What's new
What's new

Trying to figure out automation for machine shop

DLevo

Plastic
Joined
Jun 25, 2022
Hey community, been hearing a lot about automation recently to increase productivity, reduce costs, etc and was wondering how machine shops are going about implementing things related to it (cobots, workflow management software, etc.).
Where should someone start to figure out automation in their machine shop, is there any decent guide or something out there on how to find and evaluate solutions, vendors, financing etc.?
 

Orange Vise

Titanium
Joined
Feb 10, 2012
Location
California
Where should someone start to figure out automation in their machine shop

Right here on PM.

The world of automation is a minefield, and most folks' automation needs are shifting targets. Your first venture into automation should be something proven with a broad range of applications, i.e. NOT a heavily customized one-trick-pony that only one person in the world can fix.

The simplest forms of automation are barfeeders for lathes and factory (non-aftermarket) pallet pools for mills. These have zero customization.

Barfed lathes can be enhanced with 3-axis milling, subspindles, and parts catchers. I see these setups as the gold standard to which everything else should be compared. Dead reliable and very easy to troubleshoot if/when something goes wrong.

Pallet pools for mills are also very reliable. The automation doesn't care about what goes on top of your pallets. Its only concern is the zero-point interface between the pallet and the machine. Things can still go wrong, but pallet pool recovery is generally straightforward. Furthermore, most pallet pools are attached to multiaxis machines like 4-axis horizontals and 5-axis machines, so you're getting the benefit of reduced setups as well.

The next step in complexity is direct part handling, and it can range from slightly more complex to extremely complicated. A factory gantry loader for a lathe is about as simple as it gets in this realm but may lack versatility. 6-axis robots are more versatile but can be much more complex. We split the difference in some of our lathes; we use a barfeeder to load and cobot to unload when the parts are too large or delicate to get unloaded into the parts catcher, e.g. shafts. Robot unloading is significantly more forgiving than robot loading. Way less precision required to pull something out than to push something in ;).

As for robots and mills, the robots are generally very reliable but there are many more pieces to the puzzle, like chip handling, tool life management, coolant management, and workholding cleanliness. All of these need to be addressed before slapping a robot in front of the machine, and many machines simply aren't equipped to handle these challenges.
 
Last edited:

Comatose

Titanium
Joined
Feb 25, 2005
Location
Akron, OH
I'll just chime in to add this. "Machine shop" is pretty non-specific. Do you run 1 part per machine in quantity 10,000 per year, or 10,000 parts in quantity 1 per year? How big? How complicated? How precise? What materials?

You can automate any of that, but the right solution for one permutation of the above will not necessarily be the right solution for a different one.
 

DLevo

Plastic
Joined
Jun 25, 2022
Right here on PM.

The world of automation is a minefield, and most folks' automation needs are shifting targets. Your first venture into automation should be something proven with a broad range of applications, i.e. NOT a heavily customized one-trick-pony that only one person in the world can fix.

The simplest forms of automation are barfeeders for lathes and factory (non-aftermarket) pallet pools for mills. These have zero customization.

Barfed lathes can be enhanced with 3-axis milling, subspindles, and parts catchers. I see these setups as the gold standard to which everything else should be compared. Dead reliable and very easy to troubleshoot if/when something goes wrong.

Pallet pools for mills are also very reliable. The automation doesn't care about what goes on top of your pallets. Its only concern is the zero-point interface between the pallet and the machine. Things can still go wrong, but pallet pool recovery is generally straightforward. Furthermore, most pallet pools are attached to multiaxis machines like 4-axis horizontals and 5-axis machines, so you're getting the benefit of reduced setups as well.

The next step in complexity is direct part handling, and it can range from slightly more complex to extremely complicated. A factory gantry loader for a lathe is about as simple as it gets in this realm but may lack versatility. 6-axis robots are more versatile but can be much more complex. We split the difference in some of our lathes; we use a barfeeder to load and cobot to unload when the parts are too large or delicate to get unloaded into the parts catcher, e.g. shafts. Robot unloading is significantly more forgiving than robot loading. Way less precision required to pull something out than to push something in ;).

As for robots and mills, the robots are generally very reliable but there are many more pieces to the puzzle, like chip handling, tool life management, coolant management, and workholding cleanliness. All of these need to be addressed before slapping a robot in front of the machine, and many machines simply aren't equipped to handle these challenges.
Thanks for the response! This is really helpful.
I sense that there is a wide range of automation options available depending on cost/benefit, usage and other requirements.
PM is a great resource! Though I was hoping there would be some kind of a structured automation guide online that I could use to compare different options and equipment providers.
Moreover, along with equipment, also interested in affordable software to monitor and manage production workflow, inventory, quality control, data, etc., any recommendations on this would also be helpful.
 








 
Top