All 8 channels RAILED with headset

tiger-chengtiger-cheng
edited December 2016 in Cyton
I use the 8 channel OpenBCI board, and everything goes well until I wear my headset. Suddenly, 8 channel turned into  "near railed"  then "railed". And all channel share the same wave shape (time scale and frequency scale).

I use the single electrode to ensure if the board work well and it give me really good wave shape but the Vrms is at 130uV. My friend tell me it's not ok since that the proper Vrms should be under 20uV.  (In some cases, Fp1 and Fp2 works well but they are still at several hundreds uV)

Is there anyone who can help me to find out why my board Vrms so high? 

Comments

  • wjcroftwjcroft Mount Shasta, CA
    Do you have some cup electrodes and paste? How does the OpenBCI perform when you follow the steps in the tutorial?


    Is this the Ultracortex you are using? Which model? Do you have your ear lobe connections to Reference (SRB2) and Ground (Bias)?

    Whenever you encounter unusual situations like this, it is always best to simplify your steps as much as possible and try one thing at a time, as in the tutorial.

    William

  • tiger-chengtiger-cheng
    I have the cup electrodes and I test every channel one by one, they are all about 70uV. And I used MARK_III_NOVA. both SRB2 and Bias is connected to my ear lobe. What I am wondering most is that why all 8 channels share the one wave shape.
  • wjcroftwjcroft Mount Shasta, CA
    There is likely something wrong with your Reference connection to SRB2. Are you using the pin closest to the board? Are you saying the RAILED condition goes away with the cup electrodes? 70 uV still seems on the high side.

    Check your wires and connections to the ear lobe clips.

    With the cup electrodes, did you also use a cup electrode at the ear lobes?

  • tiger-chengtiger-cheng
    We use the SRB2 and BIAS (bottom) connected to left ear and right ear, and  N1-N8 connect to 8 different brain zone (with spring electrode). sometimes all just work, but suddenly, all channel become railed.
  • tiger-chengtiger-cheng
    Could you tell me the mechanism of showing railed? 8 channels tend to turn into railed altogether, even some channel show the right wave shape after "railed"
  • wjcroftwjcroft Mount Shasta, CA
    It sounds like your reference lead is intermittent. When it is connected solidly, you get good signals. When it becomes intermittent, all channels will RAIL, as the differential amplifiers have no reference point. Similarly, if a given signal pattern is showing up on all channels simultaneously, it HAS to be an issue with your reference because that is the lead which is shared with all channels.

    Have you tried swapping your left and right clip-on ear electrodes? One of them may be bad / intermittent. Intermittent electrodes usually go bad at the connector points where there are solder or crimp joints. If you have a DVM ohmmeter, you can check the electrode while you are simultaneously wiggling the one then the other end point of the wire. Good electrodes will show low ohms resistance regardless of how the cable is moved.

    William

  • tiger-chengtiger-cheng
    We change a board and finally the data turns into normal range. But new problem occured: when I use OpenBCI_Node_SDK, the board is 8 channels, but when I use "ourBoard.numberOfChannels()" , it shows "16".
  • wjcroftwjcroft Mount Shasta, CA
    I assume you are referring to,


    AJ Keller @pushtheworld is the Node SDK developer.

    Can you also clarify your solution to the RAIL issue? You replaced your OpenBCI 8 channel board with a spare that you had? You are not using the Daisy, correct?

  • pushtheworldpushtheworld
    edited December 2016
    It seems there is a bug wth the daisy detection with the node. Not sure if firmware is not picking it up or what! It's pretty clear to me that if you want daisy you are going to need to explicitly state it, other wise it's 8 channel.

    See issue here: https://github.com/OpenBCI/OpenBCI_NodeJS/issues/131

    I'm going into HQ after I finish my coffee and breakfast and will get a daisy. I'll be testing this later tonight and will release a patch for OpenBCI Node late tonight early morning.

    Why is this in headware haha @wjcroft ??
  • tiger-chengtiger-cheng
    @wjroft The RAIL issue is still there in the old board. I just replace it with a new one. Maybe it's the hardware problem. I am not using Daisy, its just the default 8 channel board.
  • tiger-chengtiger-cheng
    And, I found that the biggest value of sampleNumber is 127. But it should be 255 isn't it?
  • tiger-chengtiger-cheng
    When I use "ourBoard.impedanceTestChannels(['n','N','n','n','n','n','n','n']).catch(err => console.log(err));" it shows “Array length mismatch, should have 16 but array has length 8”
Sign In or Register to comment.

© Vanilla Keystone Theme 2026