Posts tagged 'EEG'RSS Feed
It's our birthday!
Doesn't time fly by? A lot has happened in four years...
Headline: EEG Dev Kit Nearly Ready
Sub-headline: EEG D.K. Will ship in July
Boot-loader issues are finally behind us. Firmware is being finished now. Some small cuts to get this out the door are likely. Update process is working smoothly so cut functionality will be rolled out remotely in due course. Its a monster of a device with so many sensors, settings and features.
Next, we are thrilled to tell you about a new suite of EEG software to help you get the most from your EEG hardware.
Focus have created a data processing platform that lets you manipulate the raw brain data into something more meaningful. You can apply filters, Fast-Fourier Transforms (FFT's) and visualize alpha, beta and delta bands. We call it focus EEG.ai
The name is a little previous, but the techniques we are developing to determine sleep stages, emotional response and focus use machine learning. So it's not ai yet, but it will be soon.
As a token of apology, all Early Bird 1st Batch customers will receive 6 months free access to foc.us EEG.ai.
The second batch of Dev Kits are selling fast at only
$999 $899 so don't miss out if you're on the fence.
Hardware < Firmware < Software
Anyone who has updated the firmware on their v2 will probably agree that it was not a fun process. Part of the problem is that bluetooth low energy has very small data transfer rates and a 300k firmware means a lot of 251byte packets. Drop any one and the process is dead.
So we wanted to make sure firmware updating for EEG DK was butter smooth and simple. It should be over-the-air and pain-free.
Frustratingly it is not yet and that is the cause of the delay. The devices are all sat here waiting to ship, but we don't want to give you something that is hard to update.
I am very sorry for the continued delay but it won't be much longer now. As soon as they can update themselves they will be shipped. We will also share some good news about some free stuff we will offer.
Here's a quick look at the final stages of the focus EEG Dry Active Electrode production.
We clean the pad area of the active PCB inside the electrode.
Then we attach the conductive silicone to the electrode.
Finished! Ready to box and ship.
Here is a look at the foc.us EEG Dev Kit production process.
First, we X-ray the EEG main PCB.
The X-ray let's us check the micro-via's and PCB layers. The dot array shown is one of seven BGA components. The pitch from dot to dot is 0.65mm.
Solder is applied to the PCBs using a stencil.
The components are then loaded onto the PCBA pick and place machine.
The pick and place machine is essential when using 0201 components that are just 0.125mm wide.
A close-up of the PCB inside the pick and place machine. The white "X" pcb's are NG (No Good) and failed probe testing. The machine is set to ignore NG PCBs.
Once all of the components are placed onto each of the PCBs on the panel the PCB moves into the oven, and we wait. Patiently.
Until the cooking is finished and they appear, oven fresh!
Then its back in the X-ray machine to check component alignments.
Each panel is then broken up into individual PCBs. Each PCB is then flashed and tested. We have no photographs of these stages because they were performed by the photographer.
The main PCB is then inserted in the plastic case and connected to the sockets PCB ready for final testing.
Final testing ensures that the device is functional and ready to ship. We have fully tested each and every EEG and they are now ready to ship. Monday and Tuesday is a holiday in China so they will ship to London on Wednesday.
DHL tracking numbers will be created on Friday and all EEG orders will be shipped as fast as we can physically manage.
The first production run for foc.us EEG Dev Kit is now entering the third week. At times it has felt like a Sisyphean feat but Monday always brings new optimism and it feels like we are nearly finished.
The active electrodes are custom cables with a 3.5mm 4 pin audio socket.
Each cable is soldered to two active electrode PCBs. Pin 1 is the shorter length, Pin 2 is the longer and pins 3 and 4 are 5v power lines.
This process is repeated by everyone on the small production line.
There are two PCBs inside the EEG Dev Kit. The main PCB as shown in a previous blog post and the "sockets" pcb shown above. The sockets PCB is relatively simple to assemble and test, so we let a Monkey do it.
The original estimated ship date was March. Obviously we failed to hit this date for a bunch of reasons, some good (extra functionality), some bad(under-estimating testing time).
We are thinking about what gift we can give to make up for these delays.
Further updates to follow.
At foc.us we like to make our products as small as possible.
Small products start life as small printed circuit boards - PCBs.
We use every possible trick to minimize the surface area of the PCB.
For example we use 0201 components. These components are tiny at only 0.25mm by 0.125mm.
Wherever possible we use BGA packages. BGA is an acronym for Ball Gate Array. The bottom of a BGA component is covered with an array of tiny copper spheres (balls).
These are two ways to save space but they come with a price tag.
BGA components can only be tested using an x-ray microscope. This is a little specialist and a little slow.
0201 components take a little longer to test because of their small size.
The focus EEG Dev Kit has ~400 components to test. We are testing as fast as we can.
The EEG active electrodes by comparison are much easier to assemble and test with only four components.
We think the benefit of a small product is worth the price of some harder testing.
Tuesday update on the EEG Dev Kit testing.
This is how we test the foc.us EEG Dev Kt PCB. We use a Netech Minisim 330 EEG Simulator to generate a controlled test EEG signal.
This is fed through our prototype boards either directly (as shown) or via the electrodes. It is possible to then compare the downstream output from the device with the known input and calculate noise levels, catch any problems. Good news is there are no problems.
This is a comparison of the output from passive (top) vs active electrodes after FFT.
Our tests show a peak to peak noise of 2μV.
You can also see how the active electrodes are practically immune to the 50Hz mains noise. The big spike at 1Hz is the DC element. This can be filtered out if desired.
These are probably the two most important tests for an EEG device, but they are not the only tests.
We have to test a whole lot of other stuff to make sure that everything works. The fNIRS, the wi-fi, the microphone, the storage, the charging, the 6-axis inertial sensor, the tDCS/tACS/tRNS/tPCS, the IoT.
There are a lot of components, there are a lot of circuits and there is a lot of firmware. This means there are a lot of tests.
It also means it's one hell of a device.
The focus dry active electrodes bundled with the foc.us EEG Dev Kit appear to be a very popular inclusion. Not surprising knowing how messy conductive paste can be.
A lot of you have asked if they can be purchased separately and as of today you can order them here.
They require a 5V DC supply on pins 3 & 4 to power the internal amplifier so a female socket is included to make the connection and setup (relatively) easier.
Each cable has two electrodes so can be used as either two EEG channels or as a single bio-differential channel.
Each focus EEG Dev Kit has up to 8 EEG channels, but did you know you can combine multiple devices for higher channel counts?
By connecting devices with our daisy-chain connectors, the clocks are synced and a reference electrode sshared. With two devices that's 16 channels. With five that's 40!