How to become "DE - RAILED" ??

jeyemjeyem
edited February 2018 in Headware
****
This thread aims at collecting tips on how to get rid of "railed" channels. (As displayed in OpenBCI_GUI.) Please, anyone, who has experience on that write a comment!
****

The GUI shows "Railed" when microvolt magnitudes for the channels are off the top end of the scale. In other words, generally it means something wrong with the differential voltage measured between the channel and the reference. Either one could have a poor connection leading to "Railed".


Help:
- Check wires.
- Check electrode contacts.
- Check connectors.
- Check board function: Connect SRB, BIAS, and railed Channel using jumpers / y-cables / ... This should give you roughly 0.00 uV. See picture (1).

1: https://drive.google.com/open?id=1VWtcAfGOxWQuRP6JoSkLQtwhT_LGXAmI

*****
I spend roughly 45 minutes getting all channels de-railed. That's quite a bit more than the 30 seconds mounting time in the marketing video. When I released the headset and put it on again, it took me some other 30 minutes to get it run. Certainly, it needs some experience on how to best mount the headset. Could anyone please share his implicit knowledge? My goal is to make it in less than 5 minutes. Thank you.

Comments

  • wjcroftwjcroft Mount Shasta, CA
    Jeyem, thanks for creating this thread.

    I'll add one.

    If you are using a headset, and see "Railed" on more than one channel: suspect your Reference lead, ear clip, connections. You could try swapping with the Bias ear clip lead to see if it makes any difference. However you need good connections on both Reference and Bias.

    The times you state to get "de-railed", 30 or 45 minutes, seem outside normal. Are you seeing just one or a few channels doing this, or many? What are you doing after the rail to get back to normal? Do your impedances look ok? Does the headset behave this way on other people? Etc.

    I know that with QEEG caps (Electro-cap), it is always advised to NOT use conditioner on the hair, as the residue left behind by the conditioner can interfere with clear readings.

    William

  • logon13logon13
    Also takes me 1-2 hours to get everything DeRAILED and does not last long (< 5 mins)
    Greatly appreciate anyones help
  • wjcroftwjcroft Mount Shasta, CA
    @logon13 & @jeyem, hi.

    Can either of you answer the questions I asked in the thread above on Feb 12? Also: When did you purchase your headset? Has it always been like this, or degraded over time, etc.?

    William

  • logon13logon13
    Usually 1 or 2. I re adjust the headset then it is a different one or two. While wearing it one of them quite often derails after awhile (A random one not consistent)

    The headset is new I got it in december

    I will get back to you on impedance when I get home

    Other people wearing the headset have the same results

    Thanks @wjcroft I appreciate the help
  • wjcroftwjcroft Mount Shasta, CA
    Are you adjusting the electrodes that do not make contact, so as to be closer to the head? Do you have full adjustment range or are you hitting either the inside or outside limit? When you get the railing one to work, which one goes out? Use 10-20 or 10-10 labels.


    Are the electrodes that go out consistently the same ones, or do they move around?
  • ConorConor
    Hi @logon13, can you attach a photo of a few of your headset's electrode tips?

    I suspect it has to do with degraded electrodes.

    -Conor
  • logon13logon13
    edited September 2018
    @Conor sorry for the late reply. I got very sick. I would love to chat on skype or discord is you are free at some point
  • logon13logon13
    picture 2. For some reason this forum is glitching out on me a bit

    I've tried running the software on both Windows and Linux
  • ConorConor
    Sorry to hear you got sick. Hope you're feeling better!

    OK. It's hard to tell for sure, but it does look like there is some white (or light gray) discoloration at the tips of the electrode in picture 2. This represents a wearing away of the Ag-AgCl plating. We will soon be selling replacement electrodes from our store. In the meantime, your best place to find replacement electrodes that will fit right away is Florida Research Instruments: 
    https://fri-fl-shop.com/product/package-of-5mm-longer-spike-dry-wet-disposablereusable-electrodes-with-48-inch-lead-wires/

    Other questions I have:
    - How thick is your hair? Certain types of hair (mainly folks who have lots of thick bushy hair) will be very difficult for recording clean EEG
    - Can you send along a screenshot of what the data looks like in the OpenBCI GUI. I take it most of the channels are RAILED but some are not?

    One thing I like to do when using the headset, is:
    1) With the headset on, adjust each node until it is almost uncomfortably tight (slowly work each electrode in and the move to a different electrode, don't do one all at one)
    2) Take the headset off to let the electrodes untangle from your hair
    3) put the headset back on, once you know all of the electrodes are at the correct depth

    This will mitigate the effects reduced signal quality due to hair being tangled up in the electrodes.

    C
  • codelyok13codelyok13
    edited June 2018
    According to the OpenBCI tutorial 6, I should  be seeing a steady beat but my data is all over the place. I don't know what  to do to fix this problem and many of my channels will start to rail for no reason and start up again after a while. I wanted to know if the information being given is wrong or if something is wrong with my device. It would be big help if I can get some answers. I have the data saved as well so if looking at it will help, just let me know. I really want to make sure I not having any major issues before I go to deep.
  • wjcroftwjcroft Mount Shasta, CA
    Codely, hi.

    There are two Search boxes in right column, for example looking for 'railed' produces:


    Check out this recent thread with various tips,


    Regards,

    William

  • codelyok13codelyok13
    Thaks for the help all my railing issues are gone.
  • wjcroftwjcroft Mount Shasta, CA
    Great, glad to hear that!

    What particular tip or trick did you find solved the problem?

