���������ϲ�

When teaching setup reduction, one of the first principles I try to explain is related to what I call the “core tasks” required to make a setup. Core tasks are the physical tasks a setup person must perform in order to get the machine ready to run production.

For a typical CNC machining center, these tasks can include (among others):

• Tear down the previous setup and put everything away.
• Make the workholding setup.
• Determine program zero assignment values.
• Enter program zero assignment values (typically into fixture offsets).
• Assemble the cutting tools.
• Measure the cutting tools.
• Enter offsets for the cutting tools.
• Load the program.
• Run the first workpiece.
• Inspect the first workpiece and, if necessary, make adjustments.
• Repeat the last two steps until the workpiece passes inspection.

Of course, some of these tasks can be done off-line while the machine is running production for the previous job. For now, I’m concerned only with online tasks—tasks that add to the length of time that the machine is down between production runs. I’m only talking about the time it actually takes to perform these tasks; not the time it takes to get ready to perform them.

Almost all core tasks are performed at the machine. If you are timing your setup people while they perform core tasks, you only count time they spend at the machine during setup.

The important point is that you compare the time it takes to complete core tasks to the total time it takes to complete the setup. I define setup time as the total length of time a machine is down between production runs.

What is the difference between the core task time and the total setup time? For companies that have not addressed setup reduction issues, it is not uncommon for the total setup time to be several times greater than the time it takes to perform the core tasks. The initial goal in any setup reduction program must be to reduce total setup time to the time it takes to perform the core tasks. While this may not be completely achievable, most companies can make great improvements in this area.

What causes the difference? In my experience, the major difference stems from two problems: disorganization and poor program verification techniques.

While performing core tasks, you’ll probably find that your setup people spend a great deal of time away from the machine. What are they doing? They’re trying to find the components needed for the setup (hand tools, fixtures, cutting tools, inserts, holders, collets, extensions, gages, and so on). Can you eliminate these tasks from the time a machine is down between production runs? You can if you can justify moving them off-line. Someone (probably other than the setup person) can gather all components needed for upcoming setups, which is commonly referred to as kitting. If you can’t justify kitting components, you must strive to make all of the components needed for setups as accessible as possible. Minimize the time it takes to get ready to make setups.

By poor program verification techniques, I’m talking about verifying the correctness of the first workpiece. It is not uncommon, for example, that a setup person spends more time trying to get the first workpiece to pass inspection than completing the core tasks (including running the first workpiece). Some critical workpiece attributes have close tolerances and just aren’t coming out on size. The setup person makes the needed adjustment(s), runs another workpiece, and inspects the workpiece again. If this must be repeated several times, it can take a great deal of time, especially if the CNC cycle is long. This is not to mention the cost of the scrap workpieces being produced.

If your setup people are having a difficult time getting workpieces to pass inspection, you must do something to address the issue. They might just need more training in trial machining techniques. You could provide trial machining commands in the program to help them. (Trial machining has been the topic of previous CNC Tech Talk columns. See ���������ϲ�Online, or you can find information about trial machining on my Web site). Possibly the process being used to machine the workpieces is flawed. Whatever the reason, don’t give up until your setup people can consistently make the first workpiece being machined a good one.