PCBA INDUSTRY – SMT PLACEMENT MACHINE ANALYSIS

Using an SMT Manufacturing Analytics Software to Maximize Your Component Placement Performance

February 24, 2022
Introduction
Component placement is arguably the most important part of the SMT assembly process. While machines get better and better every year, every SMT manufacturing factory still experiences some small errors from time to time: mispicks, not placed component, etc. Considering the high throughput of modern SMT placement machines even a small percentage of errors can add up to significant waste of time, material and money. One way to reduce this waste is to use an SMT-specific analytics software to identify the root cause of these problems. Is it due to the components or the machine, a specific feeder, a nozzle?
Factory intelligence analytics historical analysis
In this short article, we’ll cover two concrete examples that demonstrate how to best maximize your component placement performance using an in-depth SMT placement analytics software. In order words, we’ll explain how to use drill-down analyses to get best out of your SMT lines.

Problem #1: High Mispick Rate (Feeder Problem)

As we all know, mispicks are a significant issue for every factory. SMT Manufacturers are always exploring ways to eliminate them. In the example below, taken from the analytics software Factory Intelligence, you can see the line with the most mispicks is LINE_4.
cogican data analytics first pass yield
Therefore, we’re going to click on this SMT line to isolate it (or drill down on it). Upon drilling down, we notice that our biggest offending machine is PNP4. Again, we’re going to drill down on this machine. As indicated on the bottom left of the analytics software screen, we can quickly see that we have a big problem with one of our feeders since FEEDER_3 has almost double the amount of mispicks compared to the other feeders.
cogican data analytics first pass yield
What to do now? In this case, you need to seek out this feeder, calibrate it, lubricate it or maybe even remove it from the production line in order to reduce your mispicks as much as possible.

Problem #2: High Not Placed Component Rate

Another issue that SMT manufacturers face to an efficient manufacturing line is “Not Placed Components”. Since those end up needing to be manually placed by a technician, they waste a significant amount of precious time and resources. Using the same analytics software, we can see that 893 out of our 1,487 Not Placed Components happened on LINE_4.
cogican data analytics first pass yield
That tells us there’s likely a problem on this specific line’s placement machine. By drilling down on LINE_4, we see our top 5 Not Placed Components. In this specific case, COMP_PN12, COMP_PN7 and COM_PN4 are the biggest offenders for Not Placed Components (see below).

In fact, if we isolate these three components we are able to see that they account for 710 (out of 893) for this entire SMT production line!

Another interesting thing is that almost 80% of those Not Placed Components are associated with two nozzles – #707 and #608 – telling us that we might be using the wrong nozzle type OR that we need to optimize our placement program.

Conclusion

In conclusion, everyone understands that an analytics software will not solve all your problems by itself. However, it’s most certainly the best tool you can find to illuminate (or reveal) what’s hindering your SMT production in order to take the right actions to make it run more efficiently. As we like to say: it’s better to take actions on a small problem than to wait for it to get out of hand. Check out Factory Intelligence today!
You might also like