<![CDATA[Prof Eric Bogatin: Office Hours with Prof Eric Bogatin]]>https://profericbogatin.substack.com/s/office-hours-with-prof-erichttps://substackcdn.com/image/fetch/$s_!lDd4!,w_256,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fedf2693a-aafc-49e9-94ad-f30cb59d9828_500x500.pngProf Eric Bogatin: Office Hours with Prof Eric Bogatinhttps://profericbogatin.substack.com/s/office-hours-with-prof-ericSubstackTue, 21 Jul 2026 00:18:09 GMT<![CDATA[Office Hours with Prof Eric Bogatin, 2026-01-05 ]]>https://profericbogatin.substack.com/p/office-hours-with-prof-eric-bogatinhttps://profericbogatin.substack.com/p/office-hours-with-prof-eric-bogatinMon, 05 Jan 2026 14:03:26 GMTWelcome to Office Hours with Prof Eric. This is a periodic newsletter from my Substack website. Subscribe to the free newsletter here.

Subscribe now

Each newsletter has five parts:

  • a Feature presentation

  • links to content always available

  • a look forward to my events coming up

  • a look back at recently released content

  • a look farther back into my archives

The feature for this week is Financial advice for new graduates.

Feature Presentation: Financial advice for new graduates

Motivation

At CU, Boulder, I teach every senior in the EE department in my capstone senior design lab course. This is a 2-semester project-focused course. We try to structure this to prepare them for their first job in the industry.

When they graduate, 85% of my students get a job in industry. The rest go on for an MS or PhD.

I also teach graduate courses, and 100% of my graduate students go on to get jobs in industry.

I tell my students that the main difference between school and work is the direction money flows. In school, they pay to learn and work on projects. In the industry, they will be paid to learn and work on projects.

Their first job may be the first time in their lives that they have positive cash flow in their pockets. When students are contemplating this transition, they often come to my office to ask for my advice. I was poorly prepared for life when I graduated and started working. I learned the hard way by making mistakes.

I always say that an expert is someone who has made all the mistakes possible. But you don’t have to make all the mistakes yourself. You can learn from others.

Of all the sources I found for financial advice, the best lessons I learned were from a talk I heard Suze Orman give on financial security. Since this is one of the topics I am asked about in my office hours at school, I thought I would include my recommendations here, on financial advice for new graduates. I think it applies to all of us at every stage of our lives.

Suze Orman says there are five steps to take and habits to develop to secure your financial future. Securing your financial independence for your future is one part of peace of mind. Here is her advice:

Step 1: Live below your means.

Ensure that you spend less than you earn. This is so obvious, but it requires discipline. Just because you want something is not a reason to get it, unless you can afford it. The larger the gap between your revenue coming in and your expenses going out, the larger the profit you have available to secure your financial security. It is a lifestyle and quality-of-life balance. To achieve a higher quality of life, you must work for it. This is the basis of living in the real world. The current phrase I hear from my students is “adulting.”

To achieve Step 1 means you must pay attention to both your revenue and expenses. Being aware of what you spend, and on what, is the first step. Making conscious decisions about your choices so you have money left each pay period is the essential step.

Step 2: Pay off your highest-interest-rate loans first.

Any interest you pay on a loan, such as a credit card, means an expense for which you get little value. It is not to say that all loans are bad, but if you do borrow money, such as a student loan, you want to ensure you are paying the lowest interest rate you can get. The highest-interest-rate loans are generally credit card loans. If you live below your means, you will have cash available at the end of each month. Use this cash first to pay off all high-interest loans.

How high is too much? Any loan with an interest rate above 7% may be too much. You can probably get a bank loan if you have good credit for this rate. Car loans are usually less than this amount. Depending on the fed interest rate, house mortgage rates are generally less than this.

Consolidate all your loans into the one with the lowest interest rate you can obtain. Consider checking with your bank to see what they can offer you.

Not all loans are bad. The most significant expenses we often incur during our youth are student loans, car loans, and home loans. First, try to consolidate all the loans into one with the lowest rate. Second, focus on making the largest payments on the loan with the highest interest rate. Pay off each loan as quickly as you can. Celebrate when you eliminate each loan.

Taking on more debt is always a balance between the increase in risk and decrease in peace of mind, with the potentially higher quality of life the extra money from the loan can bring. This balance is a personal decision and is different for different people. Pay attention to how much risk (debt) you are willing to take on.

Step 3: Set aside some money for emergencies.

After you have completed steps 1 and 2, deposit some of your surplus cash each month into a savings account at the bank to build a security fund. This would be used in emergencies, such as if you lose your job, have to make an urgent trip to visit a sick relative, or experience a health crisis.

How much should you put aside for an emergency? It depends on how much risk you want to take. At a minimum, it should cover your expenses for 3 months. This is the typical time you may be out of work before you can find another job. Having a larger emergency fund might reduce your stress if you need more than 3 months to find a new job. If you can cover all your monthly expenses for $3,000, then you might put aside as much as $10,000 in cash in a savings account.

This is money you do not draw on, even for a short period. It will contribute to your peace of mind.

Step 4: Invest in a 401k especially if there are corporate matching funds

Once you have accomplished steps 1-3, it is time to consider growing your financial nest egg for the future. But this is done only after steps 1 to 3. It may take you 5 to 10 years to reach this point. This is why habits are so important. They help you in the long term, the marathon that is life.

Whatever cash you have available at the end of the month, you can put into investing. This is different from savings. You should consider this as early as possible in your life journey, but only after you have accomplished steps 1 to 3.

The best way to approach long-term investing is to treat it as a monthly expense. Set aside a fixed amount from your monthly profits and deposit it into your investment account. This becomes a long-term habit. The amount you take out each month is largely determined by how quickly you want to grow your future and how much you’re willing to spend in the short term. Have a plan.

This is especially important if your company has a 401k plan. This means that money you invest in this account is taken out of your salary before taxes, and any income you earn from the growth of the stocks, even if they are traded, is not taxed until it is removed from your account.

Many companies offer some matching of investment funds. This is free money; take advantage of it. If you can afford it, max out your 401k contributions each month.

If you have the option of what to invest your 401k money in, consider this important observation from the Motley Fool: in the long term, 92% of professional fund managers underperform against the S&P 500.

As a rough rule of thumb, investing in the S&P 500 stock fund has historically returned approximately 7-9% per year, compounded annually over 20 years.

If you don’t have a company 401k plan you can open up a self-managed account with a firm like Charles Schwab or Fidelity. You put cash into the account every month. With it, you buy shares of an index fund.

Which one do you pick? It depends on how much time you have and your risk. The S&P 500 tracks the top 500 stocks in the US market. The plot of the S&P 500 value since 1975 is shown here on a semi-log scale. An exponential growth, which is what compounded annual growth is, would be a straight line on this plot.

The straight line I drew has a slope of 8.9% CAGR. Over the long term, you can get as much as a 9% annual growth rate. But this is deceiving. If you put all your money in starting in 1970, there would have been no growth for 10 years. The market was flat.

If you put all of your money in in 2000, it would have taken you to 2015 before you saw any growth in your money. This is why investing in an index fund should be viewed as a long-term investment, spanning 20 years or more.

There is also the QQQ, also known as the NASDAQ 100 fund. Its long-term CAGR is 14%, as shown below.

But, again, you have to be willing to ride it out for the long term. If you invested in 2000, your money would have lost money until 2015.

