Signal Integrity Expert Eric Bogatin on Best Measurement Practices

August 22, 2018 Judy Warner

Download this episode (right click and save)

 

Signal integrity expert, Dr. Eric Bogatin, shares why best measurement practices have become his go-to topic when speaking with PCB designers around the world. As Signal Integrity Evangelist at Teledyne LeCroy, a leading provider of oscilloscopes, protocol analyzers and related test and measurement solutions, Eric lectures around the world and he will be one of the keynote presenters at AltiumLive 2018: PCB Design Summit. Listen to Eric and Judy talk about the importance of best measurement practices and where to learn more — from webinars to conferences to the Signal Integrity Journal and Rule Number 9. Eric has valuable insights for PCB designers, so tune in and learn more in this episode of the OnTrack Podcast.

Signal Integrity with Eric Bogatin Video:

Show Highlights:

  • The OnTrack Podcast is in 84 countries! Congrats to Daud Zoss who was the closest guess at 37 countries. He gets a free pass to AltiumLive as Judy’s guest.
  • Dr Eric Bogatin will be a keynote speaker at AltiumLive in October 2018
  • Best measurement practices - how do you get the answer to the performance, root cause, characterization, etc. as quickly as possible?
  • How do you know what the performance of your instrument is, so that you know its capabilities and what the device is doing compared to your measurement instrument? It’s important to know what the properties of your scope in the probe is, to know the properties of the device you’re testing.
  • Measurement data: Such as the rise time, frequency or figure of merit must be excavated to give you useful information. How do you get the information so it’s high quality and can be trusted, how do you turn it into information that you can turn into action?
  • Eric is also the Editor of Signal Integrity Journal, working with Janine Love and Patrick Hindle.
  • Expert content - if anyone is interested in writing a technical article for Signal Integrity Magazine, please write*: Eric or Judy.
  • Janine Love manages the EDI CON coming up in Santa Clara in October (a couple weeks after AltiumLive). Part of this is EDI CON University offering tutorials by industry experts.
  • Industry Experts on the Editorial Advisory Board: Bert Simonovich, Yuriy Shlepnev, Larry Smith and Steve Sandler, Rula Bakleh, Jay Diepenbrock, Vladimir Dmitriev-Zdorov, Alfred Neves, Istvan Novak, Doug Smith, and Lisa Ward.
  • Rule #9 - Before you do a measurement or simulation, think about what you expect to see ahead of time, and if it’s not what you expect, there’s always a reason for it. You need to identify the reason why it’s not what you expect.
  • Hands on learning is a necessity for students. Eric and Mike Horowitz put together a five-week, standalone crash course on how to design a board.
  • Designing for connectivity is just about driving the board to enable finding the parts and laying them out for assembly. Really simple.
  • The lack of experience with Oscilloscopes is surprising because nobody has ever taught these students the correct method.
  • Hands on experience is giving students an edge in the marketplace. There isn’t enough of the ‘real world’ activity in most Universities.

 

Links and Resources:

LeCroy webinars

Upcoming events at LeCroy

SI Journal webinars

Upcoming SI Webinars

Hyperloop competition

AltiumLive 2018: Annual PCB Design Summit

 

*If anyone is interested in writing a technical article, please write: Eric Bogatin or Judy Warner

 

Hi everyone, this is Judy Warner with Altium's OnTrack Podcast. Thanks for joining again, if you would please connect with me on LinkedIn I like to share lots of information relative to PCB designers and engineers who are laying out boards and on Twitter I'm @AltiumJudy and Altium is on all the usual places; Facebook, LinkedIn and Twitter.

So please let us know what you'd like to hear about on the podcast and we will do our best to get it done. So today I have a rock star with us and he needs no introduction . But before we get going with Dr. Eric Bogatin; I wanted to say that a few weeks back I had put a challenge out there to see if you guys could guess how many countries the On Track podcast has reached; and we have a winner! So congratulations to Daud Zoss. He's a Senior Staff Engineer at Dexcom; he guessed 37 countries and he was the closest one, unfortunately it was only about half, because we've actually reached 84 countries, I kid you not! So anyways, thank you for listening and engaging with us and all across the world. We really appreciate it.

