Robot arm vs flip up part bin

[TAIWA NUMBA WAAN]

For live tool lathes and mill turn applications, I'm wondering about the utility of these two things. A robot arm will add about 50k to the price of the machine for a customer, another 30k for an auto part and material stocking drawer system, the kind of thing where you put the saw cut bars in the part stocker and the robot feeds them into the machine after unloading the old one into the part stocker.

I'm asking the more experienced lights-out production people here, what types of downsides do these have vs a basic part bin that pops up to catch during part-off?

 

[youngwerth]

Depends on a lot of things. Bar feeders and parts catchers make for great automation but they have limitations. For larger parts, the parts catcher might not have enough outfeed capacity to be very useful for unattended production. Lots of different types of parts catchers out there. Bar feeders are generally best for 2.5" diameters and less. "Part integration" time for a bar feeder (time it takes to get it working the first time) is generally fast for bar feeders. "Part changeover" time is generally under an hour for bar feeders

Robot arms can handle both small diameters and much larger diameters but are limited by the payload capacity of the robot. A dual-headed gripper (unload finished part, load next raw material without going back to infeed) will increase efficiency a lot. Lots of choices in infeed/outfeed with robots. Consider what you plan to run over the life of the system. If you only plan to run short parts in a "puck" configuration, a drawer or stacker infeed can be a good choice. If you need to run "shafts" or taller parts, a table infeed provides a lot more flexibility and can often achieve infeed densities equal to or greater than drawers or stackers by stacking parts in the infeed.

There are big differences between robot systems in terms of "part integration" time and "part changeover" time. On the mechanical side, what needs to be engineered to add a new part? Gripper fingers? Part locators? Those can add cost and time to part integration. Some systems require a lot of engineering for part integration, others require none.

Some robot systems are pre-programmed. Just fill out a form, save the configuration, validate the configuration, and you're ready to call up that job anytime. Typically less than an hour for part integration. In other systems, you are on your own and every part integrated is a new robot program. If you're a solid robot programmer starting from scratch, first part integration might take 1 to 5 days, subsequent parts less. If you're not a competent robot programmer... I come across many robots sitting in the corners of shops collecting dust.

For part changeovers, look at how much needs to be changed during a part changeover. Some systems only take a few minutes, and others can take a few hours up to a day or more.

In general, lots of small parts, with small diameters, bar feeders work well. With more variety in part sizes and shapes, a robot system will typically work better. Just make sure you know what you're getting into with the robot system and if it is really designed for the type of production you're doing.

 

[thunderskunk]

Citizen Swiss turn, copper barstock with 40 second cycle times and canola oil for cutting fluid: had conveyors that dropped into pans. We had systems retrofitted with robots off the conveyors which needed a lot of attention. I opted to have our cell use a gripper installed by citizen thinking it would alleviate issues with orientation…

It was awful. Constant leaks, chip wrap issues, dropped parts, etc. we should have never installed it. I never saw the robot work as the grippers had too many issues to start with.

I’m not saying grippers or robots are good or bad, but certainly application specific. If you’re cycle time centric and a few seconds makes a huge difference, maybe a robot isn’t the right choice, or at least one that opens the door. That was our case.

 

[alek95]

I run bar fed mill-turn centers with subspindles.

For the right parts, these machines are amazing.

The company I work for is in the process of developing a few new products, so I and my colleague are programming these parts. The machines were purchased to make these new products.

Some parts are very easy to automate. Others are a huge pain in the ass.

For example, some of the parts are made from tube material. These parts can not be automated because chips travel down the ID of the material and get deposited in the bar feeder, causing it to jam and error out.

Delrin parts are also some of the hardest to automate because chip control can be inconsistent. All it takes is for one drill to get wrapped in chips and break, and then you have to stop. The machines have air blast, which is actually a huge help.

But the idea is that once these products launch, production will be very low on labor.

 

[Orange Vise]

How about both?

You can have a machine with a barfeeder and parts catcher as well as an external robot. That's what we're doing to unload parts that don't fit the parts catcher.

 

[DanielG]

For example, some of the parts are made from tube material. These parts can not be automated because chips travel down the ID of the material and get deposited in the bar feeder, causing it to jam and error out.

Does plugging the end of the tube not solve the problem? Or do the chips in the tube cause recutting issues?