Advance Your Company Through Automation
During IPC APEX EXPO 2019, Yash Sutariya discusses the labor shortage he has experienced in the Detroit area, the impact automation can have in the manufacturing process, and other strategies to advance your company.
Patty Goldman: Yash, it’s good to see you again this year. Let’s start with a bit about yourself and your company.
Yash Sutariya: I run two companies in Detroit: Saturn Electronics, which is rigid PCB manufacturing, and Saturn Flex, which is rigid-flex. I’ve been in the business since I was 14, I believe. I grew up in the back of my father’s shop. I left after college and said I’m never coming back, but my mom made me come back. Since 2001, I’ve been in management. I saw the downfall of our entire industry, which is when I came back, so apparently it was my fault back in ‘01 (laughs). Since coming back, we’re bigger than we used to be, but instead of making boards for automotive, we’re in everything else, including production as well as prototype for what seems like the whole world.
Goldman: We were talking yesterday about automation and the labor shortage, and I found your perspective on that very interesting. I’d like you to share that with our readers.
Sutariya: Sure. Especially being in the Detroit area, when automotive gets a cold, we all get sick. We’ve never really had a big shortage of labor. With the rebound that started in late 2009 and early 2010, I’ve been expecting a recession since 2014 or 2015, and it’s getting stronger and stronger to the point where we never thought it would happen in Detroit, but we have a labor shortage. We can’t find people, and the people you do find should not have been found in the first place. There’s a reason there is chronic unemployment—because some people just aren’t made to work.
I read about Alex Stepinski’s work at Whelen [now GreenSource Fabrication] and he invited me to see what he’d done. It was a big inspiration—his vision of making circuit boards without people. I think the way he explains it now is, “Take out the labor but let’s also gain huge leaps in terms of the knowledge and data we can get from the process. Let’s learn what we can about materials and make everything more accurate.” That really inspired me, not necessarily from the data standpoint or a cutting costs standpoint, but because it focuses on consistently making product. I’ve had days where I couldn’t run lines because an operator decided not to show up, they quit, etc.
So, we started taking a look at automation to keep a process running regardless of available operators because my machines don’t have bad days. They don’t get sick, they don’t have sick relatives, and they’re not lazy. We have contacts in our work with Glory Faith, a large overseas manufacturer that we support, and they return that favor by supporting us at finding vendors and sources. We’ve been working for the past two years to identify overseas manufacturers of automation because, quite frankly, everybody’s gone here. You have some European options to choose from, and some domestic companies offer automation, but it’s the same Chinese stuff marked up by two to three times.
By having these resources, we’ve been able to establish contacts and do development work with automation manufacturers out of Asia. We brought in some units, put them by our worst processes like hot air leveling and etching to see how bad we could beat them up, gave feedback, made revisions, and then made different size loaders and unloaders. That’s where we see our future because as we’ve implemented one set after another, it has smoothed out production; mishandling has also gone down, and it’s put a little bit of fear into people at the factory so that they’re not as complacent. We’re taking it a step further. It’s a shame more of our industry doesn’t support IPC APEX EXPO because of what you can learn with the vendors on our side. We represent maybe 15% of the entire floor space.
Goldman: And if you’re talking about PCB companies, I’m saying even less than 15%. But if you’re talking about the suppliers, that adds a lot to it.
Sutariya: Yes, the suppliers. The lack of PCB fabricators exhibiting this year shows you the nearsightedness of our industry. At what other venue are you going to be able to get in front of 200–500 individual companies? And even if you get a 1% close rate, that’s probably more new customers than most companies in our industry have gotten in the past 12 months, and it costs you a few thousand dollars.
But let’s talk about our supply base; they still struggle to come out here, they’re doing their thing, and they might get five visitors, and that’s it because we fabricators don’t come out here. But if you did come out here, you’d see some new vendors. When it comes to my next focus—wet processes—everybody has done this and said, “We have materials. If you even have a loader, let it load, convey through, and catch it at the other side.” But how many people struggle with thin materials?
And I’m not talking one-mil cores; for some companies, even eight-mil cores get chewed up in their lines. Even if you’re the most rudimentary shop, you’re seeing layer counts go up for the most basic product. So, what do we do? We pull the materials—tape them to leaders, have one person in the front loading them one at a time, one person at the end unloading them one at a time and untaping them, and you’ve more than doubled your labor content for that process. What I’m looking to do is making my shop leaderless. Alex talked a lot about that, but he’s talking about 25-micron substrates. I’m talking about four- or five-mil cores. I need it dumbed down a little bit.
