Sign in to follow this  
Followers 0
moosetracks

3 axis stacking

4 posts in this topic

I’m developing hardware and software to stack widgets on a pallet. I have some ideas on how to accomplish this using the traditional PLC (AB) programming I have been doing, but it will need to be augmented with some additional PC programming to keep the PLC program easier to manage. I am going to pick the widget up with a vacuum system on a 3 axis gantry system and then pass it over a Gocator (measurement laser) to determine its size and orientation and then adjust the drop off point to place it where it needs to be. Then the program will need to remember where all the pieces are dropped off so it can continue to operate until the stack is complete. When I started to think about the programming I was curious what combinations of software (languages and types) are typically used in these types of applications? Any feedback is appreciated. Thanks!

Share this post


Link to post
Share on other sites
My experience with palletizing is limited to Fanuc Robotics using Pallettool which is a collection of teach pendant programs and karel functions that carry out the tasks of servicing multiple "infeed stations" and multiple pallet stations, with a 4 zone vacuum gripper, and pallet and slip sheet handling, all the tools are there, you just customize the code for your work cell, much of which is menu driven, and create and download pallet patterns with PC based software or via your thumbs and the teach pendant. Even without Pallet Tool the base robot software based on Cartesian coordinates supports accumulating heights, and there are functions written for you to call if you need to find which obstacle (be it a pallet stack or a tall skid of product) is tallest. Collision avoidance has to be added on to the base logic. Pallet tool supports the ability to pass 8 bit numbers through "group i/o" which can be from a communication channel, digital robot input, digital expansion input or whatever, and uses some archaic logic to facilitate the ability to transfer pallet tool info back and forth with even the most ancient of PLCs. So, the code is heavy on the extra logic you'll never need side, but has all the tools to pretty easily set up and palletize rectangular cartons. Their newest software (including camera based vision) can pick from a moving conveyor and palletize randomly shaped rectangular objects. I expect that most any Cartesian based system would be standard stuff. The tools to design the pallet layouts and how much of that you automate at the controls level probably varies. When I first took over this job, i inherited robots that were commissioned by some folks who apparently had no understanding of the custom gripper. They would only column stack 2 of our more than a dozen sausage products. We were hand stacking 80% of the product by the time I was hired. No one knew what would happen when I tried to start doing interlocking patterns, and I even I was surprised by my first trial. It took me weeks to get the mess straightened out. Pallet Tool is complicated, too complicated in my opinion, but its effective. They give you every thing to palletize or depalletize, purge out infeed lanes, make request to upstream equipment for hte number and orientation of cases, they've been palletizing for so long, "there's an app for that" applies, and much of them are TP programs which are concise and fit well within the model, can be copied modifed..pretty clean code overall well threshed. PalletTool is structured well for the sake of ease of customization, and it would be a simple and safe task to remove the depalletizing calls and functions to tidy things up, perhaps impact performance if you can remove a layer of program calls based on known constants. At one point, I was ready to roll my own pallet tool. i figured about 100 lines of teach pendant code to follow a series of PLC commands with all the motion paths in PLC variables! I could whip this out in a day i muttered under my breath as I opened up notepad++ to compare what they'd done with the "stock" code. But, I finally got things working pretty well, by removing everything that didn't jive, setting the custom gripper properties correctly and then dividing the other messy spots into manageable chunks ...now it works as designed and is customized as Fanuc would have intended, so it works with PalletTool PC and/or RoboGuide where the real handy work lies. Our new engineer has mastered RoboGuide (3-D emulator) with Pallettool and has gotten some recent cycle time gains by breaking the layer barrier for a 4 case gripper on 6 unit layer patterns. So much of the math and calculations are done automatically with PalletTool PC offline development software to determine the best patterns for a layer based on your constraints, it would be interesting to see that logic pushed down to the controller level. Edited by OkiePC

Share this post


Link to post
Share on other sites
Thanks for all the info. I have been scouring the web for someone that does the Cartesian based systems that include stacking software, but so far I haven’t found anything. I hate to think I have to reinvent the wheel and write this stuff myself, but that looks like that is where I am headed for right now.

Share this post


Link to post
Share on other sites
All the big players have their packaged palletizers. I am only familiar with Fanuc, but highly impressed by what I see and hear on the web from ABB, then there's Kawasaki, Kuka, there are many others. Edited by OkiePC

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0