Communication, trust, process: The three pillars every co-design team must get right

Stephen V. Chavez (Siemens) (00:03.192): Hi everyone. Thanks for tuning in to the Printed Circuit Podcast where we discuss trends, challenges, and opportunities across the printed circuit engineering industry. I’m your host, Steph Chavez. Today we’ll address the evolution of PCB design methodology of concurrent design and co-design. I’m very excited to welcome today’s guests who are two very dear friends of mine and who are amazing engineers in today’s industry. We have Charlene McCauley, a senior PCB designer who is president and owner of McCauley Design Group, LLC. We also have Terrie Duffy, principal electrical engineer of AI Automation with Dell Technologies. Welcome to the podcast, ladies.

Stephen V. Chavez (Siemens) (00:51.682): Great, great, glad to have you here. So Charlene and Terry, could each of you give our audience a brief introduction of your background, experience, and education as senior engineers in PCB design, respectively? So Charlene, you want to start with you?

Charlene McCauley (01:03.23): OK, so I’m Charlene McCauley. I own McCauley Design Group. And I’ve been designing boards for going on about 45 years. I also have been a teacher at Austin Community College, where I taught PCB design and electrical electronic drafting.

Charlene McCauley (01:40.246): And so I’ve been around a while. I’ve been doing some conferences and podcasts. And I appreciate you inviting us here. So my name is Terrie Duffy. And my background was originally software. And then I took Charlene’s class at Austin Community College. And she hired me straight out of class.

Charlene McCauley (02:08.813): I’ve enjoyed it. I love board designs. Great. And so now I work, I worked for Charlene for nine years. Yeah. And then I just recently started working for Dell. So. And she’s teaching the class. yeah. And I’m teaching the class now at Austin Community College.

Stephen V. Chavez (Siemens) (02:08.813): Awesome.

Stephen V. Chavez (Siemens) (02:13.538): Yeah, that’s right. Yep, that’s right. You took over the awesome community college course that Charlene initially initiated. And I know because you’ve had myself and a few others as guest speakers into that course as well. So you guys are doing some great, great stuff there in awesome Texas. You know, I would tell you, I’ve seen you guys do, or you guys, I’ve seen you ladies do your presentation, you know, at multiple conferences and it’s, the content’s amazing and.

Charlene McCauley (02:24.55): Yes, and that’s been awesome. Thank you for coming to my class.

Stephen V. Chavez (Siemens) (02:41.702): and what we’ll get into today. So I’m really excited to talk about concurrent design. So let’s get right into it. On our topic today, which I’ll reiterate again, it’s concurrent design and co-design, specifically when it comes to the overall PCB design process. So first, let me start with a question. What do you think the people misunderstand most about concurrent design versus how it actually works in practice? So Charlene, I’ll start with you.

Charlene McCauley (03:07.142): Well, I understand it as being where multiple people work on a design together. Now, I have had that happen with me. I also worked at Dell for nine years. I was in the servers, and we were laying out huge planers for the servers. And we had a program called Petition to where we could petition people to work on this project at the same time.

Stephen V. Chavez (Siemens) (03:18.638): Mm-hmm.

Charlene McCauley (03:34.75): which is very helpful because you had a boundary, you had to kind of follow that. Somebody could be working at the same time. They all could be working at the same time. So that’s what I understand it as being. And also you can do it one-on-one, like maybe daytime, nighttime, where multiple people can work on a project at the same time.

Stephen V. Chavez (Siemens) (03:40.13): Mm-hmm.

Stephen V. Chavez (Siemens) (03:56.365): Mm-hmm.

Charlene McCauley (04:00.83): That’s kind of my take on it too, is you’re talking about multiple people working on the same design. Or like Charlene said, what is that called? Chasing the sun, where you have one person work on it during this day, and then they pass it off to the person working their day and stuff like that.

Stephen V. Chavez (Siemens) (04:01.218): Terrie, would like to add anything on there?

Stephen V. Chavez (Siemens) (04:16.46): Yeah

Stephen V. Chavez (Siemens) (04:22.082): Yeah. Yeah. Yeah. That also, you know, many refer to as like a staggered shift where you start someone early in the morning and then you have a midday shift and a night shift where the design is working around the clock. But, know, when we think about the misconception, you know, you know, it’s not just people working one at a time. I mean, you guys can attest to that. And I’ve seen some of the very, very complex designs you’ve done there at Dell over the last decade. And it’s

Stephen V. Chavez (Siemens) (04:54.082): It’s cutting edge stuff. Can you guys elaborate a little bit more when you think about the reality of how you were able to achieve success, even though there are times when we both have shared, you know, the true frustrations and sometimes the lack of information doesn’t flow quite steadily. And it’s like a hurry up and wait, hurry up and wait. then, or, you know, one step forward, two steps backwards, you know, with the reality and then, yeah, let me know, give me your take on that.

Charlene McCauley (05:10.206): Yes.

Charlene McCauley (05:19.25): Well, I mean, the major thing that has to happen is communication. You have to be able to communicate with the engineering, with manufacturing, if you’re able to do that. Like our presentation at the PCB conference is about the CAM2 and DDR5. So through that process, where we became a world standard for this connector,

Charlene McCauley (05:51.25): We were talking to manufacturers of the connectors, manufacturers of the DRAMs, engineers at Dell, and also, you know, I was always the liaison between the people that worked for me too. So communication, understanding what has to be done so you don’t have to back up, you know, and keeping that, and keeping your schematic connected with your board, you know all of the things have to link together for it to flow and you know multiple engineers too, so that is you know, they’ve got to keep their side concurrent as well. So and changes are big deal

Stephen V. Chavez (Siemens) (06:14.51): It’s gotta be in sync.

Stephen V. Chavez (Siemens) (06:30.504): yeah. Yeah, I think we, the three of us would agree that change is one thing that is constant throughout the process. Change is inevitable and how we respond to change and how we adapt to change, you know, it makes a difference as well as, you know, our ability to cope with that change and how do you adapt to it? How do you rebound to it and how do you respond?