The way around this problem is twofold:

Invest a constant amount each month, such as $100 or $1,000. As the index goes down, you are buying more shares. As it goes up, you are making more money. This is called dollar cost averaging. It will enable you to achieve returns that exceed the straight-line CAGR.

Second, you have to leave it in for the long term. This is your retirement account.

You can use this online calculator to explore the impact of compound interest rates and the long term.

For example, if you put away $500 each month, with a CAGR of 10%, possible in the QQQ, at the end of 20 years, you have $386k total money available:

This is the power of compound interest rates. There will be many ups and downs, but over the long term, you will reap significant benefits.

This may be all you ever need to do to secure your retirement. The more you can put away each month on a steady basis, the faster your investment will grow. Experiment with the calculator and consider the long-term implications.

Once you have a steady flow in an index fund, you might consider investment in individual stocks. There is more risk, but more reward. I personally find this a very exciting activity. If you want to get started, here are three valuable resources from which to learn:

The Motley Fool

Investor’s Business Daily and the approach of Bill O’Neil

CNBC and Jim Cramer

There are plenty of others, but these are the ones I use all the time. They have a wealth (pun intended) of resources from which to learn about stock investing.

Step 5: Strike a balance between quality of life and peace of mind.

Once you have accomplished steps 1-4, it’s time to consider the balance of how you want to live your life. Steps 1-4 establish peace of mind that you have secured your financial future. Now it is time to consider your quality of life. What do you want to spend your money on? How do you want to allocate your time? This part is up to you, but achieving peace of mind involves steps 1-4.

Always keep in mind that there is more to life than making money and securing financial security. However, if you consciously and strategically plan this part of your life, it frees you up to decide how you want to spend the rest of your time, reducing one element of stress.

Summary

For many of us, life is not about making more money. However, we all have to live in the world we are placed in. Financial security may not be an end in itself, but it can be one pillar on which we are able to balance quality of life and peace of mind. While it might not seem to be the case, it is never too early in your career to think about how to achieve financial security. If followed, these habits will provide long-term benefits.

It is never too late to begin forming these valuable habits.

Websites with a collection of my videos, articles or courses always available:

· My faculty website with links to many of my publications:

· Electronics 101 series of courses on the Fedevel Academy

· Master Classes in Signal Integrity at the Eecosystem

· Video courses at the Signal Integrity Academy (use promo code CUQ425 for a complimentary three-month subscription)

· Video series at the Signal Integrity Journal

· My pod cast series with the SI Journal: SI Fundamentals

· Other publications at the Signal Integrity Journal

· 2024 Webinar Series on Best PCB Design and Measurement Practices with Teledyne LeCroy

· 2025 Webinar Series on Best PCB Design and Measurement Practices with Teledyne LeCroy

Coming events

New series of courses released on the Fedevel Academy: Electronics 101. This is a new course based on the sophomore circuits course I taught at the University of Colorado, Boulder. It is a foundation building course upon which to base your electrical engineering career.

Feb 12, 2026 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 18: Best practices for using 10x scope probes

Feb 24-26, 2026 Technical talks and Chip Head Theater presentation at DesignCon 2027, Santa Clara, CA

Recently released content in Q4, 2025:

Articles:

Podcasts:

Live Presentations:

Oct 24, 2025: My presentation to Freshman: What I wish I knew when I was your age.

Oct 22, 2025: 5th Annual IEEE Midwest PCB Fest 2025!! Part 1 “SIPI Open Source Boards for learning best design and measurement practices. View the recording here. Download the slides here.

Oct 17, 2025: I was in Austin to present an afternoon of SI and PI demos.

Broadcasted webinars:

Dec 11, 2025 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 17: When to use Ferrites in the PDN.

Oct 9, 2025 Webinar: Part 16: “The Role of Vias in Reducing Crosstalk.”

Featured publications from my archives (all of my publications can be found here)

These days I am focusing on creating video content. I started doing this for the lab classes I taught to provide guidance to students on best measurement practices for scopes and other instruments. I initially called these skill-building workshops (SBW). Here are four of the playlists on YouTube about best measurement, design or assembly practices:

My YouTube channel: Prof Eric Bogatin - YouTube

SBW-7: Best measurement practices with the Keysight 4024 scope

SBW-6: How to use the Digilent AD2 scope

SBW-3 Secrets to Great Soldering

SBW-16: The SAMD21 Xiao microcontroller module

]]><![CDATA[OH with Prof Eric Bogatin 2025-12-15]]>https://profericbogatin.substack.com/p/oh-with-prof-eric-bogatin-2025-12-f40https://profericbogatin.substack.com/p/oh-with-prof-eric-bogatin-2025-12-f40Mon, 15 Dec 2025 14:01:50 GMTWelcome to Office Hours with Prof Eric. This is a periodic newsletter from my Substack website. Subscribe to the free newsletter here.

Each newsletter has five parts:

  • a Feature presentation

  • links to content always available

  • a look forward to my events coming up

  • a look back at recently released content

  • a look farther back into my archives

The feature for this week is two free S-parameter viewers an SI engineer can’t live without

Feature Presentation: Two free S-parameter viewers that an SI engineer can’t live without

Why should I care about S-parameters

S-parameters have become the de facto standard way of describing the electrical properties of interconnects. One way to electrically describe interconnects is with combinations of ideal equivalent-circuit models, such as ideal R, L, C, and transmission-line elements. These are scalable ideal circuit elements that will run in most circuit simulators such as SPICE.

An alternative way to describe the electrical properties of an interconnect is to use its S-parameters. We sometimes refer to this as a behavioural model, in that it describes the behaviour of how sine wave signals interact with the interconnect. In principle, everything you ever wanted to know about the electrical properties of an interconnect is contained in its S-parameters.

The output of an electromagnetic field solver, which takes geometry and materials information, solves for the E and B fields given the boundary conditions, is in terms of the S-parameters of the interconnect.

An S-parameter model is a point solution in that it is a model for a specific interconnect and is not scalable. If you want to know what happens if you were to change a design parameter, like the length of a line, you have to re-solve for that new condition.

These behavioral models can come from simulation and from measurements. They have the same format and can be used in the same way in system-level simulations.

If you are, or want to be, an expert in signal integrity engineering, you must become an expert in S-parameters. One of the problems in learning S-parameters is that, at their heart, they have some complicated math behind them. If you are not comfortable with matrix math and electromagnetics math, it is easy to get lost in the math and lose sight of the underlying principles.

If you want to get up to speed from the ground floor, here are three resources to review:

E. Bogatin, “S-parameters for high-speed digital engineers: A beginner’s guide,” IEEE Electromagnetic Compatibility Magazine, vol. 12, no. 1, pp. 34-41, 2023. Available: Link

“Reading S-parameters like a book”. Part 2 in the Teledyne LeCroy webinar series, Beyond the TDR. Free registration here.

“Master class in S-parameters” Teledyne LeCroy webinar series, parts 1, 2, and 3, free registration here.

If you already feel you understand the basics of S-parameters, and want to become an expert, here are two references to tackle:

D. M. Pozar, Microwave Engineering, 4th ed. Hoboken, NJ, USA: Wiley, 2012.. One of the best graduate-level textbooks on S-parameters is available here.

P. J. Pupalaikis, S-Parameters for Signal Integrity. Cambridge: Cambridge University Press, 2020. Available to purchase here.

Subscribe now

Viewing S-parameters

Once you are over the hurdle of understanding how to interpret the valuable information inside S-parameters, you will want to open up the “black box” of the behavioural model and explore the details of the S-parameter models you use. Never just accept them and use them blindly; always look at them and perform the important consistency tests.