So today, as I mentioned, we have Dr. Eric Bogatin with us who needs no introduction; who is a signal integrity guru. You might know him from many conferences in North America and I suppose around the world Eric has has presented, and I'm lucky enough to be here in California, where I've seen him present many times on Be The Signal and now the Be The Signal and Eric Bogatin brand is flying under the Teledyne LeCroy flag. So he has lots of lectures and demos and things and I'll let him tell you more about that. So Eric, welcome, we are glad to have you.

Hey thanks Judy, I'm happy to be here with you today and tell you about all the things I've got going on.

Well, we're super excited to have you as a keynote at AltiumLive, so we really appreciate you coming out for that and we've done some neat things together with students, so we'll talk about more about that. So why don't you start off by telling us a little bit about your day job at Teledyne LeCroy?

Sure yes, so many of you may know, and I know you - - I knew you back when I had my own company it was Bogatin Enterprises, and I literally went around the world and did training classes. And about seven years ago, my training company was acquired by LeCroy and we continued the training classes and then began to make a slight transition to, most of what I've done over the years has been best design practices.

How to get the design right the first time and LeCroy is in the measurement business; we are the third largest manufacturer of oscilloscopes and some of the highest end oscilloscopes; and our CTO Dave Graef, he likes to say that that in designing, the goal is to get it right the first time, but if you don't get it right the first time then the goal is to get it right the second time; and the way you get it right the second time is, you have to find the root cause of the problem and invariably that involves some measurements. So that's kind of the connection with LeCroy, is we're number three in the  scope world and have the highest end performance scopes out there.

We really specialize in the business of helping customers get it right the second time; kind of a faster time to insight. And so we started out when I joined them seven years ago, doing the same Best Design Practices presentations and classes I used to do, and then over the years since then, I've been working on this new area of Best Measurement Practices and so, with my day job at Teledyne LeCroy, I am still Signal Integrity Evangelist, but I spend more time now going around talking to folks about, and doing presentations on what are some of the best measurement practices. How do you kind of get the answer to either the performance, or the root cause, or the characterization, or get the Figure of Merit? How do you get that as quickly as possible?

And recently, in fact, I've got a couple of live events in the Bay Area coming up - actually next week - in last week in August, and then in Boston in September. And you can check the Teledyne LeCroy website for the events page to see where I'm coming next, but those presentations are really focused on, how do you - I call it kind of two aspects of in best measurement practice - one is situational awareness. How do you pay attention to - how do you know what the performance of your instrument is, so that you understand what its capabilities are, so what your device is doing, compared to your instruments. So you make sure that you are not seeing an artifact in the measurement.

Wow that's interesting.

Situational, because I find in talking a lot of folks about measurements; gosh there's a lot of confusion about what's the scope doing. And unfortunately there's no such thing as the ideal instrument; they're always - - or ideal probe, for that matter. There are always interactions of the probe and the scope with the device we're looking at, and it's important to understand what the properties of your scope in the probe are, to know how far away you are from the properties of the device you're testing so that you're getting good quality information about the device you care about and not an artifact of how you're doing the measurement. So that's the first piece of what we try to present and teach - those principles.

And the second piece is - and I see this with my students all the time - that they sometimes feel that just getting the data, just getting the measurement is enough. So they, push the right buttons and they get a screenshot and say: okay , here's my data. And I see a lot of engineers doing that as well, and the data is just the starting place. That's not - you're not done with the data - you need to take that data, the measurement and turn it into information. So you need to extract out, what's the few pieces of valuable information.