Charlene McCauley (06:53.95): Because one little change can cause a board to be broken. If you don’t get it communicated properly and made properly, you’ve got to be really careful when you’re working with multiple people.

Stephen V. Chavez (Siemens) (07:10.638): Yeah, think, you know, and I’ve heard you guys, you know, state this before that the success depends on communication, trust and process discipline. Is there anything else that you want to elaborate on those three bullets? I know that those are three things that you talked about quite frequently at different, you know, different conferences regarding your success.

Charlene McCauley (07:32.444): Well, when we talk trust, like I have my own company doing this work for Dell. I have to trust the people like Terrie that worked for me. And it had to, you reach a point where I can trust what she turns in to the customer.

Stephen V. Chavez (Siemens) (07:52.587): Mm-hmm.

Charlene McCauley (07:53.596): That’s a very nice thing that can happen. But then you’ve got others that maybe come through. You’ve got to see what they can do. You’ve got to create a trust between you and the other designers and make sure that they’re on track with what we need to do. that takes a lot of reviewing, checking, making sure that person that’s designing work for you is following the process too.

Stephen V. Chavez (Siemens) (08:25.134): I would agree. I think the biggest thing is knowing the team players, understanding the stakeholders and the requirements. And then how do you convey manufacturing requirements or manufacturing within the pocket of manufacturing and you left shift that and move that into engineering. So you have like a manufacturing driven design. So that way your content that you produce is manufacturable, but as well as it meets your engineering needs.

Stephen V. Chavez (Siemens) (08:55.134): I think that’s the key. I would say that when it comes to concurrent design, in my perspective, I think I know you two share the same, is that the days are gone where you can do things in silos and you’re not doing things in serial anymore. Everything is done in parallel for the most part, especially with today’s complexity. And I’ve seen some of the complex stuff you guys produce and it is off the charts. It’s really high-end and you can’t deliver that kind of content.

Stephen V. Chavez (Siemens) (09:32.98): you know, the first time without having clear communication, you know, especially when you’re doing things in parallel. I think you guys would agree with me.

Charlene McCauley (09:32.98): yeah. There is one more thing too, is that the expertise. Like Terry has her programming mind and I’ve been a designer for a long time and it was very, I watched that, I wanted that. Her program mind, the PCB stuff can come as we learn the process, but having the expertise involved.

Stephen V. Chavez (Siemens) (09:38.19): Mm.

Stephen V. Chavez (Siemens) (09:50.734): Mm-hmm.

Stephen V. Chavez (Siemens) (09:55.394): Mm-hmm.

Charlene McCauley (10:00.542): and help get the design down the road faster and again, the communication between us to get that project done.

Stephen V. Chavez (Siemens) (10:11.394): Yeah, I think that’s a critical aspect that we think about is the skill set, understanding the skill set and having the expertise to complete the task that’s required. And this is why it’s important to make sure our key stakeholders understand what they’re up against and that we do have the right people doing the right job. And in some cases, it may not be so clear, but the fact that you’re communicating and I’ve seen you two work in tandem and it’s effective, and I think you guys definitely help set the baseline of how teams work together. Sometimes it’s not always pretty, but the fact is you’ve got to grind it out sometimes. And as designers, that’s what we do. I think Charlene, you mentioned about being the focal. I think that’s one of the key roles of a PCB designer today is you are the focal of all the disciplines when it comes to designing that circuit board. And you’re bringing all those specialties together.

Charlene McCauley (10:44.35): We communicated a lot.

Stephen V. Chavez (Siemens) (11:10.69): to eventually produce a circuit board that’s going to now get down in the lab and then you really do the magic of bringing it to life.

Charlene McCauley (11:20.926): Yeah, the trust and making sure, I forgot what I was gonna say. It’s also the stress points. There are stress points and you got that person that you’re trusting that works beside you.

Stephen V. Chavez (Siemens) (11:28.97): Stephen V. Chavez (Siemens) (11:37.294)

Charlene McCauley (11:43.004): You know, she’s got to understand or they have to understand that I’m going to be under a lot of stress. We are going to be on a lot of stress and understand that there could be some stress points that, you know, you might lose your cool a little bit, you know, because it’s and that’s what made it great for us to work together. You know, it’s we kind of make it fun to and so that helps.

Stephen V. Chavez (Siemens) (11:58.649): Sure,

Stephen V. Chavez (Siemens) (12:06.958): Sure. I mean, I think that’s the key. When you’re working on a team, have to understand the key here is in the end, you’re after the team result. And yes, you tried to have a nice cohesive culture within the team, but sometimes it does get stressful. And there are stress points, but nothing’s personal. Everyone’s trying to do the best they can for the intent of the overall team. And I think

Stephen V. Chavez (Siemens) (12:46.958): as engineers, especially as we get seasoned, you get a little like thick skin to where, you know, you don’t take it personal. It’s just, you know, all of us are alphas and we’re trying to deliver the best we can. yeah, and in the end, it’s a team effort. You know, with that said, you know, when I think about teams and let me move on to organizations. So when we think about in the real organizations, whether it’s SMB or enterprise level, where does concurrent design typically break down?

Stephen V. Chavez (Siemens) (13:26.958): where do you guesstimate where it usually goes south? It doesn’t flow or doesn’t go right for them.

Charlene McCauley (13:30.514): Well, it’s all those things, all those things that we just talked about. Making sure my team’s all on track and following the steps that we need to follow. Because if I got somebody that’s kind of going off in some other direction and not really following what the engineer is asking us to do, that’s going to cause issues.

Charlene McCauley (14:00.514): And that’s on my part too. I I’ve got to make sure I can trust what’s happening because that’s where it breaks down. If they’re going off on another track, we’re stopped almost if something is not going the right direction. And that’s where the concurrent design, there has to be a watchful eye on what’s happening.

Charlene McCauley (14:29.774): unless there’s so much trust like what I had with Terry, you know, I know it’s okay, but then we know, we both know when it’s time for us to both to look at what we’re doing. I mean, her look at what I’m doing, you know, and kind of create a partnership to where we trust each other and have another eye on the project.