Having a simple-to-use tool to explore S-parameters is critically important for all SI engineers. I rate the various tools available that understand the Touchstone file format of S-parameters into three categories:

  • Pro-level tools that have features that allow you to explore S-parameters and perform system-level simulations with them.

  • Free tools that require some scripting or circuit building to get to the insight you want from the S-parameters.

  • Free tools with a simple-to-use GUI with a < 20 min learning curve that allow you to view the S-parameters.

Most pro-level simulation and analysis tools, such as Keysight ADS, Keysight PLTS, Siemens’ HyperLynx, Ansys HFSS, Simberian Simbeor, MathWorks SerDes Designer, NI Labview, and CST Microwave Studio, have features to help you explore S-parameters and incorporate these behavioural models into system simulations. If you have access to any of these tools, it is worth the investment in learning to use them effectively.

There are three powerful free tools I recommend that require more than a 20-minute learning curve: QUCS Studio or uSimmics, Signal Integrity PY, and PYBERT. In addition to exploring S-parameters, these free simulation environments rival those of their pro-level counterparts. Because they are so powerful and so useful, but have some learning curve, I am planning a few online courses focusing on how every engineer can become an expert using these tools to accelerate up the learning curve. If you do not have a pro-level tool, I recommend exploring these free tools.

If you want to invest no more than 20 minutes to learn how to use a tool to explore S-parameters and quickly gain insight, I recommend the following two tools.

Download some S-parameter Files I am sharing

To explore any S-parameter viewer, you need some S-parameter files to view. If you do not have any handy, you can download some of mine from here. These are a few files of measured structures, both as .s2p and .s4p files.

The two most valuable free S-parameter viewer tools

There are many free tools that can display the S-parameters contained in a Touchstone file in the frequency domain. Even Excel can do this. But if you are going to use just one tool, I recommend a tool that meets these criteria:

  • easy to use (less than 5 min learning curve)

  • always free, not just in a trial version

  • some support and not orphaned

  • can display S-parameters:

    • in the frequency domain

    • in the time domain

    • as single-ended

    • as Mixed mode

  • can compare multiple S-parameter files

  • a plus: can synthesize an eye diagram with simple conditions

I wasted way too many hours viewing all the free S-parameter tools I could find online. I found two that meet these criteria. Every SI engineer should have at least one of these available at their fingertips whenever they want to just view a touchtone file in its various forms. Here is my list in order of preference:

In the following sections, I provide a brief review of each tool and a short video that walks you through how to use it. The perfect tool meeting my criteria above does not exist, but these two are pretty close.

Teledyne LeCroy WavePulser iX user interface

Full disclosure, I am still a Fellow with Teledyne LeCroy and continue to produce webinars with my buddies at LeCroy. Even so, in my professional opinion, the interface used with their WavePulser Network Analyzer instrument, part of the MAUI Studio download, and completely free, is one of the best free tools to import a Touchstone file and view it in the time or frequency domains, as a single-ended or mixed mode. It will read from any touchstone file from other VNA instruments or from outputs of any EM simulator.

A very important note: I have never found a single good reason to save mixed-mode S-parameters. If you have the option, ALWAYS save your S-parameters as single-ended. These are the S-parameters used in a system simulation. Most tools can convert the single-ended S-parameters into mixed-mode or differential S-parameters.

The WavePulser interface is part of the MAUI Studio package. You do have to register before you can download the MAUI Studio software package, but it is completely free.

Step 1: Start on this page and select Download MAUI Studio software, the third option from the bottom.

Step 2: You will be asked to create a free account before you can get to the download page.

Step 3: Once you download the installer, you will be asked which features to install. Just select the WavePulser, shown here:

Step 4: There is no other licensing or registration required to install the interface. After you install the software, you must reboot your computer.

Step 5: When you run it, it will come up and say, No WavePulser attached, Running in stand-alone mode. This is what you want. Be patient. It may take 20 seconds to get to this point. Click ok.

Step 6: To import a Touchstone file, select from Files/recall Touchstone and browse to your file. If you do not have any S-parameter files handy, you can use some of mine. These are a few files of measured structures, both as .s2p and .s4p files.

Step 7: Adjust the display to the grids you like, and select up to 16 different plots to graph. An example of what I sometimes do with frequency domain on one side and time domain on the other is shown below.

In the live display, you can only select single-ended or mixed mode at one time. If you want to compare both, you have to save selected waveforms and then change the set up.

Step 8: to show the mixed mode, you first click the Convert button, which opens up a dialog box to assign the ports.

Step 9: Then you have to click the Recalculate button, and the terms shown are mixed mode.

To master this tool, you will want to explore:

  • Changing the display format

  • Selecting the specific S-parameter that is displayed in each register

  • Moving traces around to different windows

  • Selecting different time domain views.

  • Changing the vertical and horizontal scales

  • Saving a waveform to then display against another file

  • Adjusting the port assignment for mixed mode

Each of these features is very briefly reviewed in the following video. If you want to master this tool, I recommend viewing some of the webinars mentioned at the beginning of this post, especially this one: “Master class in S-parameters,” Teledyne LeCroy webinar series, parts 1, 2, and 3. Free registration here.

View this video for all the details.

Subscribe now

Huwin SnpView

This tool is online only. The company that sponsors this tool provides consulting and training in rf. They offer this tool as a way of attracting leads to their website. In the top menu of their home page is the link to the SnpView tool, located on its own website. The user interface is relatively clean. However, not all the terms are clear. Here is a quick cheat sheet:

Step 1: After you open their web page, you first select if you have a .s2p file which is called “Single” or a .s4p file, which is called a “Differential” file. These are in the top buttons:

Step 2: First, we look at a .s2p file. Click the Single button and then load the file button. You are given a menu to identify the ports. Click ok.

Step 3: The screen is still blank. You need to select a chart. There are two terms used by this tool: Chart plot+ and add chart. The Chart plot+ button will add a new graph window. In this window, you can add multiple traces using the Add chart button. A little confusing. The Chart plot+ adds new windows. The Add chart button adds more traces to the open window.

Step 4: select the specific S-parameters you want to plot with the buttons on the left middle pane:

Step 5: for each selection you want to add to the chart window, click the Add chart button. Here is an example with S11 and S21 plotted:

Step 6: To look at the TDR plot, you select the TDR radial button and a new Chart plot+ window. It takes a few seconds to compute the TDR, but a new window opens up. You can add another TDR trace to this window with the Add chart button.

Step 7: access all the controls for the TDR plot by clicking the TDR advanced options. When you make a change, it applies to the next plots you add.

Step 8: select a .s4p file by pushing the Differential button. Then press the load file button and select the .s4p file. In the menu that opens, you can define the port assignment. This is a very confusing feature in this tool.

For a thru differential pair, the tool assumes the connectivity is 1 to 3 and 2 to 4. This means 1 and 2 are on the left and 3 and 4 are on the right. This is the port assignment the tool assumes when it converts the single-ended into the mixed mode. If this is NOT the port assignment in your file, the mixed-mode S-parameters are calculated correctly. To correctly plot the mixed-mode S-parameters, you have to remap the port assignments.

When you load a .s4p file, the port assignment dialogue box opens.

In the diagram, the hard-coded port assignments, in the grey box on the outer edges, are the port assignments the tool uses to define the ports. It assumes the connectivity for the differential calculation is as shown 1 to 3 and 2 to 4. The inside port numbers that you can select are the ports in your touchstone file.

