“Let’s go to the whiteboard.”
“Let’s go to the whiteboard.”
Setpoint was asked to retrofit a two up pick and place to a four up pick and place on a leak check machine. This would be no problem, but due to space constraints we were only allowed to use one gripper. One gripper gripping four pars is not trivial due to the fact that the world is not perfect and every thing is not exactly the same size and shape. We quickly made a prototype to see if this was even going to be possible.
When we got the grippers in we tested them on some parts given to us by the customer. The test showed potential. We did, however, see that some parts if rotated became loose. To solve this, we made some UHMW inserts that would give some compliance to the grippers. After bench testing the UHMW gripper inserts on the parts given to us it looked as if the grippers would work. The machine that we were retrofitting ran parts with three different heights. After retrofitting the machine and running it we found that for the two taller parts the grippers worked fine, but for the smaller one the gripper did not work.
To make the grippers work on the smaller parts Warren suggested using O-rings. Instead of using O-rings we went one step further and made the UHMW gripper inserts out of polyurethane. The polyurethane were more compliant and we could shape it to hug the whole part unlike an O-ring. This gave us our best result yet. The only problem with the polyurethane is that it would wear too fast and the gripper inserts were being replaced every three days. This was unacceptable so we went back to the drawing board.
We decided to make the gripper a three-point gripper. This helped, but then we were told that we could not use polyurethane any more due to its poor wear properties. We were so close to making the grippers work, but if we could not use polyurethane then this 4 up gripper tooling was not going to work. Then it just so happened that I found an O-ring that worked perfectly with the current configuration of our gripper. I quickly put them on and found they worked great. I knew that I was going get an “I told you so” from Warren… and I did. I took the grippers up to the customer and put them on and they ran fine. I guess the moral of all this bla bla bla is listen to your shop guys.
Saturday night I sat down on my nice futon in front of a good movie. Changing an inner tube can be a very mundane experience unless accompanied by a series of unexpected events. I had my road bike tire (flat), the new inner tube, tire tool, my pump, and I even had a great movie on. No hurry, I will just take my time and enjoy the movie while I’m working.
The inner tube I put in was one of those goop filled tubes. They build these inner tubes differently than the “Non-goop” innertubes, because there has to be a way to get the stuff into the tube. The presta valve has a removable core in it that is threaded into the valve stem.
So. . . I get this goop filled inner tube into the tire. I’m sitting on my futon with this tire in my lap occasionaly peering through the spokes of my wheel at my movie. Before I pump, I recall the advice Steve Nuetzman gave me to keep the stem up on top so the goop won’t leak out if air does escape. I turn the wheel until the valve is up and proceed to attach the pump clamp to the presta valve. At about 85 psi, the clamp starts to leak heavily – so I unclamped it, forced it on further and reclamped it. At about 95 psi the air began leaking so excessively that I knew something was wrong. The pump clamp was stuck and refused to disengage but was still bleeding air.
Everything happened quickly after this and became a blur, so I can’t remember the exact reason why – but somewhere in the process of trying to remove the pump clamp I rotated the wheel back down so the valve stem was at the bottom. I couldn’t get the clamp off with just one hand, so I leaned the wheel spokes against my forehead and employed both hands to remove this stubborn clamp. Well, it came off.
In the split second it took to blink – the valve core shot out with a loud blast. My forehead was still on the spokes, eyes down. That valve core shot out like a bullet and caught me square in the forehead. If it was only the valve core it wouldn’t have been a problem. But by now all the goop had worked its way back down to the bottom. Don’t let them lie to you when they tell you how much goop they put into the these tires. “It’s only about a tablespoon, barely enough to lightly coat the tube.” I am willing to bet money on about 1/2 cup. This yellow stuff came spewing out of the stem like the snorkel of an angry diver that just swalled salt water.
Trying to protect my face, I lost grip of the wheel and it fell to the carpet. It didn’t just fall, it rolled and danced like a quarter on a table top. The first sensation I remember is the taste. Classic Elemers school glue flavor. I looked down at the wheel on the floor, still slowly spitting this yellow goop onto my carpet with no remorse. I got up and walked to the bathroom to wipe off my face. It was all over, my eyes, my hair, my ears, my teeth. When I saw myself in the mirror, it was mixed emotions. Next, I saw the carpet. Glad my wife wasn’t home. Next I noticed the futon, and the ceiling, and the wall, the tv, and the speaker up in the CORNER of the room, all spotted with this evil stuff.
I woke up at about 2:30 am, and brushed my teeth again because every breath still hinted of rubber cement and Elmers school glue.
