" apart. I want to control this with either a stepper or servo so that when the batch counter sends a pulse the belt will advance to the next paddle. The paddles and a side rail will form a U to catch the cards coming off the press. This will effectively stack the cards in between the paddles for a worker to remove and place on a card to go to out guillotine cutter for further finishing then eventually to our packaging line.

    So my question is, will a servo or stepper controller accept a pulse to advance a set distance at high speed? Do I need a PLC? I'm new to this whole automation thing and need a little help. I plan to make the paddles removable so they can be adjusted for various width cards as in some cases we have two or even three rows of cards coming off the machine. As always any help is greatly appreciated.

  • #2
    Good news is, it's easy enough with a little hardware and programming experience. You're going to need a minimum of a stepper motor driver (probably bipolar half-step) and control logic. The control logic can be a single chip solution. All it's gonna do is accept a pulse from the PE sensor, then issue the requisite number of pulses to drive the stepper motor the required revolutions.
    If you have even the vaguest idea in programming, might I suggest
    www.picaxe.co.uk/
    Usual disclaimer.
    Software and extensive tutorials are available free from the site. Programming hardware consists of 2 resistors and a D9 plug. Any old windoze box will suffice for programming. Language is dedicated RISC basic compiler. Simple to learn, easy to program.
    Stepper drivers are available off the shelf from sources like xylotex and gecko.
    Regards, Lin
    Just got my head together
    now my body's falling apart

    Comment


    • #3
      There is no need to get that fancy. Besides, if it loses count you lose position. Use the paddles to break a sensor beam or otherwise trigger a sensor to stop the drive. Override the stop function with a one shot timer just long enough to advance the paddle out of the beam so it will drive until the next paddle interrupts it and stops the belt. An ordinary motor will do. Some allowance will need to be made for coast.

      To make the belt track put a slight crown on the drive pulley. A few wraps of tape in the center
      will do for a test.


      Note: 23 years experience with paper handling equipment.
      Last edited by Evan; 11-15-2006, 01:51 AM.
      Free software for calculating bolt circles and similar: Click Here

      Comment


      • #4
        Fair call evan
        If you do go that route, I'd suggest braking resistor and contacts across the motor to cut the 'coast' problem.
        Just got my head together
        now my body's falling apart

        Comment


        • #5
          By friction drive do you mean the standard fare rubber belt with traction and idler rollers?

          If your tracking a count and packeting,is the count critical and does it flush out count wise now,I.E. you get 50 cards no more no less per packet?
          I just need one more tool,just one!

          Comment


          • #6
            Ok, I'm intrigued. I'd rather not build a controller though if at all possible. Is there some sort of off the shelf micro controller that will do the job?

            Evan, I thought of something similar but I'm more interested in a stepper or servo for the positive start and stop, coast with the friction belt has been a problem. By using a link belt and stepper/servo I can eliminate the coast issue altogether. Like watching my CNC rapid to position then stop and perform an operation. The link belts are gear driven and should work well in this application I would think. The current setup gets us by but there are problems with the card stacks falling over when the belt advances as well as not getting a good stack due to only having a back stop. I figure with the paddles as sides and a back stop I should get a nice stack that a worker can easily remove or I could eventually even automate that part as well.

            The real problem here is temp help. We don't run this setup all the time so it's usually a temp worker removing the cards from the belt and they are at best unreliable. The more I automate the less temp help I need and the more I can pay our full time staff as productivity will increase and therefore we can handle more orders. We could just pay to have a system built but we just shelled out $300k for a packaging line so money is a bit tight for a custom conveyor system and I have the knowledge and tools for the mechnical and electrical part of the assembly, it's the electronics and programming that I'm not so knowledgeable of. Aside from that the learning experience would benefit us in other areas.

            Then there is the fact that any time I spend building this thing will be paid play time in the shop.

            Comment


            • #7
              Originally posted by Swarf&Sparks
              Fair call evan
              If you do go that route, I'd suggest braking resistor and contacts across the motor to cut the 'coast' problem.
              Brakes can be had for cheap surplus,spring actuated,release on make(releases brake when motor sees voltage) Youngs surplus usually has them in the 40-60$ range can be had in 56c face motor through mount so the unit fits between the motor and speed reducer.
              I just need one more tool,just one!

              Comment


              • #8
                Wire wound resistor from a few cents to free(junk box or wind your own, anything from guitar string to 316 MIG wire) and relay contacts which you probably already have.
                Christian, if you need some help with a picaxe solution, email me at [email protected]
                Rgds, Lin
                Just got my head together
                now my body's falling apart

                Comment


                • #9
                  Use an ac motor and stop it by shooting it DC (the AC half waved through a beefy diode). It will stop so fast that you had better lockwire the nuts holding things together.