If you remap to obtain the correct mixed-mode S-parameters, the single-ended S-parameters will appear incorrect. In this port remapping, an insertion loss is S31. If you only want to look at the single-ended S-parameters, then do not remap your ports. Match the touchstone file ports to the tool’s ports.

If your touchstone file has S31 as an insertion loss, then you do not have to remap the ports.

Step 9: View the mixed-mode or single-ended S-parameters using the mixed-mode checkbox, which is unchecked. When the box is checked, the single-ended touchstone file is converted to mixed-mode S-parameters, and you can plot the SDD, SCC or other mixed-mode terms.

Step 10: You can add any S-parameter into any existing chart. For example, I compared the S21 term with the SDD21 term here:

Step 11: The TDR controls work the same way for the mixed-mode S-parameters.

Step 12: This tool also gives the option to run a PRBS signal through the interconnect and display the expected eye diagram. Select the eye diagram radial button, and a new set of options opens up. You can run the signal through the S21 or the SDD21 channel. After you click Chart plot+, it may take as long as 60 seconds to perform the simulation and plot the eye.

Here is an example of an eye generated with this tool:

There is mention of a future feature that will include equalization, but this is in beta and not stable. It may become available in the future.

To see this tool in action, I walk you through how to get started using it in this video:

METAS VNA tool.

METAS is the Swiss Federal Institute of Metrology, an organization similar to NIST. They provide standards, measurements, and calibrations. As a service to their customers and the industry in general, they offer a tool to visualize VNA measurements (and simulations). All of their tools are free, but there is a registration process they ask you to go through. Alternatively, you can download their METAS VNA Data Explorer from here.

This tool is really designed for professional, experienced engineers. It is not as simple as other viewer tools, but is free and capable of all the features I listed above. It has a considerable learning curve. I decided not to include any other features of this tool because it does not offer any advantages over the other tools I reviewed, and it requires a considerably higher investment of time to learn. I list it here if someone wants a challenge.

Summary

These two tools are must-haves for any serious SI engineer. The advantage of the WavePulser tool is that it is supported by a large, stable vendor and is used in their mainstream products. It runs locally on your computer.

The SnpView tool is online. It is a Korean company. There may be some security concerns in sharing your touchstone files in an online tool. However, it runs in a browser with zero setup and is easy to manipulate plots.

If you have your favorite free tool that meets my criteria above, drop me a note and tell me about it. I will include it in some future articles.

Websites with a collection of my videos, articles or courses always available:

· My faculty website with links to many of my publications:

· Electronics 101 series of courses on the Fedevel Academy

· Master Classes in Signal Integrity at the Eecosystem

· Video courses at the Signal Integrity Academy (use promo code CUQ325 for a complimentary three-month subscription)

· Video series at the Signal Integrity Journal

· My pod cast series with the SI Journal: SI Fundamentals

· Other publications at the Signal Integrity Journal

· 2024 Webinar Series on Best PCB Design and Measurement Practices with Teledyne LeCroy

· 2025 Webinar Series on Best PCB Design and Measurement Practices with Teledyne LeCroy

Coming events

New series of courses released on the Fedevel Academy: Electronics 101. This is a new course based on the sophomore circuits course I taught at the University of Colorado, Boulder. It is a foundation building course upon which to base your electrical engineering career.

Feb 12, 2026 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 18: Best practices for using 10x scope probes

Feb 24-26, 2026 Technical talks and Chip Head Theater presentation at DesignCon 2027, Santa Clara, CA

Recently released content in Q4, 2025:

Articles:

Podcasts:

Live Presentations:

Oct 24, 2025: My presentation to Freshman: What I wish I knew when I was your age.

Oct 22, 2025: 5th Annual IEEE Midwest PCB Fest 2025!! Part 1 “SIPI Open Source Boards for learning best design and measurement practices. View the recording here. Download the slides here.

Oct 17, 2025: I was in Austin to present an afternoon of SI and PI demos.

Broadcasted webinars:

Dec 11, 2025 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 17: When to use Ferrites in the PDN.

Oct 9, 2025 Webinar: Part 16: “The Role of Vias in Reducing Crosstalk.”

Featured publications from my archives (all of my publications can be found here)

Many of you know that I write science fiction. I have two novels published. They make great Christmas gifts (hint, hint)

I am working on the third in the six-novel Sciquest Legacy series. Here are the first two:

Bogatin, Eric, and Lori DeBoer. Shadow Engineer: Book One in the Sciquest Legacy Series. Amazon Kindle, 2015.

Summary: Unbelievable even to a tech savvy young Silicon Valley engineer who stumbles into a company hiding in the shadows with advanced technologies that could change the world. At the same time, he learns that this new company, his life, and the rest of civilization are threatened by a force even more advanced. And the opposition has a head start. Will his skills as a Ninjaneer be enough? What can one engineer, a beautiful astrophysicist and an old math professor do to save the Earth? Whatever it takes.

Bogatin, Eric. S is for Space: Book Two in the Sciquest Legacy Series. Amazon Kindle, 2019.

Summary: A powerful, secret organization has infiltrated the highest levels of our world. All that stands in their way is a retired Navy SEAL commander, a full-blooded Navajo ex-SEAL Princeton graduate warrior, a beautiful and deadly ex-Mossad assassin, and an MIT dropout hacker, all connected with a shadow high-tech Silicon Valley company with beyond state-of-the-art technology. From an attack sub deep in the Pacific Ocean, to low Earth orbit in a converted space-plane to a secret Moon base, the battle rages in this non-stop action technothriller.

I hope you enjoy reading them as much as I enjoyed writing them.

]]><![CDATA[OH with Prof Eric Bogatin 2025-12-01]]>https://profericbogatin.substack.com/p/oh-with-prof-eric-bogatin-2025-12https://profericbogatin.substack.com/p/oh-with-prof-eric-bogatin-2025-12Mon, 01 Dec 2025 14:03:56 GMTWelcome to Office Hours with Prof Eric. This is a periodic newsletter from my Substack website. Each newsletter has five parts:

  • a feature presentation

  • links to content always available

  • a look forward to my events coming up

  • a look back at recently released content

  • a look further back into my archives

The feature for this week is an introduction to four free animation tools to help you understand signal integrity principles

Feature Presentation: the dynamic nature of signals illustrated by four free SI animation tools.

First two important full disclosures:

After my company, Bogatin Enterprises, was acquired by Teledyne LeCroy in 2011, I worked for them for 11 years until I transitioned to full-time on the faculty at CU, Boulder. I am still a fellow at Teledyne LeCroy and periodically present webinars and live events with my buddies there.

I wrote about this topic, free animation tools illustrating signal integrity principles, in the September 2024 issue of the Signal Integrity Journal, for which I am the technical editor. While this article was very useful, the links to the animation tools were not permanent, and I did not include videos to illustrate how to use them.

In this feature issue of my Office Hours, I want to provide a permanent home for these tools so that everyone can take full advantage of them. With these tools, you will strengthen your engineering judgment about how signals propagate on interconnects.

This topic is an example of one of my interests: curating high-value, free, and accessible content for all engineers. Check out my SI J article for a list of other columns I‘ve written introducing high-value, free content.

As I state in my SIJ article,

“Learning does not end when we leave school and start our professional careers. The only thing that changes is the direction of the money flow. When we are students, we pay to learn and work on projects. When we are professional engineers working in industry, we are paid to learn and work on projects.”