Like what's the rise time, or what's the frequency, or what's the jitter? It's a figure of merit that takes a lot of data and gives you one or two numbers that you can do something with. I was giving a talk at one of my events a couple weeks ago, and as I mentioned, that we have this huge amount of data in a scope. I mean, one acquisition can be we can take up to five Giga samples worth of data - but you know stupidly maybe - 10 - 20 mega samples but that's 10 or 20 million data points in one acquisition. It's a huge amount of data but you only want one or two numbers out of it. And so I used to call it data mining, and someone said: hey with all that data there it's not mining, it's excavating.

So it's kind of excavating the data for useful information, and then the third piece - once we have the information - is this: so what? It's how do you turn that information into action? How do you use the information you've got, to tell you is this good or bad? Should I, raise the line width or decrease the line width? What do I want to do with that information now? How am I going to use that to influence a decision? So it's those three steps that we talk about in our workshops; of how do you get the information, do you have high quality of confidence for the data, do you have high confidence in it? How do you turn that data into information, extract a couple of figures of merit, the nuggets of valuable information and how do you take that information and turn it into action?

So that's what I'm focusing on these days, the idea of best measurement practices. In addition to the stuff I've done forever, of best design practices. So that's kind of what I'm involved in now, spending a lot of time going around, doing live demonstrations, incorporating them in my workshops. Now we've got some really cool scopes and bring a lot of test vehicles and structures, so we can do live measurements of various signals. And so it's always a lot of fun when you can have...

A physical scope there, right.

-yeah a working device and the scope, and then people that come to these; you know I love working the crowd, and we talk about: well, if that's really what's going on, if you made the the rise time shorter, what what will you see? Or if I expanded the time base, what's the signal going to look like? And so we can do that as a live experiment in the group. So they're a lot of fun, very interactive activities. So that's that's what I do is my day job now.

Well, that's a lot, and it sounds - you make it sound really fun and engaging. So also, Teledyne LeCroy will have a table at AltiumLive, I hope we can talk you guys into bringing an oscilloscope so we'll let you work our crowd and I'm sure.

You know, another thing - oh by the way - I would encourage people that are listening to connect with Eric on LinkedIn, or connect with me I've been sharing those classes that Eric is teaching, so you'll be able to pick those up and see the different locations that he's teaching those courses. And we will also add those links below here in the show notes.

So if you're in those areas you can hop into one of Eric's classes, and he's super fun too it's a very plain spoken - and like I, can learn things from Eric Bogatin, and I am not, my technical prowess is limited, so I really appreciate that about Eric. The other thing you do, we have some friends in common which are Horizon House, the publishers of Microwave Journal, have published a new magazine called the SI Journal, which I am very excited about and you are also the editor of SI Journal, and we have friend in common Janine Love and Pat Hindle and the whole group. I used to write a blog for Microwave Journal that is put out by the same publishers on their website, talking about making RF boards and all the fun that goes along with that. And so now, Eric is editor of SI Journal, so you can also subscribe to that online. We will also share that link. So how's that been so far? Tell us about your job - how long has it been now? It hasn't been too terribly long?

You know what, I think it was about - - it's almost 2 years now, so I'm just going to...

Wow! I was gonna say a year and it's like two years. Wow.

So I think it was it'd be - - between Pat and Janine they kind of came up - they've been focusing on the Microwave Journal which has been around for 30-some years and this is one of the - I think it's the top...

I think it's like 60 years or something.

Is it 60 years?

I don't know I might be...

It's a long time - it's been around forever.

And it's been a real icon in the industry for good quality articles about microwave technology and with Janine's experience with the Design Con and in the signal integrity world; I think between she and Pat they realized: hey, the industry could really use another kind of curated source of high value information and so many of the magazines that we're used to getting have - - the print magazines have disappeared and they're all online, and so Pat and Janine decided to create this as an online journal initially. And they asked me to come on board as the editor; really the technical editor right at the beginning - about two years ago -

and so since then we've been kind of planning it out, putting together the editorial review board - of really some industry heavyweights and kind of creating a lot of new content, soliciting content from other experts in the industry, in fact, while I have a captive audience here, if anybody out there listening, is interested in writing technical articles for us, that'd be great.