Stephen V. Chavez (Siemens) (14:29.774): Mm.

Charlene McCauley (14:49.214): Sometimes she would even send me a zoom and she says, okay, I’m doing this. Watching from the other side as she’s doing something. She would watch me on the zoom and watch me make the change. And I’ll say, this is how we’re, you know, wait, you missed one. So, it’s, uh, you know, that’s doing, going down another path, not following the directions of the engineering.

Stephen V. Chavez (Siemens) (15:03.245): Hmm.

Stephen V. Chavez (Siemens) (15:07.927): Hahaha

Charlene McCauley (15:18.494): what manufacturing needs, you know, that takes it, that just stops the project really when you’re not following the right path for each other, you know, for the project.

Stephen V. Chavez (Siemens) (15:26.979): Yeah.

Charlene McCauley (15:32.284): Yeah, and I see that like version control who has the master copy, you know, where does the master copy sit at all times because you’re going to duplicate effort. You’re going to waste hours. You’re going to lose a lot of productivity if you don’t know, you know, where the master is and where to start. Because, yeah, I’ve taken one off the site and started editing it. And they’re like, wait, that wasn’t the most current.

Stephen V. Chavez (Siemens) (15:38.094): Mm.

Stephen V. Chavez (Siemens) (15:38.094): Yes.

Charlene McCauley (16:02.208): We figured out a good way to, I think we’re gonna talk about version control, but we figured out a good way to kind of eliminate issues with that. If you’re a small company and you don’t wanna spend lots of money on a version control software or something.

Stephen V. Chavez (Siemens) (16:02.488): Been there, done that, yep.

Stephen V. Chavez (Siemens) (16:12.014): Mm-hmm.

Stephen V. Chavez (Siemens) (16:19.426): Yeah. I mean, when you think about version control right away, that jumps to mind is like PLM system or your product life management system. How are you going to control your data? How are you going to lock it down? Using servers, you know, small companies I’ve been there, you know, I use a server and somehow or another the wrong version gets updated or worse, the wrong version gets gets fabricated and if things get missed because control who’s controlling the data or the master has been deleted from the server and nobody knows what happened.

Stephen V. Chavez (Siemens) (16:55.006): So I think that’s a strong case of implementing a good PLM system into your organization, regardless of whether you’re… Yeah.

Charlene McCauley (16:55.006): And it’s a big deal getting boards made to, you know, making sure they have the right files when they’re getting ready to make the boards.

Stephen V. Chavez (Siemens) (17:02.508): Yeah. Yeah, when you think about the main problems I see are inconsistent design details. Do you understand all your requirements? Are all the requirements captured? And you think about design constraints. Did you get all the constraints in? Are they designed appropriately? Did you communicate with your fabricator upfront?

Stephen V. Chavez (Siemens) (17:42.508): and get your stack up locked in so when you have your design constraints all locked down, everything’s in sync and you can’t have that without communication. This is why, as you mentioned, we think about just summarizing the requires, strong upfront planning, ongoing reviews throughout the different phase gates. And Charlene, you mentioned that you’re the focal between the two of you. Or when you’re on the task is that you’re kind of the oversight. So you need leadership oversight.

Stephen V. Chavez (Siemens) (18:07.324): not just within your realm of the PCBs, but also the project lead and the project managers need to be all in sync so that way there are no phase gates missed.

Stephen V. Chavez (Siemens) (18:07.324): So that leads me to, go ahead. Oh yeah, I mean, we all know sometimes, God bless the double Es, but sometimes they will deviate and they’ll wanna try something out or they’ll wanna add a parallel circuit to the design because they’re not sure which one is gonna work. And now it’s our job to try to squeeze it all in there and it’s a challenge. It’s a challenge indeed, so.

Charlene McCauley (18:07.324): Yeah, and even keeping the engineer on track too.

Charlene McCauley (18:07.324): Yeah.

Charlene McCauley (18:29.982): And if they ask you to make something that doesn’t make sense, like I give you a change and you know, call them, say, hey, this is exactly what you want, right? Don’t waste your time doing all the changes before you verify. They always think in black and white, we’re looking at color and multiple layers. And we got to kind of get them on track to say, okay, this is what this looks like. Are you sure you want to do this?

Stephen V. Chavez (Siemens) (18:55.918): Yeah, no, you’re exactly right. It’s even worse, I think, when you have a young engineer or less experienced, I’ll just say that, less experienced engineer who gives you a database back and he says, I went ahead and did the route, some routing and implemented the change.

Charlene McCauley (19:09.822): It’s not always the young ones that do that either. Yeah, I always tell them to stay in your place. If you’re going to do something, don’t go over there in that other part of the board. Stay right there in your corner.

Stephen V. Chavez (Siemens) (19:14.094): It’s always scary.

Stephen V. Chavez (Siemens) (19:21.814): Yeah. Yeah, so that leads me to our next question. So how do teams actually divide up a PCB so that multiple designers can work on it at once? When you think about whether you partition it or depending on whatever tool you use, how are you going to partition the work up? And in some cases, I know in Xpedition, we can have multiple designers attacking the design simultaneously, whether you’re in the front or the back end.

Stephen V. Chavez (Siemens) (20:01.814): I think your guys’ ecosystem, you can do the same. It just depends on the ecosystem. And I also believe that it’s the culture. You got to buy into that kind of work. And if not, I don’t care how good your tool is, if you don’t have the mindset that that’s the way you’re going to work, it’s going to fail every time. So I’m curious, what is your take on that? How do you see the teams actually divide up to work?

Charlene McCauley (20:12.028): Well, OK. OK, so like I said, when I was at Dell, we did have the partitioning software where four or five designers could work on the design. But you always had a dividing line. So you had to kind of figure out what can be done in different areas of the board. But as a small company,

Stephen V. Chavez (Siemens) (20:20.696): Mm.

Stephen V. Chavez (Siemens) (20:33.325): Mm-hmm.

Charlene McCauley (20:36.484): I don’t have the luxury of getting a partition software. So then we have to look in because the software we use, you have to be able to clip in things. So you have to kind of divide the board up in ways that you can easily clip it together. So you’ve got multiple things happening on a planer or a motherboard.

