high impedance with cup electrodes? poor SSVEP results. [resolved]
Hasni
Pakistan
I have used silver and gold cup electrodes and also bought a fresh Ten20 Paste still my electrode impedence is 100k to 500k.. Why it is too high ???
What is the best way to check impedance ? (Maybe i am using wrong method to check)
Board Used: Cyton
Project : SSVEP Data Collection
Comments
Hello Hasni,
You can check the validity of your impedance measurements, by using a glob of paste to connect one cup (say channel 1) with reference (SRB2), and include Bias as well. When you click on the 'ohm' [Ω] sign for that channel. Result should be close to zero ohms. This would verify your cabling and Cyton operation.
William
You can only check impedance on one channel at a time for Cyton due to a limitation of the ADS1299 chip, as it injects a small amount of current into the channel. Doing this for multiple channels will distort the impedance readings.
Take Care,
RW
Thankyou for response.. i did same as you told... still i am getting 30k to 50k ohms impedance... where could the possible error ??
I use two earlobes (for SRB and BIAS... is this right ??/)
i am not getting any prominent ssvep frequecy peaks in my spectrum
@wjcroft I think @hasni is close. Can you recommend or rephrase a test that can be done to ensure the board is functional?
For basic tests .... i** always start with eye blink test and Jaw Clench test... they come out to have good results** in Open BCI GUI Plot
But SSVEP is never recognized or never seen in my plots even after processing in MATLAB
Electrodes I am using for SSVEP : O1 O2 Oz PO3 PO4
BIAS and Referance are on Earlobes
@hansi Please, let's focus on getting good impedance signals first. Thank you for the additional information though.
Try checking impedance on the forehead: Fp1 and Fp2 first. These sites do not normally have hair.
sure brother... i will check and let you know... so you mean FP1 and FP2 should show impedance less than 5k ? while hairy scalp has more than 5K ? (for my case 30k to 50k)
Fp1 and Fp2 should have the lowest impedance. Also, can you do this in the OpenBCI GUI send a screenshot of the GUI before and during impedance check, so I can have a look at the raw signals.
sure brother... i will share with you...
Did you try the test mentioned? This verifies that your wiring and connections are all good.
Yes i got exactly zero...
Read this regarding why you may be having difficulty with SSVEP. It's because your screen flashing is not 100% accurate. Some tests are mentioned that you can perform with a photodiode and provided program.
https://openbci.com/forum/index.php?p=/discussion/3052/not-getting-expected-results-from-vep-visual-stimuli-eeg-data-resolved
Thanku> @wjcroft said:
Thankx
These videos are helpful, in showing how cups/paste are placed on scalp areas with hair. The hair in those areas should be pushed aside, so that the paste has optimal direct contact with skin. If this push-aside is not done, you will have higher impedance.
https://openbci.com/forum/index.php?p=/discussion/176/placing-electrode-sensors-on-the-scalp
Read this regarding how the high amplifier INPUT impedance (1 gigaohm), means that skin impedances in the ranges you mentioned, are not actually an issue of any importance.
https://openbci.com/forum/index.php?p=/discussion/comment/13328/#Comment_13328
Thankx alot brother> @wjcroft said:
this is of impedence at FP1
This is when sitting idle with electrode at FP1
This is pic showing mulitple eye blinks at FP1
Why is your notch filter set at 60 Hz, when your mains power frequency is 50 Hz?
I just needed impedence check..that why i didnot set the filter of 50hz
Signals look acceptable. Would have been better to look at the noise floor with 50Hz Notch filter.
The impedance sensing mode of the ADS1299 injects a small current at 31 Hz, which you can see on the FFT.
But... I don't believe the algorithm / signal processing that displays the impedance on the Time Series graph, is bandpass filtering at 31 Hz. Hence, any other strong signals, such as your 50 Hz mains, will distort the actual calculated impedance value. Making it seem higher than it really is.
Ok got you point thankyou