What Every PCB Designer Needs To Know About Track Impedance With Eric Bogatin

PCB design starts off being a relatively easy affair — you create a rectangular outline, assign some component footprints, run some traces, and dump out some Gerber files to send to the fab. Then as you get more experienced and begin trying harder circuits, dipping into switching power supplies, high speed digital and low noise analog, things get progressively more difficult; and we haven’t even talked about RF or microwave design yet, where things can get just plain weird from the uninitiated viewpoint. [Robert Feranec] is no stranger to such matters, and he’s teamed up with one of leading experts (and one of this scribe’s personal electronics heroes) in signal integrity matters, [Prof. Eric Bogatin] for a deep dive into the how and why of controlled impedance design.

RG58 cable construction. These usually are found in 50 Ω and less commonly these days 75Ω variants

One interesting part of the discussion is why is 50 Ω so prevalent? The answer is firstly historical. Back in the 1930s, coaxial cables needed for radio applications, were designed to minimize transmission loss, using reasonable dimensions and polyethylene insulation, the impedance came out at 50 Ω. Secondarily, when designing PCB traces for a reasonable cost fab, there is a trade-off between power consumption and noise immunity.

As a rule of thumb, lowering the impedance increases noise immunity at the cost of more power consumption, and higher impedance goes the other way. You need to balance this with the resulting trace widths, separation and overall routing density you can tolerate.

Another fun story was when Intel were designing a high speed bus for graphical interfaces, and created a simulation of a typical bus structure and parameterized the physical constants, such as the trace line widths, dielectric thickness, via sizes and so on, that were viable with low-cost PCB fab houses. Then, using a Monte Carlo simulation to run 400,000 simulations, they located the sweet spot. Since the via design compatible with the cheap fab design rules resulted often in a via characteristic impedance that came out quite low, it was recommended to reduce the trace impedance from 100 Ω to 85 Ω differential, rather than try tweak the via geometry to bring it up to match the trace. Fun stuff!

We admit, the video is from the start of the year and very long, but for such important basic concepts in high speed digital design, we think it’s well worth your time. We certainly picked up a couple of useful titbits!

Now we’ve got the PCB construction nailed, why circle back and go check those cables?

Continue reading “What Every PCB Designer Needs To Know About Track Impedance With Eric Bogatin”

Circuit-less PCB Featured As Faceplate For A Digital Clock

If there’s no circuitry on a printed circuit board, does it cease being a “PCB” and perhaps instead become just a “PB”?

Call them what you will, the fact that PCBs have become so cheap and easy to design and fabricate lends them to more creative uses than just acting as the wiring for a project. In this case, [Jeremy Cook] put one to work as the faceplate for his “742 Clock,” a name that plays on the fact that his seven-segment display is 42 mm tall, plus it’s “24/7” backward.

In addition to the actual circuit board that holds the Wemos ESP32 module and the LEDs, a circuit-less board was designed with gaps in the solder mask to act as light pipes. Sandwiched between the boards is a 3D printed mask, to control the light and direct it only through the light pipes. [Jeremy] went through a couple of iterations of diffuser and mask designs, finally coming up with a combination that works well and looks good. He mentions a possible redesign of the faceplate board to include a copper backplane for better opacity, which we think is a good idea. We’d also like to see how different substrates work; would boards of different thickness or using FR-4 with different glass transition temperatures work better? Check out the video below and see what you think.

We’re seeing more and more PCBs turn up as structural elements, from enclosures to control panels and even tools, and we approve of this trend. But what we really approve of is what [Jeremy] did here by making this clock just a dumb display that gets network time over NTP. Would that all three digital clocks in our kitchen did the same thing — maybe then they wouldn’t each be an infuriating minute out of sync with the others.

Continue reading “Circuit-less PCB Featured As Faceplate For A Digital Clock”

Art of 3D printer in the middle of printing a Hackaday Jolly Wrencher logo

Forget The UV Resist Mask: Expose Custom PCBs Directly On Your SLA Printer

For the enterprising hobbyist and prototyping hardware developer, creating custom PCBs remains somewhat of a struggle. Although there are a number of approaches to go about this, they usually involve printing or drawing a mask that is used to expose the photoresist layer on the to-be-etched PCB. Here [Andrew Dickinson]’s Photonic Etcher project provides an intriguing shortcut, by using the UV source of an MSLA 3D printer directly after converting the project’s Gerber files into a format the MSLA printer can work with.

The concept is as simple as can be: since MSLA printers essentially function by creating a dynamically updated UV mask (either via an LCD panel or DLP system), this means that an MSLA printer can be used to expose the PCB’s UV-sensitive photoresistive coating, effectively making the mask there insoluble during the etching step. This can be done with negative as well as positive photoresistive coatings, depending on the use case.

The obvious advantage of this approach is that you don’t need an additional UV source or any kind of separate mask, only an MSLA printer with a large enough work area to fit the PCB you wish to expose. One limitation of [Andrew]’s project at this point is that it can only convert Gerbers to PWMS (Photon Mono) files, but this can presumably be fairly easily extended to support more MSLA printers.