Stephen V. Chavez (Siemens) (20:40.302): Sure.

Stephen V. Chavez (Siemens) (20:47.903): Mm. Mm.

Stephen V. Chavez (Siemens) (21:03.893): Mm-hmm.

Charlene McCauley (21:04.284): like Dell’s motherboards, you’ve got, you’re thinking long ways, not just corners. And so you got to think about where you’re routing, what layers you’re routing on. Can we easily clip in what, maybe what Terry is working on, onto my board. And then we also figure out who’s the master.

Stephen V. Chavez (Siemens) (21:09.997): Mm.

Charlene McCauley (21:24.978): who’s going to have the master design, and everybody else will submit their portions of the board. And you have to do it at least maybe in the middle of the day or in the end of the day. But you’ve got to keep track of what’s happening when you don’t have something like a partitioning software. Right.

Stephen V. Chavez (Siemens) (21:25.006): Mm.

Stephen V. Chavez (Siemens) (21:41.507): Mm-hmm.

Charlene McCauley (21:46.024): So if I finished a section, I would say, hey, it’s ready to clip. And then you clip it in, make sure the master is up to date. And then I would take a new copy of the board and start something else. And it does help to do it in different shifts because once we get something clipped in, then we can kind of look at it again and figure out maybe we move on to somewhere else.

Stephen V. Chavez (Siemens) (21:49.902): Mm.

Stephen V. Chavez (Siemens) (22:05.39): Mm-hmm.

Charlene McCauley (22:09.266): But again, it’s about communication, trust, and everything. You’ve got to stay on track with what is happening on the design. And we’re talking motherboards here that we’ve been working on that you’ve got 12, 14 layers and then lots of high speed lines.

Stephen V. Chavez (Siemens) (22:22.572): Yeah, no.

Stephen V. Chavez (Siemens) (22:28.429): Mm-hmm.

Charlene McCauley (22:32.19): and lines to connectors. So there’s a lot of areas that you can share, but you got to be able to work it in into the design.

Stephen V. Chavez (Siemens) (22:38.157): Mm-hmm.

Stephen V. Chavez (Siemens) (22:43.832): Yeah, I think the key here is, you you have it cleanly partitioned on who’s going to do what and when, and how you’re going to implement whatever changes. And if something is going to affect another team member’s area, I think it needs to be quickly communicated. And, you know, from my experience within my ecosystem in the past, you know, I’ve been able to do this all on the fly. I’ve had three designers, you know, in one, one design, one shift, and then

Stephen V. Chavez (Siemens) (23:33.832): When I hand off in the evening, so everyone works eight hours, I hand it off to another design team in another region of the world, and they’ll attack it, three designers. So in one shift, if you think about like one day shift, sun up to sun down, and then sun up again, I’ve just produced 48 hours of man effort in one physical day. And when I hear people say, oh, it doesn’t work, well, I’ve got almost two decades of using that format, and it’s worked.

Stephen V. Chavez (Siemens) (24:13.832): very effectively for me. So, and I know that you two have done the same, you know, in that level of success that you guys have achieved. And, but don’t get me wrong, I’ve done my share of, you you know, in a serial format where I worked on early shift or late shift or night shift and handing it off. But in the end, you know, you know, it requires communication, communication, communication, constant communication. You know, when you think about the benefits, which enables you know, parallel work on the fly, know, and the risks, you know, merging complexity, it’s always a challenge. then inconsistency between sections, you know, like because each of us have our own artistic characteristic of being designers, we each have our own flair of how we design boards. And you could see it in the work. And I think you guys have I’ve heard you compliment or I’ve heard you state that in some of your conferences where you could see who does what. You know what I mean?

Charlene McCauley (24:45.786): And I can tell if somebody sticks it in an autorouter too.

Stephen V. Chavez (Siemens) (24:49.901): Yeah.

Stephen V. Chavez (Siemens) (24:49.901): Yes, yes, exactly. I mean, there is a difference between automation and just using automation without any control or whatsoever. And you get what you get versus an interactive approach where you are controlling the automation, you’re controlling the interactive routing. In the end, I today’s evolution of when you say manual, it’s not actual manual. I think you guys agree with me. It’s all constraint driven.

Stephen V. Chavez (Siemens) (25:27.334): you’re controlling the automation of your routing. You’re not just freely just using the automation and the tool clicking the button.

Charlene McCauley (25:27.334): You know, that’s a good way to explain it is, you you think of anybody that has worked with an autorouter, they know what it does. It’s on its own path.

Stephen V. Chavez (Siemens) (25:33.859): Mm-hmm.

Charlene McCauley (25:37.412): It follows the net list. It goes its own way. Well, that can be a person too that goes their own way. And so you can kind of do similarities between an autorouter and a new person. You’ve never seen their work. And they’re not as educated or have the years in. They’ll be like an autorouter. They’ll just go off in their own way. And that’s why you’ve got to really watch what’s happening on the design.

Stephen V. Chavez (Siemens) (25:38.786): Ahem.

Stephen V. Chavez (Siemens) (25:44.322): Yeah.

Stephen V. Chavez (Siemens) (25:57.047): Hmm.

Stephen V. Chavez (Siemens) (26:08.216): Yeah, I think oversight is the key word. You use that a couple of times already in this podcast. You gotta have oversight and especially if you’re working with an unknown variable or a team member you haven’t used before or even someone who’s new to the field and are fairly new and you don’t quite have the experience yet. And so you gotta definitely keep them, you know, like bumper rails. You gotta keep them down the path, which is, what constraints do constraints help us keep us down that path. But

Stephen V. Chavez (Siemens) (26:48.216): The constraints are only as good as the human who input those constraints, and how well they relate back to the design requirements. So with that said, let’s move on to the next question I have. So why is upfront setup and planning so critical when multiple designers are working on the same PCB?