As a professional engineer, a project is a task with an end goal, as part of the process of producing a product that can be shipped for revenue. Rarely is the goal in a project to understand some SI or PI principles. The goal is to solve a problem and get a product out the door that meets an acceptable level of performance, constrained by a cost budget and a schedule.

However, often part of the product development strategy is to understand the essential principles of signal and power integrity. By understanding, we can establish robust, cost-effective design rules and avoid problems early in the design cycle, saving time and money and reducing risk. It is a more efficient strategy than the “build it and test it” approach.

This is one of the values of curated, free, high-value content that contributes to your understanding. The four tools introduced in this feature will not directly help you design your 112 Gbps system, which requires 1,000 A per GPU, but they will help you establish a firm foundation for understanding how signals propagate and interact with interconnects.

The most important principle these tools illustrate is the dynamic nature of signals. This is a concept at the foundation of how signals interact with interconnects. This is a concept I focus on in my online master classes.

If you do not have a clear understanding of the dynamic nature of signals, you will never develop engineering judgement, which is your most powerful design tool. Engineering judgement is what will differentiate your design decisions from an AI’s, an important value-add for every human designer.

The dynamic nature of signals is hard to describe in just words. It cannot be easily illustrated on a flat, static piece of paper.

As I am still a Fellow with Teledyne LeCroy, the third-largest manufacturer of scopes, I can say this in the most loving way: if you learn signal integrity only by looking at waveforms measured by a scope, your intuition (engineering judgement) about signal propagation will be screwed up. A scope only measures a voltage between two points. It says nothing about the signal's direction of propagation.

This is why it is important to distinguish the voltage measured by a scope, which is a scalar quantity, from the signal propagating in a specific direction along the interconnect. The signal is the voltage difference between the signal and its return conductors that is propagating. The signal is a vector quantity in that it has both magnitude and direction. This is the main topic in this webinar I did with Teledyne LeCroy.

Once you understand the dynamic nature of signals, you can interpret the scalar scope voltage measurements as the combination of signals propagating past the same point in the interconnect in opposite directions.

This dynamic nature of signal propagation is best illustrated in a dynamic animation. Enter these animation tools created by Yoshi Tsuji, a former Application Engineering Manager at Teledyne LeCroy-Japan. Teledyne Lecroy’s value in making them freely available is the brand association, as there is a prominent Teledyne LeCroy logo on the page for these tools. This is the only commercial aspect of the tool.

I use these tools when teaching the essential principles of signal integrity. In the following sections, I illustrate how to use each tool and provide an example of the value they offer in building your engineering judgement about signal integrity. It is important to keep in mind that these tools will not “teach” you signal integrity. You still need to study textbooks, lectures, presentations, videos, and articles. But they will help you exercise your understanding.

Subscribe now

Download all four tools here. They were written in Flash but are executables and will play on any Windows PC. No other software is required. Anyone is free to use them for personal use and to show others. Treat them with the respect they are due.

Return Currents on Transmission Lines

This tool illustrates the basic nature of signals propagating on transmission lines and their return currents.

Watch this video to learn how to use this tool.

An Animated TDR

This tool illustrates how propagating signals interact with impedance changes. This is the most important behaviour of signals on interconnects.

Watch this video to see how you can set up and use this tool to exercise your understanding of signals and reflections from impedance changes.

Near and Far End Cross-talk

Without considering the dynamic nature of signals, cross-talk can never be intuitively understood. This tool clearly shows why there is a different signature in the near-end and far-end cross-talk in a microstrip, and how the capacitive and inductively coupled currents cancel out at the far end in stripline.

Watch this video to learn how to use this tool to explore how cross-talk depends on the dynamic nature of signals.

How Return Loss Gets Its Ripples

The fundamental feature we always see in the return loss from through-interconnects is ripples. This is fundamentally due to the interference between waves reflected from the front and back interfaces of a transmission line with a characteristic impedance different from its port impedances. What we see at the receiver on port 1 is interference from these two reflected waves.

This is why the return loss at low frequency is always very small, independent of the interconnect, as long as it is electrically short. And this is why the insertion loss of every interconnect is nearly 0 dB at the lowest frequency, independent of its nature.

Watch this video to learn how to use this tool to show the behaviour of return and insertion loss.

Summary

These tools will not contribute to the design of your high-end, advanced product, but they will help you accelerate up the learning curve and strengthen your engineering judgement to tackle your most advanced high-speed signal systems.

Download all four tools here.

Subscribe now

Check out these websites with a collection of my videos, articles and courses always available:

Coming events

New series of courses released on the Fedevel Academy: Electronics 101. This is a new course based on the sophomore circuits course I taught at the University of Colorado, Boulder. It is a foundation building course upon which to base your electrical engineering career and future projects.

Dec 11, 2025 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 17: When to use Ferrites in the PDN.

Feb 12, 2026 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 18: Best practices for using 10x scope probes

Feb 24-26, 2026 Technical talks and Chip Head Theater presentation at DesignCon 2027, Santa Clara, CA

Recently released content in Q4, 2025:

Articles:
Live Presentations:

Oct 24, 2025: My presentation to Freshman: What I wish I knew when I was your age.

Oct 22, 2025: 5th Annual IEEE Midwest PCB Fest 2025!! Part 1 “SIPI Open Source Boards for learning best design and measurement practices. View the recording here. Download the slides here.

Oct 17, 2025: I was in Austin to present an afternoon of SI and PI demos.

Broadcasted webinars:

Oct 9, 2025 Webinar: Part 16: “The Role of Vias in Reducing Crosstalk.”

Aug 14, 2025 Webinar: Part 15: “Three Ways to Measure Current Draw and Inrush Current in Power Rails.”

Featured publications from my archives

(all of my publications can be found here)

I’ve written six books relevant to signal and power integrity:

Eric Bogatin, Bogatin’s Practical Guide to Best Measurement Practices for Digital Oscilloscopes , Artech, 2025.

Practical Guide to Prototype Breadboard and PCB Design (2021)

Practical Guide to Transmission Line Design and Characterization for Signal Integrity Applications (2020)

L. Smith and E Bogatin, Principles of Power Integrity for PDN Design–Simplified: Robust and Cost Effective Design for High-Speed Digital Products (2017)

Signal and Power Integrity – Simplified (Signal Integrity Library, 3rd edition) (2017)

M. Resso and E. Bogatin, Signal Integrity Characterization Techniques, 2nd ed. Chicago, IL, USA: International Engineering Consortium, 2009. ISBN: 978-1-931695-93-0.

— 30 —

]]><![CDATA[Office Hours with Prof Eric for Nov 17, 2025]]>https://profericbogatin.substack.com/p/office-hours-with-prof-eric-for-novhttps://profericbogatin.substack.com/p/office-hours-with-prof-eric-for-novMon, 17 Nov 2025 14:03:09 GMT

Subscribe now

Welcome to Office Hours with Prof Eric. This is a periodic newsletter from my Substack website. Each newsletter has five parts:

  • links to content always available

  • a look forward to my events coming up

  • a look back at recently released content

  • a look farther back into my archives

  • a feature presentation

The feature for this week is an introduction to three free and useful SI tools on the Sierra Circuits website.

Check out these websites with a collection of my videos, articles and courses always available:

Coming events

New series of courses released on the Fedevel Academy: Electronics 101. This is a new course based on the sophomore circuits course I taught at the University of Colorado, Boulder. It is a foundation building course upon which to base your electrical engineering career and future projects.