Drop me an email or send it through Judy, and I'd be happy to take a look at what you like to do. We created this and our focus is to provide high value content that's curated. That there's so much information out there online right now. If you do a google search on Signal Integrity or Power Integrity it's not that you don't find anything, you find like 10 gazillion different sources.

You get flooded.

Yeah it's hard to know what's the good stuff and what the stuff is that I should waste my time with and so I think that's really the value of having an online publication or portal that is curated, and that's what we try to do is between myself and Janine and Pat and the editorial advisory board; we try to curate the content so that it's in our opinion what we would consider to be high-value content. And so we don't want to waste people's time or our own time and so there's, we think, a lot of really good valuable content. We've done the traditional stuff of short columns, of feature-length articles. Janine manages the annual conference EDI CON, which is now coming to the Santa Clara area in October. I think it's a couple weeks after AltiumLive.

Yah, it is, it's really close.

And part of that is now I think Janine's calling it the EDI CON University which is going to be tutorials by industry experts that are available for all the attendees. And then she also manages webinars, and if I can just plug a previous webinar. So we had Rick Hartley do a webinar...

Which we love and you know as I mentioned you and Rick are just so well respected and the SI field so I'm glad you snatched him up.

Yes we got him to do one a couple months ago and then that's recorded and posted on the...

Oh great!

-and then I did one a couple weeks ago that's also up there. So we have maybe it's 20 or 30 different webinars and they're all free and all available for anybody if you go to the SI Journal.com website, and you can look under videos and webinars, anybody can access all the content on the SI Journal is free as well.

So, there's some other people that are dear friends Bert Simonovich I know is on your team on the magazine who - - I think is Yuriy on that team as well?

Yeah Yuriy's been involved Istvan Novak has been on the Editorial Advisory Board. We just brought on Steve Sandler - - let's see; Larry Smith who is, he's my buddy, we worked on a book together that came out last year on Power Integrity and he now is at Micron; used to be at Qualcomm, he's maybe the one or two world expert on power integrity. So I learned a lot working with Larry. Let's see - - so yeah those are them.

They're all heavy hitters I mean, all really, they are the industry experts you really have, kudos to Horizon House for putting together such a crack team with you at the helm, which is just incredible, and like you said curating that content. Because there's so much noise out there. How do we bring the noise down and just cherry-pick, the best pieces? And I was kind of around before and as they were launching EDI CON and I was really glad to see them, as the high-speed digital and the RF world kind of moved together and some of the challenges were kind of overlapping to launch a show like EDI CON I think is really exciting and this magazine so, yay! Very excited about that so I can't... and again we'll put all these links below.

I'll even I'll see if I'll go pluck out some of those webinars and put those links in too if you didn't send those to me already.

So while I'm plugging webinars can I plug one other webinar too-

Yeah,

-that I should have mentioned. So I've been spending a lot of time, too much time, traveling doing these live events but also doing webinars; I mentioned the one with SI Journal. I've also been doing some through LeCroy, and we have a whole landing page on what LeCroy has done.

Yeah there's a lot there.

There's a lot of high value content that's all free. Anybody can view them and I'm hoping you'll put up a link to the webinar page from from LeCroy as well.

Okay.

I've put a series together on, a little bit about fundamentals of measurements, part of this best measurement practices series that I mentioned earlier they're one-hour webinars on various scope measurement principles and I'm doing them on a regular basis. I think we have two or three more scheduled for the rest of this year and then we'll have another series starting up in January.

Exciting, I like the idea of this best measurement practices, it's like really practical.

Yeah and it's the same thing with design practice. There are accepted practices that you want to follow unless you have a strong, compelling reason. Otherwise these are the right ways of doing things.

Right.

And same thing with measurement; there are just as many ways of screwing up a measurement as there is a design and so you've got to pay attention to both of them.

And there's a human in the loop too besides your probe and all that, so.

Oh absolutely.

