

DevBoard
Open Source EEG project
- Size
- 134 × 127 mm
- Copper layers
- 2
- Thickness
- 1.6 mm
- Parts
- 20
- Nets
- 50
- Pads
- 205
- Tracks, vias
- 348, 6
- Routed
- Yes
- KiCad
- KiCad 6 (file 20211014)
- Created
- 22 Apr 2021
- Updated
- 31 Jul 2022
Overview
DevBoard is an open-source printed circuit board by VolksEEG, from the VolksEEG/VolksEEG repository on GitHub, published under the GPL-3.0 licence. It is a 2-layer board of 134 × 127 mm with 20 parts on 50 nets, routed with 348 tracks and 6 vias.
Its parts include 6 connectors, 6 switches and 4 ICs. It belongs with the microcontroller boards boards and the nRF52 and nRF5x boards of this library.
README
README.mdVolksEEG: Acheiving the Promise of Open Source Medical Devices
In a nutshell...
-
The VolksEEG project is creating an open-source electroencephalogram (EEG) machine, fully-functional for clinical or research use.
-
The final device will obtain clearance by the USA FDA, and CE marking (the equivalent of FDA clearance in the EU).
-
Currently, no other open-source electronic medical device (that we know of) has full regulatory clearance
-
All hardware and software designs will be freely available (free as in speech and as in beer).
VolksEEG will be the first in a line of open-source medical devices that will enable a quantum leap in device safety, cost reduction, and innovation.
To sign up for updates about the project, head to volkseeg.org.
Here is a video showing the real-time display of 32 waveform channels, from a UI proof-of-concept application:
https://user-images.githubusercontent.com/58418/132131736-32bca075-f591-47dc-b3fa-c07dddc95de6.mp4
(The final device will display EEG channels rather than sine waves, of course.)
Qs & As
You might have a few questions; we have answers, click on the links below to read the:
Why are open-source medical devices a good idea?
Open-source medical devices will provide increased safety, increased innovation, and reduced device costs.
Increased Safety
Like all devices, medical devices have become much more complex over time. Unlike all devices, medical devices are often safety- or life-critical.
As complexity increases, it becomes ever easier to overlook small design flaws during development, often leading to recalls later – each year, hundreds of thousands of devices are subject to Class I recalls, the FDA’s most serious class of recall, reserved for issues that could cause serious injury or death.
Design flaws are detected by design reviews. A core challenge of medical device safety is ensuring the adequacy of design reviews – both in the person-hours spent, and the expertise of the reviewers. While design reviews of proprietary medical devices typically involve only a handful of reviewers, open-source devices can leverage the eyeballs of myriad reviewers with a wide variety of expertise; as stated in The Cathedral and the Bazaar, "given enough eyeballs, all bugs are shallow".
Increased Innovation
Innovation in medical devices is painstakingly slow – compared to devices from decades ago, most of today’s devices may have more convenience features, but by-and-large their core functionality is the same as it was decades ago.
There are several structural issues that impede device innovation. One of these is the barrier between device companies and clinicians/researchers. By building proprietary systems using a conventional profit-only model, manufacturers make it difficult or impossible for researchers to try new techniques and technologies that could lead to breakthroughs.
Open-source designs will accelerate innovative research in hospitals and universities. Investigators will be able to access data from devices, and even make changes to devices to test new concepts. (To ensure safety, the FDA and all hospitals have strict rules around how altered devices can be used.)
In fact... VolksEEG will contain essentially the same circuitry as will be used in our clinical EEG, and will directly fill a currently-unmet need for research into making anesthesia safer.
Reduced Device Costs
Medical devices are expensive. There are various reasons for this – some of these contribute to better patient care, but many don’t. A simple illustration of this can be seen in the title of a recent journal article Prices for Cardiac Implant Devices May Be Up To Six Times Higher In The US Than In Some European Countries[1]. Frankly, if device companies couldn’t profitably sell devices for one-sixth the cost… they wouldn’t.
Open-source designs will democratize medical devices, resulting in substantially lower prices. While relatively few companies have the ability to design devices which meet regulatory standards, many more can manufacture them. A verified design can be adopted by multiple vendors who otherwise would not be able to develop devices, and competition reduces pricing. Open-source designs will also enable poorer countries and NGOs to manufacture their own devices, cutting out costs added by middle-men.
Do You Actually Know What You Are Doing?
The founding members of VolksEEG are:
-
Bryan Glezerson, a neurosurgical anesthesiologist (Bryan add stuff here).
-
Alan Cohen, an engineer with decades of experience in medical device development (including obtaining FDA clearance and CE marking). Alan is also the author of the book ProtoType to Product: A Practical Guide for Getting to Market.
VolksEEG development will result in a safe design which conforms to US and EU regulations – no different than any other commercial medical device. Project participants will gain some knowledge and experience in the medical device design/development processes.
Why start with an EEG machine?
There are two major reasons for an EEG machine being our first product:
Immediate Need
(Bryan writes this)
Low-hanging fruit
