Homebrew Network Card With No CPU

April 10, 2024 by Al Williams 16 Comments

A modern normal network card will have onboard an Ethernet controller which, of course, is a pre-programmed microcontroller. Not only does it do the things required to keep a computer on the network, it can even save the primary CPU from having to do certain common tasks required for communicating. But not [Ivan’s]. His homebrew computer — comprised of 7 colorful PCBs — now has an eighth card. You guessed it. That card connects to 10BASE-T Ethernet.

There’s not a microcontroller in sight, although there are RAM chips. Everything else is logic gates, flip flops, and counters. There are a few other function chips, but nothing too large. Does it work? Yes. Is it fast? Um…well, no.

He can ping others on the network with an 85 ms round trip and serve web pages from his homebrew computer at about 2.6 kB/s. But speed wasn’t the goal here and the end result is quite impressive. He even ported a C compiler to his CPU so he could compile uIP, a networking stack, avoiding the problems of writing his own from scratch.

Some compromises had to be made. The host computer has to do things you normally expect a network card to do. The MTU is 1024 bytes (instead of the more common 1500 bytes, but TCP/IP is made to expect different MTU sizes, which used to be more common when more network interfaces looked like this one).

Even on an FPGA, these days, you are more likely to grab some “IP” to do your Ethernet controller. Rolling your own from general logic is amazing, and — honestly — the design is simpler than we would have guessed. If you check out [Ivan]’s blog, you can find articles on the CPU design, its ALU, and even a VGA video card all from discrete logic. The whole design, including the network card is up on GitHub.

We love the idea of building a whole computer system soup to nuts. We wish we had the time. If you need a refresher on what’s really happening with Ethernet, our [Arya Voronova] can help.

How DEC’s LANBridge 100 Gave Ethernet A Fighting Chance

April 9, 2024 by Maya Posch 17 Comments

Alan Kirby (left) and Mark Kempf with the LANBridge 100, serial number 0001. (Credit: Alan Kirby)

When Ethernet was originally envisioned, it would use a common, shared medium (the ‘Ether’ part), with transmitting and collision resolution handled by the carrier sense multiple access with collision detection (CSMA/CD) method. While effective and cheap, this limited Ethernet to a 1.5 km cable run and 10 Mb/s transfer rate. As [Alan Kirby] worked at Digital Equipment Corp. (DEC) in the 1980s and 1990s, he saw how competing network technologies including Fiber Distributed Data Interface (FDDI) – that DEC also worked on – threatened to extinguish Ethernet despite these alternatives being more expensive. The solution here would be store-and-forward switching, [Alan] figured.

After teaming up with Mark Kempf, both engineers managed to convince DEC management to give them a chance to develop such a switch for Ethernet, which turned into the LANBridge 100. As a so-called ‘learning bridge’, it operated on Layer 2 of the network stack, learning the MAC addresses of the connected systems and forwarding only those packets that were relevant for the other network. This instantly prevented collisions between thus connected networks, allowed for long (fiber) runs between bridges and would be the beginning of the transformation of Ethernet as a shared medium (like WiFi today) into a star topology network, with each connected system getting its very own Ethernet cable to a dedicated switch port.

Ethernet For Hackers: Transformers, MACs And PHYs

March 7, 2024 by Arya Voronova 21 Comments

We’ve talked about Ethernet basics, and we’ve talked about equipment you will find with Ethernet. However, that’s obviously not all – you also need to know how to add Ethernet to your board and to your microcontroller. Such low-level details are harder to learn casually than the things we talked about previously, but today, we’re going to pick up the slack.

You might also have some very fair questions. What are the black blocks near Ethernet sockets that you generally will see on boards, and why do they look like nothing else you see on circuit boards ever? Why do some boards, like the Raspberry Pi, lack them altogether? What kind of chip do you need if you want to add Ethernet support to a microcontroller, and what might you need if your microcontroller claims to support Ethernet? Let’s talk.

Transformers Make The Data World Turn

One of the Ethernet’s many features is that it’s resilient, and easy to throw around. It’s also galvanically isolated, which means you don’t need a ground connection for a link either – not until you want a shield due to imposed interference, at which point, it might be that you’re pulling cable inside industrial machinery. There are a few tricks to Ethernet, and one such fundamental Ethernet trick is transformers, known as “magnetics” in Ethernet context.

Each pair has to be put through a transformer for the Ethernet port to work properly, as a rule. That’s the black epoxy-covered block you will inevitably see near an Ethernet port in your device. There are two places on the board as far as Ethernet goes – before the transformer, and after the transformer, and they’re treated differently. After the transformer, Ethernet is significantly more resilient to things like ground potential differences, which is how you can wire up two random computers with Ethernet and not even think about things like common mode bias or ground loops, things we must account for in audio, or digital interfaces that haven’t yet gone optical somehow.