The scara robot Setpoint will be integrating into station 1 on the new MGG machine has a spindle (z axis) with an internal diameter of Ø11mm. All cables, tubes, and wires routed to this tooling must pass through the spindle. Specifically, the tooling must have one sensor cable, one Ø5/32 airline, and two Ø1/4 air lines – one for vacuum and one for exhausting air.
Obviously, this will not all fit through the spindle – we recognized this issue early in the design stage. This resulted in a drastic redesign to develop an end effector that can function under either 80 psi pressure or 10mmHg vacuum while being plumbed through the same port.
We have now designed, prototyped, and tested a configuration of the tooling that is fully functional and all wiring/tubing fits through the unforgiving 11mm spindle shaft. Central to the success of our first solution is vacuum generator M20A6-BN made by PIAB. All prototype components were printed on Setpoints 3D printer out of natural ABS plastic. Initial testing of this prototype unit is complete and ready to mount to the Scara robot. We are now refining the design further to improve upon those points of the design that required compromise during critical redevelopment phase.
Through the years here at Setpoint we have seen automation companies come and go, and here are some of the things that we have learned as an industry leader. Anybody who is looking for automation should know the following 10 things – but if you don’t, that’s okay; we’ll guide you through it.
1. Automation can be a practical alternative to overseas labor.
Automating processes can decrease labor costs associated with producing products by combining multiple steps into one compact machine, making it cost effective to remain at home.
2. Costs twice as much and takes twice as long as you think.
When automating processes, the inclination is to look at the big components like robots or electrical control systems, add up their costs, and then factor in some engineering hours. What is forgotten, however, are important items like light curtains, safety guarding, brackets, pneumatic valves, hoses, cables or rails to move the robot back and forth as well as the items that are required to make all components work together. Lead times on robots can be 16-18 weeks, with most lead times on major items averaging around 8 weeks. With the lead-time on components being so long, a project can take twice as long as you are anticipating it will.
3. Cheap and fast, or good and reliable: which one do you want?
If you don’t want to pay a lot and still get a Mercedes Benz product, chances are it will be broken and rusting away in a corner. If you want a fast solution, you’re rushing through the design and assembly, risking the chance of excluding some critical safety or quality measures that you need. The good and reliable solution is like getting the Mercedes Benz at the proper cost and in a timeframe that allows for proper lead times and a good design to be built.
4. Technologies are constantly changing.
Just because you bought something 5 years ago, and it works well, it doesn’t mean that it’s still the best technology on the market. Think about how many updates and upgrades there are with the typical computer; there is always some upgrade or new version that makes it function better. The same holds true for automation, products are improving and new technologies are constantly emerging.
5. It’s difficult for companies inexperienced in automation to articulate and visualize what is needed.
If you can’t articulate what you want, you don’t need it. For companies just starting down the automation path, there are a number of processes and capabilities that they do know, but hundreds more that they don’t. A good automation company can help suggest better options to make your process run smoothly.
Reveals flaws in processes and part consistency — inconsistent parts don’t work well with automation.
6. Processes and tolerances must be tightened up
When automating a process, tolerances, dimensions, and part accuracy are critical. Processes must be consistent. If not, the automated process won’t work and you will hate the machine.
7. Just because you can build your parts with a hammer and an anvil, it doesn’t mean the process can be automated.
Although we would love to be able to automate every process, sometimes the cost of the machine versus the payback makes it unwise.
8. Chances are that your automation vendor will be out of business within 2 years.
Most automation companies are small businesses that design machines one at a time and then pass on all intellectual property to the customer along with the machine. When purchasing machines, 60-65% of the price is for materials needed to build the machine. If the machine is not paid for until the end of the project the automation company runs the risk of paying for the machine and the design of it until it is complete. This practice, along with taking on machines that are risky for the company because they may not specialize or have experience in certain areas, can put companies out of business. Be prepared to put money down on a custom automation project.
9. Lean automation still means you need labor.
Automating processes does reduce your labor force, but does not eliminate it. Lean automation still needs someone to run the machine that has the ability to stop the production line if parts are being processed incorrectly or if an error is occurring.
10. A full automation effort will require maintenance and spare parts to sustain reliability and a high uptime.
All machines need to have routine preventative maintenance performed on them. Spare parts need to be maintained to replace those parts that wear out. Sensors and vision systems need to be adjusted and tightened back into place to maintain the reliability of the machine.
11. Automation will never call in sick, show up late, or create a sexual harassment lawsuit for your company.
This one just speaks for itself.
I know that was actually 11, but since we’ve automated most of our processes around here, I had to do something with my extra time.