An EEG machine is relatively simple compared to many other medical devices, making it a good candidate for a first effort in open-source devices:
-
Chips are available, e.g. the ADS1299 that we’re using, which are purpose-built for EEG use; they are essentially EEGs-on-a-chip that greatly reduce the effort needed to develop a full device. These components are at the heart of many proprietary EEGs.
-
EEG machines are a lower safety risk than other sophisticated medical devices, such as vital signs monitors and infusion pumps. A fundamental principal of medical device development is that effort should be based on risk. At the start of a project, potential risks are identified then addressed in development. Fewer potential risks translates into less effort to assure safety.
What needs to be done?
VolksEEG development will have several phases. The first will result in a “protoduction”[2] device to be used for EEG research. The second will result in a reference design for a full, FDA-clearable device.
A basic block diagram of the prototype system architecture is below:
Other relevant information for nerds ☺️:
-
MCU is a nrf52840. Firmware is currently written in Arduino, but we’ve not yet decided what it will ultimately be. C? C++? Arduino? Mbed? Also we might want to add an RTOS. Or we might not.
-
The MCU and electronics support Bluetooth and NFC – we’re not particularly planning to use these initially, but we like the idea of being able to enable it later by flashing a firmware update.
-
The SD card will contain EEG waveforms as standard *.edf files.For test puposes these files can be read over USB and will appear to the PC application as "real" EEG waveforms.
-
The PC software is being developed in C#, a language that makes good tradeoffs between performance and software reliability.
How can I help?
Development of this prototype breaks down into the following major tasks:
-
Electronics: this is currently staffed
-
Firmware: we can use some help
-
Software: Several proof-of-concept applications have been written to demonstrate concepts such as reading the serial port at sufficiently-high speeds, writing 32 real-time waveforms to the screen in C#/Windows, and so forth. The next step is to build the Windows-based spplication software using best practices, and we would definitely appreciate some help with this.
-
Cybersecurity. We could use expertise/help, mainly on the embedded side of things.
-
Mechanical: designing and fabricating enclosures. This will be fairly low effort to start, probably 3D printed. Later we'll likely use injection molding.
-
Regulatory: we have this reasonably under control, but welcome experienced help.
-
Quality: this activity will ramp up over time, and we can use some help in writing documentation.
On the software, writing code work we can use a hand in:
-
Working with doctors and technicians to understand and document workflow so our software can work the way our customers want.
-
UI/UxD design.
Some non-engineering help we could use:
- Web site development
- Social media
- Fundraising
- Business development
- Legal
- Business issues
- IP issues
Where are the detailed requirements and specs?
The short story is that our initial effort will be a highly-functional prototype, not a medical device.
When a medical device is cleared by the FDA, it's required that there be detailed requirements and specifications, and that development and testing has proceeded deliberately to meet those requirements and specifications.
Medical device development uses relatively-heavyweight development processes (i.e., there's a good bit of documentation, traceability matrices, etc) that the FDA generally refers to as design controls. If development only proceeds after writing and reviewing formal requirements/architecture/specs etc., we run the risk of finding technical surprises that will cause us to back up and update those requirements/architecture/specs that we worked so hard on. So, to avoid wasted effort in creating requirements/architecture/specs that might get changed later, it often makes sense to start by building a fairly-functional prototype without using full medical-grade design controls, then once that's working to circle back and redo the process using design controls. This usually ends up reusing the designs produced during the prototype build, but now we document and otherwise control them properly.
Endnotes
-
Wenzl, Martin, and Elias Mossialos. “Prices For Cardiac Implant Devices May Be Up To Six Times Higher In The US Than In Some European Countries.” Health Affairs, vol. 37, no. 10, Oct. 2018, pp. 1570–77, doi:10.1377/hlthaff.2017.1367.
-
Protoduction refers to a device that is somewhere between prototype and production devices: far enough along that it can be used for its intended use, but not a fully-developed commercial product. Production quantities are typically low, in the ones or tens.
For more information on VolksEEG (beyond this readme)
Parts
20| Qty | Value | Footprint | References |
|---|---|---|---|
| 5 | Tact-Switch | VolksEEG:Sparkfun Tactile Switch | SW1, SW2, SW3, SW4, SW5 |
| 2 | Conn_02x06_Odd_Even | Connector_PinHeader_2.54mm:PinHeader_2x06_P2.54mm_Vertical | J1, J2 |
| 1 | ADS1299EEGFE-PDK | VolksEEG:ADS1299EEGFE-PDK | U3 |
| 1 | Adafruit-4650-FeatherWing-128x64 | VolksEEG:nRF52840_Feather_Sense | U4 |
| 1 | Adafruit_SD_Breakout | VolksEEG:Adafruit_SD_Breakout | U1 |
| 1 | Conn_01x07 | Connector_PinHeader_2.54mm:PinHeader_1x07_P2.54mm_Vertical | J5 |
| 1 | Conn_01x10 | Connector_PinHeader_2.54mm:PinHeader_1x10_P2.54mm_Vertical | J4 |
| 1 | Conn_02x01 | Connector_PinHeader_2.54mm:PinHeader_1x02_P2.54mm_Vertical | J3 |
| 1 | Conn_02x06_Odd_Even | VolksEEG:6-Jumper Header | J6 |
| 1 | Reset Switch | VolksEEG:Sparkfun Tactile Switch | SW6 |
| 1 | nRF52840-Feather-Sense | VolksEEG:nRF52840_Feather_Sense | U2 |
The 16 most common of 20 parts.
More from VolksEEG/VolksEEG
1- 4 layers

