Brainflow: why are my signals huge?
This is a photo of 10 seconds worth of EEG data in brainflow, even without filtering it should not have such high voltage. I filtered it anyway:
As you can see, there is a difference but still the signal is ridiculously high.
I have checked with the OpenBCI GUI to make sure all the electrodes are ok and not railed or anything. But when I used brainflow this is what happens.
What's going on?
Comments
https://openbci.com/forum/index.php?p=/discussion/201/large-millivolt-data-values-fbeeg-full-band-eeg
Hmmm that makes sense but I tried it and these are the results, still bad. Even though I filtered it (one on the left) it doesn't look like EEG signals and the magnitudes are still too high.
I have a low pass of 50Hz and highpass of 1Hz with a butterworth filter at order 4
Mango, hi.
If the signal looks ok in the GUI, but weird in your plots, then suspect your filtering is not doing what is expected.
More typical would be a bandpass from say .5 Hz to 45 Hz. And a notch filter at your mains frequency. Setting a "low pass at 50 Hz" will generally not be enough to block mains effects. As the power at 50 Hz is huge and the band edges are a gradual slope; band edge is NOT sharp.
William
Notch filter did the trick, you're an unsung hero on this forum mate!
The help is very much appreciated.
Thanks. The DSP 'filter order' is what determines the sharpness of the cutoff edges. But even with higher numbers, you cannot beat the efficacy of a notch.