NimbusNimbus
PersonalizerStudioMCPTeamDocs
Sign InSalesContact SalesGet Started
← Back to all posts

Alpha You Can See: Checking Your EEG Signal Before You Trust It

By Nimbus Team·November 17, 2026
Key Takeaway

Run a live EEG signal check in the browser: per-channel contact verdicts, 20-bit range, dry-electrode offsets, and the eyes-closed alpha rhythm at 2.5-3x - see your brain signal before you build anything.

The first thing everyone does after putting on an EEG headset is squint at the raw trace and ask the same question: is this my brain, or is this noise? Every first-time BCI user needs a good answer before they will trust anything built on top. So we made the answer a first-class screen: connect a device, and Studio shows you its live signal with a per-channel verdict — before you build anything.

The check anyone can run

Open My Devices, connect your headset straight in the browser (no install — see our recent post on browser-direct hardware), and the Signal Check panel goes live: four traces scrolling in real time, each with its own verdict. Green means good contact. If a channel needs attention, the screen says so in plain words — check the fit on that side, moisten the electrodes, adjust the strap.

The whole flow takes under a minute, and it catches the one problem that ruins more BCI sessions than any algorithm ever will: bad electrode contact. Hair, dry skin, a band worn too loose — the check sees it before it wastes your calibration.

Why raw EEG looks so jumpy

A dry electrode sitting on skin carries a standing contact offset — often a couple hundred millivolts, drifting slowly as the contact settles. On the BrainBit Headband we measured offsets around 270 mV on dry electrodes: most of the way to the edge of the input range.

That is why the input range matters. This headset digitizes at 20-bit resolution across a ±400 mV range — sub-microvolt steps, with more than enough headroom that a settling contact never bumps against the ceiling before your brainwaves, which live in the tens of microvolts, even show up. A narrower input would flatten the interesting part of the signal entirely.

The standing offset itself is not signal, so it is removed before samples ever reach the pipeline: filters, quality checks, and models all start from the brainwave part of the trace, not the drift. And the check stays honest about contact quality: it wants to see each channel settle — under roughly half a millivolt of wandering on moistened electrodes, in our measurements — before it calls the contact good.

The alpha moment

Here is the payoff, and it is something you can see with your own eyes. Alpha is the rhythmic brainwave that appears at the back of the head when you close your eyes — it is the classic first signal in EEG, described almost a century ago.

Close your eyes during the check and the traces at the occipital sites (O1 and O2 in the standard naming) rise sharply: in our sessions, two and a half to three times their eyes-open level, sweeping up in unison. Open them, and it falls back. Nobody has to take our word for anything — the rhythm is on the screen, and it answers when you do.

That single moment converts a squiggle into a signal. It is your brain, verifiably, in a browser tab.

From green check to good decoder

Everything downstream depends on this moment. Calibration starts only after the check passes, because a decoder trained on channels with bad contact learns the contact, not the brain. With the check green, the tens-of-microvolts rhythms you saw respond to your eyes are the same signal your model will train on.

This is also why we put the check before everything else in the flow — before building pipelines, before games, before research sessions. Trust the input first; then everything downstream has a fighting chance.

Try it

Open studio.nimbusbci.com, go to My Devices, connect your headset, and run the live check — eyes open, eyes closed, and watch the alpha answer. We put a short video of the whole journey — connect, pair, first live signal — in our post on connecting EEG hardware in the browser.

Build with Nimbus

Ready to run real-time Bayesian BCI?

Design and simulate pipelines visually in Nimbus Studio, or install the sklearn-compatible Python SDK for sub-20ms local inference.

Try Nimbus Studio Free →Nimbus PersonalizerMCP for AI Agents

Related Articles

November 16, 2026

No Install, No Middleware: Connecting EEG Hardware Directly in the Browser

Twelve EEG device types connect straight from a web page: Web Bluetooth for BLE headsets (BrainBit family, Muse, OpenBCI Ganglion), Web Serial for the OpenBCI Cyton dongle. The browser is the driver.

November 3, 2026

Tokenizing the Brain: Vector-Quantized EEG Representations with NeuroRVQ

How residual vector quantization turns continuous EEG representations into discrete token sequences — and what NeuroRVQ's multi-scale tokenizer means for generative brainwave models and downstream BCI research.

Nimbus BCI

Stop writing boilerplate. Start publishing papers. Built by researchers, for researchers.

LinkedInXFeatured on Better Launch
Navigation
ProductPersonalizerStudioMCP ServerDeviceTeamDocsPaperResources
Nimbus Studio
Download for WindowsDownload for macOSComparisonFeaturesPricingFAQ
© 2026 Nimbus BCI Inc. All rights reserved.
Get startedSign inPrivacyTermsCookies