Koh Young: Getting Equipment In-Line and Customers Online

Koh Young: Getting Equipment In-Line and Customers Online

Feature Interview by Nolan Johnson

With machine installs somewhere in the range of 500 new annually and award-winning customer and field service, Koh Young has a commanding perspective on the evaluation, purchase, and installation processes for new equipment. We spoke with Mitchell Kim, applications engineer manager; Brent Fischthal, senior manager of Americas marketing and regional sales; and David Nemeth, Koh Young’s service manager for the U.S. and Canada, on what they see as the best practices for acquiring the right capital equipment.

Nolan Johnson: As experts in designing, manufacturing, selling, and servicing capital equipment for PCB manufacturing, what’s your best advice for customers? How should they move from the decision to purchase a piece of equipment to installing it?

Mitchell Kim: The biggest thing that customers need to oversee is having dedicated resources to implement these systems. People are shorthanded as is, but that kind of planning really benefits any system. You need to have qualified personnel and resources to be part of the implementation process, so that should get developed early on. From there, you move into internal screening and try to keep information flowing internally so that the process can be sustainable.

Johnson: On the application side, how often is Koh Young involved in guiding a customer through what they need to do?

Kim: Most of the time, the decision-making process is taking place on the front end. Afterward, there’s a follow-up on the process improvement side, because after the purchase, there’s a phase of installation and training that’s along the lines of, “Just how do I utilize the system?” From there you can move to the bigger picture; you take that training and you apply it from a process improvement standpoint.

David Nemeth: It’s fair to say that the applications team is very heavily involved in the pre-sales. After a trade show, the customer might want to see some action on their own product. Mitchell’s team will often receive boards and data, then use those to create board reports to show what our machine can do for the customer. That might be one of the first steps after a trade show. The customer wants to see something more on their product without doing a full-blown evaluation on site. Mitchell and his team will do quite a bit of that work, and then maybe do a virtual meeting to discuss how to utilize the equipment.

Brent Fischthal: Usually, before the show, they’ve done all their paper benchmarking and have spec’d all the machines; at the show, they’ve actually seen them all. The next step is to run some sort of product, either a benchmark test board or something similar. I’ll elaborate later about the boards we created and the benefit of those to actually running on their own product. It’s just a quick demo, maybe at one of our demo facilities, or something in Korea if it’s demanding enough, or even just something online like a webinar.

Johnson: The process starts with the customer realizing that they probably need some new equipment, so they start “paper benchmarking,” as you said. I assume that’s looking up datasheets, that sort of “seeing that what’s out there” for their application, the vendors, the models, etc. They figure out what they want and then see that equipment, which is where something like a trade show comes into play. What are some of the alternatives? If you know a prospective customer is looking at this mid-year and it’s still six to eight months until you get to a big show where they can see the “big iron,” what do they do then?

Kim: They can always reach out to us at any time. The doors in our office here in Atlanta are always open to host anybody who’s able to come. We have partner sites that have our equipment as well, so if logistically it makes more sense to go to Chicago, or the San Jose area, we can make that happen. We have equipment there and can have somebody onsite to support them there as well.

Johnson: So, customers could visit one of your demo centers in lieu of going to a show?

Fischthal: Yes, I think there are a couple aspects to what Mitchell is talking about here. Instead of the show, going to a Koh Young demo facility is absolutely an option. We’ve got them in Atlanta and in Guadalajara, plus many of our sales partners have a demo room with our gear onsite. You can go to one of our industry partners that has our gear in their facility, like Panasonic, Fuji, Universal, and ITW EAE. ASYS has gear as well. You can check out different machines at these facilities almost anytime. We have some equipment at universities like Binghamton University where we are focused on smart factory initiatives and at the Rochester Institute of Technology (or RIT) where we are helping the next generation of engineers. We even have gear at solder paste and materials suppliers too, so if needed we could take them into a materials company and show the machine there. Our customers have lots of options; they don’t need to wait for a show to see our machine, to touch it and feel it.

Johnson: All right, so they’ve had a chance to see the machines either on the show floor or onsite in an application environment, and the next step is, “Let’s see how it works.” Customers want to know how it works with their own projects and their own product. That’s where you start getting test cases from them, right? Let’s dive into that a little bit deeper, Mitchell. What makes for a successful test run and benchmark? What do you need to make that work?

Kim: We obviously need some samples and their respective files, everything to put a program together that will let the customer know, “Here’s what we should expect to need to move forward with this system.” On the front end, you need to do the necessary preparation. Oftentimes, it’s very beneficial to the customer to have a test plan where they say, “We want to run these specific tests,” and they provide their reasoning behind those decisions. It’s not just pulling things out of the air, but some sort of gauge, preferably something that they’ve created around whatever system it is that they’re looking for, because the gauge will vary depending on what product you’re looking at. Having somebody on the inside who’s familiar with that product is also very beneficial. That way, that person can oversee and understand what’s happening throughout the entire process.

