Custom Robotic Machining Cell
- brian7412
- Jul 9, 2021
- 3 min read
Updated: Jul 14, 2021
We had a job that had such a long cycle time that to meet the weekly quota, it needed to run 24/5 each week. This required a robot to load and unload parts even when we were lights out. We installed and integrated a robot to load and unload the parts. Our founder, Brian, handled this entire project, from design and some machining, assembly, programming, and finally integration.
It uses components from several companies that provide excellent products and support such as #VekTek #miteebite #AutomationDirect #YaskawaRobotics #Okuma

Integration wasn't as easy as you would think, there was a lot of issues to contend with they were:
Part access
The parts needed to be machined with the least amount of transfers as possible, as each transfer has the potential to introduce errors in the final product. this part also needed to be machined on all 6 sides. After trial and error, we found that we could do it in 3 operations using special hydraulic clamps from MiteeBite and Vektek on a custom-designed machining fixture.

Laying out the parts for clamp locations using MiteeBite OK vise clamps

Plumbing up the hydraulic body block pull cylinders from Vektek
Custom Hydraulics and controls
Once the hydraulics were plumbed up, we had to build our own hydraulic controls for the robot to control and to interface with the machining center.
One requirement was the ability to receive text messages should the robot or mill go down, should the coolant stop or a tool break, material run out, or any other issues. This required an ethernet-enabled PLC from Automation Direct with the ability to send emails. With this in mind, the PLC was integrated into the hydraulics control cabinet.


Integrating the controls and hydraulics
Custom Valve Block
After searching, it appeared that the type of hydraulic valve block we needed was unavailable, so we needed to design and machine our own.

Custom valve block machined by us.
Robot Programming
Brian had very limited experience with robot programming and had to learn to program the Motoman robot. He studied and learned the programming, interface, and operation of our robot. The program to machine these parts required over 300 individual motions, input/output switching, and timer commands. and was trimmed from a 6 minute cycle time to a 2 minute, 23 second time.

Brian programming the Motoman Robot
Robot gripper interference
Once the hydraulics, Robot mounting, first draft machining program, and PLC programming were completed, we needed to try out the load/unload cycle. During the cycle, it was determined that the original grippers needed to be modified due to the limitations of the robot's movement and collision potential with the fixture and parts. New Grippers were designed that required the relocation of the open and close ports of the gripper assembly. Custom air lines and fittings were made to integrate with the gripper and the robot end-effector mount.

Custom air fittings and gripper mount
CHIPS!
Finally, we were getting close, we began running parts, however after a few parts in, we discovered that chip buildup couldn't be handled using directed blasts of coolant during and after machining the parts. The chips would build up on the locating features of the fixture causing the parts to shift and be machined out of tolerance. After some research, it was decided that utilizing air blades directed to these locating fixtures along with a Chip fan at the end of the machining cycle gave us the best chance of success. Custom air blades were machined into the fixture and extras were added where they couldn't be integrated. Along with the air blades came the addition of 6 new air solenoids and control integrations.

#$%& CHIPS!!!

Custom adjustable air blades prior to deburring and installation
Success! and failure...
After a lot of work and tuning, we were getting good parts and everything was working as it should, and we began delivering parts when COVID hit. The manufacturer of the parts we were machining for requested a pause because their customers had reduced orders due to the pandemic. The lull in production still hasn't resumed. So for now, the fixture was unpinned and set on a shelf and the mill was upgraded to add a 4th axis as well as automated probing for in situ QC. The robot is in a waiting position to allow us to use the mill and everything is waiting for a large run project. If you have a part our machining cell can make for you, please contact us!
A video of the load/unload cycle can be found here:
Comments