unexpected noise on Cyton channels, varies over time

lucasbaldezzarilucasbaldezzari Uruguay
edited August 2021 in Cyton

Hello, how are you? I was looking for in the forum trying to understand why in some cases, the OpenBCI Gui doesn'y show channel activity.
I'm using an ultracortex with 8 channel connected to the Cyton Board but in some cases, there are no voltage variations. I've attached two figures, you can see the the activitiy for 8 channels for one trial -4 seconds-. The first figure called show the eeg for 8 channels coming from the Cyton and the second one show the FFT for channel 6.


The next figure show a 8hz spike -I was watching a flickering led-

Eerythin goes well. Then I stop working, I took off my ultracortex. A few hours later, I put on the ultracortex and test the data using the OpenBCI Gui and there were no voltage variations.
The next figure show the "activity" coming from the Cyton for one trial -4 seconds-.

I don't know what's going on. I tryed changing the ultracortex for gold cup electrodes and I have same results, not voltage variations.

Could you help me, please?

Regards,

Lucas.

P/D: The data for the plots was acquired using BrainFlow.

Comments

  • wjcroftwjcroft Mount Shasta, CA

    Lucas, hi.

    I don't understand your 2nd set of graphs. For the 1st set, signal amplitudes look normal. For the 2nd set, the microvolts range has become huge. Surely the 2nd set of graphs has not filtered out the source of these huge noise signals.

    What do you see with the OpenBCI_GUI when the 2nd case is in effect? Are you using some type of filtering? Recommended would be say a bandpass from .5 Hz to say 45 Hz. And a notch at your mains frequency, I assume 50 Hz.

    Because electromagnetics conditions can change, depending on environmental variables, you always must filter out signals outside the EEG normal range of frequencies.

    Also check the integrity of your Bias / Ground, and Reference (SRB2) electrodes, cables. If any of these come in contact with local metallic surfaces, those conductive circuits can totally garbage the EEG signal.

    William

  • lucasbaldezzarilucasbaldezzari Uruguay
    edited August 2021

    Hello William, how are you? Thank you for your fast reply.
    For the second graph I use the same conditions as the first one, same room conditions, same filter params (bandpass in 5Hz - 38Hz, Notch at 50Hz). On the other hand, some sometimes when checking the EEG channels using the OpenBCI Gui -befor use python and brainflow in order to try a BCI- some channels are flatten -in some ocassions all of them-. Is this due to high electrode impedance?
    I will check the integrity of Bas and SRB2 electrodes cables.

    Thank you so much, William.
    Greetings from Uruguay.

    Lucas

  • wjcroftwjcroft Mount Shasta, CA

    On the other hand, some sometimes when checking the EEG channels using the OpenBCI Gui -before use python and brainflow in order to try a BCI- some channels are flatten -in some occasions all of them-. Is this due to high electrode impedance?

    Generally 'flat' channels, channels with no signal, imply an issue with your connectors, wires, electrodes not making contact. You can use a digital voltmeter / ohmmeter, to continuity check the path from Cyton pin to electrode. While wiggling wires / connectors to determine if you have an intermittent connection.

    You can additionally check one channel's impedance at a time, by using the 'ohm' symbol at the left of each channel's time series graph. Next time you are getting the unreal ~ 200,000 microvolts stuff in Brainflow, instead use the GUI to determine what it is seeing.

    Regards, William

  • lucasbaldezzarilucasbaldezzari Uruguay

    Generally 'flat' channels, channels with no signal, imply an issue with your connectors, wires, electrodes not making contact. You can use a digital voltmeter / ohmmeter, to continuity check the path from Cyton pin to electrode. While wiggling wires / connectors to determine if you have an intermittent connection.

    Great William. I will check in a while.

    You can additionally check one channel's impedance at a time, by using the 'ohm' symbol at the left of each channel's time series graph. Next time you are getting the unreal ~ 200,000 microvolts stuff in Brainflow, instead use the GUI to determine what it is seeing.

    Yeap, befor use my BCI in Brainflow I check the channels state with the OpenBCI Gui.
    I will tell you if the problem is solved in the next hours.

    Thank you so much for your time, William.
    Regards,

    Lucas.

  • lucasbaldezzarilucasbaldezzari Uruguay
    edited October 2021

    Hello, how are you @wjcroft?
    I'm using the ThinkPulse active electrodes with the Cyton board, I can see very nice signals from channels of my interest in the OpenBCI GUI (reducing the gain less than 8x). But, when I recorded the EEG coming from the cyton board using Brainflow API I see a flat line, like the channels were saturated (or railed?), but when I use the gold cup electrodes or the wet headset, the signals are fine.

    Next pic is the signal coming from channel 1. Like you can see, is like the gain was setted 24x (I see the same flat signal when I use the OpenBCI GUI with 24x gain). I think is not a wire connection problem because in the OpenBCI GUI with 1x gain, the signals over channels are fine.

    So, need I reduce the gain before starting the eeg acquistion? If the answer is yes, how can I do this without using the OpenBCI GUI?
    Best regards,

    Lucas

  • wjcroftwjcroft Mount Shasta, CA

    Lucas, hi.

    You can definitely send commands to the Cyton board, via Brainflow, that will allow you to adjust the gain. See,

    https://docs.openbci.com/Cyton/CytonSDK/#channel-setting-commands

    The SDK commands are sent to Cyton, with the config_board function of Brainflow,

    https://www.google.com/search?as_q=config_board&as_sitesearch=openbci.com

    William

  • wjcroftwjcroft Mount Shasta, CA

    https://openbci.com/forum/index.php?p=/discussion/comment/15536/#Comment_15536

  • lucasbaldezzarilucasbaldezzari Uruguay

    Thank you so much, William!!!
    Greetings from Uruguay.

Sign In or Register to comment.

© Vanilla Keystone Theme 2026