Johnson: Is it a requirement for you to have somebody from the company onsite to assist with that, or is this something you can usually do on your own?

Kim: Most of the work takes place remotely. They’ll essentially send us their samples and files, and then we’ll set up a virtual meeting where we can do a live demo of the system. They aren’t physically seeing the system, but they are seeing what’s happening on the system. I’d say that’s at least 90% of the interaction at that stage.

Johnson: Getting customers to give you the pieces and parts you need, as well as a test plan or some objectives to accomplish, is pretty critical to deciding whether you’ve got what they want.

Kim: It also makes it fair, especially if they’re looking at other vendors. They can hold everybody to the same standards.

Johnson: How do you handle a situation (if you’ve even experienced this) where your customer really doesn’t quite know what they want? Say they’re a little unsure about their goals, but they’ve reached out to get a sample test done anyway. How do you guide them through that?

Kim: We always try to help educate the customer. For example, in the measurements world, there are certain standards used to gauge how you verify how reliable or repeatable a system is. Things like running a Gauge R&R, for example, as that can be performed on all the equipment, which makes things a little more binary—you know which system is performing better than the other systems.

Fischthal: There are a couple different options here. The customer brings their own board that they could use against multiple companies, or you could use something more objective, some sort of Gauge R&R test to make things a little more objective.

Johnson: What’s your overall advice for a prospective customer interested in working with you on this? We’ve just talked through what they should do here, but do you have a sense for how much pre-work this really is?

Fischthal: The amount of work they must do is tied directly to how dedicated they are to putting the machine through its paces. They could walk in, take a look, kick the tires, and feel like it’s good to go. Or they could really go through it and have their own evaluation criteria prepared, where they add weights and values to everything from user friendliness to accuracy and repeatability. Each of these aspects has the potential to affect their process, so what they want to accomplish determines how much work they will need to do. I think this goes back to what Mitchell said earlier: You’ve got to have somebody dedicated to this project. You can’t “share” it as a committee; you need one person driving it, and they must put in the work to really figure out how in-depth they want to go. They could do it in a couple of hours, or spend a couple of weeks deciding whether they want to bring a machine on site for a larger evaluation or whether they want to make an investment.

Johnson: Let’s say they’ve selected you and now it’s time to prep the system, get it delivered, uncrated, and installed. Is there a correlation between the amount of detail in the evaluation process and the ease of installation later?

Kim: From the programming aspect, we can provide a lot of the work we do during the evaluation directly to the customers, so all that data isn’t wasted and it puts them a little bit ahead of the game when they first get the system. So, yes, there’s some added benefit there.

Johnson: Is that true whether it’s done with test cases or Gauge R&R?

Kim: I would say so. With the test cases, they have those product programs and libraries already built, so some of that work is already done with any other gauge testing. It’s one way they can validate that the system they received is performing equally to the system that the original testing was conducted on. I think you can see benefits all around in that.

Johnson: How much time do you typically set aside to install a piece of Koh Young equipment?

Nemeth: The answer to that is: It depends. Is it a single machine install? Is it a first install? Is it a new customer? Are we adding to existing machines? Assuming it’s a new customer, the installation part is fairly straightforward, but again, it depends on the system. A single SPI purchase is quite simple; it’s three to four hours to actually install the machine,assuming it’s uncrated, has been moved to the location, and all the facilities are ready—we provide all that information up front to a customer. We like to give them facilities requirements and what they need for data for programming. Even if they didn’t go through that phase with Mitchell so they already know, I always provide that information: “This is what we need, and these are our network requirements. This is what you can expect when it comes to how we work with antivirus. Here are all those our policy documents and expectations.” All that is shared with the customer prior to getting the machine. Then we put in a good half of the first day after travel to install the system, and get it connected and talking to the network. Assuming everything is there and good to go, after that, we spend most of our time training and helping the customer get ready to utilize the equipment.

Johnson: What’s typical for post-install training?