So if the human isn't 'tuned up' -

Yeah

- now so one of the principles that I teach my graduate students and at University and also engineers I talk to, is I call it rule number nine and... have I talked to you about rule nine? Okay I'm definitely gonna be mentioning it at the at my keynote because I think it's one of the most important rules for any engineer and basically it says: before you do a measurement or simulation, you want to first anticipate. Think about what you expect to see and I have found that to be the most valuable kind of habit to get into, because just like you said, when there's a human involved it's easy to make a mistake. And how do you know that you don't have the connector connected where it should be, or how do you know: I think I'm looking on channel two, but I'm really going on channel three? Or I typed in 17 but I meant 71? How do you know?

You can check yourself but there's a limit to, how well you can check yourself and so, if you think about what you expect to see ahead of time, whether measurement or simulation and you look at the result and it's not what you expect, there's always a reason for it and you shouldn't proceed with that information until you've identified how come it's not what I expect. And when I do these live demos in front of groups, I'm constantly making mistakes because you know, it's under pressure. I get a screw in that connector and I'm not sure which demo am I on right now, and so I'm always looking at the screen to see, is it what I expect to see, and I can tell instantly when I've done something wrong because I use rule number nine. And I sometimes play a game with the audience, the engineers there, and say: okay, we expect to see this waveform go up and then down and it's flat - how come?

And it's good experience, good practice, that thinking of what could go wrong in the debugging process because that's what we all end up doing and the more experienced we can become at finding the root cause and why it's not what we expect I think, the quicker we can get to a good answer and move on to the next problem. So it's an incredibly powerful habit that I use all the time and I try to teach all my students.

This is what I love about your classes and things you teach Eric. I've sat in a few of them over the years is, that they're insanely practical and intuitive and memorable. Like rule number nine, I can remember that right, so I really have to say that about you.

Of course, don't forget I also reinforce good behavior with chocolate so that...

Oh yeah he does! He throws chocolate out at his classes so yeah it's like Pavlov's dog, yeah it's so true. Well I wanted to jump into the way that you and I started working together, is I think a month or two ago Iconnect007 came out with a magazine with an empty pair of shoes walking down the street, and it said, who's gonna fill your shoes? And everybody seems to get on this bandwagon about all the people that really, fundamentally understand PCB design in regards to, not just designing but manufacturing, assembly, the whole, all the stakeholders that are kind of implied in that process are greying and gonna retire, and so you know, there's been studies out by UP Media saying - by a pretty large sample - saying that in under 10 years half of PCB designers are going to be gone and so everyone has sort of gotten to this hysteria about it seems like the unanswerable question.

What I appreciate that you've done is I'm going to call you professor now - he wears lots of hats - professor Bogatin called me up and said, Judy, you know, I'm gonna do this program, he's used different tools right now that this - I think the students were sort of driving, or somebody was driving one at Altium Designer, so you kicked off this amazing semester-long course at the University of Colorado Boulder and you - I think co-teach that right Eric?

Yeah so I can give you the quick history.

Okay let's hear it.

So I've been teaching a graduate signal integrity class at CU Boulder for a number of years, based on my textbook and in talking to folks there, we realized that our students - so CU Boulder tends to be very project oriented very hands-on we believe in that, you know you learn from textbooks, you learn from studying, but you understand by doing. And it's the hands-on part that you really - everything comes together. And there were a number of classes that required building circuit boards and I would get called in as a consultant to help them in designing the circuit boards and there's relatively simple boards, two layer boards. But these kids had absolutely no idea. They could push the buttons on the tool, but they had absolutely no idea how the performance was influenced by it by what they do in the layout. And so it became really clear that, boy it sure would help if they had a little bit of guidance in how to design boards correctly. And so a number of us got together and realized: hey, we need this more formal training and a buddy of mine Mike Horowitz, who is an expert at design of circuit boards, we got together and put this course together which was - and it's kind of a funny organization too - we're also trying another experiment.