  • ai20ai20
    I bought a Mark IV headset (large) and  the cyton/daisy board. At the beginning both using the GUI and the python script me and a couple of collaborators we had no problem in recording all the 16 channels. Now it seems the system it's becoming very sensitive to railing: randomly some channels rails and only after a lot of fitting on the headset the channels not rail in the GUI. I'm wondering if you have any suggestion because it's making the whole device barely usable. Thanks
  • wjcroftwjcroft Mount Shasta, CA
    Hi, @ai20,

    I merged your new thread into this existing thread. See some of the previous posts.

    It's also possible that there is some wear or corrosion on the tips of the dry electrodes. Can you experiment with swapping or installing newer electrodes?

    Regards,

    William
  • ai20ai20
    Hi,


    I just bought the device in June... but yes I will try with different electrodes. Anything else I can do?
  • wjcroftwjcroft Mount Shasta, CA
    Hi @ai20. In this thread, you can see the names of the previous people posting. For example, @codelyok13. You can click on the name in any of these posts, and send a private message to that person (using the button you see). Codelyok was able to resolve his issues, and might have some extra tips for you; or could chat on the phone. Same for the other posters.

    One other area to look at, might be regarding particular hair conditioners or oils or substances / shampoos, that your user base is employing. Has that changed since when you began? Do you notice any difference when trying the headset after a shower / shampoo, but with using no additional conditioners or additives? 

    In the field of EEG QEEG, it is always advised to NOT be using conditioners before a recording session. Because some of these interfere with the electrodes / gel / etc. Dry electrodes should be less prone to this, but still a precaution.

    Regards,

  • ai20ai20
    @wjcroft thanks. Just to update here what happens in the following scenarios:

    To anyone,
    Is there any suggested way to clean the electrodes?

  • wjcroftwjcroft Mount Shasta, CA
    I would suggest something easy like a terry cloth washcloth, which has been dipped in clear alcohol. Such as the 70% isopropyl you find in drugstores. 100% would be fine also. Then gently massaging each electrode head.

    Such a solvent should help to remove any coatings which may have adhered to the electrodes.

    Do the electrodes "look the same" as when you first got them? Can you compare the appearance of the non-working ones to the working ones, etc. Please followup on the previous suggestion and network with others in this thread who have solved the issue.
  • ai20ai20
    @wjcroft Do the electrodes "look the same" as when you first got them? Yes I am using the the electrodes provided for the Mark IV so no gold cup. At sight I cannot see any difference. I can say one thing though: I can see less railing using only 1 earclip but when I put 2 on, many
    channels start railing and it takes a lot of times to adjust everything.
    The adjustments do not last.
  • kunheekunhee London
    im having  issues with derailing as well. just leaving a comment so i can follow this  thread. don't see a  thread button
    thanks  for all your contribution you absolute heroes. will  try  all your  suggestions!
  • wjcroftwjcroft Mount Shasta, CA
    edited September 2018
    ai20 said: "I can see less railing using only 1 earclip but when I put 2 on, many channels start railing and it takes a lot of times to adjust everything."

    If you are seeing 'many' channels misbehave, based on a change in the earclip, then I would begin to suspect that the earclip may be involved. Were you able to try 'cleaning' some of the electrodes, including the earclips? When you look at the electrodes and earclips, they should have a uniformly grey coating. This is the chloride coating on top of the silver plating. These electrodes are plastic, first plated in silver, then chloride-ed. If the chloride wears off, you will generally see a different color underneath. Either silvery, or of the silver has worn also, then the plastic underneath.

    You said "but when I put 2 on". Explain further. You should be using separate bias and reference electrodes. Where is your bias connected? Reference? Sometimes (especially neurofeedback / QEEG), reference electrodes are "linked" with a Y-cable, then connected to both ear lobes. In that case the Bias electrode is placed on the scalp, generally on the centerline. However the Ultracortex instructions show Bias and Reference connected to opposite ear lobes.

    If you see issues suddenly affect many channels, you must suspect your Reference connection. Since that is shared with all channels.

  • ai20ai20
    @wjcroft 'However the Ultracortex instructions show Bias and Reference connected to opposite ear lobes.' that's what I did. But if I put both the reference and the bias in my earlobes many channels get railed on the GUI. If I just put on the bias not as much.

    I also read that the earclips need to be changed after 25-50 times? If so it may be cause but the usage of the headset becomes expensive (earclip are expensive). Do you have any test to verify if the earclips have been worn out?
    Any idea appreciated.
  • wjcroftwjcroft Mount Shasta, CA
    The OpenBCI CANNOT WORK AT ALL, if you only connect the Bias / Ground and leave off the Reference. The Reference electrode is what provides the other polarity for your electrodes on the scalp. All EEG is done 'differentially', measuring the microvolts difference between two electrodes. One plus and the other minus.

    The fact that your Reference connection Rails many channels, indicates what I said previously. There is something wrong with your Reference ear clip connection. Have you tried swapping the Reference and Bias clips? (Both connected.) This might show that one ear clip is more flakey than the other.

    I don't see replacement clips listed in the OpenBCI store, but FRI has some. You would need to replace the connector or use an adapter cable. Mentioning @Conor and @openbci and @biomurph.


    Regards,

    William

  • ai20ai20
    I just did a test on the electrodes. Apparently the screws were a bit loose and the channels were railing. I now tight the screws and the railing quite decreased (or it is negligible).

    Here is a question is there any suggested 'conductive paste' to enhance conductivity between the base of the electrodes and female base for the screw? Or any nut washer to use so that we can force a bit the screws on the female base?

    Thanks in advance
Sign In or Register to comment.

© Vanilla Keystone Theme 2026