Signal quality difference between Cyton and Ganglion

I'm working on a project to read EEG and get the band powers, specifically Alpha and Beta.  I've been using a Ganglion, but have been having very noisy signals.  I switched to a Cyton I had for a previous project and had much much cleaner signals.  Here's my setup: I have four electrodes across the forehead (AF7, Fp1, Fp2, AF8), and two earclip electrodes.

On the Ganglion, the four electrodes are connected to the +1 ... +4 pins and the earclips are attached to D_G and REF.  When holding absolutely still, I'm getting around 100µV on average.  The channel switches are set off, so the channels will use the reference pin, and not the corresponding -1 ... -4 pins.
image
image

Immediately after, I moved the electrodes to the Cyton board.  Here, the four electrodes are connected to the 1P ... 4P pins and the earclips are attached to BIAS and SRB pins.  When connected, I set the SRB1 to "Yes" for the first four channels in the "Hardware Settings" menu.  Now, I'm getting around 3.5µV on average.
image
image

So my question is why is the Cyton board so much better than the Ganglion?  Am I wiring things up correctly?

Thanks,
-Pete

Comments

  • wjcroftwjcroft Mount Shasta, CA
    Pete, hi.

    What could be the source of the huge 40 hz noise in the Ganglion picture? Are you sitting in exactly the same spot as with the Cyton test? Any proximity to noisy electronics, power conduits, extension cords, wall warts, etc., can distort the EEG.

    Has your Ganglion been like this since you received it, or only recently? The 40 hz spike in the FFT is remarkable. That would have been of note immediately the first time you turned it on?

    Curious also why you don't use the default setting with the Cyton, with using all the pins closest to the board? Though your SRB1 IN*P pins will work with the settings adjustment.

    Cyton DOES have somewhat better common mode noise rejection, since it is injecting a counter signal on the Bias lead. Ganglion does not actually have DRL driven right leg / bias, only ground.

    Still with this level of noise, there must be some anomalous source.

    Mentioning Joel @biomurph.

    William

  • wjcroftwjcroft Mount Shasta, CA
    You might also try an impedance check of the electrodes on both Ganglion and Cyton. What type of electrodes? You mention ear clip, so it sounds like FRI with a velcro band?
  • biomurphbiomurph Brooklyn, NY
    @petewall
    This is a known issue. It is happening because you are using a mac. There is something to do with the Bluetooth hardware on the mac.
    We have solved this with the latest GUI v3.3.0, which is in beta. You can download it here
    It will require you to use a BLED112 dongle.
     
  • wjcroftwjcroft Mount Shasta, CA
    Joel @biomurph, thanks wow.

    Do we know which Mac models are affected? It must just be a certain small group? Otherwise it seems there would have been more users mentioning this.

    Regards,

    William

  • @biomurph I downloaded the 3.3.0-beta1 release.  Is this dongle the same one that was bundled with the Ganglion (which says "CSR 4.0" on top)?  I cannot see the Ganglion device now.

    Here is the device information for the USB dongle:
    image
    And here's the error I get when scanning for BLE devices:
    image
    If I click "START SEARCH", it alternates between "Code 412" and "Code 411".
  • wjcroftwjcroft Mount Shasta, CA
    Pete, no, you have to buy the new BLED112 dongle. Not the same as the CSR dongle.
  • wjcroftwjcroft Mount Shasta, CA
    If you have another Windows laptop, the CSR can work with that. Do you have another Mac? This glitch does not affect all Macs.
  • Worth mentioning.  I tried the Ganglion on a Windows machine I have here and it looked a lot like how the Cyton looked on my Mac.  It does appear to be an issue with Macs and Ganglions.
  • @petewall there was a bug in that build that i was unaware would be a problem, please redownload the release and try again!
Sign In or Register to comment.