Similar Result in each channel (Time Domain and Frequency Domain)

nugroho_budinugroho_budi
edited July 2018 in Cyton
Hi, my friend and I already tried to record data using open BCI. We are new to this kind of research. 
Here are our setup:

System Settings : Windows 8, 64-bit
Open BCI Board : Cyton 32 bit, 8-Channel
openBCI_GUI : 3.3.0 (maybe) - let me check it tomorrow in our computer in the laboratory.


Headware (Ultracortex Mark IV) and channel's connection:
+ N1P-N3P >> not connected 
+ Oz - N4P (spikey electrode)
+ P7 - N5P (spikey electrode)
+ P8 - N5P (spikey electrode)
+ O1 - N7P (spikey electrode)
+ O2 - N8P (spikey electrode)
+ SRB2 - Left Ear-Clip - (BOTTTOM)
+ BIAS - Right Ear-Clip - (BOTTTOM)

Problem:
After recording data in a few minutes, we got something strange about the signal. Signal from Oz, P7, P8, O1, O2 are very similiar in time domain and frequency domain. Although the raw data each channel are different. 

Question:
(a) Was it because we recorded the Oz channel using spikey electrode?
(b) Any hints/guidance/another setting that we missed from the tutorial? 


Troubleshoot that We've done:
- Check cables' continuity using multimeter,
- Check each spikey electrodes' continuity,
- Connecting each spikey electrodes to a Function Generator and display it to the openBCI_GUI. The result, everytime we shift the frequency of sinusoidal signal from the function generator, the "FFT Plot Widget" in openBCI_GUI also displayed at the same frequency. In short, frequency from Function generator = frequency that displayed in "FFT Plot Widget". We also adjusted the amplitude of the sinusoidal signal to 0,005 Vpp.


Here another picture from another recording (Channel 4 upto 8).


Thank you,
nug

P.S.: 
Please excuse any mistakes as English is my second language.

Comments

  • wjcroftwjcroft Mount Shasta, CA
    Nug, hi.

    Usually when you see the same kind of noise on each channel, it means your reference (SRB2) lead is the one that is suspect. This is because the reference lead is in common with all channels. Have you tried swapping your ear clips?

    Regards,

    William

  • nugroho_budinugroho_budi
    Hi, William, thank you for the fast reply.

    i forget to mention that we also measure the connectivity of the ear-clips. we got up to 0.6 - 0.8 Ohm.
    We will try to swap the ear clips, we hope it will works. 

    thank you.

    regards,
    nug

  • wjcroftwjcroft Mount Shasta, CA
    By clicking in the 'ohm' sign column on the GUI, you can check each channel's impedance. But this again is using SRB2 as common reference, so includes that as well.

  • nugroho_budinugroho_budi
    Alright, we will try it. thank you.
  • nugroho_budinugroho_budi
    Hi,

    Today's setup:
    Spikey electrode connected to N1P, N5P, and N6P. 
    Ear-Clips are swapped (as you suggested). White Cable to BIAS Pin, and Black cable to SRB2.

    After we swap the ear clips, we still got a similar data in channel 1, 5, and 6.

    First Trial data and screenshot: https://1drv.ms/f/s!AlKXtcDKkmi2ghgjjJdxJuHYkhkM
    Second Trial data and screenshot: https://1drv.ms/f/s!AlKXtcDKkmi2ghP9oejlOfInpr3Z

    Do you have another suggestion? What should we do?

    Regards,
    nug
  • wjcroftwjcroft Mount Shasta, CA
    The signals you are showing in all the GUI screenshots (current and previous post) are all very noisy. Hundreds of microvolts in amplitude. Normal EEG in only in the tens of microvolts.

    What happens when you go through the steps of the tutorial EXACTLY. Do you get the ECG, EMG, and EEG (alpha waves)? Can you try gold cups and paste?


    Also read through this thread,

  • nugroho_budinugroho_budi
    Hi, 
    We haven't tried to measure ECG and EMG. 
    We don't have any gold cups. and yes, we use the paste when we record the EEG.

    Thank you for your help. we'll come back here if we are stuck. 

    regards,
    nug


  • wjcroftwjcroft Mount Shasta, CA
    Actually paste has been known to corrode / corrupt the dry sensors used with the Ultracortex. I was only suggesting gold cups because that is what the tutorial uses.
  • nugroho_budinugroho_budi
    We will try using the snap electrode, and post the result here asap. 
  • nugroho_budinugroho_budi
    Update:
    using the snap electrodes using channel 1, 2, and 6. Still got the same result.

    Here are the screenshot:

    And Here are the pictures of our setup:

    What should we do now? any suggestion?

    P.S.:
    - is it because of the firmware of the cyton board only 1.0.0? : https://1drv.ms/u/s!AlKXtcDKkmi2giWWMDyXyO0u2J9N
  • wjcroftwjcroft Mount Shasta, CA
    Firmware version should not be affecting the signal quality.

    You have several channels with rather high impedance (136Kohms). You should try to get that down somewhat, as low as you can. How is it that you have some channels at 1K and 4K ohms, but others very high impedance?

    Have you tried running your tests in another lab or area? The noise spikes you see on all channels could be from other radio frequency / microwave gear nearby.

    I don't understand your pictures 4 and 5? I thought you were testing with the electrodes on someone's skin? It wont work just laying them on the table like this.

    Were your Ultracortex tests done with the headset on the head?? Or on the table?

    I still suggest getting a set of cup electrodes and going through the tutorial link I posted previously.


  • nugroho_budinugroho_budi
    Q: I don't understand your pictures 4 and 5? I thought you were testing with the electrodes on someone's skin? It wont work just laying them on the table like this.

    A: Yes, of course, We've done it with the headset on the head. Those pictures only to show electrodes, cyton, and the connection that we used. 

    ----------

    In Indonesia, we kind of have a local amazon/e-bay named tokopedia. this electrode (here) suitable/compatible with open BCI, right? Of course we have to change the connector. 
  • wjcroftwjcroft Mount Shasta, CA
    Yes, any cup electrodes will do. Use them with Ten20 paste.
  • nugroho_budinugroho_budi
    Alright. Thank you for your help, William. Maybe we'll try it next week. Thank you.
  • nugroho_budinugroho_budi
    Hi @wjcroft, 

    we haven't tried to use the gold electrode yet. let's ignore it for a while.

    I am just curious. when we do the previous setup (SRB2 and BIAS, behind left and right ear and then ch 1, 2, 6 in my forehead), why we got the same signal pattern? please see picture 1 and 2 from my previous post. Is it only because the cable? 

    regards,
    nug
  • wjcroftwjcroft Mount Shasta, CA
    Do you notice a correlation between the noise pulse and your eye blinks? If so that is just normal EMG noise from the eye muscles. It is especially strong on the forehead and frontal sites.
  • nugroho_budinugroho_budi
    i think there is no correlation, because i move one of my channel to my neck (channel 1, 2 on my forehead, channel 6 on my neck) and still got the same result: the shape of signal on channel 1,2, and 6 still the same. 

  • wjcroftwjcroft Mount Shasta, CA
    Try the gold cups with paste when you get them, going through each step of the tutorial.

    You either have electrical noise pulses (EMF, EMI) in your environment, or there is something odd going on with your reference lead. Try another location / room / building if you can, away from AC power conduits, wall transformers, wifi, cell towers, etc.
  • nugroho_budinugroho_budi
    Alright, then. Thank you, @wjcroft. 

    nugroho
Sign In or Register to comment.

© Vanilla Keystone Theme 2026