About this project

This one wasn’t planned. It came out of sheer frustration while helping run labs for COGS 189: Brain-Computer Interfaces at UC San Diego. One of our core tools—the OpenBCI Cyton board—just... kept disconnecting. At the worst possible moments. Mid-recording. Mid-demo. Mid-everything.

The problem...

The OpenBCI Cyton is an 8-channel board used to collect neural signals. It’s built to be simple and safe—it sends data wirelessly via a USB dongle, keeping your laptop safely separated from the participant’s body. Good in theory. But in practice?

It kept losing connection.

And not just a little bit. Random disconnections, corrupted data streams, boards hopping between dongles during a session, chaos if more than one group was recording at once.

Time to investigate

Since Cyton is open source (bless), I cracked open the firmware and started digging. It turns out… the issue wasn’t random at all.

There’s a chunk of firmware that makes the board constantly look for a stronger dongle to connect to—even in the middle of data collection. Which means if you’ve got multiple dongles in the same room (like, say, a classroom), the boards will happily switch horses mid-race.

Even worse, turning them on in order or spacing them out didn’t help. They still switched during recording. And that’s when it clicked: this wasn’t user error. It was firmware behavior.

The solution: a physical one

Since rewriting the firmware for every lab station wasn’t scalable, I took a different route: electromagnetic isolation.

I ended up designing a little 3D-printable enclosure for the Cyton dongles:

• Lined the inside with copper tape to block unwanted signals
• Covered it in metal mesh (from Amazon) to allow heat to escape while still shielding from EM interference
• Left intentional vents and gaps for airflow to avoid overheating

Basically: a Faraday cage, but DIY.

Why mesh?

Originally, I thought about enclosing it fully—but Cytons can run warm. The mesh allowed us to keep signal shielding while letting heat escape. Plus, it was cheap, easy to install, and kind of looked cool in a rugged cyberpunk way.

Final thoughts

This wasn’t a project I expected to do, but it ended up solving a very real lab problem. Now our Cytons stay locked to their dongles, we have cleaner recordings, and our students aren’t pulling their hair out mid-lab.

It’s a good reminder that sometimes, the fix isn’t in the code—it’s in a bit of copper tape and creative problem-solving.

Let me know if you want pictures or STL files—I’ve got those too.