Charlene McCauley (27:11.6): One thing that Charlene mentioned, I’ll let you talk about it, but go ahead, was getting the board set up from the beginning to be able to Gerber out. You want to start with all your constraints and your views and all your setups correct before you even start. Because if you don’t have that, there’s no need to partition it really and try to do anything with it because you’re just going to have to go change it all again.

Stephen V. Chavez (Siemens) (27:37.026): Mm.

Stephen V. Chavez (Siemens) (27:37.026): Yeah.

Charlene McCauley (27:40.072): Well, like she’s saying, we had some boards that were similar and we always had to, you know, Gerber out. So I would go in and set it up, get the Gerber set up, you know, and be able to use the design we’ve done before. And you eliminate a lot of distress at the end of the design. When you’re, they say, we got to push the button, we got to send it out.

Stephen V. Chavez (Siemens) (27:55.298): Mm.

Charlene McCauley (28:05.118): and get boards made. And that’s where the stress point comes. OK, are we set up? Am I going to have to spend a couple of days so I can get it set up to Gerber out? So I try to do as much as I can up front on the design and make little changes in the end. But the stackup is the big thing.

Stephen V. Chavez (Siemens) (28:15.822): Mm.

Stephen V. Chavez (Siemens) (28:26.211): Mm.

Charlene McCauley (28:29.654): But a lot of times our projects, we didn’t have the proper stackup. We had it because we did a lot of feasibility stuff and the stackup would constantly be changing while we’re working on the design, which is another problem sometimes when you’re routing in stackup, you have to add two more layers or the line width changes, the space changes, then you got to back up again.

Stephen V. Chavez (Siemens) (28:33.794): Mm-hmm.

Stephen V. Chavez (Siemens) (28:52.493): Mm-hmm.

Charlene McCauley (28:56.674): and make sure everything’s

Stephen V. Chavez (Siemens) (29:01.474): Yeah, that’s where you’re banging your head against the desk and like what? You want me to add what 120 on canvas? What you want me to add? You know, DD at what? You’re smacking your head. You want me to increase the trace list?

Charlene McCauley (29:04.286): Mm.

Charlene McCauley (29:09.17): you

Charlene McCauley (29:14.204): But even with all that, I look at it like we’re making something happen here. These changes are going to get us to that end point.

Charlene McCauley (29:28.118): And so that’s what’s amazing. It’s a lot of stress to make changes, but it’s all going to lead to something amazing in the end. And you just got to keep your cool. I mean, have lost my cool before. It comes with the job. Well, and feasibility, it is a constantly changing target.

Stephen V. Chavez (Siemens) (29:37.966): Sure.

Charlene McCauley (29:52.254): So, I mean, when we were helping with the cam stuff, I mean, that even just the design of the connector changed. I can’t even remember how many times that changed. Different sizes, different pins, different… Different pin placements, yeah. The way the pins go. I mean, we’re just changing it almost weekly. But you still have to route the whole board to Yeah, because we gotta see if it’ll route, yeah. So, yeah, was… It was interesting.

Stephen V. Chavez (Siemens) (29:52.366): Mm.

Stephen V. Chavez (Siemens) (30:20.11): You know that that’s

Charlene McCauley (30:21.564): But in the end, we have a connector that’s a world standard, so like our cam connector.

Stephen V. Chavez (Siemens) (30:26.018): Yeah, I would tell you, that leads me into the next question. And I would discuss a little bit before, but why does version control become such a major pinpoint in concurrent design? mean, we kind of danced around it, why is that? Why is version control become such a major pinpoint from concurrent design? Charlene, I’ll let you.

Charlene McCauley (30:46.59): You want to start? Sure. I just know if you’ve been burned by starting with the wrong one and then you put a whole day’s worth of work in and it was worthless because you didn’t get the right version. Yeah, you mean you get way more concerned about what version you’re getting. And you have to.

Stephen V. Chavez (Siemens) (30:54.67): Hmm.

Charlene McCauley (31:10.044): I mean, you have to have a convention. Like Charlene has her own naming convention. And we have the key areas where we store the master and keep it identical to the schematic. And you’ve got to have your steps in place. But even with your steps in place, we still occasionally get the wrong one. Or the name doesn’t get changed. Or the engineer doesn’t put the right schematic back on the site. So.

Stephen V. Chavez (Siemens) (31:15.021): Hmm.

Stephen V. Chavez (Siemens) (31:32.652): Hmm.

Stephen V. Chavez (Siemens) (31:38.296): Sure.

Charlene McCauley (31:39.054): It just can make a huge mess if you are not starting from the correct board. You’ve got to watch when an engineer will send you a schematic and he says, okay, here’s the new schematic. You load it up on your board and everything disappears. know, or things start popping off the board and going,

Stephen V. Chavez (Siemens) (31:56.417): Yeah.

Charlene McCauley (32:00.126): I don’t think you had the right revision on that. So we were constantly watching what the engineers did, what we’re doing, and we devised a way where you had the date of the project, because it’s changing daily, and it’s changing hourly.

Charlene McCauley (32:21.026): So we would use a date, we would use a letter like A, B, C, D, E, F, G, and then the person’s initials on the project.

Stephen V. Chavez (Siemens) (32:33.806): Mm.

Charlene McCauley (32:49.035): Because a lot of times we’re clipping back and forth. So everybody’s keeping their revision forward, going forward. But if we reach a point where something’s not working and we got to back up, we can go back to A that day, or B, or wherever we need to go. So it’s the problem with having a small business.

Stephen V. Chavez (Siemens) (32:49.035): Mm-hmm.

Charlene McCauley (32:55.122): We don’t have the luxury of some of the big softwares that’s out there. I don’t have the luxury to buy those type of software. But we did pretty good. We probably had a glitch here and there. it really flowed really well. And it’s just keeping track, making sure who has what, who’s the master again, making sure who’s the master.

Stephen V. Chavez (Siemens) (33:00.192): Mm-hmm.

Charlene McCauley (33:20.838): and who’s working on what. Even with the revision software, you can still get burned. mean, nothing’s perfect. So you still have to keep a little more vigilance on am I getting the right software? Am I starting at the right board? Am I got the right engineering changes that need to be made? But the nice thing with boards, can.