Historically, we had lots of vendors to choose from in the U.S. and Europe, but now we have two U.S. companies left and maybe one or two wet process manufacturers left in Europe. But in Asia, there are tons of suppliers with rough pricing of 50,000–60,000 RMB per meter. If you convert that out, that’s wet process lines for under $200,000 where you’d have to pay $500,000 stateside. For under $150,000–200,000, overseas companies are fully thin-core capable and often have advanced features like vacuum etching, ultra-thin core handling, and can customize the build to what you want.
There are so many things you can do to advance your shop by taking advantage of suppliers. Whenever I find a supplier in Asia, the first thing I do is introduce them to one of my vendors that sells or services equipment because now I’m creating my own supply chain. I don’t need the sales part, but I sure do need the service and parts supply chain. I get them factory trained here and say, “Go sell to everybody else,” because I need to have competition in this industry. We can’t survive if everybody goes out of business. If there aren’t more domestic options, there’s no point for us to exist because it’s all the more reason for buyers to go overseas if we can’t support competitively.
Goldman: It’s a critical mass thing. So, you’ve started with automation mainly for load-unload?
Sutariya: Yes, we started it with just simple load-unload—nothing complex—because anything that’s complex has a higher likelihood of breaking, such as a robotic arm, which people are pushing for $60,000. When you have a simple inclined loader and unloader—everybody’s probably thinking about the old Loehr & Herrmanns and the AMTECHs made back in the ‘90s—the new ones are actually quite nice; they’re delicate with the materials. They have double panel sensing, PLC controls where you can program dwell times, and exit speeds that vary once the material reaches a certain spot. They’re fantastic. In-house maintenance can address most issues if there are any.
The set I had for hot air leveling has been running eight months straight. I had two problems: one was because we installed it crooked, and the other was because an operator smashed the PLC screen with a panel when taking panels off manually instead of pulling the cart. We’re not used to having the ability to pull the cart out of the unloader because we haven’t had new automation, just what we picked up at various auctions. We started with that, then it’s getting into wet processes, which are easier and cheaper to fix. Then, your ROI is less than a year.
Goldman: And have you improved production throughput?
Sutariya: Throughput per man-hour doesn’t change because the throughput is dictated by the machine, but if I have to have an operator loading and unloading panels, not showing up, falling asleep at the wheel, and crunching up panels, it doesn’t do me any good; that means I’m not making product. This automation has enabled me to make product consistently. Now, as we get into wet processes, we’ll be taking more handling and labor out, so you’re talking the dual effect of lower cost and higher quality plus less scrap.
Goldman: So, that becomes part of your ROI.
Sutariya: Absolutely. I only count ROI on simple stuff. I don’t believe the models that companies give me with all this fluff in there to juice the ROI. Give me the hard costs, and I’ll take the fluff as a back-end benefit. I used to say I need three-year ROIs, but five years is fair. Five makes me a 20% ROI, and typically, equipment is outdated after five years in our industry anyway, so that’s a break even. Now, if I get some yield improvements, that’s additional ROI. So, if we’re bringing that 20% up to a 25% or 30%, now I’m talking serious dollars above and beyond what I could get in the bank, stock market, etc.
Goldman: You said wet processes are next. What’s your plan there? Are you thinking a DES line load-unload?
Sutariya: Right now, we’re working on metallization. We just had a two-hour meeting with Circuitech, and it’s their first time at the show. They’re great. We started talking last year, but it’s a little slower process for the Shadow LE—the new chemistry from MacDermid Alpha Electronics Solutions—so there are very specific design requirements. MacDermid has been great working with us, using their Asia counterparts to interface with the factory. We finalized it in our meeting, and we’re probably going to issue that order within the next few weeks; then, it’s about a six-month lead time until it gets here. We’ll check on the progress in three months. It’s looking good, and everybody is confident. Our schedule is Shadow, SES, DES, and then oxide; those are the biggest bang for our buck, as well as my oldest equipment.
Goldman: Are you planning on putting in conveyorized equipment and pulling out tanks?