                  By sensing the paddles you eliminate slip problems, variable spacing problems with different setups and the entire effort is much simpler. I have seen just about every possible way there is to move paper and card stock around, sort it, stack it, staple it, bind it, cut it etc. Simple wins every time and by using an absolute reference to the paddles it can't get out of position and it doesn't need to be initialized or reset.

                  Note that the paddle sensing system doesn't have to be anywhere near the stacking part of the belt. It can be anyplace on the belt that is convenient.
                  Last edited by Evan; 11-15-2006, 02:18 AM.
                  Free software for calculating bolt circles and similar: Click Here

                  Comment


                  • #10
                    with the proviso that it's an AC indution motor
                    Just got my head together
                    now my body's falling apart

                    Comment


                    • #11
                      oops, an induction motor would prob be better
                      Just got my head together
                      now my body's falling apart

                      Comment


                      • #12
                        Resistors are great,until you need a positive stop and hold.If ever you need to stop a conveyor that runs on an incline you will find out what I mean.Believe it or not a 40:1 wormgear reduction will backdrive and pile up the belts contents on the tracking roller end,that's why on large conveyors backstop sprags are used.

                        DC won't matter either,a PM motor will back generate as it coasts,you can dump a resistor across the leads and do the same thing.
                        I just need one more tool,just one!

                        Comment


                        • #13
                          "DC won't matter either,a PM motor will back generate as it coasts,you can dump a resistor across the leads and do the same thing."

                          zackry my point weird. Tho, as you say, if there's much load on a belt it can be driven back. Don't think that's the case in handling paper/card tho.
                          Just got my head together
                          now my body's falling apart

                          Comment


                          • #14
                            Originally posted by wierdscience
                            By friction drive do you mean the standard fare rubber belt with traction and idler rollers?

                            If your tracking a count and packeting,is the count critical and does it flush out count wise now,I.E. you get 50 cards no more no less per packet?
                            Yes rubber belt with drive and idler rollers. The drive motor spins continous and the drive roller is engaged with a clutch.

                            Cards are counted by a PE sensor as they drop onto the belt, when the count reaches 50 the clutch is turned on for 2 seconds to get the stack out of the way. Currently the acceleration without paddles causes the stack to fan out so someone has to restack them. Also on card always winds up between the stacks as it takes a split second for the clutch to fully engage.

                            I got the idea for steppers or servos watch the friction feeders on the new packaging line. Three servo operated feed various cards and inserts from stacks onto a servo driven chain conveyor that advances them to a wrap around heat sealer that makes a package similar to the packaged crackers you get with your soup at a restaurant. Fin seal along the bottom and end seals with cut. This feed as vacuum labeler which lifts the packages and applies a label to the bottom of the package before dropping it onto a shingling conveyor.

                            I'm building the shingling conveyor presently using a simple AC drive tied to a 0-10V output on the labeler. This way I can adjust the shingling in relation to the line speed of the packaging line using a simple potentiometer.

                            The whole process is pretty cool, we make the adhesive for the label stock then coat it using a Meyer Rod coater and laminate face stock to it. This is then coated with a photo sensitive ink that I have a patent on using the same coater. This is followed by a UV calibrated topcoat that is also water proof. From there the coated rolls are edge trimmed and slit to printing press width on our slitter. From the slitter they go to the editing machine and then to the printing press where they are first printed roll to roll and then run back through for diecut, matrix strip and sheeting at which point they drop onto the conveyor I want to replace. Once stacked they are sent to our guillotine cutter and cut into one of three different finish sizes, these are stacked and will then go into the feeders on the packager.

                            We are a small company that prefers to automate as much as possible so that we can have a limited number of well trained and well paid employees. Normally we'd just pay someone to make something like this but I enjoy the challenge and the company doesn't mind saving some money right now. I made the side frames for the shingling conveyor on my CNC today, much more fun than sitting at my desk doing paperwork (which will still be there when I'm done )

                            BTW I got my inverter drive from Automation Direct today, can't wait to hook it up to the CNC. Couldn't help but notice the Marathon motors on the coater are good for 120hz, maybe I'll have to put one on the mill to spin those 1/8-1/4" cutters faster.

                            Comment


                            • #15
                              Originally posted by Swarf&Sparks
                              "DC won't matter either,a PM motor will back generate as it coasts,you can dump a resistor across the leads and do the same thing."

                              zackry my point weird. Tho, as you say, if there's much load on a belt it can be driven back. Don't think that's the case in handling paper/card tho.
                              Agreed,it's just a matter of how quick he needs it to stop.
                              I just need one more tool,just one!

                              Comment

                              Working...
                              X