My Cyton+Daisy raw data looks broken?

I have two OpenBCI boards. A Cyton and a Cyton+Daisy.
When I look at the data from either board just running Cyton, it looks like what I expected. Here is a screen shot of the raw (unfiltered) data from a Cyton board using BrainVision LSL viewer:

When I hook up the Daisy board however, the raw data looks broken. Here is a screen shot of unfiltered Cyton+Daisy raw data.

Using either Cyton board (without Daisy), I can quite easily see alpha wave behavior in the band power graphs. Using the Cyton+Daisy, I have never seen alpha wave behavior in the band power graph.

I am reading samples from the board using brainflow library.

Is my Daisy board broken? Is there something in the configuration details I missed?

Any help is appreciated. Thanks

Comments

  • wjcroftwjcroft Mount Shasta, CA
    edited December 2020

    Graham, hi.

    First thing that pops out from these screen grabs, is the horrendous amount of mains noise that you have. Do you know about strategies to reduce this?

    Secondly, I would ask you, how do the same board setups look, with the OpenBCI GUI. If it looks fine there, but messed up when streamed through LSL, then I would suspect the LSL streaming has some glitches.

    William

  • William

    Thank you for the prompt reply. Regarding noise, this was just a quick test to get the screen capture showing the odd behavior of the raw data, and there is a lot of static electricity in my office chair ?

    I did use OpenBCI GUI software to record the data as well. You do not see graphs like this in the GUI because of the filtering, however if you look at the raw data it has the same 'nodes' in it. I will plot something in Excel and post it back here.

    It looks suspicious that the raw data is so different for 8 channel (no nodes) and 16 channel using all same conditions otherwise. Also, I have no trouble getting band power spikes at 10 Hz using the 8 channel setup, but when using 16 channels, the band powers are always 'flat'.

    Thanks

    GB

  • wjcroftwjcroft Mount Shasta, CA

    If GUI looks fine, then the artifacts you see are probably the result of the huge and unwieldy amount of mains noise you have. The nodes etc, are in the mains noise signal, as you can see.

  • wjcroftwjcroft Mount Shasta, CA
    edited December 2020

    My guess is that when your Daisy is plugged in, the electrical circuit traces that are now engaged, allow different 'beat' patterns to arise between slightly different EMF noise present in your room environment. Circuit traces and electrodes acting as 'antennas' for EMF noise. Your first priority should be to determine the noise source and separate yourself from it.

  • wjcroftwjcroft Mount Shasta, CA

    It's always best to reduce EMF noise sources as much as possible: lights, extension cords, conduits (walls, floor, ceiling), wall warts, power supplies, transformers, etc. Once you have reduced as much as you can, then any remaining mains noise can be handled with the notch filter at mains freq. But it's not wise to 'only' use a notch, and ignore close EMF sources.

  • William
    Thank you for taking the time to add these additional comments. Things make much more sense now.

    I am working with this equipment in my home office, which is chock full of laptops, mobile devices, lava lamps, routers, wifi hubs, and more Raspberry Pis than you can shake a stick at. I do power the Cyton from an individual battery, but for convenience sake, I have an old 10 foot USB cord that I fixed a JST connecter onto one end that I plug into this battery. This setup is almost certainly acting as a giant antenna for all that RF noise. I will obtain some proper small batteries for the Cyton and conduct some new testing in a quiet room. I will post back here in a while, hopefully to report a nice clean signal with the Cyton+Daisy.

    GB

  • wjcroftwjcroft Mount Shasta, CA

    If you have a phone with a magnetometer chip, there are apps that can act as inexpensive EMF meters. You can learn a lot about mains EMF sources in your immediate area with such meters.

    https://www.google.com/search?q=phones+with+magnetometer
    https://www.google.com/search?q=apps+that+measure+emf+fields

    Of course the EMF spectrum goes way beyond 50 or 60 Hz range. If you are into measuring general EMF present into gigahertz range, this is a good meter:

    https://www.google.com/search?q=cornet+88t+plus

    Generally the mains range is what you should be mostly concerned about. But very strong cellular phone or wifi router signals, can also sometimes contaminate EEG.

Sign In or Register to comment.