Continue reading “Ethernet For Hackers: Transformers, MACs And PHYs” →

Pictures of the internals of the Starlink adapter

Restoring Starlink’s Missing Ethernet Ports

February 28, 2024 by Arya Voronova 29 Comments

Internet connectivity in remote areas can be a challenge, but recently SpaceX’s Starlink has emerged as a viable solution for many spots on the globe — including the Ukrainian frontlines. Unfortunately, in 2021 Starlink released a new version of their hardware, cost-optimized to the point of losing some nice features such as the built-in Ethernet RJ45 (8P8C) port, and their proposed workaround has some fundamental problems to it. [Oleg Kutkov], known for fixing Starlink terminals in wartime conditions, has released three posts on investigating those problems and, in the end, bringing the RJ45 ports back.

Starlink now uses an SPX connector with a proprietary pinout that carries two Ethernet connections at once: one to the Dishy uplink, and another one for LAN, with only the Dishy uplink being used by default. If you want LAN Ethernet connectivity, they’d like you to buy an adapter that plugs in the middle of the Dishy-router connection. Not only is the adapter requirement a bother, especially in a country where shipping is impeded, the SPX connector is also seriously fragile and prone to a few disastrous failure modes, from moisture sensitivity to straight up bad factory soldering.

Continue reading “Restoring Starlink’s Missing Ethernet Ports” →

Ethernet For Hackers: Equipment Exploration

February 28, 2024 by Arya Voronova 28 Comments

Last time, we talked about the surface-level details of Ethernet. They are fundamental to know for Ethernet hacking, but they’re also easy to pick up from bits and pieces online, or just from wiring up a few computers in your home network. Now, there’s also a bunch of equipment and standards that you will want to use with Ethernet – easy to find whether used or new, and typically as easy to work with. Let’s give you a few beacons!

Routers And Switches

Whenever you see a box with a few Ethernet ports, it’s either referred to as a router, or a switch, sometimes people will even use the word “hub”! Fortunately, it’s simpler than it may seem. A router is a smart device, typically with an OS, that ties two or more networks together – routing packers from one network to another, and typically taking care of things like handing out local IP addresses via DHCP. A switch merely helps Ethernet devices exchange packets between each other on the same level – it’s typically nowhere near as smart as a router gets. Oftentimes, a home router will contain a switch inside, so that you can plug in multiple of your home devices at once. That’s the main difference – a switch merely transmits packets between Ethernet-connected devices, while a router is a small computer taking care of packet forwarding between networks and possibly including an Ethernet switch on the side.
Continue reading “Ethernet For Hackers: Equipment Exploration” →

Underwater Sensor Takes Single Pair Ethernet For A Dip

February 27, 2024 by Maya Posch 27 Comments

The 10BASE-T1 Ethernet standard is also known as ‘single pair Ethernet’ (SPE), as it’s most defining feature is the ability to work over a single pair of conductors. Being fairly new, it offers a lot of advantages where replacing existing wiring is difficult, or where the weight of the additional conductors is a concern, such as with the underwater sensor node project that [Michael Orenstein] and [Scott] dreamed up and implemented as part of a design challenge. With just a single twisted pair, this sensor node got access to a full-duplex 10 Mbit connection as well as up to 50 watts of power.

The SPE standards (100BASE-T1, 1000BASE-T1 and NGBASE-T1) 10BASE-T1 can do at least 15 meters (10BASE-T1S), but the 10BASE-T1L variant is rated for at least 1 kilometer. This makes it ideal for a sensor that’s placed well below the water’s surface, while requiring just the single twisted pair cable when adding Power over Data Lines (PoDL). Whereas Power-over-Ethernet (PoE) uses its own dedicated pairs, PoDL piggybacks on the same wires as the data, requiring it to be coupled and decoupled at each end.

Continue reading “Underwater Sensor Takes Single Pair Ethernet For A Dip” →

Ethernet For Hackers: The Very Basics

February 12, 2024 by Arya Voronova 92 Comments

Ethernet is ubiquitous, fast, and simple. You only need two diffpairs (four wires) to establish a 100Mbit link, the hardware is everywhere, you can do Ethernet over long distances easily, and tons of the microcontrollers and SoCs support it, too. Overall, it’s a technology you will be glad to know about, and there’s hundreds of scenarios where you could use it.

If you need to establish a high-bandwidth connection between two Linux boards in your project, or maybe a Linux board and a powerful MCU, maybe make a network between microcontrollers, Ethernet’s your friend. It also scales wonderfully – there’s so much tech around Ethernet, that finding cables, connectors or ICs tends to be dead easy. Plus, the world of Ethernet is huge beyond belief. Ethernet as most of us know it is actually just the consumer-facing versions of Ethernet, and there’s a quite a few fascinating industrial and automotive Ethernet standards that flip many of our Ethernet assumptions upside down.

Now, you might be missing out on some benefits of Ethernet, or perhaps misunderstanding how Ethernet works at all. What does it mean when a microcontroller datasheet says “has Ethernet interface”? If you see five pins on an SBC and the manufacturer refers to them as “Ethernet”, what do you even do with them? Why does the Raspberry Pi 4 SoC support Ethernet but still requires an extra chip, and what even is GMII? Continue reading “Ethernet For Hackers: The Very Basics” →