Active electrodes (TLC272)

edited September 2019 in Electrodes

Hello!
Recently I've started using Open BCI in one project.

I have a quastion about using active eletrodes

To achive higher accuracy of recieving data I decided to make active electrodes with TLC272. The idea was taken from http://uazu.net/eeg/ae.html .

I connected them same as passive electrodes. Unfortunately, I have some problems. This electrodes create some kind of noise on the channels that does not connected to anything. This noise is copying signal from channel that connected to active electrode. Is there any idea about how to solve this problem?


How to properly connect these electrodes to Open BCI? For now my connection is:
ch1 <-> out
AVDD <->AVDD (electrode)
AVSS <->AVSS (electrode)

Comments

  • wjcroftwjcroft Mount Shasta, CA

    Below is the comment I left for Vlad on an email he sent me:

    Unused channels can simply be connected to Reference. or Reference and Ground. Open channel pins not connected to any signals, are highly sensitive to noise. Or, you can turn off the unused channels in the GUI.

    Is your active circuit just a 'buffer', not doing any amplification? Any active electrode will have power connections, ground, etc. All those extra wires give opportunity for capacitive or inductive coupling to other wires. Possibly enclosing in a shielded cable could help.

    Additionally, in the last email he mentioned using AVDD AVSS to power his op amps. These should NOT be used, as they are highly related to the analog sensitivity of the ADS1299. Instead try powering your active from DVDD and GND. Any noise back coupled from the active into AVDD AVSS will show up on all channels, matching what you are seeing.

    Regards, William

  • @wjcroft said:
    Below is the comment I left for Vlad on an email he sent me:

    Unused channels can simply be connected to Reference. or Reference and Ground. Open channel pins not connected to any signals, are highly sensitive to noise. Or, you can turn off the unused channels in the GUI.

    Is your active circuit just a 'buffer', not doing any amplification? Any active electrode will have power connections, ground, etc. All those extra wires give opportunity for capacitive or inductive coupling to other wires. Possibly enclosing in a shielded cable could help.

    Additionally, in the last email he mentioned using AVDD AVSS to power his op amps. These should NOT be used, as they are highly related to the analog sensitivity of the ADS1299. Instead try powering your active from DVDD and GND. Any noise back coupled from the active into AVDD AVSS will show up on all channels, matching what you are seeing.

    Regards, William

    Thanks for your reply. Connecting dvdd and gnd to active eletrodes coused zero input on every unit. Is there any other possible suggestion?

  • wjcroftwjcroft Mount Shasta, CA

    I have not studied your active circuit schematic. How is the output from the active buffer connected to the channel and reference. Is there perhaps some anomaly here? I also don't know your op amp configuration. Does it require plus and minus supplies? In that case DVDD and GND are only plus 3.3 and 0.

  • Output from AE connected to bottom pins. AE have 100Ohm resistor before output pin for protection causes. There is no direct connection to referense, SRB2 pin connected to the head and bias to earlobe. TLC272 doesn't required plus and minus supples, and should work with 3.3 V. In this tutorial https://openbci.com/community/active-electrodes-for-openbci/ user TowerComp had connection to AVSS and AVDD. I didn't recieve any reply from him yet, but in his post he assures that schematic should work.

  • @wjcroft said:
    I have not studied your active circuit schematic. How is the output from the active buffer connected to the channel and reference. Is there perhaps some anomaly here? I also don't know your op amp configuration. Does it require plus and minus supplies? In that case DVDD and GND are only plus 3.3 and 0.

    Output from AE connected to bottom pins. AE have 100Ohm resistor before output pin for protection causes. There is no direct connection to referense, SRB2 pin connected to the head and bias to earlobe. TLC272 doesn't required plus and minus supples, and should work with 3.3 V. In this tutorial https://openbci.com/community/active-electrodes-for-openbci/ user TowerComp had connection to AVSS and AVDD. I didn't recieve any reply from him yet, but in his post he assures that schematic should work.

  • wjcroftwjcroft Mount Shasta, CA

    Did you use the shield as he suggests in that thread? Maybe your issue is just noise induce without shielding, and you could go back to the AVDD AVSS ?

  • roborobo Bavaria, Germany

    So how was the problem solved ?
    Why publish a problem when not having the happiness to solve it ... ...
    I have the feeling that the cheaper and low power ADS1298 together with active electrodes might be superior to only the ADS1299.

Sign In or Register to comment.