PCB007 Magazine

PCB007-Feb2022

Issue link: https://iconnect007.uberflip.com/i/1453746

Contents of this Issue

Navigation

Page 84 of 129

FEBRUARY 2022 I PCB007 MAGAZINE 85 everything. What are the simple controls? We get ideas from other industries. For instance, gravimetric measurement, or weight. Rather than chop up every panel to check the plat- ing thickness, there are three things that you can do. Every circuit board shop has them. You need a scale, a CMI, and a CMM. is is a great way to address legacy equip- ment that doesn't have a lot of controls. You can use some simple things in your factory to take measurements and not need a major upgrade where you only have one machine and it must be replaced, or you don't have space to put a second machine next to it. at's a big oppor- tunity. With a lot of the equipment, you can also get upgrades to get some better data collection to put into your database, and all this can be done with cloud computing. You just buy space as you need it and gradually scale up. e chal- lenge with many brownfield sites is they have fully capitalized equipment kits, much of which is not suited for Industry 4.0. You can just add on some features, so it is suited because you don't want to spend money on a brand-new machine for just a few added features. Now, if something is extremely old, you haven't maintained it, and everything is bro- ken, then replace it. But if the equipment's being maintained and you only need a few things, then just do the upgrades. at's the approach for a brownfield. You start generat- ing enough payback to do some new invest- ments. en you start to get the culture of adding sensors and collecting data. is is where we need to be in this industry. If you're a process engineer and you can't code, then it's proba- bly not for you. You need to take some classes, and if you're not liking it, it's not suited for you, and you need to hire somebody who can do it. One of the challenges is this "missing gener- ation" in our industry. ere are a lot of gray- beard folks who don't code, and a lot of young folks who do, but they don't have subject mat- ter expertise. And there are few people in the middle, so this is the thing that needs to be overcome. You need a team of subject matter experts, and you need the young folks so it's a little more effort than if you had good demograph- ics. In other places, it's the more experienced engineers who have both skills, so you must create this bridge. Nolan Johnson: It sounds like the goal is to go through a relatively simple implementation process in an existing factory to get the data needed to be more predictive, and use that to become even more efficient in the processes and reduce waste, thereby improving input and operational costs and giving you more control over what your input costs will look like. Is that correct? Stepinski: at is absolutely correct. at's the way forward in my opinion, because otherwise you're leaving a lot of money on the table. You can't outsource these things economi- cally, and here's why. A printed circuit board shop is not an engineered factory. A brownfield is not an engineered factory; it's an evolved factory. It's more like an animal. Over decades you added equipment, the varying vintages and suppliers, with different chemicals and procedures from one shop to another—all this evolved by trial and error mostly over time. A little bit of engineering, but mostly trial and error because they didn't have PLM sys- tems, or the tools do this in a different way. You need a team of subject matter experts, and you need the young folks so it's a little more effort than if you had good demographics.

Articles in this issue

Archives of this issue

view archives of PCB007 Magazine - PCB007-Feb2022