Robots in metal finishing.
When you start looking at automation the robot topic will always come up. Some will tell you they work great and some will tell you they don't work at all so what is reality? The honest answer is that it depends so lets look at it just a bit.
If you have a high volume but also I high mixture it may not be the best option. Now if your mixture is parts that are similar such as truck wheels you can work around the issue. What robots thrive on is consistency. You set the cell up to process a part and you just keep feeding it parts and it just keeps going. If you need to shut down and change fixtures or grippers after 20 parts it may not make sense. Fixtures and grippers can be designed to work with a range but there are always limits. If you have a high mix you need to pay close attention to those topics.
You need to have a good part to start with to automate. That is not true and anyone who says that is just plain wrong. You can have crap parts as long as they are consistently crap. Back to the consistency point we go, the robot will not know that there is a random scratch. If that same scratch is present on every part you can plan for it and there is no issue. If you weld a part by hand you can still plan for the variance in the weld as you know in advance. You do not need to have "perfect" parts to automate. Better parts often mean less time to finish so its not a negative to have good parts but it is not required.
You can only automate some finishing tasks. This is also for the most part not true. From tasks like knocking down a manual weld to the fine polishing of luxury watches most things are possible. There are cells that put the brushed finish on a subway car so size is not a limit. Sanding operations be it on welds or CNC machined parts is not an issue so sanding is not a limit. Mirror polishing jewelry made of precious metals is done every day so polishing is not a limit.
What is a limit is mixing some tasks. If you need to grind a weld but also need to mirror polish that means two separate cells. The dust from grinding will contaminate the buffing operations no matter what. You can add all the extraction, air knives, brushes you want and it will still be a problem. The dust will be in very crack or crevice and as soon as you spin up a buffing wheel it all goes airborne. It will cause defects in the finish and you will struggle to figure out where they came from. If you are not looking for a high quality finish you may be able to skirt the issue but it will be there.
I own many of the same tools as a seasoned auto mechanic, but I do not have the same experience. Knowing where and how to use each of those tools is part of what makes a good mechanic. The same applies to robots and automation. Yes that guy might have robots but is there the experience behind it? Most times when automation fails its never one issue as is the case with most things. A very large part is right up front when you decide who to work with. They say a doctor practices medicine not that they perform medicine. Time is not the only indication to look for when you get into automation. Diversity in applications can be worth far more in experience. The challenges that came from working with very small parts, very large parts, parts that require many steps or processes and even very high speeds adds up. That experience and diversity only comes with doing it and the problems that are overcome.
You can automate near any finishing task but you need to work with someone that just tells it how it is. Its far better to be given the truth even if its not what you want to hear from the start. It is a situation where an ounce of prevention is worth 1000 pounds of cure. If you find yourself with a "Yes" man you need to run and fast. The devil lives in the details and he is just waiting for you there.