Stephen V. Chavez (Siemens) (33:40.014): Mm-hmm.

Charlene McCauley (33:43.814): You can fix a broken, something that broke. You can pretty much get it in there quickly. That’s what I tell my students. Don’t worry about it. Just get it in there and we can make it work. We can fix it. We can fix it. We’ll it out.

Stephen V. Chavez (Siemens) (33:46.412): He he.

Stephen V. Chavez (Siemens) (33:57.475): Yeah, I think that leads us into our next question of what workflows help you avoid endless iteration loops and rework when co-designing?

Stephen V. Chavez (Siemens) (34:25.475): So when you think about your workflows, because you talked about workflows and why, and sometimes you have to take a step backwards or two step back, or as a Marine, I don’t say retreat, but instead you just fall back to a better fighting position. So you think about sometimes you got to go back to that previous revision and go to in a different direction because it did the first, that path you were on, it turns out it’s not the right path. You could just see it down the road that, hey, this isn’t going to work.

Charlene McCauley (34:40.83): Well, it’s, you know, the fact that we worked on feasibility, you are going forward and backwards all the time. I mean, it’s a constant, go a little bit forward, see if it works. If it doesn’t work, you gotta go back. So whenever we always started something, a new test, we made sure we had a file that said, this is before the test one.

Stephen V. Chavez (Siemens) (34:50.402): Mm-hmm.

Stephen V. Chavez (Siemens) (35:05.388): Mm-hmm.

Stephen V. Chavez (Siemens) (35:10.008): Yeah

Charlene McCauley (35:10.574): test one is here and we go off and we do it and this is what we have. Then we go back and say, okay, let’s try test two and see if that works. It’s like a scientist. know, we, had options, C directory, option D directory. So, know, sometimes because you may need to go back to that test one and go a little further down the road. So we were constantly testing out all different scenarios.

Stephen V. Chavez (Siemens) (35:26.638): man, my goodness.

Stephen V. Chavez (Siemens) (35:32.792): Sure.

Charlene McCauley (35:40.4): for the CAM2 boards that we did. And so, you never say it’s not gonna work. You never say that this isn’t working, so let’s go over here. But I always have to know you might have to go back to that other design and try to see if that’s gonna work. So it’s a…

Stephen V. Chavez (Siemens) (35:45.23): Mm.

Stephen V. Chavez (Siemens) (36:03.022): Sure.

Charlene McCauley (36:09.086): It’s a really just think of yourself as a scientist. Think yourself of you’re trying to get down this path and seeing if you can make it work and just be open to that. have revisions where letters starts with A. I think I got all the way to X in one day. I was like.

Stephen V. Chavez (Siemens) (36:28.909): My goodness.

Charlene McCauley (36:31.024): I’m gonna save this one. I’m gonna save this one. I’ve gone AA, after AA, AA, AA, and AB, AC. my goodness. You know, it’s, I’ve gone past X before.

Stephen V. Chavez (Siemens) (36:39.138): Mm.

Stephen V. Chavez (Siemens) (36:45.314): Yeah, you know, early in my career when I worked at a small service service bureau or even small companies, mean, we didn’t have the digital thread or the digital twin, you know, per se, as we do now. And it’s a whole other ballgame now within the ecosystem. You know, with my former company I was at, we were very effective because we had the digital thread, the digital twin of how we communicated on the fly. We were able to communicate very effectively.

Stephen V. Chavez (Siemens) (37:25.314): Even though the person’s on another state or on the other side of the ocean, we could still work as if we were sitting right next to each other the way you two are right now on this podcast. And it was very clean in how we communicated. But in the end, it comes down to the engineers. And as you two have said it before in this podcast is, you know, the experience and the expertise of knowing.

Stephen V. Chavez (Siemens) (37:55.314): what needs to happen and when it needs to happen, that it’s gotta be constant communication and you’re constantly looking. Even when the EE hands it to you, hands you content, sometimes you gotta question or you need to double check because as you mentioned, Terry, sometimes you start working on design, you realize, my God, he gave me the wrong or he or she or they uploaded the wrong database and the changes they made were to a version behind.

Stephen V. Chavez (Siemens) (38:25.314): Charlene, mentioned, sometimes you go to integrate it, those changes, all of a sudden half the circuitry rips up. Like what happened? Well, it’s because he had two revisions or she had two revisions behind. They didn’t update what they were working on to the latest and you end up losing time. and then the power goes out.

Stephen V. Chavez (Siemens) (38:31.606): Yeah, yeah, there’s alternative stresses that affect us and I call that, you know, real world stresses, you know, that hit us. It’s not just the stress of the project, it’s, you know, outside our daily lives that also affect us, you know, as well. welcome to engineering, huh? So with that said, let me go to the next question here. So can you two share some real or examples of where co-design either failed badly or

Charlene McCauley (38:45.587): Right.

Charlene McCauley (38:53.63): Yeah

Stephen V. Chavez (Siemens) (39:01.762): worked exceptionally well. know you guys done both, just give something to the audience here of something you shared. I know you shared it before some of the conferences. So that’s why I wanna ask you this question.

Charlene McCauley (39:21.762): We have had to send our work to other engineering teams, they come back and completely did it differently than what we had submitted.

Stephen V. Chavez (Siemens) (39:35.691): Mm-hmm.

Stephen V. Chavez (Siemens) (39:41.154): Mm.

Charlene McCauley (39:44.252): You know, we have to keep track of like working on the projects we work on, like we’re working with Dell a lot of times, but we’ll send it off to a team. And they wanted something from us, but then we send it to them and they do it completely different than what they’ve asked us to do.

Stephen V. Chavez (Siemens) (40:05.614): Sure.

Charlene McCauley (40:05.734): And that just causes a lot of stress. mean, you you have to kind of get in their face or in there and say, okay, what are you doing? Why are you making these changes? So we got to understand why are they going down that path when they asked us something? So that’s where it starts to break sometimes, you know, because you’re working with other complete different teams.

Stephen V. Chavez (Siemens) (40:28.365): Mm-hmm.