Similar boards
- 2 layers66.9 × 74.5 mm

- 2 layers86.4 × 64.8 mm

- 2 layers36.2 × 26.8 mm

Esp32 minimal
atopile/atopile
Design circuit boards with code! ✨ Get software-like design reuse 🚀, validation, version control and collaboration in hardware; starting with electronics ⚡️
20 parts4kMIT - 2 layers108.1 × 393.8 mm

Sweep compact ble
davidphilipbarr/Sweep
Sweep - a small promicro based keyboard inspired by the Ferris.
20 parts2.1k - 2 layers26 × 15.9 mm

TPS54340 Buck Regulator
rusefi/rusefi
rusefi - GPL internal combustion engine control unit
20 parts1.2k - 2 layers67.6 × 30.5 mm

Rgb to hdmi
hoglet67/RGBtoHDMI
Bare-metal Raspberry Pi project that provides pixel-perfect sampling of Retro Computer RGB/YUV video and conversion to HDMI
20 parts1kGPL-3.0 - 2 layers44.3 × 28.7 mm

Lin bus adapter esp32
hackclub/OnBoard
💡 Join 1,000 teenagers and make your first PCB with a $100 grant!
20 parts915 - 2 layers89.5 × 61.6 mm

FreeDeck TH 2020-02
FreeYourStream/freedeck-hardware
all the stuff you need to build your own FreeDeck
20 parts772
Questions
How big is the DevBoard PCB?
It measures 134 × 127 mm and has 2 copper layers, 1.6 mm thick.
What parts does the DevBoard use?
20 parts. The full list of its parts is on this page, with links to each part's symbol and footprint where PCBKit's library has them.
Where can I download the DevBoard KiCad files?
From the VolksEEG/VolksEEG repository on GitHub, at the commit this page was made from: the board is hardware/DevBoard-old/DevBoard.kicad_pcb and its schematic hardware/DevBoard-old/DevBoard.kicad_sch. PCBKit opens the board in the browser, with nothing to install.
Can I use the DevBoard design in my own project?
Its repository publishes it under the GPL-3.0 licence: you may reuse it on that licence's terms (keep the author's notice, and for a copyleft licence share your changes alike).