Dec 11, 2025 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 17: When to use Ferrites in the PDN.

Feb 12, 2026 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 18: Best practices for using 10x scope probes

Feb 24-26, 2026 Technical talks and Chip Head Theater presentation at DesignCon 2027, Santa Clara, CA

Recently released content in Q3, 2025:

Articles:

Jul 28, 2025 — “Roll Your Own TDR.” Signal Integrity Journal (with Supreeth Laderpet Yadubala).

July 14, 2025 The Wisdom of Grace Hopper, Relevant Today, SI Journal

Podcasts:

Sep 8, 2025 — “Revolutions in Chip Technology.” SIJ Podcast (guest: Steve Sandler). podcasts.signalintegrityjournal.com

Aug 4, 2025 — “The Future of High Reliability Precision Timing Clocks.” SIJ Podcast (guest: Piyush Sevalia). podcasts.signalintegrityjournal.com

Jul 7, 2025 — “High Speed Serial Links and Their Challenges.” SIJ Podcast (guest: Patrick Connally). podcasts.signalintegrityjournal.com

Live Presentations:

Oct 24, 2025: My presentation to Freshman: What I wish I knew when I was your age.

Oct 22, 2025: 5th Annual IEEE Midwest PCB Fest 2025!! Part 1 “SIPI Open Source Boards for learning best design and measurement practices. View the recording here. Download the slides here.

Oct 17, 2025: I was in Austin to present an afternoon of SI and PI demos.

Broadcasted webinars:

Oct 9, 2025 Webinar: Part 16: “The Role of Vias in Reducing Crosstalk.”

Aug 14, 2025 Webinar: Part 15: “Three Ways to Measure Current Draw and Inrush Current in Power Rails.”

Featured publications from my archives

(all of my publications can be found here)

I’ve presented a few papers at our Rocky Mountain Regional American Society of Engineering Education (ASEE) on some of my teaching projects:

Feature Presentation: Three Free and Useful SI Analysis Tools from Sierra Circuits

There are a number of sites offering analysis tools for PCB design or SI/PI analysis. One of the sites I find particularly useful is from Sierra Circuits. While their main business is the fabrication and assembly of printed circuit boards and assemblies, they offer this free tool to attract attention to their site. It works.

In this edition of Office Hours, I am reviewing three of their tools:

  • A transmission line impedance analyzer

  • A reflection-bounce diagram illustrator

  • A maximum current and trace resistance analyzer.

Transmission line impedance analyzer

For more information about how to understand transmission lines and what characteristic impedance means, check out these resources:

An ideal transmission line is defined by its characteristic impedance and time delay. In a uniform transmission line, the characteristic impedance depends on the stack-up, as illustrated on the Sierra Circuits website for a stripline.

When designing a stack-up to achieve a specific target characteristic impedance, we need to accurately predict the characteristic impedance from the stack-up parameters. The best tool for this job is a 2D field solver.

While there are 2D field solvers built into many high-level simulation tools, such as Keysight ADS, Siemens’ HyperLynx, Altium (which integrates a tool from Simbeor), Polar Instruments SI9000, and Ansys 2D Extractor, these tools are not free and require some expertise to use.

There are a few online calculator tools that use approximations under the hood to give you a characteristic impedance value after you enter the stack-up geometry and material properties. Be aware that these tools use approximations. Just because the output sometimes has 3-digit precision does not indicate its accuracy. Generally, approximations are accurate to about 3% near 50 ohms but can be off by more than 20% absolute accuracy for impedances more than 20 ohms from 50 ohms.

Generally, a 2D field solver calculates a characteristic impedance to better than 1% absolute accuracy. (see this article about one specific tool and this article.). If given the option, you always want to use a 2D field solver to calculate the characteristic impedance of a transmission line from the stack up. Enter the Sierra Circuits online tool.

They offer two levels of access to their tool. Without any registration, you can use a few predefined, simple, single-ended, and differential stackups. By creating a free account and registering, you can access all the stack-ups they offer.

In this video, I walk you through using the online tool.

The first step is to select the predefined, parameterized stack-up. In this simple example, I will show how to analyze a microstrip with a solder mask. This is called a coated microstrip. It is illustrated here:

There are two ways of using this tool. The first way is direct analysis. Enter all the input parameters, defined here:

If you are unsure of the parameter values to use, start with the fab vendor’s nominal values. After inputting all the fields, you can calculate the resulting characteristic impedance. In this case, I got a value of 51.2 ohms:

In addition, you get the propagation delay and the L and C per unit length parameters. The other useful term for microstrip is the effective Dk. This is based on the combined dielectric system of the core, the solder mask, and the air above. While there are some approximations for effective Dk, a field solver is the best way of getting this property.

You can modify any parameter and see its impact on impedance. For example, you can turn the solder mask off by making its Dk = 1 and see the impedance increase to 51.8 ohms, about a 1% increase.

Another way to use this tool is to design a specific line width based on the required target impedance. You would enter the target impedance and see the calculated line width.

For example, using the nominal conditions above and selecting a target trace impedance of 50 ohms, the required line width is 103.8 ohms.

The same approach applies to the other stackups.

This is the most accessible, easiest-to-use tool with field-solver accuracy I have found. For simple applications that just need a quick calculation, I recommend this tool. I use it in my classes all the time.

Subscribe now

A Reflection Bounce Diagram Simulator

If you are familiar with transmission lines and how signals reflect from impedance changes they see along the line, you will be familiar with the bounce diagram. If you are not familiar with it, and you really want to understand how signals interact with transmission lines, check out:

A bounce diagram is a way to keep track of all reflections at the ends of a transmission line due to the launch and termination. It has limited usefulness, as it is only practical for simple systems. Any more than 1 transmission line, and the bounce diagram is too complicated to use. A full transmission line circuit simulator, such as SIMetrix, QUCS, Hyperlynx or ADS is needed with more than 1 transmission line.

However, I think it is important to construct a bounce diagram at least once in your life to help you understand the dynamic nature of signals propagating and reflecting on a transmission line. Sierra Circuits has a very nice bounce diagram simulator as one of its online tools. They refer to it as a Transmission Line and Reflection Tool, illustrated here.

Watch this video where I walk you through using this tool:

I recommend thinking about six case studies:

  • Source impedance 5 ohms (driver), 50 ohms (source series termination), 500 ohms (high impedance source)

  • Line impedance = 50 ohms

  • Load impedance = 50 ohms (terminated) and 1e6, open end.

You should first think about these six cases and what you expect the signal at the output or the RX would be. Then run the simulation and see if it matches your understanding. This bounce diagram will trace through each reflection at each interface to show you the NET or total voltage you will see at the output (or input), from all the reflections. If you understand these six cases, you can really say you understand signals on transmission lines. If you are still confused, check out the references at the beginning of this section.

A maximum current and trace resistance analyzer

The DC properties of interconnects are sometimes just as important as the high-speed properties. This is when you want to know the series resistance of a trace or how much current you can put through it before it gets too hot and it may fuse open. If you want to understand the details of this behavior, you can check out

The Sierra Circuits online calculator for DC trace properties is called the Trace Width, Current Capacity and Temperature Rise Calculator

Watch this video where I walk you through using it.

The resistance of a trace is easy to calculate using the principle of sheet resistance and the number of squares along the trace. This is an incredibly powerful way of looking at trace resistance. If you are not familiar with this method, review Chapter 6 in my textbook mentioned above.