Stephen V. Chavez (Siemens) (40:32.654): Mm-hmm.

Charlene McCauley (40:32.83): and you have no control over what they’re doing. And that can cause some really a lot of stress and lot of cause things to break. But things that did work was this project we worked on that our presentation is about the CAM2. Those boards that we worked on.

Stephen V. Chavez (Siemens) (40:44.524): Yeah.

Stephen V. Chavez (Siemens) (40:55.149): Mm-hmm.

Charlene McCauley (40:58.238): It worked really well. mean, we were able to complete the design or the project, become a world standard, and it was in a shorter amount of time than typical. Like typically a world standard would take six to 10 years. We did it to four and five years. And it was a very small team that did that. But that was keeping the concurrent.

Stephen V. Chavez (Siemens) (41:20.686): Mm-hmm.

Charlene McCauley (41:27.9): design, all of the things that we’ve been talking about, staying on top of it, making sure everything’s flowing, and we were able to complete the project faster than normal.

Charlene McCauley (41:42.829): mean, the concurrent design can also mean the engineers for us were still working on things and developing things at the same time. So we would get a new schematic, and it might change the logic of what we’d already done. And it wasn’t just, you know.

Stephen V. Chavez (Siemens) (41:42.829): It’s out, it’s out.

Stephen V. Chavez (Siemens) (41:51.342): Mm-hmm.

Charlene McCauley (42:00.314): a simple change but it was actually changing the whole way it was put together sometimes. Or they would throw new pieces in there and say, we have to add this. But so they’re doing their side of the concurrent work too. Yeah, so they were.

Charlene McCauley (42:21.026): doing their studies and doing their updates. And then the mechanical guys would throw loops at us and new mounting holes and stuff. So there’s concurrent, not just layout people, but there’s all the engineers and manufacturing. the whole bit is working at the same time. that part is still going as well. And that was a big success for us, that whole thing.

Stephen V. Chavez (Siemens) (42:29.55): Mm.

Stephen V. Chavez (Siemens) (42:40.226): I think…

Charlene McCauley (42:45.03): I mean, there were times when I would post something and I didn’t save it to a new revision. And then we’d be like, you already had that one. And we didn’t pull out the new one and something had been changed. And I didn’t update the revision number or something. So yes, we have definitely gotten burned.

Stephen V. Chavez (Siemens) (43:00.365): Hmm.

Stephen V. Chavez (Siemens) (43:05.806): There’s a lot of details, lot of details, attention to the details. I think the biggest summary that I would take away from what you’ve said so far is the fact that it took a while, but once you establish the trust on both sides, not just from the Dell side, but from Charlene, your company side, when you and Terry were supporting them as basically like consultants to them, it took a while for them to settle in on the trust.

Stephen V. Chavez (Siemens) (43:45.806): And once it sounds like you settled in and proved yourself to them, then it seems like things started to go really smooth. And that’s when you really started to notice a big difference. And I think that’s one thing that…

Charlene McCauley (43:47.062): You have to see the success, but also you have to watch out for being attached to things. You gotta let somebody else take it on and see what they’re wanting to do with it. You gotta let go of that attachment to it to allow this design to reach the level that it needs to go to.

Stephen V. Chavez (Siemens) (43:56.077): Mm.

Stephen V. Chavez (Siemens) (44:13.038): Mm.

Charlene McCauley (44:14.716): So we had to do a lot of letting go and kind of say, okay, tell us what you want and we’ll go down that path with you. I would think that letting go, mean, that’s gonna be big if you’re doing the serial design, like you’re handing it off and handing it off. Because when you get it back the next day, it’s not gonna look just like you thought it should look.

Stephen V. Chavez (Siemens) (44:27.49): Yeah.

Charlene McCauley (44:40.764): You gotta let everybody have. Yeah, she would try to make it look, but, you know, it would, it takes trust that they did okay and that maybe that you have to look through all the work that came back to you. And then you start trusting that it’s correct. You know, lot of times time, time too will affect things. Where maybe you’re gonna be broken a little bit on the.

Stephen V. Chavez (Siemens) (44:47.65): Yeah.

Stephen V. Chavez (Siemens) (44:58.99): No, I…

Charlene McCauley (45:10.162): the time crunch. You have to get it out fast. So you gotta let go and you know, okay, we have to do this so we can get it done. And so it’s a luxury to have all the time in the world to get a project done.

Stephen V. Chavez (Siemens) (45:15.148): Mm.

Stephen V. Chavez (Siemens) (45:31.182): I would say, like I was lecturing a few universities a couple of weeks ago, time is one thing engineers don’t have. We never have enough time. There’s never enough time. And it’s all about mitigating risk. Our job is to mitigate risk. And how do you meet your requirement by doing that?

Stephen V. Chavez (Siemens) (45:51.182): So with that said, there’s a couple more questions I want to ask before we close this out, because I know we’re getting deep into time here. So one thing we talked a little bit about, but let’s just address it head on, is how do you two see AI and automation changing co-design workflows in PCB design?

Charlene McCauley (46:10.294): AI is the big question. A lot of the software companies are incorporating AI into their software. As far as a designer, I don’t see it taking over, but I do see it enhancing the steps that, like all these steps that we talked about.

Stephen V. Chavez (Siemens) (46:10.294): Either one of you.

Stephen V. Chavez (Siemens) (46:22.574): Mm-hmm.

Stephen V. Chavez (Siemens) (46:30.86): Yes.

Stephen V. Chavez (Siemens) (46:34.84): Mm-hmm.

Charlene McCauley (46:35.134): It’s in the memory. It’s in there where you can utilize something you did in the past. And the AI will trigger and say, oh, you did this last time. Would you want to that again this time? And more of a guide or a.

Stephen V. Chavez (Siemens) (46:51.628): Mm-hmm.

Charlene McCauley (46:55.614): something that can take away a lot of the stress maybe that you’re trying to remember. How did I do that last time? Or create checklists that you can follow. So I see it as a smooth way of getting you down the path, making sure you’re following all the steps that you need to follow, laying out a printed circuit board.