Nemeth: For an SPI, it takes about two and a half days to go through the programming operations and basic maintenance, depending on how deep they want to dive into the SPI plus tool. We give them the whole overview of all those areas, but the majority of time is spent on programming and operation with the AOI. There’s quite a bit more to that, so a new AOI installation generally comes with a two-week training plan. For the first week, we give them the raw level-one programming course and operations so that they can walk away and be able to program and utilize the equipment. Their operators know how to use the review station, how to classify defects; their programmers know how to create a solid program. After that, we give them some time to really get to know the system and do some programming. They generate a lot of questions as they go about this; they will run into some difficulties, and they’ll have questions about things that we hadn’t covered yet, because there’s just so many options. That’s why we come back several weeks later, go through their questions, and help them dive deeper into the algorithms that we didn’t get into before. We give them the skills up front so they can drive it. When we come back later they can really put it around the track. That’s the plan. It’s about a two-week implementation for an AOI, but for most of our SPI customers, we’re done after that week and they’re good to go. They may have some additional questions later, but that’s generally the plan for those two platforms.

Johnson: It sounds like, especially for a new customer, that set of documentation you prepare for them—the pre-installed worksheets, checklists, and such—is extremely valuable for them.

Nemeth: It is, but it still generates a lot of questions. Sometimes it might be multiple machines and offline programming station, or multiple review stations. I want to have a meeting with the customer to ask, “What do you really want to do? How do you want to go about this?” You’re not going to just buy a brand new AOI and replace your existing one the same week you take the old one offline. Generally, you want to do some programming, build your library, get your people familiar and comfortable with the tools so that you can move it in line and then utilize it for production. SPI doesn’t normally require that; we can usually get that machine in line with them, get them comfortable with it. We can leave it. We put it in line, walk away, and they’re functional. AOI can be that way too, but it requires a lot more commitment from the customer. That’s where it comes down to having a champion, having someone or two people really focused on learning the system and taking ownership.

Johnson: Is it effective to have some of the staff start training on specifics, particularly AOI? If the customer wants to start training on that before it arrives, can they get training from you while they’re waiting for the machine to be delivered?

Nemeth: Sometimes we do that, but it’s not as common; if they want to come to us for a week or two of training, maybe. For the install, that’s possible. We’ve done that. We don’t want the time to be too long between when the machine arrives and when it’s installed, because it’s easy to forget things.

Nemeth: Sometimes customers will have the machine installed and they’ll come for training the week after. We’ve done some of that as well. They don’t want to do it on site because of the things that happen in manufacturing, where the people that are dedicated to it get pulled away. Training onsite can be a challenge, but the overwhelming majority of the training that we do is onsite, so we have to find a way to deal with those challenges.

Johnson: In that case, do you have a preference? Is it better overall for the customer if the training is done on at your facility vs. at the customer site?

Nemeth: I always think that it’s better for the technicians and engineers to be at one of our facilities so they’re able to retain what they learn and have the chance to really dive into the work without distractions. Normally, it’s not logistically possible to do that with most of our customers, though, so we have to deal with the challenges of onsite training. Some places are good at it; for other places it’s a real challenge. It’s a mixed bag, but most of the work is done onsite.

Sometimes customers want to come later for that second week of training, or they may have some additional training that they want to do with other programmers or technicians. In that case, we suggest that they come to us if they can; the training classes run every month, but we can only accommodate so many people. There are only so many classes in a year or in each quarter so, like I said, the majority is done onsite, but we like to do it in-house when we can.

Johnson: The ongoing in-house training gives your customers a chance to send staff later. If, for example, the person that received the equipment and the initial training got promoted or shifted to a different job or something like that, customers can still train somebody new by sending them to your course.

Nemeth: We train people all the time that move on or move up; it’s one of the biggest challenges. The number one request we get on a regular basis is about additional training for new people. We’d like them to come to the office for those classes when they can.

Johnson: Is training free?

Fischthal: At the end of the day, the customer invested in a relationship with Koh Young, so we want them to succeed. As Dave mentioned, when a customer buys an SPI or AOI, we will train them. We’ll be onsite for installation and training that first week, and then we will come back. We can provide a second session to answer questions that arise after using the equipment. Of course, they have access to our help desk whenever needed to answer questions. Additionally, we offer advanced training classes either onsite or in one of our locations. But if somebody just needs to get new hires brought up on board, then we like to bring in a trainer and have that be a paid class. The paid model ensures that they’ll engage with the material, that they have some skin in the game. If we just constantly give handouts, a free training that they know they can get anytime they need it, they don’t pay that much attention. We want to make sure they understand that there’s value in what we’re doing, and it costs us money to get a guy out there. We’re asking folks to please be invested in it.

Nemeth: This also highlights the need for there to be a real subject matter champion. You really want a few people onsite that are ready to own this process. If we just move people around and no one really cares that much about it, it becomes costly having to retrain established people or train new people.

Johnson: Finding a way to train a trainer at the customer facility is cost-effective for everyone.

Fischthal: Yes, the “train the trainer” approach is always helpful, especially if the customer is going to have multiple lines.

Originally published by I-Connect007 in the February 2023 SMT007 Magazine