As a quick start, the resistance for a 1-inch-long trace created from 1 oz copper is about 0.5 mohms/square x 1/line width [in inches]. If the line width is 5 mils or 0.005 inches, then the resistance for a 1-inch length of trace is 0.5 mohms/0.005 inches = 100 mohms. It really is as simple as that.

The second important DC property of a trace is its maximum current before the temperature increases too much. Due to its series resistance, as current flows through a trace, it will consume electrical energy and convert it into heat. Its temperature will increase until the heat power flow out of the trace to the ambient matches the power flow in from electrical energy. The higher the current, the higher the temperature.

The temperature rise for a given current is very difficult to calculate from first principles because it depends on the number of layers and their stack-up in the circuit board, the metal distribution, the ambient airflow, and the laminate’s material properties. However, using empirical experiments, the IPC has created a simple reference guide in the IPC-2152 standard. This empirical calculator is built into the Sierra Circuits online calculator.

The way we use this is to input the trace width and the maximum acceptable temperature above ambient. Generally, I select a value of 40 degC. Then we can calculate the maximum current before the temperature gets too hot. For example, a 6 mil wide trace will handle as much as 1.4 A before it gets above 40 deg C. This is shown below:

This is the same value offered by other tools that implement the IPC-2152 spec. In my PCB labs, students measure the maximum current in 6 mil wide traces before they get hot. They usually find that traces can handle easily 1.5 A without any noticeable temperature rise. At 2 A, the trace is warm to the touch. At 3 A, the trace is hot. When the current exceeds about 3.7 A, the trace usually goes into thermal runaway and quickly fuses open. This is one of their favorite labs, blowing traces up. Here is an example of the test board with current through one of the traces, heating it up:

You can use this online tool to calculate the trace resistance for different line widths and the maximum current allowed through a trace. Keep in mind, this is only a rough approximation and depends on many factors. But it offers a good starting place for the maximum current to allow for a trace.

It is remarkable that a 6 mil wide trace can handle at least 1.5 A. This is a trace of copper with a cross-sectional area smaller than a human hair, and you can send more than 1.5 A of current through it before it even gets slightly warm. For a 20 mil wide trace, you can send almost 3 A as per the IPC-2152 spec. These are useful values to remember. When designing a board with current distributions even as high as 3-5 A, there is no need for power planes. Simple traces 20-40 mils wide will easily handle these levels of currents.

Summary

These online calculators are only tools. The most important tool when designing any interconnect is your engineering judgement. If you don’t understand the principles behind these tools and some of their assumptions, then you are not ready to use these tools. But when you do understand the information they can give you, they are incredibly powerful at increasing your efficiency as a hardware designer.

Check out the Sierra Circuits online calculators here.

— 30 —

Subscribe now

]]><![CDATA[Office Hours with Eric for Nov 1, 2025]]>https://profericbogatin.substack.com/p/office-hours-with-eric-for-nov-1https://profericbogatin.substack.com/p/office-hours-with-eric-for-nov-1Mon, 03 Nov 2025 13:02:48 GMT

Subscribe for free to receive new posts

Welcome to Office Hours with Prof Eric. This is a periodic newsletter from my Substack website. Subscribe to the free newsletter here. Each newsletter has five parts:

  • links to content always available

  • a look forward to my events coming up

  • a look back at recently released content

  • a look farther back into my archives

  • a feature presentation

The feature for this week is A review of the recently released Signal Integrity Journal October issue eBook.

Check out these web sites with a collection of my videos, articles or courses always available:

My faculty website with links to many of my publications:
Electronics 101 series of courses on the Fedevel Academy
Master Classes in Signal Integrity at the Eecosystem
Video courses at the Signal Integrity Academy (use promo code CUQ325 for a complimentary three-month subscription)
Video series at the Signal Integrity Journal
My pod cast series with the SI Journal: SI Fundamentals
Other publications at the Signal Integrity Journal
2024 Webinar Series on Best PCB Design and Measurement Practices with Teledyne LeCroy
2025 Webinar Series on Best PCB Design and Measurement Practices with Teledyne LeCroy

Coming events

New series of courses released on the Fedevel Academy: Electronics 101. This is a new course based on the sophomore circuits course I taught at the University of Colorado, Boulder. It is a foundation building course upon which to base your electrical engineering career.

Dec 11, 2025 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 17: When to use Ferrites in the PDN.

Feb 12, 2026 Webinar: Teledyne LeCroy webinar on Best Design and Measurement Practices for Prototypes, Part 18: Best practices for using 10x scope probes

Feb 24-26, 2026 Technical talks and Chip Head Theater presentation at DesignCon 2027, Santa Clara, CA

Recently released content in Q3, 2025:

Articles:

Jul 28, 2025 — “Roll Your Own TDR.” Signal Integrity Journal (with Supreeth Laderpet Yadubala).

July 14, 2025 The Wisdom of Grace Hopper, Relevant Today, SI Journal

Podcasts:

Sep 8, 2025 — “Revolutions in Chip Technology.” SIJ Podcast (guest: Steve Sandler). podcasts.signalintegrityjournal.com

Aug 4, 2025 — “The Future of High Reliability Precision Timing Clocks.” SIJ Podcast (guest: Piyush Sevalia). podcasts.signalintegrityjournal.com

Jul 7, 2025 — “High Speed Serial Links and Their Challenges.” SIJ Podcast (guest: Patrick Connally). podcasts.signalintegrityjournal.com

Live Presentations:

Oct 24, 2025: My presentation to Freshman: What I wish I knew when I was your age.

Oct 22, 2025: 5th Annual IEEE Midwest PCB Fest 2025!! Part 1 “SIPI Open Source Boards for learning best design and measurement practices

Oct 17, 2025: I was in Austin to present an afternoon of SI and PI demos.

Broadcasted webinars:

Oct 9, 2025 Webinar: Part 16: “The Role of Vias in Reducing Crosstalk.”

Aug 14, 2025 Webinar: Part 15: “Three Ways to Measure Current Draw and Inrush Current in Power Rails.”

From my archives

2025-03-31 A. Rao, E. Bogatin and M. Piket-May, “Metrics for Acceptable Oscilloscope Bandwidth,” in IEEE Transactions on Instrumentation and Measurement, vol. 74, pp. 1-8, 2025

2024-11-04 A. Rao, E. Bogatin, M. Piket-May and M. F. Hadi, “Non-TEM Dispersion in Microstrip Structures at High Data Rates,” in IEEE Transactions on Signal and Power Integrity, vol. 3, pp. 178-185, 2024,

Thanks for reading Prof Eric Bogatin! Subscribe for free to receive new posts.

Feature Presentation: The latest eBook released by the Signal Integrity journal

Download your copy here.

Full disclosure, I am the technical editor of the Signal Integrity Journal.

Pat Hindle, the editor in chief, and I are proud to present this latest issue of the October 2025 print version of the SI Journal. You can download a copy here. As always, all the content from the SI J is always completely free.

While I offer here a summary of the articles in this issue, you will find it entertaining to view the pdf copy of the entire ebook.

Editorial: The Impact of AI at the Singularity, by Eric Bogatin.

In this editorial, I offer my perspective on the peril and the hope of the impact AI might have on the future of our lives. The punchline is the last sentence: “As developers of the next generation of technology, we should always keep in mind that technology amplifies human nature. Let’s pay attention to the unintended consequences of what we create.”