Stephen V. Chavez (Siemens) (47:01.272): Mm-hmm.

Charlene McCauley (47:21.438): I see it enhancing design in that way. I can see some designs, maybe it will be able to do it, but it has to be a pretty simple design for AI, I think. That’s my opinion that it would take over and do the design for you. Because we just talked about all these things.

Stephen V. Chavez (Siemens) (47:40.841): Hm.

Stephen V. Chavez (Siemens) (47:46.464): sure.

Charlene McCauley (47:47.102): And can you imagine AI doing all of that and getting your design done? I mean, yeah. My analogy that I give to my students is that, because that’s one of their first questions when they come in the class is, is this job going to be available? Is AI taking over the job?

Stephen V. Chavez (Siemens) (47:52.142): Not yet. Not for a complex design, no.

Charlene McCauley (48:10.502): And I say, well, here’s my version, is that if you started back in the light table taping days and you decided that when the first CAD tools came out that you’re like, no, this process is working and I don’t need that stuff, you know, and you never learned the new CAD tools, then you were obsolete. And well, I don’t know how long it took, five, 10 years, but you would become obsolete and it would take over your job.

Stephen V. Chavez (Siemens) (48:31.842): Mm-hmm.

Charlene McCauley (48:37.528): or somebody else would take the job. So that you should embrace any tools and the AI enablements that help process faster. And I think making your process faster and making it more secure, like Charlene was saying, to maybe help with reviews or things like that.

Stephen V. Chavez (Siemens) (48:50.285): Yes.

Charlene McCauley (49:01.692): Yeah, I mean, I was there for the taping board, know, taping boards. And then the first software came out. I enjoyed taping boards. It was like a piece of artwork. we had to go with the flow. We had to go with what was there. And, you know, you can’t ever think about something’s going to take your job. You know, your expertise, you grow your expertise to where, you know, you can enhance

Stephen V. Chavez (Siemens) (49:06.051): Mm-hmm.

Stephen V. Chavez (Siemens) (49:29.421): Yeah.

Charlene McCauley (49:31.836): the design and using these new softwares, learn about them, enhance your knowledge, and do everything possible to speed up the process. Because we just mentioned time. That’s what it’s all about. And so if you have something at your fingertips that could eliminate some of the time that it takes, that’s a good thing. Or that helps it be more accurate first time. Yeah.

Stephen V. Chavez (Siemens) (49:45.995): Mm-hmm.

Stephen V. Chavez (Siemens) (49:45.995): Mm-hmm.

Stephen V. Chavez (Siemens) (49:57.613): Yeah.

Stephen V. Chavez (Siemens) (50:01.135): Yeah, I, you when I think about, you know, the evolution of AI in our, in our craft, in, uh, of being PCB designers is the fact that AI, is just allowing me to be a better version of me. But the key here is you have to embrace and use it to your advantage and understand when and how to use it to your advantage. You can’t just blindly push a button or just let it do its thing because it’s going to produce garbage.

Stephen V. Chavez (Siemens) (50:41.135): It still needs the engineer. And the bottom line is the engineer still owns the decision. Who is going to take your job is not AI. It’s the engineer wielding AI that’s going to take your job. That’s the way I see it. So with that said, yeah, go ahead, please.

Charlene McCauley (50:42.984): Can I say something else? Okay, you know, it’s just like when we have to push the button to Gerber a board. You know, everybody thinks you just push a button and everything pops out and you get a good board. You know, a lot of the higher ups think that AI is gonna be that button that they can push and it’s gonna flow out.

Stephen V. Chavez (Siemens) (50:56.237): Hmm.

Charlene McCauley (51:06.798): And it’s the same scenario. Everybody that doesn’t know what’s going on, they’re going to think that AI is going to take care of everything. But we’ve got to convince them that we have to work in parallel with this AI. We have to work together, utilize the good and bad of the AI and with the designer. And it’s all about providing something that

Stephen V. Chavez (Siemens) (51:07.864): Hmm. Yeah.

Stephen V. Chavez (Siemens) (51:21.453): Yeah.

Stephen V. Chavez (Siemens) (00:05.787): So we’ve covered quite a bit and unfortunately due to time, I think that’s all we have time for today. So I want to make sure you guys get a chance to give any parting advice to teams starting with co-design and what would it be? So can you share your any last insights or any last comments?

Charlene McCauley (00:26.294): It’s just like what we talked about. You have to have a team that you can trust and want your immediate team. But also, you have to be open and willing to allow other people to come in and have their opinions. And that’s what we’re talking about is working at multiple teams, be accepting of different ideas.

Charlene McCauley (00:56.294): And communication is everything. And be diligent about, you know, if you do have certain processes about virgin control or whatever, you know, even if you’re like, that doesn’t matter, do the steps every time because you will get burned if you don’t. And it’s worth the extra couple of minutes to do that extra step.

Charlene McCauley (01:26.294): But yeah, so be diligent about it. And if you think something’s wrong, don’t hesitate to question it either. Because you don’t want to have to go back and redo it. Because you thought it might be wrong, but you didn’t ask. And then it ended up it was wrong. It’s just communication. Follow the process. Be diligent. Diligent about all of your details. Really, details do matter.

Stephen V. Chavez (Siemens) (01:46.083): I agree. think the key is is making sure you have a process in place and it’s established and it’s locked down and then you follow the process no matter what what you’re doing. Follow the process. So Charlene, Terry, I want to thank you both for joining the podcast today and for sharing your valuable insights on concurrent design and co-design.

Stephen V. Chavez (Siemens) (02:16.083): To our audience, want to continue to have you or I want you to continue to tune in for more trends, challenges, and opportunities across the printed circuit engineering industry.

Charlene McCauley (02:19.114): And thank you for inviting us to. Yes, thank you so much. It was fun. It’s fun to talk.

Stephen V. Chavez (Siemens) (02:24.591): It was, it was, always. I could talk to you two for hours over hours, days on end. So thanks again for being on the show.

Charlene McCauley (02:28.718): That’s true.

Charlene McCauley (02:28.718): All right. Thanks. Thank you. Bye.