At CU you are semester based, and some courses are typically like 15 weeks or so. But we are experimenting with creating short five-week modules so it'll be the full regular course, that is a normal schedule of of 3 hours per week but it only lasts for five weeks. And so, Mike and I were tasked with putting a course on Printed Circuit Board Design and Manufacture together, that would have a five week beginning piece that could be a standalone so that most students, undergraduates, would take that and that'd be enough to get them going on their projects and then everybody else would continue for the rest of the ten weeks. And that would go into more detail so it gets them more experienced at circuit board design. And so that's how it got started, and Mike and I worked on it - it's every semester; so we did it twice last year and now. So that was kind of our joint development. And now Mike has gotten more involved in his CEO activities and so I'm gonna solo it this semester.

Oh okay.

So the format is basically a five-week crash course on how to design a board so you have a good chance of success when you build a two layer board. And then the other five weeks are more the same, more the technical detail about measurement technique - this idea the best measurement practices. How do you bring up a board? How do you design a board for simple tests and bring up? And then we'll do four layer boards, and then a little bit on the more high-performance systems. So it's a little bit more advanced and really you know, the way we've positioned it as: there are two levels of design. The first is if you can build a prototype and build it with a solderless breadboard and have wires going all over the place; if that works then designing the circuit board and having it work is really straightforward.

We call it designing just for connectivity you don't have to worry about performance, it's about, you want to be able to manufacture it, but performance isn't on; the interconnects don't matter, and and some of the student designs are just designed for connectivity. It's just about driving the board so you can find the parts of the library and build it in the schematic and then place them on the board and lay them out, so you can assemble it by hand - pretty straightforward.

But many of our student designs these days, are getting more sophisticated. They use a Wi-Fi connection, so you have RF on the board, they have sensitive analog to digital converters on the boards, they have components that sometimes -even BGA components - where the microcontrollers are using a really fine pitch; hard to design by hand, and some of these have a couple nanosecond edges where ground bounce is a tremendous problem. And so we're focusing our class on how to design a board. Not just for connectivity - that's the easy part - but for performance, so that you can designed so it's manufacturable, it's lower-cost reduce the - so much of it is risk management - and then kind of the basic performance issues to worry about.

And in my keynote, I'm thinking that I will probably present on what we have found to be the two most impactful design issues in designing a board, not for connectivity but for performance. If all you think about is connectivity you're gonna run into two significant problems. And so one of the topics is this idea of rethinking how signals propagate on interconnects and I've done this at PCB West and I did it at some of my other courses - and I had a couple people come up to me afterwards and tell me that it was a life-changing moment for them. Because I completely changed the way they've been working on boards for 20 years, and I completely changed how they thought about signals on boards. So I hope it will have a similar impact at AltiumLive, but it's going to be about how to rethink and how to train your intuition to think about how signals really propagate on interconnects.

I loved your Be the Signal....

so... and I'm sure this is a little bit more complex than what you're gonna present, but I remember the first time I ever sat in a course by Eric Bogatin, and he was talking about 'be the signal' and he's like: if this signal's moving from A to B what do you think is gonna happen to la-la-la... and I'm sitting there as a non-designer and he's like: no, be the signal. What would you do? And kind of helped us to frame, kind of this visual - and I'm a visual person - so I like to kind of visualize, what the things that were going on, in that signal path to create whatever it was. So I really loved that.

And that's basically what I'm going to be talking about, that Zen approach to thinking about signals propagating and and how to apply you know -  I'll probably mention it once, in my talk, about how to apply master's equations but in an intuitive way, to understand what's really going on in the interconnect.

So Eric's talk is called 'Living in the White Space' and that will be relative to signal propagation and I'm sure all of us, our brain will explode a little. I have these moments with Rick Hartley from time to time too where he says something and I'm like: nah! You know it could be that simple or whatever and I'm sure, you know Rick Hartley is a student of Eric Bogatin so, I'm sure it's more of the same.