Berlin Clock Takes Inspiration From Sci-Fi Sources, Looks Incredible

What would a HAL9000 look like if it eye were yellow and sat atop a front panel inspired by an Altair 8800? You’d have today’s feature, [Stephan]’s BerlinUhr, a gorgeous little take on a Berlin Clock.

At Hackaday, we have a soft spot for clock builds. They’ve graced our pages from early times. When we saw this ultra cool Berlin Clock, we couldn’t resist the urge to share it with all of our readers.  For those of you not familiar with a Berlin Clock, it’s a clock that consists of 24 lights, and was the first of its kind back in 1975.

[Stephan]’s build is notable because not only is it a beautiful design, but the work that went into the design and build. At several inches tall, the BerlinUhr is supported solely by a USB-C connection, although it can also be hung on a wall. The RTC is backed up by a CR1216, and an ATtiny167 provides the brains for the operation.

A neat part of the build comes with the KPS-3227 light sensor, used to adjust the LED brightness according to ambient lighting. Rather than being a straightforward part to insert into the PCB, KiCad’s footprint had some pins reversed, causing [Stephan] to learn how to correct it and contribute the fix to KiCad. Well done!

We weren’t kidding about clocks, by the way- check out the link to the Atomic Wrist Watch on this post from 2005, and this Russian VFD based clock from 2006- with video!

Do you have your own favorite clock build you’d love to see grace our pages? Be sure to submit a tip!

Panelize PCBs Graphically With Hm-panelizer

When you’re working with PCBs and making single units to knock out in those Chinese fabs, going from layout to manufacturable Gerber files is just a few button presses, no matter what PCB layout tool you prefer. But, once you get into producing sets of PCBs that form a larger system, or are making multiple copies for efficient manufacturing, then you’re not going to get far without delving into the art of PCB panelization. We’ve seen a few options over the years, and here’s yet another one that’s looking quite promising —  hm-panelizer by [halfmarble] is a cross platform Python GUI application, which leverages Kivy, so it should run on pretty well on most major platforms without too much hassle. The tool is early in development, so is restricted to handling only straight PCB edges, with horizontal mouse-bites for now, but we’re sure it will quickly grow more general purpose capabilities given time and support.

In an ideal world, open source tools like KiCAD would have a built-in panelizer, but for now we can dream and hm-panelizer might just be good enough for some people. For more choices on panelizing, checkout our guide to making it easy, and just to muddy the waters here’s another way to do it.

Cityscape Multi-Level Keyboard Really Stacks Up

Keyboards with wells like the Maltron, the Kinesis Advantage family, and everything dactyl-esque out there are great. Trust us, we know this firsthand. But if you want to build your own curvy girl, how the heck can you implement that shape without 3D printing, clever woodworking, or access to tooling and plastic molding equipment? Well, there is another way. Over on twitter (translated) (Threadreader: Japanese, English), [tsukasa_metam] has achieved the key well effect by stacking up PCBs to create a skyline of vertically-staggered keys.

The boards of Cityscape are all screwed together for mechanical integrity, but those screws are working overtime, providing electrical connections between the layers as well. We particularly like that there is an impetus for this build other than ‘I thought of it, so let’s do it’ — [tsukasa_metam] tends to typo in the double key press sense, hitting Q for instance at the same time when A was the intended target. Between the 3.2 mm of key travel, the 2.8 mm step height, and those flat F10 keycaps, that is no longer an issue.

Instead of the popular low-profile Kailh choc switches, [tsukasa_metam] went with TTC KS32s, a new switch introduced in 2020. Unlike chocs, they’ll take Cherry MX-style keycaps, as long as they’re wearing short skirts. Cityscape isn’t totally open source, but the idea is now out there nonetheless, and we happen to have an Odd Inputs and Peculiar Peripherals contest running now through July 4th.

Do stacked PCBs seem kinda familiar? Hey, it’s easier than winding transformer coils.

Via KBD #79

Electronic Catan Game Board Is Modular

Plenty of gamers around these parts require an expensive PC to play games, often spending thousands of dollars for a gaming machine. Believe it or not, though, there are entire classes of games that don’t require any electronics at all, but that doesn’t mean that they don’t benefit from the addition of some neat gadgets. This Settlers of Catan game uses custom LCD tiles with a built-in custom mesh network.

The tiles for the game board themselves are hexagonal and snap together using magnetic pogo pins in order to form a board of any size or shape. The pogo pins also allow communication for a pseudo-mesh network to operate with each tile’s built-in PCB to allow the game board to know exactly which tiles are placed where and to display the correct image on each one. Each tile contains it own RP2040 microcontroller, keeping the overall cost of each tile to a minimum.

For those regularly hosting game night, a project like this could really change the traditionally analog game’s dynamic for the better. It was mostly a project that [Colin Iuliano] built just for fun, and if he ever builds a second one he does plan on some improvements, but we’d say that it looks like a success already. For other Catan-based electronic design inspiration, take a look at this complete and non-modular electronic game board.