Cover feature: Optimization of IBIS-AMI Model Parameters with Machine Learning Algorithms by Jared James and Ambrish Varma, Ph.D., Cadence Design Systems

In this piece, the authors present a case study they did to use a machine learning algorithm to optimize the equalization settings of the TRX for three different channels. This is a feature built into Sigrity. Altogether, there were 636,804 different combinations of equalization settings. How do you find the optimized set of parameter values for a given channel performance without testing every single combination?

One of the challenges in an optimizer is that you need a gradient in “goodness” to determine the direction to move a parameter. When there are many more combinations that result is a closed eye than an open one, it is difficult to use a traditional optimizer.

They outline the methodology of using a machine learning algorithm to find the optimized equalization settings. Their conclusion was “This method found a good set of parameters in fewer simulations than if a traditional manual method had been deployed, conserving the use of limited human and computing resources. In most test cases, it was found that only 100 simulations were needed to find the best set of parameters.”

Seeing Through the Noise: Reliable Power Rail Measurements in High Current AI Systems. By Seamus Brokaw, Tektronix, Beaverton, Ore., and Steve Sandler, Picotest, Phoenix, Ariz.

In this piece, the authors, experts in power integrity, begin by defining the problem. Some AI processor boards have current transients of over 1,000 A, on a 1 V or lower power rail. When evaluating VRM design tradeoffs, it’s important to be able to measure small voltage changes in order to evaluate their advantages or disadvantages.

In an initial series of measurements, the authors found that the variation in the measurement of precision generated transient voltages can be as large as 25 mV, just by how the board was probed. They say the two biggest sources of noise are ground loops in single-ended measurements and increased probe noise. Based on their observed root causes, they recommend using probes with galvanically isolated shields to prevent ground loops and using differential probes with low tip inductance.

ManyPoint Networks: A System Co-Design Framework for 448 Gbps AI Fabrics and Beyond, by Andrew Josephson, Samtec

The author offers a new metric as a single figure of merit to describe the combination of the interconnect density, the IO routing topology, and the data rate or signal bandwidth in dense high-speed serial link networks. The essence of the bisection bandwidth is “defined as the count of real, bidirectional hardware links crossing a minimal physical bisection of the compute node cluster, normalized by the number of IO ports per compute node.”

A selection of measurement data is presented for 448 Gbps pathfinding systems, including the insertion losses of cabled and PCB interconnects. These measurements show a very linear channel response to at least 82 GHz, essential for any scaling metric.

The value of this new metric allows exploring the relative performance advantages of different physical routing architectures early in the pathfinding phase. The author concludes. “The proposed blueprint framework directly addresses this need by overlaying topology-aware co-design that scales bandwidth with node count and IO efficiency as well as link data rate, ensuring that the interconnect does not become the bottleneck in compute-intensive, communication-heavy AI systems.”

High-Speed Digital Interface Characterization Requires New Test Approach by Hiroshi Goto, Anritsu

The author starts with a very good summary of the accelerating performance of various high-speed serial standards such as USB, PCIe, and Ethernet. After highlighting the general difficulty of characterizing the performance of channels at increasing data rates, he focuses on the impact of intra-pair skew. While characterizing the channel is important, stress testing the receiver and the system’s equalization is also important. This is the value of a bit error ratio tester (BERT).

It is possible to generate and drive a differential pair interconnect channel with single-ended p and n signals. This allows precise skew stressing between the lines of the differential pair. The features of this tester are described. He concludes, “The BERT-based setup uses a new method with dual transmitters to control the phase of signals within a differential pair, enabling granular measurement of intra-pair skew. This capability is crucial for understanding and mitigating the impact of skew on BER margins at high data rates such as 64 GT/s.”

Using Ultra-Broadband Baluns to Perform Differential S-Parameter Measurements Using Single-Ended 2-Port VNA, by Madrone Coopwood, Jason Yoho Ph.D., and James R. Andrews Ph.D. of Hyperlabs

The authors provide an alternative solution to using an expensive, high-bandwidth 4-port VNA for differential pair characterization, using a 2-port VNA and a high-performance balun. What enables this low-cost alternative is the design of an ultra-wideband balun. Their clever design uses a power splitter for the p signal and an inverter with a matched delay for the n-signal. One of these baluns connects to port 1 and another to port 2 of a VNA to turn it into a 4-port differential VNA. They illustrate this method using an Anritsu 67 GHz VNA.

The 2-port VNA is calibrated as a 2-port VNA with SOLT standards connected to the ends of the differential ports of the two baluns. The differential ports were each uncoupled so just using open, short and 50 ohm loads worked well.

As a final test of the system, the authors compared the measured S-parameters of the same amplifier using a true 4-port system and this 2-port differential system. The agreement between these two measurement systems is excellent, in most cases within less than 1 dB of each other.

Material-Induced Skew in High-Speed Multilayer PCBs: Influences and Mitigation Strategies, by Bob Nurmi, John Coonrod, and Vitali Judin, Rogers.

Glass weave-induced skew has been an important issue in all high-speed serial link designs operating above 10 Gbps. While rotating the artwork to the weave axis or using zig-zag routing solves this problem, they are inefficient at board utilization.

The problem arises due to the difference in Dk the p-line sees compared to the n-line in a differential pair. This is due to the typically higher Dk of the glass yarn compared to the resin. One solution is to use a lower Dk glass yarn to “index match” the glass and the resin.

The authors present an alternative approach using a resin that has high Dk ceramic fillers added. This takes the other approach of raising the Dk of the resin to index match the resin to the higher Dk of the glass.

Using the ceramic-filled resin, they demonstrated they can reduce the glass weave-induced skew by a factor of 20 over just using unfilled resin for a variety of open-weave glass fabrics. While the authors point out that laminates with no glass yarn have virtually no skew, the glass fiber is needed for dimensional stability in multilayer boards. The use of ceramic-filled resin combines the benefit of using conventional glass weave, no requirement for special routing orientation, and a dramatic reduction in glass weave skew for systems operating at 448 Gbps.

The Imperfect Via: The Rough Truth Lurks Beneath the Surface, by Bert Simonovich, Lamsim Enterprises

The topic of anisotropic material properties has been discussed for more than 10 years. There was no doubt that the Dk of a glass weave laminate has a different Dk in the x-y or in-plane direction than the Z-direction, or out-of-plane direction. There are two important questions: how much is this anisotropic Dk, and so what? what is the impact this has on performance?

While it is possible to include an anisotropic Dk in a simulation, it is challenging to measure it, to confirm its value. Because signal traces are more sensitive to the out-of-plane Dk value, this is what is usually measured.

A via or via stub is the most sensitive structure to the in-plane or x-y Dk value. The effective Dk can be measured from the quarter-wave stub resonant frequency of a via stub. While the resonant frequency can be measured to better than 0.1%, interpreting this in terms of the anisotropic Dk value is difficult. The resonant frequency depends on the stub length, the precise dimensions of the via hole, the capture pads, the clearance holes and the alignment of the hole and the pads.

In addition, the author points out that the copper roughness and copper wicking into the drilled side wall also affect the effective Dk. He deconstructs the contributions to the resonant frequency from all the known parameters based on cross-section information in the measured structure. He concludes that the anisotropy term increases from 4.8%, just using the bulk Dk value, to a 12.5% anisotropy after including the pressed thickness and the via roughness. What is really pointed out in this analysis is that uncertainty in manufacturing variations in every feature can easily mask any effect from anisotropic Dk.

Be sure to download and review your copy of the October issue of the SI Journal.

— 30 —

]]>