So, well I really appreciate you Eric, taking on these students at university level. Here at Altium absolutely, I think I would do this part of my job for free; is to help students get equipped with not only tools - like I can give them free tools - but that's a really incomplete model, for them to learn. They're learning about electrical theory in school but really how to design a board, and how does - - I just finished a podcast today with Julie Ellis who's a Field Applications Engineer from TTM - what about stack-up? What about all these variables and how they come into play, that are not taught at university? But I love that you've brought them in at a university level because I think these are the kids - I think these are exactly who's going to fill some of those shoes, and they may be EEs laying out boards, they may end up like Rick Hartley did saying, I like just designing boards better than circuit design. Who knows? We don't know.

What has been some of your surprises by the way? What feedback from students?

So I think two things absolutely surprise me; one is, their lack of experience with oscilloscopes. That their way of using oscilloscope is first to push the autoscale button. And I slap their wrist if I catch them doing that, or pushing every button without knowing what it is until they see something, on the screen...

Until the light goes on, they're looking for the LED.

-and and so they, just the basic understanding of what an oscilloscope does and how to control the vertical/ horizontal and the trigger. You know the very basic things. A lot of these kids; nobody's ever sat them down and talked to them about it, so we focus on good - again - best measurement practices in the class as well as the design. The second thing is, there is a disconnect between what you learn in the textbook and what you see in the real world and it's the same thing, but you have to know how to apply what you learn in the textbook. And I don't think any university does enough of that hands-on, real-world activity. We try to do it a lot at CU, we have a lot of projects that students get involved in. Most of them are really about designing a little robot or designing some gadget that does something with the code that you write in there. So a lot of it is - some software, as well as the hardware.

But in our class we try to close the loop of the; here I do an estimate or calculation, and here I do a measurement. Like one of the first labs we do is blowing up traces.

That's fun! It is. Everybody likes blowing up something.

Everybody likes to blow shit up

[laughter].

And so the question, the first question I ask them is: okay, you're gonna lay out a board and you're gonna put some tracer - what line width should you use? You can select it to be anything you want.What line width should you use?

And so one of the things that surprised me is, when I asked the students is, they thought that a six mm wide line, just the narrowest that most fab shops will do. A six mm wide line, was too narrow because it's way too much resistance, or I can't put more than a couple milliamps of current through it. So I gotta use a twenty mm wide line, or fifty mm wide line. And it's the kind of thing that, the very first day in class, we calculate or we look at how do you calculate the resistance of a trace? How do you calculate what the maximum current handling it, using the IPC guidelines that Doug Brooks has been so heavily involved in.

And when you put in the numbers you realize: oh my gosh, it looks really narrow on this board it's only six mm wide but but gosh; it's resistance is going to be still in the tens hundreds of milli ohms for typical lengths. So it really isn't that high a resistance even though it looks really narrow because copper is an incredibly good conductor.

Exactly.

People don't have good calibration of that. And then, so I'll give you the number, and I hope none of my students are listening to this because they're gonna figure it out in class. But if you look in the IPC specs for maximum current handling for a six mm wide surface trace; it's like two amps or three amps and when we put two amps through, we have a test board with the different line widths on them. When you put six amp - - two amps through it, you find we can monitor the voltage across it with constant current and see the voltage increasing because it gets hotter, as you see the beginning of the runway, and RNDF around three amps, IPC's around two amps, around three amps, darned if it doesn't go to thermal runaway and we zap the trace and so, you can really get a good estimate by putting in the numbers ahead of time, of how some of these interconnects are going to behave.

But it's that habit of putting in the numbers doing simple estimates, applying what we learn in class to the real world, that the students don't have that good experience with and that's what we try to do in our class at CU.

Well to your point of hands-on, I feel like that's something that for whatever reason has left our education system too much right there's no shop at school anymore, there's no metal, there's no like just - and it's not just, what they would consider low labor skills or whatever. It's just hands-on learning the kinesthetics of it because I bet you dollars to doughnuts that kid, is gonna remember blowing up a three amp trace, more than if he read something about it in a book right?