Sutariya: No, we have all Hollmueller [conveyorized] right now. They’re gone, and we ran our stuff really hard. Even though the brochure said two mils, I don’t think it transported four mils without a leader from day one. Now, you have advancements, a knowledge base, interdigitated wheels, the wheel structures, the way you transfer from one module to the other, and the drying chambers, which have extra supports for the material so that they’re not flopping around and going to get wound up. That’s quite a few advancements since they’ve done mass production in China where they can’t leader anything. They’ve had a lot of resources to put into figuring that issue out, and having MacDermid come out here opens that door for us.
Once we go from that, we’ll start hitting the higher dollars. We already have higher speed drills coming in. Next, we’ll look at imaging. ROI is just not there yet for automated direct imaging, but it will get there. We’re a hair away from meeting our threshold.
Goldman: How about your drill area as far as automating?
Sutariya: That’s a tough one. I saw Alex’s idea about what he’s done to automate. It’s cool, but my footprint just doesn’t support it. That’s an area where we’re trying to get more throughput per square foot of floor area, which means faster drills and also getting smarter. We have fast technology software, but we don’t use it. We bought it and use some of it, but we don’t use it—not just tracking machine optimization, but cassetting and using resident tooling to the best of its ability. That minimizes my swap-out time between part numbers, so we’re starting the program to take a deeper look. We brought on a great process engineer over the past year, and apparently, I must have had more problems than I thought I did because he has been a very busy guy, but we’re only going to get smarter.
Goldman: Will you be able to collect more data and are you going to collect more data on these new, more automated processes? I know Alex always says data, data, data.
Sutariya: Alex is a lot smarter guy than I am, and he has a different end game. From my standpoint, I want to collect data, but it’s going to be more on the rudimentary level. Alex is looking at, “What’s the thickness difference of a piece of substrate every inch as I go from end to end at a very high level.” I’m still down in the nitty gritty where I just want to run my shop more efficiently. So, we’re going to look at some data. What’s my throughput? What’s my chemistry usage per panel? We’ll be tracking that because that adds to the ROI.
I’m treating these suppliers as partners. We’re taking a risk with them, and they’re probably taking a risk to launch with us in many cases. Thus, the more data that I can give them in terms of pure ROI, the easier the sell is. If they can get their equipment up into 30% ROI, they should be flat out busy for a long time if people are smart.
Goldman: Well, that’s all good information to hear.
Sutariya: Luckily, I went to business school. I wish I had gone to engineering school, but I guess it comes into play every once in a while.
Goldman: Is there anything else you would like to add? How many positions have you not had to refill since you started focusing on automation?
Sutariya: We just started, so we’re still evaluating the loaders and unloaders. We had a full-sized set brought in, and we’ve signed off on that. We also ordered another full-sized set, but along the way, we worked with them to design a half-size version because a lot of time, we just don’t have space. We may give up some of the capabilities, but it must be able to fit. That’s shipping right now. We’ll be evaluating that, and then it looks like we will have to place a bulk order—probably 40 units to outfit every process. But if I could reduce labor by one person out of every seven to 10 people, I’m in great shape.
Goldman: Perhaps another way to look at it is how much more production time do you think you can gain by not having “people gaps?”
Sutariya: Another way you could look at it is operating optimization. Like any other shop, many times, you get five panels to a process, run them, wait a half hour, another 20 come, and you run them. With automation, I can batch process because I can wait until I have a critical mass (e.g., 100, 200, or 500 panels)—something that will be worth my time to turn the machine on. Then, you stack them on a loader and unloader, kick it on, let it run, and let everything go into standby.
You’re talking about things like water consumption because the newer machines will have automatic solenoids that turn the water off. They will have smart heaters that can go into a standby mode, so you’re saving electricity and chemical consumption. Often, our old habits were to leave the line up and running because we keep coming back every 15–20 minutes with a few panels to run.
Goldman: You might need it for a hot job or something.
Sutariya: Exactly. Because I have all these smart people working for me, I don’t want people thinking anymore; let the machines do the walking. As I said, if I can get even a 10% labor reduction through automation, I guarantee I’ll match that with a productivity increase by whoever’s left, thinking “Maybe I don’t have a guaranteed job.” Not to sound rude or anything, but that’s just the way it is. And for the last three or four years, I honestly feel like a lot of my people have taken advantage of me because they know there’s a labor shortage. So, this helps bring the controls back where they should be.
Goldman: Interesting. Thank you very much. I appreciate it.
Sutariya: Thank you.