Right, and sees the smoke and it pops and, there's this feedback well, what I was telling you about is, again one of my favorite parts of my job is, we just came back about three weeks ago from filming these kids that are doing the Hyperloop Competition -

Oh I'm gonna send you this video Eric - you're gonna die because what they do is so awesome and it's just because they get to do hands-on and they make a good - - there's no way they're not making, more mistakes per minute than everybody else in their field, but because of that and because the lack of constraints they have on them as far as businesses and law and whatever, you know this one team we sponsor is from Munich Germany, they just broke the world speed record inside the tube that Elon built at 290 miles per hour.

Crushing!

There's commercial  companies with venture capitalists that haven't hit that number. Because they're young and they're hungry and they're hands-on, and they're excited. But these are the kids who I think, there needs to be much more of what you're doing. I wish every University would, hopefully you'll set an example that others will follow. And by the way, I've cited your course to a group of six universities I was invited to speak at UCSD, UC Davis was there, five other universities and I put a screenshot of your course -

Oh that's great

-and so I, so if you start getting weird phone calls...

(laughter]

-because I'm like see what he's doing - you all need to do this you know, so hopefully the word will spread but until that happens things like the Hyperloop competition, the FSEA competition, where kids get to get their hands on it and blow stuff up and do it wrong, until they do it right, and learn how to use an oscilloscope in this really hands-on manner well these kids are coming out of college and the internships of these kids, one of these kids from University Wisconsin in the Hyperloop team he's a Qualcomm right now, the team lead is going to SpaceX on internship - companies are plucking them out because of the hands-on. So I think the more we sort of beat this drum and spread this message, I again, something I'm very passionate about - I know you are too - and thank you so much for doing that course and we cannot wait to hear from you and about Living in the White Space at AltiumLive.

Thank you so much for taking time out of your busy schedule.

Well I look forward to seeing you at AltiumLive and all the other viewers that you have and I hope folks come up in and say hi while I'm over there.

Okay will do, and make sure - well not make sure - see if we can get LeCroy and company to bring out an oscilloscope so you can - -

We will definitely have one at our table.

Okay good, good I think that would be something notable and something that people, the attendees would enjoy so thank you again Eric this has been...

Thank you Judy.

Thank you again, this has been Judy Warner with Altium's OnTrack podcast and Dr. Eric Bogatin of Teledyne LeCroy. We look forward to being with you next time. Until then, remember to always stay on track.


 

Click here to view all Episodes

Or click here to access the Altium Designer Free Trial Today.

 

About the Author

Judy Warner

Judy Warner has held a unique variety of roles in the electronics industry since 1984. She has a deep background in PCB Manufacturing, RF and Microwave PCBs and Contract Manufacturing with a focus on Mil/Aero applications in technical sales and marketing. She has been a blogger, writer, contributor and journalist for several industry publications such as Microwave Journal, The PCB Magazine, The PCB Design Magazine, PDCF&A and IEEE Microwave Magazine and is an active member of multiple IPC Designers Council chapters. In March 2017, Warner became the Director of Community Engagement for Altium and was immediately tasked with the launch of Altium’s monthly On Track Newsletter. She was also instrumental in launching AltiumLive 2017: Annual PCB Design Summit in San Diego and Munich, a newly founded annual Altium User Conference. Her passion is providing resources, supporting and advocating for PCB Designers around the world and acting as brand ambassador for Altium.

Follow on Twitter More Content by Judy Warner
Previous Article
DDR3 Memory Interfaces and Topologies in PCB Design
DDR3 Memory Interfaces and Topologies in PCB Design

DDR3 memory topologies are important to know how to create for their roles in creating high-speed data stor...

Next Article
How Does Dispersion in FR4 Really Affect Propagation Delay?
How Does Dispersion in FR4 Really Affect Propagation Delay?

Are you using FR4 for your PCB substrate? Learn more about FR4-specific high-speed design issues in this ar...